2016 EdBook

Amgen
Merck & Co., Inc.
2016 ASCO EDUCATIONAL BOOK
This publication is supported by an

educational donation provided by:
AMERICAN SOCIETY OF CLINICAL ONCOLOGY
2016 EDUCATIONAL BOOK

Support for this program is funded through
Collective Wisdom: The Future of Patient-Centered Care and Research

A PEER-REVIEWED, INDEXED PUBLICATION
52nd Annual Meeting | June 37, 2016 | Chicago, Illinois | Volume 36
Vol. 36
American Society of Clinical Oncology Educational Book

The 2016 ASCO Educational Book (Print ISSN: 1548-8748; Electronic ISSN: 1548-8756) is published by American Society of Clinical Oncology,
Inc. (ASCO).
Requests for permission to reprint all or part of any article published in this title should be directed to Permissions, American Society
of Clinical Oncology, Inc., 2318 Mill Road, Suite 800, Alexandria, VA 22314. Tel: (571) 483-1300; fax: (571) 366-9550; or email:
permissions@asco.org.
All other questions should be addressed to ASCO Educational Book Managing Editor, American Society of Clinical Oncology, Inc., 2318
Mill Road, Suite 800, Alexandria, VA 22314. Tel: (571) 483-1300; fax: (571) 366-9550; or email: edbook@asco.org.
Single issues, both current and back, exist in limited quantities and are offered for sale subject to availability. For further information,
email edbook@asco.org or call (888) 273-3508.
Copyright 2016 American Society of Clinical Oncology, Inc. All rights reserved. No part of this publication may be reproduced or transmitted
in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system,
without permission in writing from ASCO.
Copies of articles in this publication may be made for personal use. This consent is given on the condition, however, that the copier
pay the stated per-copy fee through the Copyright Clearance Center, Inc. (222 Rosewood Drive, Danvers, MA 01923) for any copying
beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to any other kinds of copying, such
as copying for general distribution, for advertising, commercial, or promotional purposes, for creating new collective or derivative works, or
for resale.
ASCO assumes no responsibility for errors or omissions in this publication. The reader is advised to check the appropriate medical
literature and the product information currently provided by the manufacturer of each drug to be administered to verify, among other
matters, the dosage, the method and duration of administration, or contraindications. It is the responsibility of the treating physician
or other health care professional, relying on the independent experience and knowledge of the patient, to determine drug dosages
and the best treatment for the patient.
The ideas and opinions expressed in this publication do not necessarily reflect those of ASCO. The mention of any company, product,
service, or therapy does not constitute an endorsement of any kind by ASCO. ASCO assumes no responsibility for any injury or damage
to persons or property arising out of or related to any use of the material contained in this publication.
American Society of
Clinical Oncology
Educational Book
Editor in Chief: Don S. Dizon, MD
Associate Editor: Nathan Pennell, MD, PhD
Managing Editor: Lindsay Pickell, MFA
Production Manager: Donna Dottellis
2016 American Society of Clinical Oncology, Alexandria, VA
Contents
2016 ASCO Annual Meeting Disclosure
xiii
20152016 Cancer Education Committee
xiv
2016 Educational Book Expert Panel
xvi
2016 Annual Meeting Supporters
xx
Letter From the Editor
INVITED ARTICLES
Collaborative Practice in an Era of Multidisciplinary Care
Michael P. Kosty, Todd Pickard, and Pamela Viale
Value in Oncology: Balance Between Quality and Cost

Derek Raghavan and Mark W. Legnini
Womens Health Issues for BRCA Mutation Carriers

Mary E. Sabatini and Leif W. Ellisen
14
Collective Wisdom: Lobular Carcinoma of the Breast

George W. Sledge, Anees Chagpar, and Charles Perou
18
POINTS OF VIEW
Compassionate Use: A Modest Proposal
Arthur L. Caplan, Alison Bateman-House, and Joanne Waldstreicher
e2
Less Is More: The Evolving Surgical Approach to Breast Cancer

Laura Esserman, Etienne Gallant, and Michael Alvarado
e5
Strategies for Sustainable Cancer Care

David J. Kerr, Anant Jani, and Sir Muir Gray
e11
The Role of Nephrectomy for Kidney Cancer in the Era of Targeted and Immune Therapies
Ulka N. Vaishampayan
e16
BREAST CANCER
Breast Cancer Survivorship: Strategies for Optimal Care
Issues in Breast Cancer Survivorship: Optimal Care, Bone Health, and Lifestyle Modifications
Michelle E. Melisko, William J. Gradishar, and Beverly Moy
e22
The 2016 ASCO Educational Book is published online at asco.org/edbook. Articles can also be accessed
from the Attendee Resource Center at am.asco.org/ARC. Articles that are only available online are
denoted with an e ahead of the page number.
asco.org/edbook | 2016 ASCO EDUCATIONAL BOOK
iii
Less Is More: A Multidisciplinary Conversation on Treatment Options

Tailoring Chemotherapy in Early-Stage Breast Cancer: Based on Tumor Biology or Tumor Burden?
Domen Ribnikar and Fatima Cardoso
e31
Targeted Therapies in Hormone ReceptorPositive Breast Cancer

Improving Response to Hormone Therapy in Breast Cancer: New Targets, New Therapeutic Options
Hope S. Rugo, Neelima Vidula, and Cynthia Ma
e40
Treatment of Premenopausal Women With Endocrine-Sensitive Breast Cancer

Challenges in Treating Premenopausal Women with Endocrine-Sensitive Breast Cancer
Hatem A. Azim Jr., Nancy E. Davidson, and Kathryn J. Ruddy
23
Triple-Negative Breast Cancer: Is Change on the Horizon?

The Evolution of Triple-Negative Breast Cancer: From Biology to Novel Therapeutics
Carey K. Anders, Vandana Abramson, Tira Tan, and Rebecca Dent
34
Updates and Controversies in HER2-Positive Breast Cancer

A Value-Based Approach to Treatment of HER2-Positive Breast Cancer: Examining the Evidence
Nancy Nixon and Sunil Verma
e56
Emerging Therapeutic Options for HER2-Positive Breast Cancer

Miguel Martin and Sara Lopez-Tarruella
e64
CANCER PREVENTION, HEREDITARY GENETICS, AND EPIDEMIOLOGY

Controversies in Genetic Evaluation
How Far Do We Go With Genetic Evaluation? Gene, Panel, and Tumor Testing
Filipa Lynce and Claudine Isaacs
e72
Evolving Recommendations on Prostate Cancer Screening

Otis W. Brawley, Ian M. Thompson Jr., and Henrik Gronberg
e80
Refining Breast Cancer Risk Assessment: Additional Genes, Additional Screening

Refining Breast Cancer Risk Stratification: Additional Genes, Additional Information
Allison W. Kurian, Antonis C. Antoniou, and Susan M. Domchek
44
CARE DELIVERY AND PRACTICE MANAGEMENT

Access to Cancer Therapeutics in Low- and Middle-Income Countries
Access to Cancer Therapeutics in Low- and Middle-Income Countries
Sana Al-Sukhun, Charmaine Blanchard, Lawrence N. Shulman, and Paul Ruff
iv
2016 ASCO EDUCATIONAL BOOK | asco.org/edbook
58
Integrating Patient-Reported Outcomes Into Real-Life Medical Decisions

Patient-Reported Outcomes in Cancer Clinical Trials: Measuring Symptomatic Adverse Events With
the National Cancer Institutes Patient-Reported Outcomes Version of the Common Terminology
Criteria for Adverse Events (PRO-CTCAE)
Paul G. Kluetz, Diana T. Chingos, Ethan M. Basch, and Sandra A. Mitchell
67
Optimizing Team-Based Oncology Care: Building the Village

Electronic Patient-Reported Outcomes: The Time Is Ripe for Integration Into Patient Care and
Clinical Research
Lee Schwartzberg
e89
Oncology Advanced Practitioners Bring Advanced Community Oncology Care

Wendy H. Vogel
e97
Quality and Value: Measuring and Utilizing Both in Your Practice

Why the Quality Oncology Practice Initiative Matters: Its Not Just About Cost
Anne C. Chiang
e102
The Oncology Care Model: The Man Behind the Curtain

The Oncology Care Model: A Critique
Christian A. Thomas and Jeffrey C. Ward
e109
CENTRAL NERVOUS SYSTEM TUMORS

Multidisciplinary Management of Brain Metastases
Immune Checkpoint Inhibitors in Brain Metastases: From Biology to Treatment
Anna S. Berghoff, Vyshak A. Venur, Matthias Preusser, and Manmeet S. Ahluwalia
e116
Targeted Therapy in Brain Metastases: Ready for Primetime?

Vyshak A. Venur and Manmeet S. Ahluwalia
e123
The Future of Immunotherapy in Glioblastoma

Current State of Immune-Based Therapies for Glioblastoma
Michael Lim, Michael Weller, and E. Antonio Chiocca
e132
Treatment of Low-Grade Gliomas in the Era of Genomic Medicine

Treatment of Adult Lower-Grade Glioma in the Era of Genomic Medicine
Susan M. Chang, Daniel P. Cahill, Kenneth D. Aldape, and Minesh P. Mehta
75
DEVELOPMENTAL THERAPEUTICS AND TRANSLATIONAL RESEARCH

Biomarkers, Blood-Based Testing, and the Heterogeneous Tumor
Tumor Evolutionary Principles: How Intratumor Heterogeneity Influences Cancer Treatment
and Outcome
Subramanian Venkatesan and Charles Swanton
e141
Biosimilars: Here and Now

Steven J. Lemery, Francisco J. Esteva, and Martina Weise
e151
Clinical Trial Eligibility Criteria: Tradition Versus Reality

Transforming Clinical Trial Eligibility Criteria to Reflect Practical Clinical Application
Edward S. Kim, Jennifer Atlas, Gwynn Ison, and Jennifer L. Ersek
83
Immunotherapy: Beyond Checkpoint Inhibitors

Arming the Immune System Through Vaccination to Prevent Cancer Recurrence
Diane F. Hale, Timothy J. Vreeland, and George E. Peoples
e159
Overcoming Therapeutic Resistance by Targeting Cancer Inflammation

Gregory L. Beatty
e168
Pharmacokinetics, Dynamics, and Genomics in the Era of Immunotherapy and Small Molecules
Pharmacogenomics, Pharmacokinetics, and Pharmacodynamics in the Era of Targeted Therapies
Emiliano Calvo, Christine Walko, E. Claire Dees, and Belen Valenzuela
e175
GASTROINTESTINAL (COLORECTAL) CANCER

Potentially Resectable Metastatic Colorectal Cancer: A Multidisciplinary Discussion
Liver Metastases in Colorectal Cancer
Gunnar Folprecht
e186
Questions on Treatment of Locally Advanced Rectal Cancer

Multimodal Rectal Cancer Treatment: In Some Cases, Less May Be More
Julio Garcia-Aguilar, Rob Glynne-Jones, and Deborah Schrag
92
GASTROINTESTINAL (NONCOLORECTAL) CANCER

Biliary Tract Cancer: The New and the Old
Biliary Tract Cancer: Epidemiology, Radiotherapy, and Molecular Profiling
John A. Bridgewater, Karyn A. Goodman, Aparna Kalyan, and Mary F. Mulcahy
e194
Designing Clinical Trials to Achieve Breakthrough Results in Upper Gastrointestinal Malignancies

The Past, Present, and Future of Pancreatic Cancer Clinical Trials
Lynn M. Matrisian and Jordan D. Berlin
e205
HER2, VEGFR, and Beyond: Genomic Profiling of Upper Gastrointestinal Tract Cancers and the Future of
Personalized Treatment
Gastric Adenocarcinoma: An Update on Genomics, Immune System Modulations, and Targeted Therapy
Jeeyun Lee, Adam J. Bass, and Jaffer A. Ajani
vi
104
Localized Pancreatic Cancer: Multidisciplinary Management With Incorporation of ASCO Guidelines

Localized Pancreatic Cancer: Multidisciplinary Management
Andrew L. Coveler, Joseph M. Herman, Diane M. Simeone, and E. Gabriela Chiorean
e217
GENITOURINARY (NONPROSTATE) CANCER

Elderly Patients With Bladder Cancer: Perspectives From a Surgeon, Medical, and Radiation Oncologist
Treatment of Muscle-Invasive Bladder Cancer in Older Patients
Eila C. Skinner
e228
Renal Cell Carcinoma: Systemic Treatment, Evolving TKIs, and Immuno-Oncology

The Evolution of Systemic Therapy in Metastatic Renal Cell Carcinoma
Thomas E. Hutson, Gregory R. Thoreson, Robert A. Figlin, and Brian I. Rini
113
GENITOURINARY (PROSTATE) CANCER

Contemporary Active Surveillance for Prostate Cancer: Do We Need Better Imaging and Molecular Testing?
Active Surveillance of Prostate Cancer: Use, Outcomes, Imaging, and Diagnostic Tools
Jeffrey J. Tosoian, Stacy Loeb, Jonathan I. Epstein, Baris Turkbey, Peter L. Choyke,
and Edward M. Schaeffer
e235
Oligometastatic Disease in Prostate Cancer: Treating the Patient or the Scan?

Approach to Oligometastatic Prostate Cancer
Brandon Bernard, Boris Gershman, R. Jeffrey Karnes, Christopher J. Sweeney, and Neha Vapiwala
119
Precision Medicine in Advanced Prostate Cancer: Understanding Genomics, Androgen Receptor Splice
Variants, and Imaging Biomarkers
Emerging Molecular Biomarkers in Advanced Prostate Cancer: Translation to the Clinic
Himisha Beltran, Emmanuel S. Antonarakis, Michael J. Morris, and Gerhardt Attard
131
GYNECOLOGIC CANCER
Paradigm Shift in the Management Strategy for Epithelial Ovarian Cancer
Paradigm Shift in the Management Strategy for Epithelial Ovarian Cancer
Keiichi Fujiwara, Jessica N. McAlpine, Stephanie Lheureux, Noriomi Matsumura, and Amit M. Oza
e247
Divide and Conquer: Epithelial Gynecologic Cancers Beyond BRCA

Gynecologic Cancers: Emerging Novel Strategies for Targeting DNA Repair Deficiency
Rebecca S. Kristeleit, Rowan E. Miller, and Elise C. Kohn
e259
Intraperitoneal Chemotherapy for Ovarian Cancer: Trials and Tribulations

Update on Intraperitoneal Chemotherapy for the Treatment of Epithelial Ovarian Cancer
Charlie Gourley, Joan L. Walker, and Helen J. Mackay
143
vii
Neoadjuvant Chemotherapy: Location, Location, Location

Primary Surgery or Neoadjuvant Chemotherapy in Advanced Ovarian Cancer: The Debate
Continues
Alexandra Leary, Renee Cowan, Dennis Chi, Sean Kehoe, and Matthew Nankivell
153
Practical Challenges in Endometrial Cancer

Treatment of Older Women With Endometrial Cancer: Improving Outcomes With Personalized Care
Linda Duska, Armin Shahrokni, and Melanie Powell
164
Symptom Management for the Patient with Gynecologic Cancer

Enhancing Care of the Survivor of Gynecologic Cancer: Managing the Menopause and Radiation
Toxicity
Linda Van Le and Mary McCormack
e270
HEAD AND NECK CANCER

Best of the Rest: Top Abstracts on Head and Neck Cancer From 20152016 Oncology Meetings
Whats New in Head and Neck Cancer: Key Findings in 20152016 From ECCO/ESMO, ASTRO,
and the Multidisciplinary Head and Neck Cancer Symposium
Sue S. Yom, Apar K. Ganti, and Andreas Dietz
176
Integrating Immune Checkpoint Inhibitors and Targeted Agents With Surgery and Radiotherapy for Patients
With Head and Neck Cancer
Immunotherapy and Checkpoint Inhibitors in Recurrent and Metastatic Head and Neck Cancer
Nooshin Hashemi Sadraei, Andrew G. Sikora, and David M. Brizel
e277
Multimodality Management of Locoregionally Recurrent or Second Primary Head and Neck Cancer
Locoregional Recurrent or Second Primary Head and Neck Cancer: Management Strategies
and Challenges
Stuart J. Wong, Dwight E. Heron, Kerstin Stenson, Diane C. Ling, and John A. Vargo
e284
HEALTH SERVICES RESEARCH AND QUALITY OF CARE

Defining and Measuring Quality
Challenges and Opportunities in Delivering High-Quality Cancer Care: A 2016 Update
Patricia A. Ganz, Michael J. Hassett, and David C. Miller
e294
Removing Barriers to Clinical Trial Participation

The Role of Clinical Trial Participation in Cancer Research: Barriers, Evidence, and Strategies
Joseph M. Unger, Elise Cook, Eric Tai, and Archie Bleyer
185
Using Social Media, Wearables, and Electronic Medical Records to Improve Quality of Cancer Care
Using Technology to Improve Cancer Care: Social Media, Wearables, and Electronic Health Records
Michael J. Fisch, Arlene E. Chung, and Melissa K. Accordino
viii
200
HEMATOLOGIC MALIGNANCIESLEUKEMIA, MYELODYSPLASTIC SYNDROMES,

AND ALLOTRANSPLANT
Evolving Therapies in Acute Myeloid Leukemia: Progress at Last?
Evolving Therapies in Acute Myeloid Leukemia: Progress at Last?
Daniel J. DeAngelo, Eytan M. Stein, and Farhad Ravandi
e302
Molecularly and Phenotypically Defined Subtypes of Acute Lymphoblastic Leukemia: Implications

for Management
Advances in the Genetics and Therapy of Acute Lymphoblastic Leukemia
Sabina Chiaretti, Valentina Gianfelici, Susan M. OBrien, and Charles G. Mullighan
e314
Progress in Myeloproliferative Neoplasms: Are We Ready?

Individualizing Care for Patients With Myeloproliferative Neoplasms: Integrating Genetics, Evolving
Therapies, and Patient-Specific Disease Burden
Ruben A. Mesa and Francesco Passamonti
e324
Why Are Myelodysplastic Syndromes So Difficult to Cure?

Integrating Frailty, Comorbidity, and Quality of Life in the Management of Myelodysplastic
Syndromes
Gregory A. Abel and Rena Buckstein
e337
Searching for a Light at the End of the Tunnel? Beyond Hypomethylating Agents in Myelodysplastic
Syndromes
Rami S. Komrokji
e345
HEMATOLOGIC MALIGNANCIESLYMPHOMA AND CHRONIC LYMPHOCYTIC LEUKEMIA

Current Management Concepts: Primary Central Nervous System Lymphoma, Natural Killer T-Cell Lymphoma
Nasal Type, Post-transplant Lymphoproliferative Disorder
Current Management Concepts: Primary Central Nervous System Lymphoma, Natural Killer T-Cell
Lymphoma Nasal Type, and Post-transplant Lymphoproliferative Disorder
Tracy T. Batchelor, Lim Soon Thye, and Thomas M. Habermann
e354
PET-Directed Strategies in Lymphoma

Functional Imaging Using 18-Fluorodeoxyglucose PET in the Management of Primary Mediastinal
Large B-Cell Lymphoma: The Contributions of the International Extranodal Lymphoma Study Group
Franco Cavalli, Luca Ceriani, and Emanuele Zucca
e368
Risk-Adapted Treatment of Advanced Hodgkin Lymphoma With PET-CT

Ryan C. Lynch and Ranjana H. Advani
e376
Chemoimmunotherapy Versus Targeted Treatment in Chronic Lymphocytic Leukemia: When, How Long,
How Much, and in Which Combination?
Chemoimmunotherapy Versus Targeted Treatment in Chronic Lymphocytic Leukemia: When,
How Long, How Much, and in Which Combination?
Jennifer R. Brown, Michael J. Hallek, and John M. Pagel
e387
ix
HEMATOLOGIC MALIGNANCIESPLASMA CELL DYSCRASIA

Consolidation and Maintenance: Moving Beyond Autotransplant
Moving Beyond Autologous Transplantation in Multiple Myeloma: Consolidation, Maintenance,
Allogeneic Transplant, and Immune Therapy
Amrita Krishnan, Ravi Vij, Jesse Keller, Binod Dhakal, and Parameswaran Hari
210
Defining and Redefining Myeloma

Future Directions in the Evaluation and Treatment of Precursor Plasma Cell Disorders
Salomon Manier, Karma Z. Salem, David Liu, and Irene M. Ghobrial
e400
The Role of Imaging in the Treatment of Patients With Multiple Myeloma in 2016
Evangelos Terpos, Meletios A. Dimopoulos, and Lia A. Moulopoulos
e407
Updated Diagnostic Criteria and Staging System for Multiple Myeloma

S. Vincent Rajkumar
e418
Multiple Myeloma: Are We Ready for Personalized Therapy?

Minimal Residual Disease by Next-Generation Sequencing: Pros and Cons
Herve Avet-Loiseau
New Targets and New Agents in High-Risk Multiple Myeloma
Ajay K. Nooka and Sagar Lonial
Next-Generation Sequencing Informing Therapeutic Decisions and Personalized Approaches
Raphael Szalat and Nikhil C. Munshi
e425
e431
e442
LUNG CANCER
Immunotherapy: Beyond AntiPD-1 and AntiPD-L1 Therapies
Immunotherapy: Beyond AntiPD-1 and AntiPD-L1 Therapies
Scott J. Antonia, Johan F. Vansteenkiste, and Edmund Moon
e450
Local Therapies in the Management of Oligometastatic and Metastatic NonSmall Cell Lung Cancer
Local Therapy for Limited Metastatic NonSmall Cell Lung Cancer: What Are the Options and Is
There a Benefit?
Puneeth Iyengar, Steven Lau, Jessica S. Donington, and Robert D. Suh
e460
Lung Cancer Screening and Prevention

Lung Cancer Screening With Low-Dose CT: Implementation Amid Changing Public Policy at One
Health Care System
Abbie Begnaud, Thomas Hall, and Tadashi Allen
Lung Cancer Screening, Cancer Treatment, and Addressing the Continuum of Health Risks Caused
by Tobacco
Graham W. Warren, Jamie S. Ostroff, and John R. Goffin
e468
223
Lung Cancer, Small Cell Lung Cancer: On the Move (Again?)

Seeking New Approaches to Patients With Small Cell Lung Cancer
Marie Catherine Pietanza, Stefan Zimmerman, Solange Peters, and Walter J. Curran Jr.
x
e477
Treatment of Lung Cancer in Medically Compromised Patients

Jeffrey Crawford, Paul Wheatley-Price, and Josephine Louella Feliciano
e484
MELANOMA/SKIN CANCERS
Clinical Conundrums in Melanoma Therapy
Biomarkers for Immunotherapy: Current Developments and Challenges
Kristen R. Spencer, Jianfeng Wang, Ann W. Silk, Shridar Ganesan, Howard L. Kaufman,
and Janice M. Mehnert
e493
Curing High-Risk Melanoma: Are We There Yet?

Surgical Management and Adjuvant Therapy for High-Risk and Metastatic Melanoma
Alexander C.J. van Akkooi, Michael B. Atkins, Sanjiv S. Agarwala, and Paul Lorigan
e505
PATIENT AND SURVIVOR CARE

Cancer Survivorship: Is Every Patients Journey the Same?
The Cancer Survivorship Journey: Models of Care, Disparities, Barriers, and Future Directions
Michael T. Halpern, Mary S. McCabe, and Mary Ann Burg
231
Cancer Treatment as an Accelerated Aging Process

Cancer Treatment as an Accelerated Aging Process: Assessment, Biomarkers, and Interventions
Arti Hurria, Lee Jones, and Hyman B. Muss
e516
New and Improved? Use of White Blood Cell Growth Factors in Oncology Practice
Real-World Conundrums and Biases in the Use of White Cell Growth Factors
Thomas J. Smith and Bruce E. Hillner
e524
Issues on the Use of White Blood Cell Growth Factors in Oncology Practice
Gary H. Lyman
e528
Optimizing Palliative and End-of-Life Care: Evidence-Based Practice Improvement

Discussing the Evidence for Upstream Palliative Care in Improving Outcomes in Advanced Cancer
Myles S. Nickolich, Areej El-Jawahri, Jennifer S. Temel, and Thomas W. LeBlanc
e534
Rehabilitation of Patients and Survivors: Seizing the Opportunity

Developing High-Quality Cancer Rehabilitation Programs: A Timely Need
Catherine M. Alfano, Andrea L. Cheville, and Karen Mustian
241
PEDIATRIC ONCOLOGY
Controversies in Germline Genetic Testing and Disclosure in Pediatric Oncology
The Advantages and Challenges of Testing Children for Heritable Predisposition to Cancer
Chimene Kesserwan, Lainie Friedman Ross, Angela R. Bradbury, and Kim E. Nichols
251
xi
Exploiting Laboratory Insights to Improve Outcomes of Pediatric Central Nervous System Tumors
Exploiting Laboratory Insights to Improve Outcomes of Pediatric Central Nervous System Tumors
Giles W. Robinson, Hendrik Witt, and Adam Resnick
e540
The State of the Art in Neuroblastoma: From Biology to Survivorship

Neuroblastoma: A Tough Nut to Crack
Frank Speleman, Julie R. Park, and Tara O. Henderson
e548
PROFESSIONAL DEVELOPMENT
Curing Burnout in Oncology: Mindful Self-Compassion, Communication, and Practice
Addressing Burnout in Oncology: Why Cancer Care Clinicians Are At Risk, What Individuals Can Do,
and How Organizations Can Respond
Fay J. Hlubocky, Anthony L. Back, and Tait D. Shanafelt
271
SARCOMA
Is There a Future for Immunotherapy in Sarcoma?
Immunotherapy for Soft Tissue Sarcoma: Tomorrow Is Only a Day Away
Alex Lee, Paul Huang, Ronald P. DeMatteo, and Seth M. Pollack
Manipulating the Immune System With Checkpoint Inhibitors for Patients With Metastatic Sarcoma
Sandra P. DAngelo
281
e558
Noncytotoxic Approaches to the Treatment of Metastatic Soft Tissue Sarcoma

Local Ablative Therapies to Metastatic Soft Tissue Sarcoma
Alessandro Gronchi, B. Ashleigh Guadagnolo, and Joseph Patrick Erinjeri
e566
TUMOR BIOLOGY
Expanding the Clinically Actionable Cancer Genome
Interrogating the Cancer Genome to Deliver More Precise Cancer Care
Joaquin Mateo and Johann S. de Bono
e577
Molecular Diagnostics in Cancer

Considerations for Implementation of Cancer Molecular Diagnostics Into Clinical Care
Daniel F. Hayes
292
Tumor Heterogeneity and Therapeutic Resistance

Christine M. Lovly, April K.S. Salama, and Ravi Salgia
e585
Whats Next in Cancer Immunotherapy?

Basic Overview of Current Immunotherapy Approaches in Cancer
Vamsidhar Velcheti and Kurt Schalper
xii
298
2016 ASCO Annual Meeting Disclosure

As the CME provider for the 2016 Annual Meeting, ASCO is committed to balance, objectivity, and scientific rigor in the management of
financial interactions with for-profit health care companies that could create real or perceived conflicts of interest. Participants in the
Meeting have disclosed their financial relationships in accordance with ASCOs Policy for Relationships with Companies; review the policy at
asco.org/rwc.
ASCO offers a comprehensive disclosure management system, using one disclosure for all ASCO activities. Members and participants in
activities use coi.asco.org to disclose all interactions with companies. Their disclosure is kept on file and can be confirmed or updated with
each new activity.
Please email coi@asco.org with specific questions or concerns.
xiii
20152016 Cancer Education Committee

The Cancer Education Committee assesses the need for, plans, develops, and initiates the education programs for the Annual Meeting.
Apar Kishor Ganti, MD, MBBSChair

Michael A. Thompson, MD, PhDChair-Elect
John Vernon Cox, DO, FACP, FASCO, MBAImmediate Past Chair
Lillian L. Siu, MDBoard Liaison
BREAST CANCER
Hope S. Rugo, MDTrack Leader
Judy Caroline Boughey, MD
Jennifer A. Brown, MD
Raquel Nunes, MD
Debra A. Patt, MD, MPH, MBA
Rinaa S. Punglia, MD
Elizabeth C. Reed, MD
Tallal Younis, MBBCh, FRCP

Vincent J. Picozzi, MDTrack Leader
Tanios S. Bekaii-Saab, MD
Jimmy J. Hwang, MD
Se Hoon Park, MD
Manish A. Shah, MD
William Small, MD

Stephen Boyd Riggs, MDTrack Leader
Thomas E. Hutson, DO, PharmD
Jose A. Karam, MD
Ulka N. Vaishampayan, MD
CANCER PREVENTION, HEREDITARY GENETICS,

AND EPIDEMIOLOGY
GENITOURINARY (PROSTATE)
CANCER
P. Kelly Marcom, MDTrack Leader

Margreet Ausems, MD, PhD
Monique A. De Bruin, MD
Jeri Kim, MD
Howard L. McLeod, PharmD
Electra D. Paskett, PhD
Surendra Srinivas Shastri, MD, MBBS
Himisha Beltran, MDTrack Leader

Chris Parker, BA, BM Bchir, MD, FRCR, FRCP
Edward M. Schaeffer, MD, PhD
Nicholas J. Vogelzang, MD, FASCO

Jeffery C. Ward, MDTrack Leader
Aditya Bardia, MBBS, MPH
Kelly Bugos, RN, ANP, MS
Moshe C. Chasky, MD
John Emmett Hennessy, CMPE, MBA
Lee Steven Schwartzberg, MD, FACP

Manmeet Singh Ahluwalia, MDTrack Leader
Eric L. Chang, MD
Nicole A. Shonka, MD
Roger Stupp, MD
GERIATRIC ONCOLOGY
Andrew S. Artz, MD, MSTrack Leader
William P. Tew, MD
William Tse, MD
GYNECOLOGIC CANCER
Helen Mackay, MDTrack Leader
Elise C. Kohn, MD
Linda Duska, MD
Linda Van Le, MD

Joseph Kamel Salama, MDTrack Leader
John Truelson, MD
John A. Ridge, MD, PhD
Irina Veytsman, MD
CLINICAL TRIALS
Sumithra J. Mandrekar, PhDTrack Leader
Janet Dancey, MD
DEVELOPMENTAL THERAPEUTICS AND TRANSLATIONAL

RESEARCH
Howard A. Burris, MD, FASCOTrack Leader
Julia A. Beaver, MD
Francisco J. Esteva, MD, PhD
John Charles Morris, MD
HEALTH SERVICES RESEARCH AND QUALITY OF CARE

Bruce Lee Jacobs, MDTrack Leader
Ethan M. Basch, MD
Daniel Jacob Becker, MD
Dawn L. Hershman, MD
Nicole Maria Kuderer, MD
Jennifer W. Mack, MD
HEMATOLOGIC MALIGNANCIESLEUKEMIA,
MYELODYSPLASTIC SYNDROMES,
AND ALLOTRANSPLANT
Donald A. Richards, MD, PhDTrack Leader

Chris R. Garrett, MD
Chi Lin, MD, PhD
Ashwin Reddy Sama, MD
Henry Q. Xiong, MD
Jorge E. Cortes, MDTrack Leader

Steven M. Devine, MD
Jonathan Michael Gerber, MD
Hanna Khoury, MD
David Leibowitz, MD
xiv
HEMATOLOGIC MALIGNANCIESLYMPHOMA
AND CHRONIC LYMPHOCYTIC LYMPHOMA
Grzegorz S. Nowakowski, MDTrack Leader
Matthew Alexander Lunning, DO
Jeffrey Matous, MD
John M. Pagel, MD, PhD
Michael A. Thompson, MD, PhD
HEMATOLOGIC MALIGNANCIESPLASMA
CELL DYSCRASIA
Sagar Lonial, MDTrack Leader
Asher Alban Chanan-Khan, MD
Parameswaran Hari, MD
Irene M. Ghobrial, MD
Ravi Vij, MD
LUNG CANCER
Martin J. Edelman, MDTrack Leader
Arkadiusz Z. Dudek, MD, PhD
Shirish M. Gadgeel, MD
Puneeth Iyengar, MD, PhD
Craig H. Reynolds, MD
David R. Spigel, MD
PEDIATRIC ONCOLOGY
Tara O. Henderson, MD, MPHTrack Leader
Gregory T. Armstrong, MD, MSCE
Stewart Goldman, MD
Joel A. Weinthal, MD
Anne S. Tsao, MDTrack Leader
Kristin Anderson, MD, MPH
Kelly J. Cooke, DO
Jill Gilbert, MD
Laura Goff, MD
Roberto Leon-Ferre, MD
SARCOMA
Gary K. Schwartz, MDTrack Leader
L. Johnetta Blakely, MD
Robin Lewis Jones, MD, BSc, MBBS, MRCP
Min S. Park, MD, MS
TUMOR BIOLOGY
Kimryn Rathmell, MD, PhDTrack Leader
Ravi Salgia, MD, PhD
Eliezer Mendel Van Allen, MD
Vamsidhar Velcheti, MD
Sanjiv S. Agarwala, MDTrack Leader
Jason John Luke, MD, FACP
Janice M. Mehnert, MD
Ahmad A. Tarhini, MD, PhD
Alexander Christopher Jonathan Van Akkooi, MD, PhD

Nagendra Tirumali, MDTrack Leader
Deborah Mayer, PhD, RN, AOCN, FAAN
Kathi Mooney, PhD, RN
Maria Alma Rodriguez, MD
Louise Christie Walter, BS, MD
LIAISONS
Sana Al-Sukhun, MD, MScInternational Affairs Committee
Howard Bailey, MDCancer Prevention Committee
Gregg Franklin, MD, PhDClinical Practice Committee
Melissa Johnson, MDCancer Research Committee
Thomas Leblanc, MD, MAEthics Committee
Leonard Saltz, MDValue Task Force
Manish Shah, MDClinical Practice Guidelines Committee
Charles Shapiro, MDCancer Survivorship Committee
William Tew, MDGeriatrics Special Interest Group
Gina Villani, MDClinical Practice Guidelines Committee
Ann Von Gehr, MDClinical Practice Guidelines Committee
xv
2016 Educational Book Expert Panel

The Expert Panel is a group of well-recognized physicians and researchers in oncology and related fields who have served as peer
reviewers of the ASCO Educational Book articles.
Ghassan K. Abou-Alfa, MD
Memorial Sloan Kettering Cancer Center,
Weill Cornell Medical College
Donald I. Abrams, MD
San Francisco General Hospital
Manmeet S. Ahluwalia, MD
Cleveland Clinic
Fabrice Andre, MD, PhD
Institut Gustave Roussy
Christina Annunziata, MD, PhD
National Cancer Institute at the National
Insitutes of Health
Stephen Maxted Ansell, MD
Mayo Clinic
Frederick Appelbaum, MD
Fred Hutchinson Cancer Research Center
Saro Armenian, DO, MPH
City of Hope
Emily K. Bergsland, MD
University of California, San Francisco
Franco Cavalli, MD
University of Bellinzona
Axel Bex, MD, PhD

The Netherlands Cancer Institute
Andrea Cercek, MD
Memorial Sloan Kettering Cancer Center
William Blum, MD
The Ohio State University
Anthony Chan, MBBS, MD, FRCP, FHKCP,

FHKAM
Sir YK Pao Centre for Cancer, Prince of Wales
Hospital
Sharon Bober, PhD

Dana-Farber Cancer Institute
Diane Bodurka, MD, MPH
The University of Texas MD Anderson
Cancer Center
Joann Bodurtha, MD, MPH
Johns Hopkins University School of Medicine
George J. Bosl, MD
Alba Brandes, MD
Bellaria Hospital
Alice Chen, MD
Institutes of Health
Allen M. Chen, MD
University of California, Los Angeles
Stephen K.L. Chia, MD
BC Cancer Agency
Laura Quan Man Chow, MD
University of Washington
Jessica Clement, MD
University of Connecticut Health Center
Gregory Armstrong, MD, MSCE

St. Jude Childrens Research Hospital
Eduardo Bruera, MD
The University of Texas MD Anderson Cancer
Center
Anthony Cmelak, MD
Vanderbilt-Ingram Cancer Center
Christina Baik, MD, MPH

Seattle Cancer Care Alliance
Alan Haruo Bryce, MD

Mayo Clinic
Ezra Cohen, MD
University of California, San Diego
Lodovico Balducci, MD
Moffitt Cancer Center
Jan C. Buckner, MD
Mayo Clinic
Robert Coleman, MD
Center
Carlos H. Barrios, MD
Pontifical Catholic University of Rio Grande
do Sul School of Medicine
Howard A. Burris, MD, FASCO

Sarah Cannon Research Institute, Tennessee
Oncology
Brigitta G. Baumert, MD, PhD, MBA

University of Bonn Medical Center
Harold Burstein, MD, PhD, FASCO

Tanios S. Bekaii-Saab, MD
The Ohio State University Comprehensive
Cancer Center
Toby Christopher Campbell, MD

University of Wisconsin Carbone Cancer
Center
Himisha Beltran, MD
Beverly E. Canin
Breast Cancer Options
Daniel Costa, MD, PhD

Beth Israel Deaconess Medical Center
Robert S. Benjamin, MD
The University of Texas MD Anderson
Cancer Center
Lisa Carey, MD
The University of North Carolina at
Chapel Hill
Carien L. Creutzberg, MD, PhD

Leiden University Medical Center
Michael F. Berger, PhD

William L. Carroll, MD
NYU Langone Medical Center
P. Leif Bergsagel, MD, FASCO

Mayo Clinic
Richard Carvajal, MD
Columbia University Medical Center
xvi
Frances Collichio, MD
The University of North Carolina at Chapel
Hill
Roisin Connolly, MBBCh
The Sidney Kimmel Comprehensive Cancer
Center at Johns Hopkins
Rena M. Conti, PhD
The University of Chicago
Sia Daneshmand, MD
University of Southern California
Molly S. Daniels, MS, CGC
Center
Neil N. Hayes, MD, MPH

Hill
Jonas De Souza, MD
James M. Foran, MD
Mayo Clinic
H. Joachim Deeg, MD
James Ford, MD
Stanford University Medical Center
Egidio Del Fabbro, MD

VCU Massey Cancer Center
William Foulkes, MBBS, PhD

Jewish General Hospital
Pierre-Yves Dietrich, MD
Hopitaux Universitaires de Geneve
`
Jonathan Freidberg, MD, MMSc

University of Rochester
Mary L. Disis, MD
University of Washington
Richard R. Furman, MD
Weill Cornell Medical College, New YorkPresbyterian Hospital
Tanya B. Dorff, MD
USC Norris Comprehensive Cancer Center
Mitchell Dowsett, PhD
The Royal Marsden NHS Foundation Trust
Martin Dreyling, MD
University Hospital Grosshadern
Dan Duda, DMD, PhD
Massachusetts General Hospital
Leena Gandhi, MD, PhD

David E. Gerber, MD
The University of Texas Southwestern
Medical Center
Peter Gibbs, MBBS, FRACP, MD
The Royal Melbourne Hospital
Reinhard Dummer, MD
University Hospital Zurich
Sharon Hermes Giordano, MD, MPH

Center
Linda R. Duska, MD
University of Virginia Health System
Robert G. Gish, MD
Stanford Medicine
Grace K. Dy, MD
Roswell Park Cancer Institute
Bernardo H.L. Goulart, MD, MS

Craig Earle, MD, MSc

Institute for Clinical Evaluative Sciences
Ramaswamy Govindan, MD
Washington University School of Medicine in
St. Louis
Martin J. Edelman, MD
University of Maryland School of Medicine
Lawrence H. Einhorn, MD
Indiana University Melvin and Bren Simon
Cancer Center
Timothy Graubert, MD
Gordon Hafner, MD, FACS
Inova Health System
Bryan Hennessy, MD
Center
John Emmett Hennessy, MBA, CMPE
Sarah Cannon Research Institute
Norah Henry, MD, PhD
University of Michigan
Peter Hillmen, PhD
Leed University
Jeffrey Hoch, PhD
Cancer Care Ontario
Christine Holmberg, DPhil, MPH
Charite Universitatsmedizin Berlin
Gabriel Hortobagyi, MD, FACP
Center
Clifford A. Hudis, MD, FACP
Kevin S. Hughes, MD, FACS
Harvard Medical School, Massachusetts
General Hospital
Kelly Hunt, MD
Center
Maha Hussain, MD, FACP, FASCO
Thomas Hutson, DO, PharmD, FACP
Texas Oncology, Baylor Charles A. Sammons
Cancer Center
Hatem El Halabi, MD
Cancer Treatment Centers of America
Peter Hammerman, MD, PhD

Marwan Fakih, MD
City of Hope
Hans Hammers, MD, PhD

Lesley Fallowfield, DPhil, BSc

Sussex Health Outcomes Research and
Education in Cancer
Nasser H. Hanna, MD
Indiana University Melvin and Bren Simon
Cancer Center
David Jablons, MD
Michelle A. Fanale, MD
Center
Michelle Harvie, PhD

Nightingale and Genesis Prevention Centre,
Wythenshawe Hospital
Sekwon Jang, MD
Inova Comprehensive Cancer and Research
Institute
Adele K. Fielding, MBBS, PhD

University College London
Hege Haugnes, MD, PhD

University Hospital of North Norway
Michalina Janiszewska, PhD

Paul Graham Fisher, MD

Stanford University
Axel Hauschild, MD
University of Kiel
Connie Jimenez,
PhD
Vanderbilt University Medical Center
Gini F. Fleming, MD
University of Chicago Medical Center
Gillian Haworth, MBBS

Great Ormond Street Hospital for Children
Maxine S. Jochelson, MD
David H. Ilson, MD, PhD

Syma Iqbal, MD
University of Southern California
xvii
Joseph Jurcic, MD
Rami Manochakian, MD
Case Western Reserve University, Louis
Stokes Cleveland VA Medical Center
Matthew Katz, MD, FACS

Center
Mara-Victoria Mateos, MD, PhD

University Hospital of Salamanca-IBSAL
Steven Katz, MD, MPH

University of Michigan Medical School
Rana McKay, MD
Stanley B. Kaye, MBBS, MRCP, MD, FRCP,

FRCPS, FRSE, FMedSci
The Institute of Cancer Research, The Royal
Marsden NHS Foundation Trust
Usha Menon, MD
University College London
Andrew Kennedy, MD, FACRO

Gretchen Kimmick, MD
Duke University Medical Center
Hedy Kindler, MD
Heidi D. Klepin, MD, MS
Wake Forest University School of Medicine
Theresa Koppie, MD
Oregon Health & Science University
Thomas Krivak, MD
Magee-Womens Hospital of UMPC
Ian Krop, MD, PhD
Geoff Ku, MD, MBA
Shaji Kumar, MD
Mayo Clinic
Stephen Yenzen Lai, MD, PhD
Center
James M.G. Larkin, MD, PhD
Jonathan Ledermann, MD, FRCP
University College London Cancer Institute
Stuart M. Lichtman, MD
Nancy Lin, MD
Tracy Lively, PhD
Ravi Madan, MD
xviii
Paul A. Meyers, MD
Ray D. Page, DO, PhD

The Center for Cancer and Blood Disorders
Sumanta Pal, MD
City of Hope
Alberto Pappo, MD
Chris Parker, FRCR, BA, MBBCh, MD
Donald Williams Parsons, MD, PhD
Texas Childrens Cancer Center, Baylor
College of Medicine
Linda R. Mileshkin, MBBS, MD, MBioeth

Peter MacCallum Cancer Centre
Lynne Penberthy, MD, MPH

Matthew Milowsky, MD
Hill
Vincent J. Picozzi, MD
Virginia Mason Medical Center
Shanu Modi, MD
Blase N. Polite, MD, MPP

Ana Molina, MD
Sandro V. Porceddu, MD
Princess Alexandra Hospital
Alison Moliterno, MD
Johns Hopkins University
Michael Andrew Postow, MD

Silvio Monfardini, MD
Istituto Palazzolo Fondazione Don Gnocchi
Halle C.F. Moore, MD
Cleveland Clinic
Kathleen N. Moore, MD
The University of Oklahoma Health Sciences
Center
Philippe Moreau, MD
CHU de Nantes, Hotel DieuHME
Lindsay Morton, PhD
Michael Neuss, MD
Lee Nisley Newcomer, MD
United Health Group
Grzegorz S. Nowakowski, MD
Mayo Clinic
Olatoyosi Odenike, MD
Francesco Onida, MD
University of Milan
Eric Padron, MD
Thomas Powles, MD
Barts Health
Kumar Prabhash, MD, DM
Tata Memorial Centre
Amanda Psyrri, MD, PhD
Attikon Hospital National Kapodistrian
University of Athens
Cornelis J.A. Punt, MD, PhD
University of Amsterdam
Mark J. Ratain, MD
John Ridge, MD, PhD
Fox Chase Cancer Center
Danny Rischin, MBBS, MD, FRACP
Peter MacCallum Cancer Centre
Miriam B. Rodin, MD, PhD
Saint Louis University School of Medicine
Julia Rowland, PhD
David P. Ryan, MD
Gilles A. Salles, MD, PhD
Hospices Civils de Lyon, Universite Claude
Bernard
Hanna Kelly Sanoff, MD, MPH

University of North Carolina at Chapel Hill
School of Medicine
Gary K. Schwartz, MD
David W. Scott, MBCHB, PhD
BC Cancer Agency
Tanguy Y. Seiwert, MD
Charles L. Shapiro, MD
Ichan School of Medicine at Mount Sinai
Zsofia K. Stadler, MD
Stephan Stilgenbauer, MD
University Hospital Ulm
Richard Sullivan, PhD
Kings Health Partners Integrated Cancer
Centre
Joel Tepper, MD
Hill
Liran Shlush, MD, PhD

Princess Margaret Cancer Centre
William P. Tew, MD
Marc Shuman, MD
USCF Helen Diller Comprehensive Cancer
Center
Mike Thompson, MD, PhD

Aurora Clinic
Sonali Smith, MD
Andreas Ullrich, MD, MPH

World Health Organization
John David Sprandio, MD

Consultants in Medical Oncology and
Hematology
Saad Zafar Usmani, MD

Levine Cancer Institute/Carolinas Healthcare
System
Catherine H. Van Poznak, MD

University of Michigan Comprehensive
Cancer Center
Brian Van Tine, MD, PhD
Washington University School of Medicine in
St. Louis
Anna Varghese, MD
Dian Wang, MD, PhD
Rush University Medical Center
Lari B. Wenzel, PhD
University of California, Irvine
William Wierda, MD, PhD
Center
Peter Yu, MD
Palo Alto Medical Foundation
Andrew Zelenetz, MD, PhD
Clive Zent, MD
University of Rochester
xix
2016 Annual Meeting Supporters*

MISSION ENDOWMENT FOUNDING DONORS
Genentech BioOncology
GlaxoSmithKline Oncology
Novartis Oncology
Sanofi Oncology
MISSION ENDOWMENT SUSTAINING DONORS

AbbVie, Inc.
Amgen
Astellas
AstraZeneca
Celgene Corporation
Eisai Inc.
Gilead Sciences, Inc.
HELSINN
Incyte Corporation
Lilly
Onyx Pharmaceuticals
Sanofi Oncology
Takeda Oncology
ENDOWED FUND DONORS

Anonymous
Sally Gordon YIA Endowment Fund
Conquer Cancer Foundation of ASCO Young Investigator Award, in memory of Sally Gordon
American Society of Clinical Oncology
Jane C. Wright, MD, YIA Endowment Fund
Jane C. Wright, MD, Young Investigator Award
Friends, Family, and Colleagues of Dr. Ronald Beller and Mrs. Judith Beller
Bradley Stuart Beller Endowment Fund
Bradley Stuart Beller Merit Award
Breast Cancer Research Foundation
Evelyn H. Lauder YIA Endowment Fund
Conquer Cancer Foundation of ASCO Young Investigator Award, in memory of Evelyn H. Lauder
Celgene Corporation
John R. Durant, MD, YIA Endowment Fund
Conquer Cancer Foundation of ASCO Young Investigator Award, in memory of John R. Durant, MD
GlaxoSmithKline Oncology
Gianni Bonadonna Award and Lecture Endowment Fund
Gianni Bonadonna Breast Cancer Award and Fellowship
HELSINN
Anna Braglia YIA Endowment Fund
Anna Braglia Young Investigator Award in Cancer Supportive Care, supported by HELSINN
Friends, Family, and Colleagues of Dr. James B. Nachman
James B. Nachman Pediatric Oncology Fund
James B. Nachman ASCO Junior Faculty Award in Pediatric Oncology
Aaron and Barbro Sasson
Bertil Eriksson YIA Endowment Fund
Ake
Bertil Eriksson Endowed Young Investigator Award
Ake
* This list reflects commitments as of April 18, 2016.
xx
CANCERLINQ MAJOR SUPPORTERS

Amgen
Astellas
AstraZeneca
Bayer HealthCare Pharmaceuticals Inc.
Boehringer Ingelheim Pharmaceuticals, Inc.
Cancer Treatment Centers of America
Chan Soon-Shiong Family Foundation
HELSINN
Janssen Oncology
Lilly
Raj Mantena, RPh
Novartis Oncology
Pfizer Oncology
Thomas G. Roberts Jr., MD, and Susan M.
DaSilva
Susan G. Komen
EDUCATIONAL SUPPORT
Ambry Genetics
Cancer Prevention, Hereditary Genetics,
and Epidemiology Track
Pre-Annual Meeting Seminar: Genetics
and Genomics for the Practicing
Clinician
Amgen
Annual Meeting Proceedings and
Educational Book Bundle
Best of ASCO Meetings
Gastrointestinal (Colorectal) Cancer
Track
Hematologic Malignancies Tracks
Bundle
Melanoma/Skin Cancers Track
Pre-Annual Meeting Seminar: How to
Integrate Tumor Immunotherapy Into
Your Clinical Practice
Pre-Annual Meeting Seminar: New Drugs
in Oncology
Pre-Annual Meeting Seminar: The
Economics of Cancer Care
Astellas
Genitourinary Cancer Tracks Bundle
Astellas and Genentech BioOncology
Lung Cancer Track
AstraZeneca
Lung Cancer Track
Integrate Tumor Immunotherapy Into Your
Clinical Practice

Lung Cancer Track
Bristol-Myers Squibb
Pre-Annual Meeting Seminar: Hematology
for the Oncologist
Celgene Corporation
Breast Cancer Track
Gastrointestinal (Noncolorectal) Cancer
Track
Hematologic Malignancies Tracks Bundle
Lung Cancer Track
Conquer Cancer Foundation & Conquer
Cancer Foundation Mission Endowment
Cancer Survivorship Symposium
Ferring Pharmaceuticals
Breast Cancer Track
Lung Cancer Track
Development Therapeutics and Translational
Research/Clinical Trials Tracks Bundle
HELSINN
Patient and Survivor Care/Health Services
Research and Quality of Care Tracks Bundle
Incyte Corporation
Clinical Practice
Janssen Oncology
Patient and Survivor Care/Health Services
Research and Quality of Care Tracks Bundle
Kidney Cancer Association
Genitourinary Cancers Track
Lilly
Breast Cancer Track
Gastrointestinal (Colorectal) Cancer Track
Head and Neck Cancer Track
Lung Cancer Track
Pre-Annual Meeting Seminar: New Drugs in
Oncology
Sarcoma Track
Merck & Co., Inc.

Annual Meeting Proceedings and
Educational Book Bundle
Breast Cancer Track
Gastrointestinal (Noncolorectal) Cancer Track
Head and Neck Cancer Track
Lung Cancer Track
Clinical Practice
Merrimack Pharmaceuticals
Track
Novartis Oncology
Breast Cancer Track
Track
Pre-Annual Meeting Seminar: Hematology
for the Oncologist
Clinical Practice
Oncology
Patient Access Network Foundation
Annual Meeting Patient Advocate
Scholarship Program (2)
Pharmacyclics LLC
TAIHO Oncology, Inc.
Oncology
TESARO
Gynecologic Cancer Track
GENERAL SUPPORT
AbbVie, Inc.
Annual Meeting iPlanner Mobile Application
and Website Bundle
ASCO Live
Young Investigator Award
xxi
Adenoid Cystic Carcinoma Research

Foundation
Career Development Award in Adenoid
Cystic Carcinoma Research
Bristol-Myers Squibb
Annual Meeting Onsite Connectivity
Resources Bundle
Patient Advocate Scholarship Program Bundle
American Cancer Society

ASCO/American Cancer Society Award and
Lecture
Celgene Corporation
Merit Awards
Amgen
Career Development Award
Diversity in Oncology Bundle
International Innovation Grant
Long-Term International Fellowship
Merit Awards
Oncology Trainee Travel Award
Young Investigator Award (2)
The Cholangiocarcinoma Foundation

Young Investigator Award in
Cholangiocarcinoma Research (1)
ASCO and Conquer Cancer Foundation

Boards of Directors
Young Investigator Award, in memory of
Jane C. Wright, MD
ASCO State Affiliate Council and ASCO
Clinical Practice Committee
Astellas
ASCO Meeting Library Bundle
AstraZeneca
Merit Awards
Avon Foundation for Women
International Development and Education
Award in Breast Cancer
Award
Resources Bundle
Annual Meeting Special Attendees Lounge
Bundle
Breast Cancer Research Foundation
Advanced Clinical Research Award in Breast
Cancer
Career Development Award in Breast Cancer (2)
Comparative Effectiveness Research
Professorship in Breast Cancer
Young Investigator Award in Breast Cancer,
in honor of Susan Hirschhorn and in
memory of her mother
xxii
Coalition of Cancer Cooperative Groups

Clinical Trials Participation Award
CRDF Global, National Cancer Institute, and
National Science Foundation
Award in Palliative Care
Conquer Cancer Foundation
Cancer.Net
Guidelines USB
Journal of Global Oncology Fellows Program
Long-Term International Fellowship (2)
Medical Student Rotation for
Underrepresented Populations
Merit Awards (2)
Patient Advocate Scholarship Program
Patient Materials
Survivorship Forum
Conquer Cancer Foundation Mission
Endowment
ASCO Community Research Forum Award
ASCO Policy Fellowship
ASCO University Maintenance of
Certification Application
Cancer.Net
Journal of Clinical Oncology Figures Program
Journal of Global Oncology
Journal of Oncology Practice, Special Series
Long-Term International Fellowship (2)
NCI-ASCO Teams in Cancer Care Delivery
Project
patientACCESS
Quality Measurement System
Resident Travel Award for
Eisai Inc.
Resources Bundle
EMD Serono Inc.

Fibrolamellar Cancer Foundation
Young Investigator Award in Fibrolamellar
Research
Florida Society of Clinical Oncology
Young Investigator Award Florida-Based
Recipient
Gateway for Cancer Research
Career Development Award (4)
Bundle
Resources Bundle
Merit Awards
Guardant Health, Inc.
and Website Bundle
HELSINN
and Website Bundle
Foundation General Mission Support
Infinity Pharmaceuticals Inc.
and Website Bundle
Janssen Oncology
Diversity in Oncology Bundle
Merit Awards
Janssen Scientific Affairs, LLC
Career Development Award in Deep Vein
Thrombosis
Young Investigator Award in Deep Vein
Thrombosis
Journal of Clinical Oncology
Kidney Cancer Association
Annual Meeting Merit Awards in Kidney
Cancer
Young Investigator Award in Kidney
Cancer
The John and Elizabeth Leonard Family
Foundation
Lilly
Bundle
Medical Student Rotation for
Merit Awards
Bundle
Lung Cancer Alliance
Young Investigator Award in Lung Cancer
J. Edward Mahoney Foundation
McHenry and Lisa Tichenor Fund of
Communities Foundation of Texas
Young Investigator Award in Sarcoma
Research, in memory of Willie Tichenor
Medivation
Bundle
Merck & Co., Inc.
Annual Meeting Program Publications
Bundle
Strike 3 Foundation
Young Investigator Award in Pediatric
Oncology (2)
Takeda Oncology
Award
Merit Awards
Oncology Trainee Travel Award
Teva Oncology
Bundle
Triple Negative Breast Cancer Foundation
The WWWW Foundation Inc. (QuadW) and
The Sarcoma Fund of the QuadW
Foundation of Communities Foundation of
Texas
Young Investigator Award in Sarcoma
Research, in memory of Willie Tichenor
NF Nimeh, MD
Novartis Oncology
Merit Awards
Susan K. Parsons, MD, and Walter
Armstrong
Pfizer Oncology
Roche
Long-Term International Fellowship
Reid R. Sacco Adolescent and Young Adult
Alliance
Sanofi Foundation for North America
Foundation General Mission Support
Sanofi Oncology
Drug Development Research Professorship
Spectrum Pharmaceuticals
Merit Awards
xxiii
Letter From the Editor
n behalf of my Associate Editor, Dr. Nate Pennell, I welcome you to the 2016 American Society of Clinical Oncology (ASCO)
Annual Meeting. It is also my distinct honor to present the 36th volume of the NLM-indexed ASCO Educational Book. This
years theme, Collective Wisdom: The Future of Patient-Centered Care and Research, sits at the heart of what we aim to do as
oncologists: improve the lives of patients with cancer. Harnessing our combined knowledge, not just the expertise of one, is how
we must deliver the highest quality of personalized care to meet the diverse needs of our patients.
Dr. Julie Vose, 20152016 ASCO President, challenged us to bring our readers a volume that reflected the multidisciplinary
aspects of cancer care, representing not only the roles of the many different types of health care providers and their specialties,
but also topics such as quality, cost, and survivorship. We answered by requesting articles authored with this interdisciplinary
view in mind, and I am pleased to say that we have more co-authored articles than ever before represented in this years
volume. To be sure, the level of effort required for this impressive collaboration of minds was no small feat, and I humbly thank
the more than 100 authors who generously took the time to write and, in some cases, revise the articles in this volume.
In the same vein, the Invited Articles in this volume illustrate how, as our knowledge of cancer evolves, we must include
various areas of expertise to transform how we deliver care to our patients, such as enlisting genetic counselors and enhancing
the role of advanced practitioners. Beyond that, applying our collective wisdom into practice also means expanding communication; we must incorporate results of molecular biology into treatment plans and converse honestly with patients about
treatment costs and care expectations, among other strategies. I graciously thank the authors of these articles for their
contributions to this volume.
This year, our readers will also find a new type of article that we are calling Points of View. Articles published under this
banner describe emerging and/or controversial topics in oncology care. I believe these articles are a most valuable addition to
the volume and highlight important care considerations for our patients as well as for the field of oncology.
Lastly, I want to recognize and thank our truly remarkable expert panel who dedicated their time and effort to careful,
thorough, and thoughtful reviews. Your commitment to education and ASCOs mission cannot be highlighted enough and is
tremendously underscored by this years theme.
It is my honor to invite you to read through the exceptional contributions that comprise the 2016 volume, only a selection of
which are found in the print edition. The print edition includes a curated selection of articles that most embody the power our
collective wisdom. For access to all of the 2016 ASCO Educational Book articles, as well as access to past volumes, please visit
www.asco.org/edbook.
Nate and I welcome your feedback and suggestions on how we can improve the content, so please contact us at edbook@
asco.org with your comments.
Sincerely,
Don S. Dizon, MD
Editor in Chief
INVITED ARTICLES
This years invited articles represent the 2016 ASCO Annual Meeting theme, Collective Wisdom: The Future
of Patient-Centered Care and Research. These important contributions to the 36th volume of the ASCO
Educational Book focus on the significance of a collaborative approach to how we provide care for our
patients. The authors represent a diverse, multidisciplinary set of expertise and backgrounds.
AUTHORS
Michael P. Kosty, MD, FACO, FASCO
Scripps Clinic
La Jolla, CA
Derek Raghavan, MD, PhD, FACP, FRACP, FASCO
Levine Cancer Institute, Carolinas HealthCare System
Charlotte, NC
Mary E. Sabatini, MD, PhD
Boston, MA
George W. Sledge, MD, FASCO
Stanford University School of Medicine
Stanford, CA
INVITED ARTICLES
COLLECTIVE WISDOM
Collaborative Practice in an Era of Multidisciplinary Care

Michael P. Kosty, MD, FACP, FASCO, Todd Pickard, MMSc, PA-C, and Pamela Viale, RN, MS, CNS, ANP
he clinical practice of oncology has become increasingly

complex. An explosion of medical knowledge, increased
demands on provider time, and involved patients have
necessitated changes in the way many oncologists practice.
What was an acceptable practice model in the past may
now be relatively inefficient. The American Society of Clinical
Oncologys (ASCOs) annual National Oncology Census demonstrates that with each successive year there is a shift
away from small independent group practice to larger independent groups or hospital-based affiliations.1-3 It is
unclear whether there will be further consolidation and/or
shift, or whether the rate of change will accelerate or
continue at its current pace. Ultimately, understanding the
practice environment surrounding providers who care for
patients with cancer is important to ensure that increasingly
complex therapies are delivered in an efficient, value-based
setting and that all patients have access to high-quality
cancer care.
The 2015 ASCO State of Cancer Care in America report
emphasized practice trends: workforce composition, health
systems innovation, regulatory compliance, and the financial
realities of cancer care today.4 Among the most pressing
issues the report highlighted was a growing population of
patients with cancer, increased complexity of care provided,
and an oncology workforce projected to fall short of the
expected demand (Fig. 1). In a recent survey of 22,000
oncologists, 11,700 medical oncologists were estimated to
provide direct care, managing the majority of patients with
cancer over extended periods of time.5 Factors contributing
to this predicted shortfall of providers and increasing
complexity of cancer care delivery are shown in Sidebar 1.6
Efficient multimodality therapy requires coordinated care
delivery among many diverse groups of clinical providers as
patients move along the continuum of cancer carefrom
risk assessment, prevention, diagnosis, treatment, and surveillance to survivorship and care of advanced cancer.
Within any given clinic, increasingly complex therapies
require clinicians with advanced, specialized training.
Clear communication and transparent, defined roles and

responsibilities help ensure that care needs are addressed
and timely decisions are made. Lasting change can only come
from explicitly helping to transform individual clinicians and
separate groups into a team that works together.
Teams are defined as two or more people who interact
dynamically, interdependently, and adaptively to accomplish a shared goal.7 The field of primary care has actively
engaged in reinventing care to form a patient-centered
medical home.8-10 Experiments are underway to apply the
same concepts to specialty care delivery.11-16 At the same
time, public and private payers and ASCO are proposing
payment models that would move away from payment
based on specific procedures and physician contributions
toward an approach that provides bundled payments for
comprehensive care and allows greater flexibility in how
care is organized and delivered.17 A team-based approach
can potentially leverage these changes, provide an opportunity to reexamine clinician roles and responsibilities, and
may enable the most efficient delivery of high-quality health
care services. Ongoing education, training, mentorship,
networking, and communication are necessary to cultivate
and maintain a collaborative team-based practice model.
Integration of resources from each practice setting, community organizations, e-health technologies, and advocacy
groups is essential. Human factors, health system factors,
situational factors, and socioeconomic factors are constantly
changing within the continuum of care, and they must be
considered in designing tailored patient and caregiver
support (Fig. 2).6 The National Cancer Institute (NCI) and
ASCO have collaborated to bring clinicians, patient advocates, and researchers together to explore application of the
evidence of team effectiveness to clinical practice. This
project fits into the larger context of the transformation of
health care delivery and payment models. The NCI-ASCO
Teams in Cancer Care Delivery project was presented at
the NCI-ASCO Teams in Cancer Care Delivery Workshop
in February 2016. Their findings will be published in the
From the Scripps Green Cancer Center, Scripps Clinic, La Jolla, CA; The University of Texas MD Anderson Cancer Center, Houston, TX; Department of Physiological Nursing, University of
California, San Francisco, San Francisco, CA.
Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook.
Corresponding author: Michael P. Kosty, MD, FACP, FASCO, Scripps Green Cancer Center, Scripps Clinic, 10666 North Torrey Pines Rd., MS217, La Jolla, CA 92037; email: mkosty@
scripps.edu.
2016 by American Society of Clinical Oncology.
KOSTY, PICKARD, AND VIALE
FIGURE 1. Baseline Projected Supply of and Demand

for Oncologist Visits, 2005 to 20205
Journal of Oncology Practice. In the context of increasingly

complex care requirements, increasing demand, and decreasing physician and economic resources, how do we create a
collaborative team-based practice model that allows for the
provision of high-quality, value-based oncology care?
INTEGRATING ADVANCED PRACTICE PROVIDERS

INTO ONCOLOGY PRACTICE
The cost of care is a primary concern in oncology today.
Collaborative practice models that provide mechanisms for revenue generation and reduce unnecessary
costs to patients through application of clinical practice
guidelines will promote patient and provider satisfaction.
Shulman18 suggested several goals for integrating advanced

practice providers (APPs) into collaborative practice
in oncology: (1) improved patient care, (2) increased
clinical productivity, (3) improved access for patients,
(4) urgent care patient treatment, (5) long-term care of
patients with cancer, and (6) coverage for the academic
physician. 6 The oncology APP is a licensed health care
professional who has completed advanced training in
nursing or pharmacy or has completed training as a physician
assistant.
The use of APPs to help meet the workforce demands and
improve the quality and coordination of care has been one of
the proposed solutions since the ASCO board of directors
approved the initial workforce strategic plan in 2008. The
utilization of APPs in oncology practices can increase
practice efficiency, productivity, and professional satisfaction of collaborating oncologists.19 The APPs role is diverse
and includes both inpatient and outpatient settings. Recent
publications have highlighted the practice of physician assistants who provide intensive palliative care on inpatient
units through an interdisciplinary team model.20 Similarly,
nurse practitioners have published their experiences as an
inpatient malignant hematology nurse practitioner service
working within a collaborative and multidisciplinary model.21
The utilization of APPs on an allogeneic stem cell transplant unit
resulted in decreased length of stay and reduced hospital
charges with similar patient outcomes.22 Towle and colleagues23 suggested similar roles for the APP in a collaborative
practice model. These included (1) assisting patients during
treatment visits; (2) pain and symptom management; (3)
follow-up care for patients in remission (survivorship care); (4)
patient education and counseling; (5) end-of-life care; and (6)
SIDEBAR 1. Factors Affecting the Demand and Complexity of Cancer Care6

1. Implementation of the Affordable Care Act has resulted in an increasing number of individuals gaining access to health care.
2. An aging population has Medicare as their primary insurance.
3. The number of cancer survivors is growing related to improvements in cancer detection, risk-adapted treatment strategies,
supportive care, and palliative care.
4. Increasing costs of care are resulting in a shift in practice models and the integration of formalized programs for preauthorization
and reimbursement.
5. The oncology workforce is aging (50% over the age of 50), with a shift toward group practices in urban settings (. 90%).
6. Cancer care initiatives are being set as standards of care or required for certification necessary to achieve designation or improve
revenue.
7. Meaningful Use benchmarks are being developed for the use of electronic health records and patient-reported outcomes and their
impact on physician productivity.
8. The American College of Surgeons Commission on Cancer published Cancer Program Standards 2012: Ensuring Patient-Centered
Care, which established new requirements around patient-centered needs and is expanding the focus on improving the quality of
care and patient outcomes. More recently, the Commission on Cancer has set a standard for distress screening for every patient
with cancer and their caregivers across the continuum of care.
9. Survivorship care: The Institute of Medicine, ASCO, and the Commission on Cancer have set guidelines for survivorship care. Cancer
survivors are projected to exceed 19 million by 2024.
10. Palliative care: The Institute of Medicine released its report, Improving Palliative Care for Cancer, in 2000. ASCO published
a provisional clinical opinion in 2012, recommending that palliative care be integrated into the care of every patient with cancer at
the time of diagnosis. The National Consensus Project put forth its Clinical Practice Guidelines for Palliative Care, a set of nationally
recognized guidelines that include quality measures and the eight domains of palliative care. The National Comprehensive Cancer
Network published the first clinical practice guidelines for palliative care in 2013.
COLLABORATIVE PRACTICE IN AN ERA OF MULTIDISCIPLINARY CARE
FIGURE 2. Collaborative Practice Relationships
Collaborative practice in oncology is a dynamic process focused on

interdisciplinary support of patients and their caregivers with a broad range of
health care providers. The AP in oncology plays a critical role in the collaborative
treatment of patients and their caregivers. Ongoing education, training,
mentorship, networking, and communication are necessary to cultivate and
maintain a collaborative practice model. Integration of resources from each
practice setting, community organizations, e-health technologies, and advocacy
groups is essential. Human factors, health system factors, situational factors, and
socioeconomic factors are ever-changing within the continuum of care and must
be considered in designing tailored patient and caregiver support.6
Abbreviations: AP, advanced practice provider.
Reprinted with permission from the Advanced Practitioner Society for
Hematology and Oncology.
ordering chemotherapy. The underlying theme in these

publications is that a collaborative practice model, with oncologists and APPs in oncology working together to the extent
of their training and licensure, can improve patient and provider satisfaction and safety and increase productivity and
revenue.24,25 However, the degree to which the use of APPs
can help meet future demand is unclear because the current

and projected workforce of APPs in oncology has been
challenging to determine. Based on census data from 2013,
the American Academy of Physician Assistants (AAPA) reported that there were an estimated 2,140 practicing
physician assistants in adult medical, surgical, and radiation
oncology subspecialties.2 This represented a 25% increase
compared with 2010 census data.3 However, important
characteristics such as age, geographic distribution, education,
years of experience, and years to expected retirement are unknown. Similar challenges are faced when trying to describe
the nurse practitioner workforce in oncology. The American
Association of Nurse Practitioners (AANP) publically reports
survey data for licensed nurse practitioners. In 2013, of the
more than 205,000 licensed nurse practitioners, approximately 2,050 worked in oncology and had been in practice for
an average of 7.7 years and had a median age of 48.25
Varied collaborative practice models are currently in use
based on the needs of the practice, patient volume, skills, and
training of the physician and the APP. Each has implications for
billing and productivity. The key to efficient integration of an
APP in oncology into a collaborative practice model is a careful
assessment of skills and knowledge about oncology practice.
At a high level, potential roles of an APP include enabling
physicians to focus on more complex and higher acuity
patient needs; extending the range of high-level services for
patients without using additional physician time; and providing follow-up, symptom management, and survivorship
care to patients. To accomplish this, three models of care
utilize APPs in the team-based care setting: (1) the autonomous or independent visit model, in which APPs predominantly see patients independently in a clinic but under a
collaborative practice agreement with their physicians; (2)
the shared visit model, in which patients are seen by both
the APP and the physician during the same clinical encounter; and (3) the mixed visit model, in which both the
independent and shared visit models are used to manage the
clinical volume but neither is the predominant encounter
FIGURE 3. Collaborative Practice Models Represented by the APSHO Practice Survey (192 patients)6,27
Abbreviation: APSHO, Advanced Practitioner Society for Hematology and Oncology.
used.26 The independent visit practice model goal is not to

supplant or replace physician involvement but to allow
physicians to delegate these tasks and medically direct
without a required physical presence at the time of service.
The shared visit model goal is to allow for combining
the individual efforts of the APPs and physicians into a
comanaged service that allows physicians to increase patient
volumes and the medical services they provide. This model
allows physicians to be continuously involved and to delegate aspects of the services provided to the APP. The collaborative practice models represented by the Advanced
Practitioner Society for Hematology and Oncology (APSHO)
practice survey are shown in Figure 3.
To examine the differences between the different models,
Buswell et al26 reported the effect of these practice models
on productivity, fees, and provider and patient satisfaction
in an academic cancer center. They found that productivity
for the independent, mixed, and shared visit models, as
measured by the number of new and established patients,
was similar for all models (6.8, 6.7, and 7.0 patients seen
per 4-hour session, respectively). Both physician and APPs
were very satisfied with the independent visit model and
reported patient-centered and productivity-based reasons
influencing the decision to use their chosen model. The
shared visit model physicians were still very satisfied with
the model whereas APPs were only moderately satisfied.
Reasons for utilizing the shared visit model were more
physician-centered, focusing on physician preferences and
perceptions. Importantly, there were extremely high levels
of patient satisfaction for both models (100% satisfaction
with care received from either model).
In a much larger study in the private practice setting, the
results of the ASCO study of collaborative practice arrangements similarly noted high levels of patient and provider
satisfaction with the different APP models.23 The most common model in the survey was the independent visit model. The
independent visit model was 19% more productive (based on
relative value units, RVUs) when the APP worked with the
entire group of physicians as compared with an independent
visit model when the APP worked exclusively with a limited
number of physicians. However, it may not be correct to
conclude that the most productive RVU model is the ideal
model in which to use APPs. Further insight into measures of
quality and continuity of care of the two models would be
important to distinguish. In addition, using RVUs as the sole
productivity measure is a limited assessment of the value an
APP might add to a practice. This study did not take into account the nonrevenue-generating activity performed for each
model, which could be important in defining the preferred
practice model. Value is not only about revenue generation;
quality of care and clinical productivity also are important and
might enhance physician quality of life.6,27
BARRIERS TO INTEGRATION
Provider and Patient
ASCOs study of the collaborative practice arrangements of
APPs identified physicians lack of interest in working with
6
APPs as the most common reason not to use them in their

practice.23 Exploring potential reasons for this lack of interest is important to determine how to best motivate attitudinal change. Because the ASCO report was primarily
physician-owned private practices (73%) with only 8% surveyed in academic practice, it is possible this lack of interest
was based on the fear of decreased physician compensation.
It has been shown that the private practice model has
significantly more oncologists compensated on an incentivebased model compared with academic models (39.3% vs.
3.1%). 28 Therefore, it may be important to focus on the
increased practice productivity when using APPs to
encourage utilization in private practice. Furthermore,
because a pure incentive-based model is associated with the
highest rate of burnout, increased professional satisfaction
when working with APPs can be another educational point to
change perceptions.
Other challenges to incorporating APPs into clinical
practice are largely historical or based more on personal bias
than fact.29 For example, data do not substantiate the belief
that utilizing APPs will negatively affect the physicianprovider relationship or that patients will not accept APPs
as part of the care team. Studies have demonstrated high
levels of patient and provider satisfaction with the collaborative practice model with increased utilization nationally.23 It is likely that the portion of the workforce nearing
retirement is also the same group that has less experience
and understanding of the physician assistant and nurse
practitioner professions and, therefore, more perceived
bias. This attitude is changing as new oncologists entering
the workforce come with experience working with APPs
during their fellowship. In a survey of fellowship program
directors in 2011, 90% of medical directors reported that
their fellows work with physician assistants and nurse
practitioners.30 What is unclear is how well prepared these
oncologists are to lead a medical team that incorporates
APPs. It will be important moving forward for oncologists to
understand the different models for APP utilization, as well
as the regulatory and reimbursement requirements to lead
the medical team effectively. Ideally, this educational need
will be incorporated into fellowship training programs before new oncologists enter the workforce and then will be
further refined at the practice level based on state laws and
institutional policies.
Productivity and Reimbursement

It is generally accepted that practices that incorporate APPs
are more productive and efficient in providing quality care to
patients than practices that do not. However, as practices
work to integrate APPs into clinical practice, they have been
challenged with accurately assessing the productivity and
value of individual APPs. Practices that use a system strictly
based on RVUs will likely underestimate the productivity and
value of APPs because of the inability to accurately measure
RVUs. For example, global surgical visits and the shared visit
model will render the time and effort of the APP invisible.31
COLLABORATIVE PRACTICE IN AN ERA OF MULTIDISCIPLINARY CARE
Even in the independent visit model, all incident-to visits and

visits for many commercial payers are billed under the
physicians national provider identifier, despite all care being
provided by the APP. In addition, numerous clinical care
activities that APPs provide are not billable encounters, but
they bring quality and value to the practice. Importantly, if
the APP does not complete these nonrevenue-generating
activities, the physician would have to complete them. The
challenges in assessing the productivity of APPs and the
limited benchmarking data available affect the ability to not
only improve productivity of the APPs, but it also will hinder
increased utilization. Practices will struggle to determine
when to hire new APPs and decide how their time should be
allocated to support the clinical enterprise. Also, practice
managers inherently will be unable to determine equitable
compensation, comparison, and accountability of APPs
within a practice.
As to reimbursement, several common myths and
misconceptions can stymie the expansion of APPs in
oncology. Impressions such as APPs cant see new patients or APPs cant bill above a certain level are easily
debunked with a little education and, if needed, support
from national advocacy organizations. However, one of
the more challenging misconceptions is the overestimation that the 85% reimbursement rate of the APPs
compared with the physician rate will have on the costeffectiveness of APPs. Numerous studies on physician
assistants and nurse practitioners have demonstrated
that APPs are cost-effective health care providers. This
can be explained by physician salaries being consistently
30% to 50% higher than APPs, incident-to billing reimbursed

at 100%, and the savings on APP reduced recruitment and
retention costs.
Currently, ASCO is exploring a project to assess the contribution of APPs in oncology practices. This project has
several goals: (1) count the number of APPs working in
oncology practices; (2) identify the mix of direct patient care
services and other work that APPs conduct in oncology
practices; (3) investigate the clinical, economic, and patient
effect of APPs in oncology practices; and (4) estimate how
current and future APP workforces will affect the shortage
that exists between demands for services and the number of
physician oncology providers.
CONCLUSION
Collaborative, multidisciplinary teambased care is essential if patients are to receive the highest quality valuebased oncology care. Achieving this goal will require effective integration of APPs into all aspects of patient care.
Although barriers remain, many are perceptual and relatively easy to overcome. Understanding the APP workforce, pipeline, and current utilizationas well as the
clinical, economic, and patient impact of APPs in oncology
practiceswill facilitate achieving the goal of optimal
patient-centered cancer care. More information on the
scope of practice of APPs, state regulations, payer policy,
and information on integration into oncology practices
can be found on the AAPA, Association of Physician Assistants in Oncology, AANP, and APSHO websites.
References
1. Forte GJ, Hanley A, Hagerty K, et al. American Society of Clinical Oncology National Census of Oncology Practices: preliminary report.
J Oncol Pract. 2013;9:9-19.
2. Yang W, Williams JH, Hogan PF, et al. Projected supply of and demand for
oncologists and radiation oncologists through 2025: an aging, betterinsured population will result in shortage. J Oncol Pract. 2014;10:39-45.
3. Hanley A, Hagerty K, Towle EL, et al. Results of the 2013 American
Society of Clinical Oncology National Oncology Census. J Oncol Pract.
2014;10:143-148.
4. American Society of Clinical Oncology. The state of cancer care in
America, 2015: a report by the American Society of Clinical Oncology.
J Oncol Pract. 2015;11:79-113.
5. Erikson C, Salsberg E, Forte G, et al. Future supply and demand for
oncologists: challenges to assuring access to oncology services. J Oncol
Pract. 2007;3:79-86.
6. Kurtin S, Peterson M, Goforth P, et al. The advanced practitioner and
collaborative practice in oncology. J Adv Pract Oncol. 2015;6:515-527.
7. Taplin SH, Weaver S, Chollette V, et al. Teams and teamwork during a
cancer diagnosis: interdependency within and between teams. J Oncol
Pract. 2015;11:231-238.
8. Kilo CM, Wasson JH. Practice redesign and the patient-centered medical
home: history, promises, and challenges. Health Aff (Millwood). 2010;
29:773-778.
9. Carrier E, Gourevitch MN, Shah NR. Medical homes: challenges in

translating theory into practice. Med Care. 2009;47:714-722.
10. Jackson GL, Powers BJ, Chatterjee R, et al. Improving patient care: the
patient-centered medical home: a systematic review. Ann Intern Med.
2013;158:169-178.
11. National Committee for Quality Assurance. Patient-Centered Specialty Practice Recognition. http://www.ncqa.org/Programs/Recognition/
Practices/PatientCenteredSpecialtyPracticePCSP.aspx. Accessed March
26, 2015.
12. Robeznieks A. At home with the specialist: oncologists and other
specialists launching patient-centered medical homes. Mod Healthc.
2014;44:24-25.
13. Innovative Oncology Business Solutions. COME HOME Model. www.
comehomeprogram.com/index.php/come-home-practices/. Accessed
March 26, 2015.
14. Oncology Management Services. Transforming Oncology. www.opcmh.
com/. Accessed March 26, 2015.
15. Community Oncology Alliance. Oncology Medical Home: Improved Quality
and Cost of Care. http://coaadvocacy.org/2014/09/oncology-medicalhome-improved-qualityand-cost-of-care/. Accessed March 26, 2015.
16. Medical Home Oncology. Commission on Cancer Accreditation. www.
medicalhomeoncology.org/coa/commission-on-cancer-accreditation.htm.
Accessed March 26, 2015.
17. Kosty MP, Bruinooge SS, Cox JV. Intentional approach to team-based
oncology care: evidence-based teamwork to improve collaboration and
patient engagement. J Oncol Pract. 2015;11:247-248.
18. Shulman, LN. Efficient and effective models for integrating advanced
practice professionals into oncology practice. Am Soc Clin Oncol Educ
Book. 2013;33:e377-e379.
19. Kosty M, Acheson A, Tetzlaff E. Clinical Oncology Practice 2015:
preparing for the future. Am Soc Clin Oncol Educ Book. 2015;35:
e622-e627.
20. Drury L, Baccari K, Fang A, et al. Providing intensive palliative care
on an inpatient unit: a full-time job. J Adv Pract Oncol. 2016;7:
60-64.
21. McNally GA, Florence KJ, Logue AC. The evolution of a malignant
hematology nurse practitioner service. Clin J Oncol Nurs. 2015;19:
367-369.
22. Shah NN, Kucharczuk CR, Mitra N, et al. Implementation of an advanced
practice provider service on an allogeneic stem cell transplant unit:
impact on patient outcomes. Biol Blood Marrow Transplant. 2015;21:
1692-1698.
23. Towle EL, Barr TR, Hanley A, et al. Results of the ASCO Study of Collaborative Practice Arrangements. J Oncol Pract. 2011;7:278-282.
24. Brown CG. Commentary: new findings substantiate the successful use
of nurse practitioners and physician assistants in collaborative practice
models. J Oncol Pract. 2011;7:285-286.
25. Welch WP, Cuellar AE, Stearns SC, et al. Proportion of physicians in large
group practices continued to grow in 2009-11. Health Aff (Millwood).
2013;32:1659-1666.
26. Buswell LA, Ponte PR, Shulman LN. Provider practice models in ambulatory oncology practice: analysis of productivity, revenue, and
provider and patient satisfaction. J Oncol Pract. 2009;5:188-192.
27. Kurtin SE, Viale PH, Hylton HM, et al. The APSHO practice survey. Paper
presented at: 3rd Annual Meeting of JADPRO Live, Advanced Practitioner
Society for Hematology and Oncology; November 2015; Phoenix, AZ.
28. Shanafelt TD, Gradishar WJ, Kosty M, et al. Burnout and career satisfaction among US oncologists. J Clin Oncol. 2014;32:678-686.
29. Tetzlaff E, Polansky M, Carr K, et al. Physician assistants in oncology.
J Oncol Pract. 2007;3:283.
30. Kosty MP. Informational itemsWAG, classic references, QOPI for
fellows. Paper presented at: Annual Meeting of the American Society of
Clinical Oncology; June 2011; Chicago, IL.
31. Pickard T. Calculating your worth: understanding productivity and
value. J Adv Pract Oncol. 2014;5:128-133.
INVITED ARTICLES
COLLECTIVE WISDOM
Value in Oncology: Balance Between Quality and Cost

Derek Raghavan, MD, PhD, FACP, FRACP, FASCO, and Mark W. Legnini, DrPH
t is a disturbing fact that the costs of health care are

spiraling upwards and that an exemplar of this troubling
trend is the domain of oncology treatment and research. By
2020, it is projected that cancer care will cost more than
$150 billion annually in the United States. Although this represents only a relatively small fraction of total health care
cost, cancer care is escalating more rapidly than most other
specialties. Many factors contribute to this unhappy situation, including the aging of the population, persistence of
risk-taking by the community at large (smoking, excessive
sun exposure, and lack of attention to industrial pollution,
to name a few), increasingly expensive research, diagnostic
tests, surgical approaches, radiotherapy techniques, and
novel systemic therapies as well as some unrealistic expectations of patients, families, and the community at large (often
predicated on false claims from our profession and/or hype
from the Fourth Estate). Of clear relevance is the consideration that another political declaration of war on cancer,
the third in 40 years, this time an amalgam of the State of the
Union address and some opportunism from the biomedical
community, is unlikely to fix the problem. Although this type
of hyperbole might increase focus, and even funding, it will
create a renewed sense that death is optional and will encourage patients and their physicians to continue futile attempts at active treatment of truly resistant disease, rather
than considering palliative care with its associated improvement in quality of life and reduction in cost.
A more sensible approach is to weave the concept of
value-based medical care into the equation, thus leveraging
resource expenditure in a more productive and sensible way
than has characterized much of modern medicine. However,
one of the more substantial challenges in modern oncology
practice is to provide true increase in value when so many
new parameters of success are being introduced into the
equation. Health care spending to treat cancer rose 55% in
the decade from 2001 to 2011, and annual retail prescription
drug expenditures for cancer quintupled over the same
period.1 Almost 40% of expenditures for cancer care in 2011
were publicly funded (34.1% Medicare and 5.3% Medicaid);

resulting federal scrutiny of health care spending meant
Washington policy makers measures of success often
became political and ideological footballs. Lobbying by the
health care industry to protect economic interests has
produced self-serving success stories, some from the
pharmaceutical and device industries and others from
providers of care locked in competition for patients and
revenues. The advocacy research behind these success
stories is frequently cited but often unvalidated and diverts
attention from a proper focus on the patient. Thus, oncologists must address the challenges of responding to the
communitys evolving requirements for greater value, without loss of innovation, in a more proactive, creative, and
structured fashion to preserve quality while attempting to
contain expense.
At its simplest, the value proposition in health care has
been defined by Porter and Teisberg2,3 with the following
equation:
Value Outcomes=Cost
This equation makes sense and is routinely used in planning
the strategy of cancer care for our health care systems in the
Carolinas and Michigan, as it helps us to consider what is
contributing to poor value in cancer care and allows resolution of those elements that are accessible. Of key importance is to critically consider the true benefits of the use of
our diagnostic and treatment modalities, defining whether
they contribute to prolongation of survival, increased quality
of life, or reduced morbidity of disease or its treatment, and
also whether the most cost-efficient options are chosen.
In addition, Porter and Teisburg2,3 have emphasized the
important principles of transparency, provision of comprehensive disease management and prevention services,
organization around medical conditions, and redefinition of
the health plan/subscriber relationship (with an end to costshifting practices) as potential solutions to the current
dysfunction in the provision of health care.
From the Levine Cancer Institute, Carolinas HealthCare System, Charlotte, NC; University of Michigan Medical School, Institute for Healthcare Policy & Innovation, Ann Arbor, MI.
Corresponding author: Derek Raghavan, MD, PhD, FACP, FRACP, FASCO, Levine Cancer Institute, Carolinas HealthCare System, 1021 Morehead Medical Dr., Charlotte, NC 28204;
email: derek.raghavan@carolinas.org.
RAGHAVAN AND LEGNINI
The prospect of cancer treatment is daunting for most

patients when one considers the nature of the treatment
being offered and its potential consequences. In addition,
asymmetry of medical information, potential fiscal toxicity,
and dislocation of patients and their families for treatment
far from their homes mean patients lose control over their
lives, compounding their anxiety over a cancer diagnosis and
the treatment that follows. The provision of symmetrical
(i.e., evenly distributed) access to high-quality care, by
overcoming barriers of poverty, language, geographical
isolation, and information while providing the key advances
in modern oncology, with a greater emphasis on palliative
care optimization, should improve national cancer survival
and morbidity figures without continuing to increase the
strain on federal and state budgets.4,5 A focus on universally
distributed access contrasts with the continued reluctance of some medical institutions to accept indigent, underinsured, Medicaid, or even Medicare patients, thus
protecting their bottom lines from loss, an approach that is
unfair to the community.3 In addition to creating maldistribution of the costs of providing care, cherry-picking of the
wealthiest, most educated, and robust patients (who also
usually have the strongest health insurance coverage) may
lead to biased representation of outcomes. These centers
publish data drawn from treatment of star patients, who
have fewer of the adverse prognosticators and comorbidities of unselected patient populations, leading to inflated
survival figures and artificially reduced morbidity and toxicity data. Sadly, many politicians and community leaders
may fail to understand this nuance when they translate the
resulting highly selected data into expectations for the
population at large, thus distorting the concept of true value.
Successful provision of symmetrical, high-quality outcomes also requires easily accessible tertiary and quaternary
level cancer care close to home, with access to clinical trials,
extensive patient support and navigation, optimal palliative
and supportive care, and the benefits of the genomic revolution as elements of routine cancer care.1,5 It also has to be
refined by the establishment of an evidence-based, systemwide set of standards for diagnosis and treatment, which are
routinely used by physicians.
In many major centers, following the innovative approach
of Fox Chase Cancer Center in the Delaware Valley, outreach
clinics have been established as a mechanism to provide
easier intake to attract new patients and provide access to
cancer trials. However, many of these ventures have not
been designed to provide all of the key resources, such as
access to translational trials with laboratory and biorepository support, genetic counseling, and patient-family
support, for these geographically isolated patients.
In the United States in recent times, less than 5% of patients have been enrolled in cancer clinical trials,6,7 and it is
clear that there are noteworthy disparities of access to highlevel cancer programs.4 We are now attempting to address
these challenging issues in the Carolinas8 and in Michigan,
two areas challenged by considerable geographical and
demographic constraints.
10
EXPERIMENTS IN THE DELIVERY OF

VALUE-BASED CANCER CARE
Toward Symmetry of Care in the Carolinas
Charlotte, the largest city in North Carolina, with a population approaching 1 million (and a referral base of around
2 million people) has not had local access to a comprehensive
cancer center. Specifically, there has been paucity of locally
available bone marrow transplant facilities and of phase I
and investigator-initiated clinical trials, although the latter
have been intermittently available on a small scale. Quaternary cancer services have been provided by outstanding
cancer centers at Emory, Duke, Wake Forest, and The
University of North Carolina at Chapel Hill, but have often
required travel of more than 3 to 4 hours for geographically
isolated patients and for those living in Charlotte who
have needed more specialized cancer services. Carolinas
HealthCare System, an amalgam of more than 40 hospitals
and medical centers spanning North Carolina and South
Carolina, employs more than 2,000 physicians and 50,000
staff and sees more than 15,000 new cancer cases per year.
As the system has expanded to be one of the largest
nonprofit health systems in North America, it made sense
to establish a tertiary and quaternary referral cancer
center that would serve its network of smaller centers
throughout the Carolinas, but which would attempt to
provide a system of cancer care closer to the patients
home, contrary to some of the more conventional models.
In design, several criteria were deemed crucial and incorporated into the creation of this multisite institute, as
summarized in the Sidebar.
Our strategy for the system throughout the Carolinas has
been to link with selected hospitals and their staffs,
establishing contractual relationships with the hospital
leadership and personal and professional relationships with
their physicians, which includes the provision of centrally
controlled Levine Cancer Institute clinical trial units. Where
needed, additional faculty members have been recruited
from a range of other academic cancer centers. Academic
and programmatic leadership has been drawn from National
Cancer Institutedesignated comprehensive cancer centers,
and trained clinical investigators have been placed at outreach centers to ensure symmetry of access to trials. We
have thus created a hybrid academic-practice model, and
this has led to dramatic increments in cancer trial accrual and
maintenance of patient care in distant geographic sites, with
improved access to tertiary and quaternary facilities, often
via telemedicine.
Strategies in Play in Michigan

The Michigan Value Collaborative provides 65 hospitals
statewide with data and tools to help them control costs for
episodes of care, including colorectal cancer resection,
esophagectomy, lung cancer resection, and prostatectomy,
up to 90 days postdischarge. This includes the initial
hospitalization plus skilled nursing, rehabilitation, home
health, transfers/readmissions, etc. The Michigan Value
VALUE IN ONCOLOGY: BALANCE BETWEEN QUALITY AND COST
SIDEBAR. Criteria for a Symmetrical System of Cancer Care
Multidisciplinary clinics with standard operating procedures and protocols of management, embedded in electronically linked,
evidence-based clinical pathways
System-wide interdisciplinary tumor boards and conferences, augmented by electronic two-way video conferencing to connect
geographically isolated oncology teams
Easy availability of clinical trials, with access close to home throughout North Carolina and South Carolina
A single, central institutional review board for cancer trials covering the whole system, facilitating swift and synchronous, systemwide activation of studies
Central protocol review and monitoring system, with initial protocol submissions via tumor-specific teams
Central oversight of a tightly controlled clinical trials unit, with central training and monitoring of staff
Centralized connections and data capture for each hospital tumor registry with ability to measure outcomes and costs of care
Extensive patient support services, including patient navigation linked throughout the system, standard operating procedures for
emergency departments throughout the system, palliative care and pain management services, cancer-focused pastoral care, and live
or video-linked genetic counseling and financial counseling
Availability of subspecialty services, such as bone marrow transplantation, phase I clinical trial units, and sophisticated radiation
techniques and equipment in as accessible a fashion as possible
A focus on translational bench research that is focused specifically on the clinical emphases of the instituteearly programs have
focused on cancer pharmacology, stem cell biology of hematologic disorders, and molecular prognostication with availability of a
cost-effective, molecular testing platform where appropriate
Optimal informatics technology support and electronic and video linkage for conferencing, tumor boards, and creation of electronic
pathways in tumor-specific team meetings
Collaborative creates a unique window for hospitals to see

charges generated for a patients care outside their four
walls, an essential tool for cost-effective care coordination.
In addition, Collaborative Quality Improvement initiatives around the state share clinical process and outcomes
data to improve outcomes for patients receiving surgical,
radiation, and medical oncology services. Together, these
programs provide clinical and claims-based registries (pricestandardized and risk-adjusted), web-based analytic tools,
and forums across the state for practicing physicians and
hospital quality improvement staff to work on individual improvement and collectively to make Michigan a
better place for patients with cancer to receive treatment.
Combining episode costs (the value equations denominator)
with clinical outcomes data from collaborative quality improvements (the numerator) gives Michigans hospitals one
of the countrys best opportunities to improve the value of
cancer care.
CHOOSING WISELY
One of the most important aspects to reduction of unnecessary expenditure in oncology is the consideration of
what is gained by the use of the available management
approaches. The Institute of Medicine has encouraged
physicians to carefully consider the benefits and drawbacks
associated, in particular, with expensive management options in the so-called Choosing Wisely campaign. In line
with this philosophy, the American Society of Clinical Oncology (ASCO) created the Value Task Force and developed
two consensus documents,9,10 based on the best available
evidence, that proposed discontinuation of standard
clinical practices that provided no patient benefit (Table 1).
In each instance, level 1 to 2 evidence clearly showed that
discontinuation of these diagnostic or management approaches would not result in reduced outcomes and in some
cases (e.g., screening for prostate cancer in elderly men with
intercurrent diseases and limited life expectancy) could
actually improve survival and/or morbidity. As can easily be
appreciated from Table 1, nationwide implementation of
these recommendations has achieved dramatic cost savings,
without measurable loss of quality or outcomes.
These items were chosen because they reflected clinical
practice among most oncologists and thus would avoid
partisan subspecialty rivalries and concerns, were broadly
applicable to large numbers of patients, and also clearly
reflected the absence of a robust evidence base underpinning patterns of standard clinical practice.
With the growing evidence on producing high value for
health expenditure, we will increasingly need to consider
clinical practices that have evolved without evidence to
support a survival or quality-of-life benefit and must focus on
whether the treatments selected have equi-active but less
costly alternatives (e.g., use of generic drugs, biosimilars,
and lower doses of cytotoxics, reduced treatment duration
of radiotherapy, and elimination of heroic but unproven
surgical procedures). In each case, the driver is cost containment or avoidance of profligate expenditure without
loss of quality or outcomes.
MEASUREMENT OF VALUE
In 2015, the ASCO Value Task Force published the Conceptual
Framework for Assessing Value in Cancer Care.11 The basic
construct is as follows:
Two major contextual categories are created, advanced
disease (e.g., treatment of metastases) and curative
intent (e.g., adjuvant therapy).
11
RAGHAVAN AND LEGNINI
Clinical benefit is given a score of 1 to 5, and then

numerical weightings are added to reflect the impact of
treatment on overall survival, progression-free survival,
or response rate. The weightings reflect the thought
that overall survival is more important than progressionfree survival, which is more important than response
rate when attempting to attribute value for the patient.
Negative scores are added to reflect the extent of
toxicity. In the advanced disease setting, bonus points
are awarded for palliation of symptoms and for longer
treatment-free intervals. Finally, the cost of treatment is
listed to broaden the context.
In the adjuvant setting, instead of the response-survival
parameters above, hazard ratio is incorporated into the
algorithm, and disease-free interval is substituted, but
with a slightly lower weighting.
The overall construct of net health benefit is incorporated to allow the patient to attribute likely value to the
treatment, balancing benefit(s) and toxicities.
Unbeknownst to the ASCO Task Force, the European
Society for Medical Oncology (ESMO) had also created a
similar task force, which produced the Magnitude of Clinical
Benefit Scale. Led by Dr. Nathan Cherny, the ESMO position
paper was published at about the same time as the ASCO
position paper.12 What is quite remarkable is the extraordinary similarity in concept and execution, achieved by two
committees working in the United States and in Europe, both
without knowledge of the work in progress of the other
team. The ESMO team also has focused on separate domains
of curative and palliative settings and attributed levels of
benefit predicated on absolute or relative increments in
survival and levels of toxicity. Of importance, they have set
time-dependent criteria, such that the value points allocated
for absolute survival increments differ for those with sustained increases versus the less fortunate patients with only
short-term increments in survival.
Interestingly, both committees used similar sets of level 1
data to develop their models and came to quite similar
interpretations of value, notwithstanding different community pressures, health care costs, and medical traditions
on different sides of the Atlantic.
These efforts are important and represent crucial first
steps by members of our profession in attempting to increase the level of self-discipline and value-based selfcriticism of what we do and increased transparency in
what we tell our patients about the benefits of treatment.
We would have preferred a higher bar with regard to attribution of value pointsit is obvious that a patient with a
prognosis of only 3 to 6 months will sustain greater relative
TABLE 1. ASCO Choosing Wisely List

Topic
Rationale
Level of Cost Savings*
Do not use cancer Rx if PS 3 to 4, no prior benefit,

ineligible for trial, and no evidence to support
more Rx
Logical, maximizes QOL, no loss of proven survival

benefit
Extensive at individual and national level
No extensive staging for low-risk early prostate

cancer
Overview of published data shows extremely low

positive yield within false-positive rate.
Extensive at national level
No extensive staging for low-risk early breast

cancer
Overview of published data shows extremely low

positive yield within false-positive rate.
No surveillance via biomarkers or imaging for

breast cancer after curative Rx (excludes
mammography)
No survival benefit from early diagnosis of metastatic

relapse
No white cell stimulatory prophylaxis for

patients with < 20% risk of febrile
neutropenia
Risk of febrile neutropenia is predictable for low-risk

patients; in high-risk patients, prophylaxis should
be considered
Avoid starting antinausea prophylaxis with most

expensive high-risk drugs for low-risk
emetogenic regimens
Effective options are available for control of regimens

at low-risk of emesis; crossover can easily be
affected if vomiting occurs
Avoid combination chemotherapy for salvage

Rx of metastatic breast cancer unless urgent
response is required
Single-agent Rx is an effective salvage option and

usually associated with less toxicity
Avoid post-Rx routine surveillance with PET or

CT-PET scanning after completion of Rx unless
evidence shows survival benefit
No data to support survival benefit of this approach

for most clinical settings
Avoid screening asymptomatic men with PSA

for prostate cancer if < 10 years of life
expectancy
Randomized data do not support survival benefit

from screening for prostate cancer and may cause
harm in elderly
Do not use targeted therapy against specific

gene aberration unless biomarker predicts
likely response
Very low response rate to most targeted Rx unless

specific target is expressed by tumor
Abbreviation: Rx, prescription; PS, pain scale; QOL, quality of life; PSA, prostate-specific antigen.
*Individual reflects potential for substantial copays or expensive self-pay. National reflects costs associated with specific service delivery as well as incidence/frequency figures at
national level.
12
VALUE IN ONCOLOGY: BALANCE BETWEEN QUALITY AND COST
benefit from a 2-month increment than someone treated

with curative or aggressive palliative intent; that said, a
2-month survival increment, offset by considerable physical
or fiscal toxicity, needs to be taken into context. This is
complex, paradigm-shifting work and will require careful
and critical attention and feedback.
TRAINING THE NEXT GENERATION OF

PHYSICIANS TO INCORPORATE THE VALUE
PROPOSITION
Training oncology residents to provide high-value care is key
to transforming the practice of oncology in the United
States.13 Current policies to improve the performance of
health care practitioners and the institutions in which they
practice depend primarily on financial incentives aimed at
changing the behavior of fully-trained physicians. This is
much less effective than intervening earlier in the workforce
pipeline to influence physicians during their training and
professional socialization.
Expressed simply, residents who do not train in high-value
care settings are less likely to become high-value physicians.13 Residency faculty must be skilled in the organization
and delivery of high-value care and in how to teach those
skills.
Exhortations without systems transformation will not
break the cycle of teaching each generation of oncologists
outmoded ways of thinking about safety, quality, and costeffectiveness during residency, when their decision making
habits, professional values, ways of interacting with patients and colleagues, etc., are shaped for the rest of their
professional lives. One way to transform training is to ensure

that teaching programs are located in care settings in which
residents see high-value care delivered every day. This will
require that funding for graduate medical education be
value-based to provide incentives for training programs to
change. Together, the federal Medicare program and a
majority of state Medicaid programs contribute approximately $14 billion annually to fund graduate medical education, based primarily on counts of trainees and residency
program accreditation. If these payments were value-based,
that would be a strong incentive for programs to ensure that
succeeding cohorts of oncologists emerge from training
imbued with the concept of value in health care and
equipped with the tools to provide safe, high-quality, costeffective cancer care.
SUMMARY
The exponential increase in costs of health care is unnecessary and reflects many avoidable factors at a community level, including poor health practices, unrealistic
expectations, corporate profiteering, and a poor medical
decision process (which often contravenes level 1 to 2 evidence). Physicians must increasingly consider true value
(outcome/cost ratio) when creating management plans and
include these considerations in transparent and realistic
conversations with patients. Attention to these issues will
dramatically reduce the burgeoning costs of cancer care in
our community while improving the quality and value of
care.
References
1. Soni A. Trends in Use and Expenditures for Cancer Treatment among
Adults 18 and Older, U.S. Civilian Noninstitutionalized Population, 2001
and 2011. Medical Expenditure Panel Survey, Statistical Brief #443.
http://meps.ahrq.gov/mepsweb/data_files/publications/st443/
stat443.pdf. Accessed February 23, 2016.
2. Porter ME, Teisberg EO. How physicians can change the future of health
care. JAMA. 2007;297:1103-1111.
3. Porter ME, Teisberg EO. Redefining Health Care. Boston, MA: Harvard
Business School Press; 2006;97-283.
4. Raghavan D. Slow progress in cancer care disparities: HIPAA, PPACA,
and CHEWBACCA... but were still not there! Oncologist. 2011;16:
917-919.
5. Goss E, Lopez AM, Brown CL, et al. American Society of Clinical Oncology
policy statement: disparities in cancer care. J Clin Oncol. 2009;27:
2881-2885.
6. Lara PN Jr, Higdon R, Lim N, et al. Prospective evaluation of cancer
clinical trial accrual patterns: identifying potential barriers to enrollment. J Clin Oncol. 2001;19:1728-1733.
7. Raghavan D. An essay on rearranging the deck chairs: whats wrong with
the cancer trials system? Clin Cancer Res. 2006;12:1949-1950.
8. Raghavan D. Costs of cancer care: rhetoric, value, and steps forward.

Semin Oncol. 2013;40:659-661.
9. Schnipper L, Smith TJ, Raghavan D, et al. American Society of Clinical
Oncology identifies five key opportunities to improve care and reduce
costs: the top five list for oncology: ASCOs top five list. J Clin Oncol.
2012;30:1715-1724.
10. Schnipper LE, Lyman GH, Blayney DW, et al. American Society of Clinical
Oncology 2013 top five list in oncology. J Clin Oncol. 2013;31:
4362-4370.
11. Schnipper LE, Davidson NE, Wollins DS, et al; American Society of
Clinical Oncology. American Society of Clinical Oncology statement:
a conceptual framework to assess the value of cancer treatment options. J Clin Oncol. 2015;33:2563-2577.
12. Cherny NI, Sullivan R, Dafni U, et al. A standardised, generic, validated
approach to stratify the magnitude of clinical benefit that can be anticipated from anti-cancer therapies: the European Society for Medical
Oncology Magnitude of Clinical Benefit Scale (ESMO-MCBS). Ann Oncol.
2015;26:1547-1573.
13. Legnini, MW. Can low-performing hospitals train high-performing
residents? Am J Med Qual. 2011;26:408-410.
13
INVITED ARTICLES
COLLECTIVE WISDOM
Womens Health Issues for BRCA Mutation Carriers

Mary E. Sabatini, MD, PhD, and Leif W. Ellisen, MD, PhD
n 2015, it was estimated that there were 1.6 million new

cases of cancer diagnosed in the United States.1 Remarkably,
more than 86,000 were diagnosed in women under the age
of 45. The majority of cancers are thought to be a result of
bad luck; that is, random mutations arising during DNA
replication in normal stem cells,2 whereas inherited mutations are thought to be causal in 5%10% of cancers. Thus,
the number of cancer cases that are attributable to inherited
mutations may seem small, however, they are clinically
important. Cancers in patients with hereditary cancer syndromes tend to occur at a younger age, and most hereditary
cancer syndromes are also associated with more than one
type of cancer.3
Approximately 20% of women with inherited breast and/
or ovarian cancers harbor a harmful mutation of one of the
breast cancer susceptibility genes 1 or 2, known as BRCA1 and
BRCA2.3,4 Mutations in BRCA1/2 are inherited in an autosomal
dominant fashion. Families that carry germline mutations may
have several women affected by breast and possibly ovarian
cancer and men with breast cancer and prostate cancer. These
mutations also predispose carriers to cancers such as fallopian
tube cancer, peritoneal cancer, endometrial cancer, pancreatic
cancer, colon cancer, and cancers in other sites. Although there
are many issues for BRCA mutation carriers who have been
diagnosed with cancer, this article will focus on the care of
women who are known carriers but who have not been diagnosed with cancer.
For BRCA1 mutation carriers, cumulative estimates of
developing cancer by age 70 vary by population, ranging
from 40% to 85% for breast cancer and 10% to 59% for
ovarian cancer. For BRCA2 mutation carriers, cumulative
estimates of developing cancer are 45% to 57% for breast
cancer and 11% to 18% for ovarian cancer.5-7 Because of
these risks, the National Comprehensive Cancer Network
(NCCN) has established guidelines for women with BRCA
mutations to reduce the risk of cancer.8 This includes breast
awareness and breast self-examinations starting at age 18,
with annual to biannual clinical examinations starting at age
25. Breast screening should begin by age 25 with MRI.

Earlier MRIs are also considered if any family member
was diagnosed with breast cancer before age 25. Annual
mammography and MRI starts at age 30. Risk-reducing
mastectomy should be offered as an option to all BRCA
mutation carriers, as it has been shown to reduce the risk
of cancer by as much as 90%, although definitive studies
showing either an overall or breast cancerspecific survival advantage are lacking. In contrast, risk-reducing bilateral salpingo-oophorectomy (rrBSO) at age 35 to 40
or whenever childbearing is completed is associated with a
lower risk of breast cancer and reduces all-cause, ovarian
cancerspecific, and breast cancerspecific mortality.9
rrBSO results in an 80% risk reduction for ovarian, fallopian
tube, and peritoneal cancers. There has been recent interest
in the possibility of performing salpingectomy only, as it
seems that some ovarian cancers may arise primarily in the
fallopian tube, but this is not recommended other than in
the setting of a clinical trial. Hysterectomy is not routinely
recommended given that the absolute risk of endometrial
cancer is low enough that the benefits do not outweigh the
risks of surgery. For those who do not elect for rrBSO,
ovarian cancer screening with transvaginal ultrasound and
carbohydrate antigen 125 (CA-125) is suggested every
6 months starting at age 30. Notably, however, there are no
data to support the efficacy of this practice.
The NCCN guidelines are established by incorporating
the best evidence available to date. However, in caring for
women who do carry a deleterious BRCA mutation, one of
the major challenges is trying to help an individual patient to
make decisions using data obtained from studying populations. Like most autosomal-dominant mutations, penetrance of a BRCA mutation is not 100%. The probability of
developing cancer for carriers is alarmingly high, yet the
morbidity associated with surveillance and/or prevention
is also not negligible. Furthermore, taking steps toward
prevention does not absolutely eliminate the risk of
cancer development. Decision making can be empowering,
From the Department of Obstetrics and Gynecology, Harvard Medical School, Massachusetts General Hospital, Boston, MA; Breast Medical Oncology, Harvard Medical School,
Massachusetts General Hospital Cancer Center, Boston, MA.
Corresponding author: Mary E. Sabatini, MD, PhD, Massachusetts General Hospital, 55 Fruit St., Boston, MA 02114; email: msabatini@mgh.harvard.edu.
14
WOMENS HEALTH ISSUES FOR BRCA MUTATION CARRIERS
as indicated by the New York Times opinion article written

by Angelina Jolie.10 Nonetheless, decisions can be anxiety
provoking, particularly when the consequences of the
wrong decision are so large and life altering.
Unfortunately, knowledge of ones mutation and the NCCN
guidelines does not necessarily enable an individual to plan
ones life accordingly. In an ideal world, perhaps every woman
who carried a BRCA mutation would have completed childbirth prior to age 35. Indeed, younger age at first birth and
breast-feeding are protective for development of breast and
ovarian cancers overall. Additionally, in an ideal world, there
would be no repercussions to mastectomy and early surgical
menopause. However, these ideals run counter to the realities
of life for many. In high-income countries, the median age of
first birth is rising,11 with many women not starting a family
until after their 35th birthday. Surgery and surgical menopause
can cause morbidity and can be life altering.
A retrospective study including 305 BRCA mutation carriers
seen in a California integrated health system between 1995
and 2012 showed that the uptake of options for BRCA mutation carriers who were informed of the NCCN guidelines was
lacking in some areas. The median follow-up during the
study was 41 months. The median time from diagnosis to riskreducing surgery was 6 months, with 74% of patients choosing
to undergo rrBSO and 44% of patients choosing to undergo
risk-reducing mastectomy. Of those choosing surveillance,
compliance was 45% in the first year but decreased to
approximately 10% at 3 years and 2% at 5 years.12
Decisions about prophylactic surgery are complicated
by concerns about appearance, sexuality, control of reproduction, and relationships. A retrospective study by
Stukey et al13 highlights that these decisions among carriers
are not, as might be anticipated, based on risk alone. Ninety
women, all of whom were carriers of BRCA mutation, were
offered prophylactic surgery. Fifty-one percent of the
women underwent surgery, 85% underwent BSO, and 28%
underwent mastectomy. Women who elected to have
surgery were more likely to be parous, married, employed,
and had a prior history of breast cancer. This implies that
those who did not elect for prophylactic surgery may be
interested in future fertility. BRCA2 mutation carriers were
more likely than BRCA1 carriers to pursue surgery. If
avoidance of developing disease were the only factor involved in decision making, BRCA1 carriers should be more
likely to undergo surgery.
Hoskins et al14 described 11 cases of unmarried women
age 26 to 35 who carry a BRCA mutation. These women
relayed challenges regarding their perception of their own
desirability and anxiety about partners reactions to disclosure of mutation status. Another web-based survey of 44
women with a BRCA mutation included 13 women who were
unmarried, and all of them expressed concerns about how
and when to disclose their status to future partners. Most of
them expressed a sense of urgency to have children.15
Considering the anxiety expressed about the effect of their
own BRCA mutation on their relationships, very little data
exist regarding the attitudes of the sexual partners of
women with BRCA mutations. One recent survey study

described the reaction of 25 male partners about their female partners mutation status.16 Twenty of the men were
married or engaged at the time of the survey, and 19 had
been in the relationship at least 5 years. Nine participants
reported changes regarding intimacy levels, and two respondents reported changes in their level of attraction.
Almost all participants noted changes in their communication after mutation disclosure, with changes in the
topics discussed and more discussions about the future.
Interestingly, of the 14 men whose partners had not undergone a risk-reducing mastectomy, five expressed concerns
about their level of attraction to their partner, but of the 11
men whose partner had had a mastectomy, all said that they
were as attracted to their partner after surgery as they were
before surgery. It is difficult to generalize the results of this
study broadly, as most of these men were in long-term or
committed relationships at the time of the survey.
In general, reproductive options for all women have vastly
expanded in the past several years, and this may also benefit
those who carry a BRCA mutation. For women who are not
yet ready to have children, in vitro fertilization (IVF) can be
used to freeze either oocytes or embryos for future use.
Embryo cryopreservation has been available since the 1980s,
and outcomes have improved dramatically in the last several
years. For couples with infertility and a female partner
younger than age 35, expected pregnancy rates after a single
IVF are upwards of 50%.17 For couples with unknown fertility
interested in banking embryos for future use, there are no
numbers to guide us, but presumably the success rate would
be at least as good if not higher. Oocyte freezing is available
to women who are unpartnered, unwilling to use donor
sperm, or who wish to maintain complete autonomy of their
reproductive potential. Historically, oocyte freezing has been
technically more challenging. Although embryo freezing
remains the gold standard, this will likely change in the near
future as success rates with frozen oocytes are approaching
those of frozen embryos.17
Although embryo and oocyte preservation provide options, IVF is expensive, arduous, and time consuming. It
comes with risks including hemorrhagic cysts, ovarian torsion, ovarian hyperstimulation syndrome, blood clots, infections, medication reactions, and surgical risks associated
with oocyte extraction. For these reasons, it is not universally available or used. Overall, IVF has not been associated
with an increased risk of development of breast or ovarian
cancer18-22 in the general population or in BRCA mutation
carriers.23,24 The data for the general population are quite
robust as a result of the length of availability of IVF and the
number of people who have used IVF. For those with BRCA
mutations, however, the numbers are smaller and not as
robust, so the conclusions are not as definitive.
One of the most challenging issues for women who carry
a BRCA mutation surrounds the choice about the potential
to eliminate this mutation from future offspring. Preimplantation genetic diagnosis (PGD) is a procedure in which
a cell or cells from a human embryo are removed and tested
15
SABATINI AND ELLISEN
for diseases and/or mutations that can result in disease (for

example, cystic fibrosis, Huntington disease, BRCA mutation).
PGD requires IVF. The resulting embryos are allowed to grow
for 3 to 5 days so there are enough cells to be biopsied for PGD.
Embryos found by biopsy to be unaffected (on average approximately 50%) can be transferred into the womans uterus
with the hope that a normal pregnancy will be established.
Affected embryos generally are not transferred and can be
disposed per a couples wishes.
The topic of PGD can raise a number of social and ethical
debates. A recent large survey study of the general population
showed that approximately 70% of respondents were supportive of PGD for diseases that were lethal in early childhood or
caused lifelong disability (e.g., cystic fibrosis), yet only 48%
supported it for screening embryos for diseases that strike later
in life (e.g., BRCA-associated cancers or Huntington disease).25
Another survey assessed the attitudes about PGD among
women with BRCA mutations and/or hereditary cancer syndromes. Of the 52 respondents, 75% thought that PGD was an
acceptable option for those with BRCA mutation.26 Interestingly,
though, of the respondents who had completed family building,
only 37.5% would have used this option for themselves, and for
those who did not yet have children, only 14% stated they would
consider using PGD. Thirty-five percent of the respondents
stated they worried about the use of PGD because they assume
that the effectiveness of screening and prevention for women
with BRCA mutation will improve with time.
Choices about the use of birth control and/or any other
hormone treatments can also be complicated for BRCA
carriers. Oral birth control pills (OCPs; containing either
estrogen and progestin or just progestin) are the most
common form of reversible contraception used worldwide.27 A very large cohort study showed that in the general
population OCPs are not associated with an increased risk
overall for cancer, and OCP users had a reduced risk of
ovarian, endometrial, and colorectal cancers.28 OCP use has,
however, also been linked to a trend toward increased risk of
breast cancer, but studies are conflicting.29-32 The literature
regarding OCP use for BRCA carriers is similar.33,34 Therefore,
the choice about use of OCPs must be individualized and
account not only for BRCA status but also other health issues
and reproductive goals.
Surgical menopause causes a more dramatic drop in
systemic levels of ovarian hormones than the gradual decline that occurs with natural menopause. Additionally, for
BRCA carriers who undergo rrBSO per NCCN guidelines,
surgical menopause happens at an earlier age than does

natural menopause. As such, menopausal symptoms, the
effects of hormonal changes, and their treatment should be
discussed with women prior to surgery. Menopause is associated with hot flashes, sleep disturbances, and changes in
mood. Women often have a decline in libido and sexual
satisfaction and have vaginal changes that result in difficulty
with intercourse; early menopause may also dramatically
impact bone and cardiovascular health.35,36
There are several medical and alternative treatments that
can be used to mitigate the effects of menopause; the most
effective of these is hormone replacement therapy (HRT)
containing estrogen. The use of HRT in women with BRCA
mutation can be somewhat of a contentious subject. Currently,
most of the literature does not show an increased risk of breast
cancer following rrBSO for BRCA carriers who use HRT.37,38
Women who use HRT report less discomfort with intercourse
and a higher quality of life than nonusers.39 Again, because of
the absence of long-term studies, the lack of randomization,
and the variations in treatments used, it is still difficult to
counsel patients about the safety of HRT. Any woman who
has a uterus should also be counseled about the use of progestin for endometrial protection. This also may impart a higher
risk of breast cancer based on previous data,40,41 but how this
applies to those with BRCA mutations is unclear.
In a recent article about her BRCA status and medical
decisions, Angelina Jolie said, Life comes with many challenges. The ones that should not scare us are the ones we can
take on and take control of.10 In that article, Jolie speaks
much about her family, her partner, and her children, which
no doubt gave her clarity on her health decisions. She did not
mention anything about PGD, so we are unclear about her
determination regarding this issue. Jolie also has vast access
to medical care and, thus, has many treatment options. For
many women, lifes circumstances do not make decisions so
clear. Decisions can be empowering, but they can also bring
about stress and anxiety. Fortunately, most of the studies
show that women who undergo risk-reducing surgery are
satisfied with their decision.42,43 Unfortunately, there are
very few studies that assess womens satisfaction with their
choices about nonintervention or appraise womens feelings
over long periods of time. Much additional work is needed to
help patients and the medical community deal effectively
with these complex issues. Even with strong data, decision
making can be complicated; but without data, it is all the
more difficult.
References
1. American Cancer Society. Cancer Facts & Figures 2015. www.cancer.org/
research/cancerfactsstatistics/cancerfactsfigures2015. Accessed November 12, 2015.
2. Tomasetti C, Vogelstein B. Cancer etiology. Variation in cancer risk
among tissues can be explained by the number of stem cell divisions.
Science. 2015;347:78-81.
16
3. American Cancer Society. Family Cancer Syndromes. www.cancer.org/

cancer/cancercauses/geneticsandcancer/heredity-and-cancer. Accessed
January 8, 2016.
4. Couch FJ, Nathanson KL, Offit K. Two decades after BRCA: setting
paradigms in personalized cancer care and prevention. Science. 2014;
343:1466-1470.
WOMENS HEALTH ISSUES FOR BRCA MUTATION CARRIERS
5. Antoniou A, Pharoah PD, Narod S, et al. Average risks of breast and

ovarian cancer associated with BRCA1 or BRCA2 mutations detected in
case series unselected for family history: a combined analysis of 22
studies. Am J Hum Genet. 2003;72:1117-1130.
6. Chen S, Parmigiani G. Meta-analysis of BRCA1 and BRCA2 penetrance.
J Clin Oncol. 2007;25:1329-1333.
7. Mavaddat N, Peock S, Frost D, et al; EMBRACE. Cancer risks for BRCA1
and BRCA2 mutation carriers: results from prospective analysis of
EMBRACE. J Natl Cancer Inst. 2013;105:812-822.
8. National Comprehensive Cancer Network Clinical Practice Guidelines in
Oncology. Genetic/Familial High-Risk Assessment. Version 2. 2015.
www.nccn.org/professionals/physician_gls/pdf/genetics_screening.
pdf. Accessed January 8, 2016.
9. Domchek SM, Friebel TM, Singer CF, et al. Association of risk-reducing
surgery in BRCA1 or BRCA2 mutation carriers with cancer risk and
mortality. JAMA. 2010;304:967-975.
10. Jolie A. My Medical Choice. www.nytimes.com/2013/05/14/opinion/
my-medical-choice.html. Accessed February 6, 2016.
11. Finer LB, Philbin JM. Trends in ages at key reproductive transitions in
the United States, 1951-2010. Womens Health Issues. 2014;24:
e271-e279.
12. Garcia C, Wendt J, Lyon L, et al. Risk management options elected by
women after testing positive for a BRCA mutation. Gynecol Oncol. 2014;
132:428-433.
13. Stuckey A, Dizon D, Scalia Wilbur J, et al. Clinical characteristics and
choices regarding risk-reducing surgery in BRCA mutation carriers.
Gynecol Obstet Invest. 2010;69:270-273.
14. Hoskins LM, Roy K, Peters JA, et al. Disclosure of positive BRCA1/2mutation status in young couples: the journey from uncertainty
to bonding through partner support. Fam Syst Health. 2008;26:
296-316.
15. Hamilton R. Being young, female, and BRCA positive. Am J Nurs. 2012;
112:26-31, quiz 46, 32.
16. Mauer C, Spencer S, Dungan J, et al. Exploration of male attitudes on
partnerships and sexuality with female BRCA1/2 mutation carriers.
J Genet Couns. Epub 2015 Aug 8.
17. SART CORS. Clinic Summary Report 2013. www.sartcorsonline.com/
rptCSR_PublicMultYear.aspx?ClinicPKID=0. Accessed January 21,
2016.
18. Rizzuto I, Behrens RF, Smith LA. Risk of ovarian cancer in women treated
with ovarian stimulating drugs for infertility. Cochrane Database Syst
Rev. 2013;8:CD008215.
19. Sergentanis TN, Diamantaras AA, Perlepe C, et al. IVF and breast cancer:
a systematic review and meta-analysis. Hum Reprod Update. 2014;20:
106-123.
20. Gennari A, Costa M, Puntoni M, et al. Breast cancer incidence after
hormonal treatments for infertility: systematic review and metaanalysis of population-based studies. Breast Cancer Res Treat. 2015;
150:405-413.
21. Zhao J, Li Y, Zhang Q, et al. Does ovarian stimulation for IVF increase
gynaecological cancer risk? A systematic review and meta-analysis.
Reprod Biomed Online. 2015;31:20-29.
22. Kessous R, Davidson E, Meirovitz M, et al. The risk of female malignancies after fertility treatments: a cohort study with 25-year follow-up.
J Cancer Res Clin Oncol. 2016;142:287-293.
23. Gronwald J, Glass K, Rosen B, et al; Hereditary Breast Cancer Clinical
Study Group. Treatment of infertility does not increase the risk of
ovarian cancer among women with a BRCA1 or BRCA2 mutation. Fertil
Steril. Epub 2015 Dec 14.
24. Perri T, Lifshitz D, Sadetzki S, et al. Fertility treatments and invasive
epithelial ovarian cancer risk in Jewish Israeli BRCA1 or BRCA2 mutation
carriers. Fertil Steril. 2015;103:1305-1312.
25. Winkelman WD, Missmer SA, Myers D, et al. Public perspectives on the
use of preimplantation genetic diagnosis. J Assist Reprod Genet. 2015;
32:665-675.
26. Menon U, Harper J, Sharma A, et al. Views of BRCA gene mutation
carriers on preimplantation genetic diagnosis as a reproductive option
for hereditary breast and ovarian cancer. Hum Reprod. 2007;22:
1573-1577.
27. United Nations. Population Facts. www.un.org/en/development/desa/
population/publications/pdf/popfacts/popfacts_2013-9.pdf. Accessed
January 14, 2016.
28. Hannaford PC, Selvaraj S, Elliott AM, et al. Cancer risk among users of
oral contraceptives: cohort data from the Royal College of General
Practitioners oral contraception study. BMJ. 2007;335:651.
29. Oral-contraceptive use and the risk of breast cancer. The Cancer and
Steroid Hormone Study of the Centers for Disease Control and the
National Institute of Child Health and Human Development. N Engl J
Med. 1986;315:405-411.
30. Hankinson SE, Colditz GA, Manson JE, et al. A prospective study of oral
contraceptive use and risk of breast cancer (Nurses Health Study,
United States). Cancer Causes Control. 1997;8:65-72.
31. Marchbanks PA, McDonald JA, Wilson HG, et al. Oral contraceptives and
the risk of breast cancer. N Engl J Med. 2002;346:2025-2032.
32. Vessey M, Yeates D. Oral contraceptive use and cancer: final report from
the Oxford-Family Planning Association contraceptive study. Contraception. 2013;88:678-683.
33. McLaughlin JR, Risch HA, Lubinski J, et al; Hereditary Ovarian Cancer
Clinical Study Group. Reproductive risk factors for ovarian cancer in
carriers of BRCA1 or BRCA2 mutations: a case-control study. Lancet
Oncol. 2007;8:26-34.
34. Iodice S, Barile M, Rotmensz N, et al. Oral contraceptive use and breast
or ovarian cancer risk in BRCA1/2 carriers: a meta-analysis. Eur J Cancer.
2010;46:2275-2284.
35. Mercuro G, Zoncu S, Saiu F, et al. Menopause induced by oophorectomy
reveals a role of ovarian estrogen on the maintenance of pressure
homeostasis. Maturitas. 2004;47:131-138.
36. Lobo RA. Surgical menopause and cardiovascular risks. Menopause.
2007;14:562-566.
37. Madalinska JB, Hollenstein J, Bleiker E, et al. Quality-of-life effects of
prophylactic salpingo-oophorectomy versus gynecologic screening
among women at increased risk of hereditary ovarian cancer. J Clin
Oncol. 2005;23:6890-6898.
38. Kotsopoulos J, Huzarski T, Gronwald J, et al. Hormone replacement
therapy after menopause and risk of breast cancer in BRCA1 mutation
carriers: a case-control study. Breast Cancer Res Treat. 2016;155:
365-373.
39. Johansen N, Liavaag AH, Tanbo TG, et al. Sexual activity and functioning
after risk-reducing salpingo-oophorectomy: impact of hormone replacement therapy. Gynecol Oncol. 2016;140:101-106.
40. Chlebowski RT, Hendrix SL, Langer RD, et al; WHI Investigators. Influence of estrogen plus progestin on breast cancer and mammography in
healthy postmenopausal women: the Womens Health Initiative Randomized Trial. JAMA. 2003;289:3243-3253.
41. Chlebowski RT, Rohan TE, Manson JE, et al. Breast cancer after use of
estrogen plus progestin and estrogen alone: analyses of data from 2
Womens Health Initiative Randomized Clinical Trials. JAMA Oncol.
2015;1:296-305.
42. Borreani C, Manoukian S, Bianchi E, et al. The psychological impact of
breast and ovarian cancer preventive options in BRCA1 and BRCA2
mutation carriers. Clin Genet. 2014;85:7-15.
43. Finch A, Narod SA. Quality of life and health status after prophylactic
salpingo-oophorectomy in women who carry a BRCA mutation: a review. Maturitas. 2011;70:261-265.
17
INVITED ARTICLES
COLLECTIVE WISDOM
Collective Wisdom: Lobular Carcinoma of the Breast

George W. Sledge, MD, FASCO, Anees Chagpar, MD, and Charles Perou, PhD
reast cancer researchers have a longstanding fascination

for infiltrating lobular carcinomas. Second in frequency
only to ductal adenocarcinomas, these tumors are characterized by unique histopathology and (among breast cancers) distinctive clinical biology, both in the primary and
metastatic settings. Early hints regarding the underlying
sources of this peculiar cancer, in particular the important
role of E-cadherin loss, have now been confirmed through
comprehensive molecular portraits of the disease. These
molecular observations, in turn, go far to explain how lobular
carcinomas play out in the clinic, both as regards local
control and therapeutic response to systemic therapy. Together, the collective wisdom of the laboratory and the
clinic paint an interesting portrait of this fascinating disease.
MOLECULAR BIOLOGY OF LOBULAR

CARCINOMAS
Invasive lobular breast carcinoma (ILC) is the second most
prominent histologic form of breast cancer and accounts for
10%15% of invasive breast tumors. On a molecular level,
lobular breast cancers are a distinct disease type and should
be considered as a unique disease. These molecular underpinnings were recently intensely studied as part of The
Cancer Genome Atlas (TCGA) effort on breast cancer,
including a recent publication focused on lobular breast
cancers.1 In this publication, 817 breast tumors from the
TCGA project, including 490 invasive ductal cancers (IDCs),
127 ILCs, and 88 samples with a mixed IDC-ILC histology,
were molecularly profiled on six genomic platforms to develop a comprehensive portrait of the genetic, epigenetic,
transcriptional, and proteomic landscape of lobular breast
cancers. Comprehensive multiplatform analyses, both supervised and unsupervised, of ILC tumors and across histologic subtypes were performed to identify genomic
drivers of ILC oncogenesis.
As expected, low expression of E-cadherin protein, as
determined by reverse-phase protein array (RPPA), and
decreased E-cadherin mRNA levels were uniformly observed

in ILC cases. In total, 63% of ILC cases had an E-cadherin
mutation, and 95% of ILC cases had loss when mutation, DNA
copy number, and low gene and/or protein expression were
summed together. It is reassuring that this modern study
reaffirmed the primacy of E-cadherin loss in lobular breast
cancer, and if a biomarker were to be chosen to identify ILC
on a molecular level, it would likely be low E-cadherin
protein expression and/or DNA mutation of CDH1. Lobular carcinomas unique histopathologic features and its
metastatic patterns undoubtedly have their origins in their
impoverished E-cadherin status.
Beyond previously reported CDH1 and PIK3CA mutations
(which occur at a 48% frequency in ILC), the TCGA study
identified a number of novel ILC-enriched recurrent mutations targeting FOXA1, PTEN, RUNX1, and TBX3. FOXA1
function is particularly intriguing, as it works with the estrogen receptor (ER) to drive the transcriptional output of
ER. Interestingly, FOXA1 also plays a similar role in prostate
cancer, but in these cancers its cofactor is the androgen
receptor.2 In ILC, we find an increased incidence (9% in ILC vs.
2% in IDC) of FOXA1 mutations, whereas in IDC, we find that
GATA3 mutations are considerably enriched in IDC luminal
tumors (19% IDC vs. 5% ILC). Within ILC tumors, FOXA1
mutations were found to cluster into a specific region of the
forkhead (FK) domain. A broader analysis of FOXA1 mutations in breast and prostate cancer, in which it is also recurrently mutated, confirms two specific hotspots in the FK
domain and the C-terminal transactivation domain. Interestingly, these mutational classes were associated with
higher FOXA1 messenger RNA and protein expression, and
with unique transcriptional changes suggesting different
functional effects.3 More work into why FOXA1 is mutated in
ILC, but GATA3 is mutated in IDC, is needed and could
possibly reveal some important underlying biology.
Another important finding concerning ILC was the increased incidence of phosphatase and tensin homologue
(PTEN)-inactivating events. When including both mutations
From the Division of Oncology, Stanford University School of Medicine, Stanford, CA; Yale University, New Haven, CT; The University of North Carolina at Chapel Hill, Chapel
Hill, NC.
Corresponding author: George W. Sledge, MD, Stanford University School of Medicine, 269 Campus Dr., CCSR-1115, Stanford, CA 94305; email: gsledge@stanford.edu.
18
LOBULAR CARCINOMA OF THE BREAST
and DNA copy number changes, ILC-luminal A showed a 13%

PTEN altered phenotype compared with 3% in IDC-luminal A
tumors. This increased mutation frequency of PTEN loss in
ILC correspond with decreased PTEN protein expression and
was largely mutually exclusive with PIK3CA mutations. Analyses of RPPA protein and phosphor-protein expression data
demonstrated increased phosphoinositide-3 kinase/Akt signaling as evident by increased levels of phosphorylated Akt (pS473
and pT308) and downstream Akt substrates including p-p27 and
p-p70S6 kinase in ILC tumors; these findings may represent a
potential therapeutic opportunity for patients with ILC.
In terms of gene-expression patterns, ILC was found to be
of the luminal A subtype in 83% of the cases. Within this ILCluminal A subset, additional expression subtypes were also
identified. These included a subset enriched for stromal and
extracellular matrix features, often referred to as the reactive subtype.4 Another ILC expression subtype was enriched for immune cell features/infiltrates, whereas the
third group showed a more proliferative expression signature and a concomitant worse patient outcome. Finally, a
multiplatform analysis of the mixed histology tumors
showed that approximately 80% of these samples could
clearly be molecularly classified as either ILC or IDC, with
only a few showing a possibly hybrid phenotype. This could
have important implications for treatment of patients with
mixed ILC-IDC histology, as it suggests they are not a unique
group, but instead that their molecular features could be
used to classify them as either ILC or IDC.
SURGICAL ASPECTS OF LOBULAR CARCINOMA

ILC is a distinct subtype of breast cancer that is deserving of
particular attention by surgeons. ILCs tend to be insidious
as a result of their lack of E-cadherin, causing noncohesive
neoplastic cells that permeate through tissue in a single-file
pattern. Given its biology, a few areas are of particular
concern for surgeons.
Preoperative Imaging
It is clear that the extent of ILCs tend to be underestimated
by conventional imaging (Table 1). Some have suggested
that MRI may be useful in this context5; however, a recent
meta-analysis found that MRI did not significantly reduce
positive margin rates in patients with ILC undergoing breast
conservation.6 The concept that MRI could also find occult
contralateral disease has also been raised; however, ILCs are
no more likely to have a synchronous contralateral cancer
than are IDCs.7 MRI is therefore not routinely recommended
in the presurgical workup of patients with ILCs.8
Breast-Conserving Surgery Versus Mastectomy

Regardless of whether neoadjuvant chemotherapy is used or
not, ILCs are more often associated with positive margins
after breast-conserving surgery.911 Patients with this histologic subtype are more likely to require re-excision, and
potentially mastectomy, for margin clearance.12 Some have
suggested that invasive lobular cancers treated with breast
TABLE 1. Imaging Concordance With Pathologic

Tumor Size (Correlation Coefficient)
Study
MRI Mammography Ultrasound Clinical Exam
Boetes et al25
0.81 0.34
0.24
26
Francis et al
0.87 0.79
0.56
Kepple et al27
0.88
0.71
Kneeshaw et al28 0.86

Munot et al
29
0.97 0.66
0.89
0.47
0.67
conservation are more likely to result in local recurrences13;

however, this has not been borne out in other studies,9,10,14
and long-term survival rates are no different between breast
conservation and mastectomy.15 Hence, both are considered appropriate in terms of surgical management for this
disease. Of note, invasive lobular histology falls into the
American Society for Radiation Oncologys cautionary
subgroup for the use of accelerated partial breast irradiation,16 given concerns regarding higher ipsilateral breast
tumor recurrences in this population.
Lymph Node Evaluation

Besides tumor extirpation from the breast, the other key task
of the breast surgeon is lymph node evaluation. Although it is
clear that sentinel node biopsy is feasible and accurate in
patients with ILC, the ability to find micrometastatic deposits
particularly on intraoperative evaluation with hematoxylin and
eosin staining alone may be challenging given the discohesive
nature of the neoplastic cells. Some pathologists have not
found this to be problematic.17 Others, however, may recommend deferral of final diagnosis to permanent sections, at
which time immunohistochemistry can be used to draw attention to deposits that otherwise could be missed.18
SYSTEMIC THERAPY FOR LOBULAR CARCINOMA:

CHEMOTHERAPY
Response to Neoadjuvant Therapy
Increasingly, neoadjuvant chemotherapy is part of the multidisciplinary approach to breast cancer, as more novel therapeutics are being evaluated in this setting. However,
physicians and patients should be aware that response rates to
neoadjuvant chemotherapy are lower for ILC than for its ductal
counterpart, as illustrated by the lower rates of pathologic
complete response and breast conservation (Table 2).
Why is neoadjuvant chemotherapy relatively ineffective for
lobular carcinomas? Is it something intrinsic to the biology of
lobular carcinomas per se? Mathieu et al19 have argued that
this relative futility is predictable, in that histologic and biologic factors predicting a poor response to chemotherapy
(low histologic grade, high ER content and bcl-2 expression,
and low proliferative rates as measured by Ki67 and negative
p53 staining) are all more frequent in lobular rather than
ductal carcinomas. These, in turn, reflect the luminal A
nature of most lobular carcinomas, discussed above.
There is a much smaller body of data evaluating neoadjuvant hormonal therapy for lobular carcinoma. Dixon et al20
19
SLEDGE, CHAGPAR, AND PEROU
TABLE 2. Response to Neoadjuvant Chemotherapy

Study
No. Patients
pCR (%)
p Value
BCS (%)
p Value
Truin et al30
IDC: 3,622
20.2
, .0001
39.4
, .0001
ILC: 466
Loibl et al31
Non-ILC: 7,969
ILC: 1,051
Lips et al32
Wenzel et al33
Tubiana-Hulin et al34
Cocquyt et al35
4.9
17.4
6.2
IDC: 601
25
ILC: 75
11
IDC: 124
20
ILC: 37
IDC: 742
ILC: 118
IDC: 101
15
ILC: 26
24.4
, .001
71.1
, .0001
59.1
.01
46
.037
33
.009
79
.002
48
.001
51
.0004
30
.0066
50
NS
38
Abbreviations: pCR, pathologic complete response; BCS, breast-conserving surgery; IDC, invasive ductal carcinoma; ILC, invasive lobular carcinoma; NS, not significant.
evaluated the responsiveness of lobular carcinomas to neoadjuvant letrozole in 61 patients treated for 3 months. Mean
tumor volume reduction was 66%, with a high rate of breast
conservation (81%). Although we lack any head-to-head comparisons of neoadjuvant endocrine therapy with neoadjuvant
chemotherapy (and may never see such a comparison), a primary endocrine approach does not seem unreasonable.
RESPONSE TO HORMONAL THERAPY

If lobular carcinoma is relatively resistant to standard chemotherapeutic agents, and if this is largely as a result of predictable
biology (i.e., higher ER and lower proliferative rates), then what
about adjuvant hormonal therapy? Recent data suggest that
not all hormonal therapies are equal where lobular cancer is
concerned. In particular, tamoxifen appears to be considerably
less effective than aromatase inhibition for lobular cancers.
Metzger Filho et al21 compared the relative efficacies of
tamoxifen and letrozole for lobular and ductal carcinomas in
the BIG 1-98 trial. This trial was among the first randomized
controlled trials of aromatase inhibitor therapy in the adjuvant setting and now has relatively long follow-up (median
8.1 years). Comparing patients by histologic subtype, patients
with ILC were far more likely to benefit from letrozole than
tamoxifen, regardless of whether patients were luminal A
or luminal B like. The 8-year disease-free survival estimate
was 66% for tamoxifen compared with 82% for letrozole in
the ILC subset (hazard ratio [HR] 0.48) and was 75% for
tamoxifen and 82% for letrozole in the IDC subset (HR 0.80).
The test for interaction was significantly positive (p = .006).
These seem, on the face of it, to represent a clinically significant difference and are paralleled by overall survival
differences (74% for tamoxifen compared with 89% for
letrozole in the ILC subset [HR 0.40]; 84% for tamoxifen and
88% for letrozole in the IDC subset [HR 0.73]).
In contrast, van de Water et al22 have examined the adjuvant TEAM (Tamoxifen and Exemestane Adjuvant Multinational) trial. This analysis differed considerably in that the
20
TEAM trial design in that early switch (2 to 3 years of tamoxifen followed by 2 to 3 years with an aromatase inhibitor)
and up-front (5 years with an aromatase inhibitor) strategies
were compared for relative benefit in lobular and invasive
ductal cancers. The TEAM analysis suggested that endocrine
therapy efficacy was similar for IDC and ILC once one had
adjusted for ER content. The early switch strategy arm might
well muzzle the treatment interactions seen in the BIG 1-98 trial.
If, as suggested by the BIG 1-98 analysis, lobular carcinomas are relatively less sensitive to tamoxifen than ductal
carcinomas, what molecular changes might underlie these
findings? Sikora et al23 have recently evaluated the response
of lobular carcinoma cell lines in vitro. Their work suggests
that the ER drives a unique program of gene expression in
lobular cancers when compared with ductal carcinomas.
Indeed, tamoxifen appears to drive the growth of these cell
lines, rather than inhibiting them, although this limited cell
line work cannot be safely extrapolated to the clinic.
In contrast to the preclinical results seen with tamoxifen,
Arthur et al24 have recently demonstrated in the neoadjuvant hormonal therapy setting that changes in gene
expression in response to letrozole were highly similar between responding ILC and IDC tumors.
CONCLUSION
Infiltrating lobular carcinoma of the breast represents a biologically distinct subset of breast cancer, a biology defined
by specific genetic aberrations in E-cadherin, the high
prevalence of ER-positive disease, the relatively low frequency of HER2-positive disease, and specific mutational
events revealed by deep sequencing of ILC genomes. This
distinctive biology, in turn, affects the presentation, treatment, andpotentiallythe prognosis of ILC. Biology affects surgery and preoperative chemotherapy results, and,
as recent data suggest, it also affects adjuvant hormonal
therapy benefits. Ultimately, an improved understanding of
ILC biology should also lead to novel targeted approaches to
the conquest of the disease.
LOBULAR CARCINOMA OF THE BREAST
References
1. Ciriello G, Gatza ML, Beck AH, et al; TCGA Research Network. Comprehensive
molecular portraits of invasive lobular breast cancer. Cell. 2015;163:506-519.
2. Jozwik KM, Carroll JS. Pioneer factors in hormone-dependent cancers.
Nat Rev Cancer. 2012;12:381-385.
3. Cancer Genome Atlas Research Network. The molecular taxonomy of
primary prostate cancer. Cell. 2015;163:1011-1025.
4. Cancer Genome Atlas Network. Comprehensive molecular portraits of
human breast tumours. Nature. 2012;490:61-70.
5. Mann RM, Hoogeveen YL, Blickman JG, et al. MRI compared to conventional diagnostic work-up in the detection and evaluation of invasive
lobular carcinoma of the breast: a review of existing literature. Breast
Cancer Res Treat. 2008;107:1-14.
6. Houssami N, Turner R, Morrow M. Preoperative magnetic resonance
imaging in breast cancer: meta-analysis of surgical outcomes. Ann Surg.
2013;257:249-255.
7. Langlands F, White J, Kearins O, et al. Contralateral breast cancer:
incidence according to ductal or lobular phenotype of the primary. Clin
Radiol. 2016;71:159-163.
8. Pilewskie M, King TA. Magnetic resonance imaging in patients with
newly diagnosed breast cancer: a review of the literature. Cancer. 2014;
120:2080-2089.
9. Fortunato L, Mascaro A, Poccia I, et al. Lobular breast cancer: same
survival and local control compared with ductal cancer, but should both
be treated the same way? analysis of an institutional database over a
10-year period. Ann Surg Oncol. 2012;19:1107-1114.
10. Molland JG, Donnellan M, Janu NC, et al. Infiltrating lobular carcinoma
a comparison of diagnosis, management and outcome with infiltrating
duct carcinoma. Breast. 2004;13:389-396.
11. Moore MM, Borossa G, Imbrie JZ, et al. Association of infiltrating lobular
carcinoma with positive surgical margins after breast-conservation
therapy. Ann Surg. 2000;231:877-882.
12. Kryh CG, Pietersen CA, Rahr HB, et al. Re-resection rates and risk
characteristics following breast conserving surgery for breast cancer
and carcinoma in situ: a single-centre study of 1575 consecutive cases.
Breast. 2014;23:784-789.
13. Hussien M, Lioe TF, Finnegan J, et al. Surgical treatment for invasive
lobular carcinoma of the breast. Breast. 2003;12:23-35.
14. Ott OJ, Hildebrandt G, Potter R, et al. Accelerated partial breast
irradiation with interstitial implants: risk factors associated with
increased local recurrence. Int J Radiat Oncol Biol Phys. 2011;80:
1458-1463.
15. Fodor J, Major T, Toth
J, et al. Comparison of mastectomy with breastconserving surgery in invasive lobular carcinoma: 15-year results. Rep
Pract Oncol Radiother. 2011;16:227-231.
16. Smith BD, Arthur DW, Buchholz TA, et al. Accelerated partial breast
irradiation consensus statement from the American Society for Radiation Oncology (ASTRO). Int J Radiat Oncol Biol Phys. 2009;74:987-1001.
17. Horvath JW, Barnett GE, Jimenez RE, et al. Comparison of intraoperative
frozen section analysis for sentinel lymph node biopsy during breast
cancer surgery for invasive lobular carcinoma and invasive ductal
carcinoma. World J Surg Oncol. 2009;7:34.
18. Cserni G, Bianchi S, Vezzosi V, et al. The value of cytokeratin immunohistochemistry in the evaluation of axillary sentinel lymph nodes in
patients with lobular breast carcinoma. J Clin Pathol. 2006;59:518-522.
19. Mathieu MC, Rouzier R, Llombart-Cussac A, et al. The poor responsiveness of infiltrating lobular breast carcinomas to neoadjuvant
chemotherapy can be explained by their biological profile. Eur J Cancer.
2004;40:342-351.
20. Dixon JM, Renshaw L, Dixon J, et al. Invasive lobular carcinoma: response to neoadjuvant letrozole therapy. Breast Cancer Res Treat. 2011;
130:871-877.
21. Metzger Filho O, Giobbie-Hurder A, Mallon E, et al. Relative effectiveness of letrozole compared with tamoxifen for patients with lobular
carcinoma in the BIG 1-98 trial. J Clin Oncol. 2015;33:2772-2779.
22. van de Water W, Fontein DB, van Nes JG, et al. Influence of semiquantitative oestrogen receptor expression on adjuvant endocrine
therapy efficacy in ductal and lobular breast cancer - a TEAM study
analysis. Eur J Cancer. 2013;49:297-304.
23. Sikora MJ, Cooper KL, Bahreini A, et al. Invasive lobular carcinoma cell
lines are characterized by unique estrogen-mediated gene expression
patterns and altered tamoxifen response. Cancer Res. 2014;74:
1463-1474.
24. Arthur LM, Turnbull AK, Webber VL, et al. Molecular changes in lobular
breast cancers in response to endocrine therapy. Cancer Res. 2014;74:
5371-5376.
25. Boetes C, Veltman J, van Die L, et al. The role of MRI in invasive lobular
carcinoma. Breast Cancer Res Treat. 2004;86:31-37.
26. Francis A, England DW, Rowlands DC, et al. The diagnosis of invasive
lobular breast carcinoma. Does MRI have a role? Breast. 2001;10:38-40.
27. Kepple J, Layeeque R, Klimberg VS, et al. Correlation of magnetic
resonance imaging and pathologic size of infiltrating lobular carcinoma
of the breast. Am J Surg. 2005;190:623-627.
28. Kneeshaw PJ, Turnbull LW, Smith A, et al. Dynamic contrast enhanced
magnetic resonance imaging aids the surgical management of invasive
lobular breast cancer. Eur J Surg Oncol. 2003;29:32-37.
29. Munot K, Dall B, Achuthan R, et al. Role of magnetic resonance imaging
in the diagnosis and single-stage surgical resection of invasive lobular
carcinoma of the breast. Br J Surg. 2002;89:1296-1301.
30. Truin W, Vugts G, Roumen RM, et al. Differences in response and
surgical management with neoadjuvant chemotherapy in invasive
lobular versus ductal breast cancer. Ann Surg Oncol. 2016;23:51-57.
31. Loibl S, Volz C, Mau C, et al. Response and prognosis after neoadjuvant
chemotherapy in 1,051 patients with infiltrating lobular breast carcinoma. Breast Cancer Res Treat. 2014;144:153-162.
32. Lips EH, Mukhtar RA, Yau C, et al; I-SPY TRIAL Investigators. Lobular
histology and response to neoadjuvant chemotherapy in invasive breast
cancer. Breast Cancer Res Treat. 2012;136:35-43.
33. Wenzel C, Bartsch R, Hussian D, et al. Invasive ductal carcinoma and
invasive lobular carcinoma of breast differ in response following
neoadjuvant therapy with epidoxorubicin and docetaxel + G-CSF. Breast
34. Tubiana-Hulin M, Stevens D, Lasry S, et al. Response to neoadjuvant
chemotherapy in lobular and ductal breast carcinomas: a retrospective
study on 860 patients from one institution. Ann Oncol. 2006;17:
1228-1233.
35. Cocquyt VF, Blondeel PN, Depypere HT, et al. Different responses to
preoperative chemotherapy for invasive lobular and invasive ductal
breast carcinoma. Eur J Surg Oncol. 2003;29:361-367.
21
POINTS OF VIEW
The section, new for 2016, contains articles describing emerging, highly debated, or controversial topics
in cancer research, treatment, and care to benefit patients and the field of oncology.
AUTHORS
Arthur L. Caplan, PhD
NYU Langone Medical Center
New York, NY
Laura Esserman, MD, MBA
San Francisco, CA
David J. Kerr, MD, PhD
University of Oxford
Oxford, United Kingdom
Karmanos Cancer Institute
Detroit, MI
CAPLAN, BATEMAN-HOUSE, AND WALDSTREICHER
Compassionate Use: A Modest Proposal

Arthur L. Caplan, PhD, Alison Bateman-House, PhD, MPH, MA, and Joanne Waldstreicher, MD
Editors Note: The following article is based on the 2016 ASCO Annual Meeting Education Session Expanded Access and the
Right to Try: Navigating the Intersection of Drug Development and Patient Access to Investigational Agents. The authors
present the approach to compassionate use through use of an independent committee to vet and review requests for one
pharmaceutical company.
atient requests for rapid access to agents that are still in

the research pipeline fall into two categories: requests
from groups of persons with the same malady and requests
by individuals. The former are often described as requests
for expanded access, the latter as requests for compassionate use. Regulatory bodies in many countries have
created programs for providing greater access to requests
from groups including the creation of expanded access
programs and emergency use waivers for patients who do
not qualify for clinical trials. Compassionate use requests for
individuals have proven to be more difficult to resolve.
Compassionate use requests can come at any time during
the research processfrom when a drug is being tested in
animals, to its first-in-human studies, to dose-finding trials,
to when the agent under investigation nears the end of
clinical trials. Requests can come from patients nearing the
end of life, but they may also come from those facing serious
disability and pain for which no approved agent has been
proven effective. Requests can come from around the world
and from parents for their children, patients on their
deathbeds, those in the midst of lethal disease outbreaks,
those newly infected as well as the chronically ill, the very
poor, and those for whom money is no object.
Until very recently, the main strategy for patients seeking
compassionate use was to try to locate a possible treatment,
frequently online, and often, though not always, with the
help (or approval) of their physician. Once a possible treatment was found, patients and/or their physicians seeking
compassionate use would try to contact the researcher or
leader of a group involved with testing the agentalmost
always at a company. Some are able to use personal connections to initiate a request, whereas others reach out
to their elected officials, shareholders in the company, or
others thought to wield influence. Sometimes patients try to
interest the traditional media in their plight. Alternatively,
they may launch campaigns using social media with the hope
that public pressure might be brought to bear on the parties
who own the drug or agent. Regardless of the approach, this

process is, at best, hit or miss and one that favors patients
with a compelling personal interest story or who are otherwise well-connected.
Many have claimed that the key obstacle for patients
seeking compassionate use access is the U.S. Food and Drug
Administration (FDA). However, estimates are that the FDA
approves over 99% of the compassionate use protocols it
receives.1 Furthermore, the FDA plays no role in responding
to requests for compassionate use until the company developing the agent has indicated its willingness to provide
the product. In contrast, companies encounter multiple
issues when faced with compassionate use requests. These
include the fact that they have no legal duty to offer access
and that they must balance requests with ongoing clinical
development programs. In addition, even if the company
wanted to respond, there may be uncertainty as to what the
response should be and how best to respond to all requests
fairly, particularly when there are those who are wellconnected and/or using social media campaigns.
In this article, we review the approach taken by one of
many companies (Janssen) that receives such requests,
which involved the formation of a third party. Internal
discussions at Janssen had focused on how to best handle
compassionate use requests consistent with its commitment to obtain regulatory approval for investigational
agents, thereby making products available to the greatest
number of persons. Ultimately, Janssen decided to undertake a first-of-its-kind partnership that enlisted a third
party to review requests made to the company for compassionate use. This novel approach was initiated as a pilot
program focused on a single investigational medicine,
daratumumab, which was being evaluated in late-stage
phase III trials at the time and had shown evidence of
efficacy in patients with refractory multiple myeloma.
Janssen approached the director of the Division of Medical
Ethics at NYU Langone Medical Center, Arthur Caplan, PhD,
From the New York University School of Medicine, New York, NY; NYU Langone Medical Center, New York, NY; Johnson & Johnson, New Brunswick, NJ.
Corresponding author: Arthur L. Caplan, PhD, New York University School of Medicine, 227 East 30th St., New York, NY 10016; email: arthur.caplan@nyumc.org.
e2
COMPASSIONATE USE REQUESTS
to propose a method for reviewing compassionate use requests that would be transparent, fair, beneficent, evidencebased, and patient-focused. Drawing upon his prior work
involving the allocation of cadaver organs for transplantation, Caplan formed a 10-person committee, the Compassionate Use Advisory Committee (CompAC), that consists
of physicians, bioethicists, patients, and patient advocates
from around the world to advise Janssen on compassionate
use requests for daratumumab. The CompAC was created
and staffed by the New York University School of Medicine
(NYUSOM) and was independent of Janssen. CompAC
members contracted with NYUSOM to remain independent and reduce the risk of conflicts of interest. Some
CompAC members accepted payment for their time;
the chair and deputy chair, however, received no direct
compensation.
HOW THE COMPAC WORKS

Compassionate use requests for daratumumab are received
directly at Janssen through a standardized patient intake
form on the companys website. Janssen physicians screen
the intake forms for medical appropriateness, and, if the
patient is eligible for a clinical trial, expanded access program, or has not yet tried an available drug or program,
Janssen informs the submitting physician of this and
does not route the request to CompAC. For all other
cases, CompAC receives requests on a weekly basis and
returns its recommendation to Janssen within 5 business
days of receiving each request. CompAC is an advisory
committee, and, legally, Janssen must remain in control
of the ultimate decision to provide access; however,
Janssen made a good faith commitment to follow CompAC
recommendations.
Although CompAC spent a good deal of time discussing
what, if any, criteria it would use in deciding who to recommend to receive access, it quickly became evident that
the key ethical requirement that the CompAC had to achieve
was fairness. The existing pathways for making requests of
Janssen and other companies were not always fair, sometimes favoring the rich, celebrities, and those who could
command attention. To work, CompAC had to ensure that it
would strengthen the fairness of compassionate use decisions to better meet the needs of patients, doctors, researchers, and the public. To do so, the CompAC:
KEY POINTS
This article describes the creation of a dedicated

committee to respond to compassionate use requests
for an investigational drug for multiple myeloma.
This article describes the importance of creating
principles of fair opportunity for those making requests
for compassionate use.
This article suggests that this model may have
application to other agents and settings.
1. worked to improve navigation of Janssens worldwide

websites so that those seeking compassionate use for
multiple myeloma would be able to more easily find
clear information on how to request daratumumab,
2. insisted that requests came from physicians and be
standardized so that all requests would have the same
information, unadorned by letters of support from
prominent individuals,
3. decided to rotate voting on requests among three of
its 10 members weekly to minimize any effort to lobby
the committee members,
4. directed Janssen to redact information pertaining to
the requesters name, location, nationality, and gender to prevent these factors from being taken into
consideration by voting members,
5. created an appeal process by which a physician whose
request was denied could reintroduce further standardized information about his or her patient (i.e., new
laboratory values), and
6. committed to respond to all requests within 5 business
days, thereby ensuring that appropriate concern was
shown to all patients and that no one would be left
without an expeditious answerthe opposite of which
is commonly a source of great patient and physician
frustration about compassionate use requests.
Soon after the creation of the CompAC in May 2015, these
policies were implemented. Further discussion among
CompAC members led to an agreement on a set of principles
that would be used to guide the committees recommendations in response to requests. These included: preventing
harm to patients, requiring that patients had exhausted all
available approved treatments before trying unapproved
agents, estimating the likelihood of obtaining an efficacious
response, and assessing patient functionality, and, at a lower
order of priority, prior participation in a clinical trial, direct
support for dependents, suffering from the disease, and age.
Decision making in response to requests commenced in July
2015, with an average of four to five requests per week.
Supply varied from week to week including some weeks in
which there was none.
LESSONS TO DATE
Although there is obvious interest both in who was selected, who was rejected, and on what basis, the major
lesson learned from this pilot program is that it is fairness,
rather than justice, that is the key to the success of the
CompAC. By having a committee with broad expertise that
is insulated from what ought to be morally irrelevant
considerations of wealth, privilege, and media interest,
the CompAC has been able to institute strategies that
minimize the ability of any patient, family, advocacy
group, or advantaged party to sway decisions. Patients,
their doctors, and Janssen officials report that they believe that the pilot approach is more equitable than what
existed beforehand.
e3
CAPLAN, BATEMAN-HOUSE, AND WALDSTREICHER
The CompAC model may be applied to other novel agents

across Johnson & Johnson. Other companies, both large and
small, patient advocacy organizations, and government officials have all expressed interest in learning about or
extending the CompAC model to other settings.
The CompAC pilot program has been resource intensive,
involving many different groups to support its activities, and is
References
1. Silverman E. The FDA Says Its More Compassionate Than You Think.
Healthcare Blog, The Wall Street Journal. 2014. http://blogs.wsj.com/
corporate-intelligence/2014/05/05/the-fda-says-its-more-compassionate-thanyou-think/. Accessed February 24, 2016.
e4
certainly not beyond criticism as to its composition, principles, and decisions. However, the CompAC has identified
ways to make very hard decisions more equitable. Those
deserve close attention and debate in a world with many
more unapproved agents that will be sought by the desperate who face constraints upon the resources available
to them.
THE EVOLVING SURGICAL APPROACH TO BREAST CANCER
Less Is More: The Evolving Surgical Approach to Breast Cancer

Laura Esserman, MD, MBA, Etienne Gallant, and Michael Alvarado, MD
Editors Note: The following article is based on the 2016 ASCO Annual Meeting Education Session Less Is More: A Multidisciplinary Conversation on Treatment Options. The authors review the indications for reducing the treatment burden for
women with breast cancer by capitalizing on the emerging opportunity to identify and recognize more indolent forms of the
disease using advanced tools, therapies, and screening methods.
OVERVIEW
Personalized medicine is emerging as an important guiding principle in diagnosis and treatment. This means not just doing
more for some, but safely doing less for others. The lessons learned about the biology of breast cancer over the last 2 decades
have enabled us to understand the incredible heterogeneity of breast cancer and its associated behavior. Although much
work remains, there is an emerging opportunity to identify and recognize more indolent forms of breast cancer, made more
prevalent through the widespread adoption of screening. With our improving systemic therapies and improved molecular
tools, we now have the opportunity to reduce the burden of treatment in women with lower-risk tumors. Our surgical
treatments have evolved, with less morbid and more cosmetic procedures. In this article, we review the indications for
further reducing local therapy, including adjuvant radiation.
he history of breast surgery is one in which we have gone

from more to less. Radical mastectomies, when first
introduced by Halsted, were an improvement over leaving
women with fungating masses. However, the field evolved
through a series of randomized trials to minimize the
morbidity of our surgical procedures. Modified radical
mastectomies have replaced radical mastectomies. Immediate reconstruction and skin-sparing mastectomy, and now
total skinsparing mastectomies, have been shown to be
oncologically safe and cosmetically better.1,2 We have reduced extended axillary procedures, and sentinel node
dissections have fortunately successfully reduced the
complications of lymphedema. Even in the setting of stage II
and III disease, with axillary disease, surgeons are finding
safe ways to reduce the burden of axillary surgery in the
setting of excellent responses to neoadjuvant chemotherapy. Especially because we are increasingly more successful
in preventing recurrence and death from breast cancer and
as women lead longer lives, we must pay more attention to
identifying the opportunities to reduce the burden of
treatment. We now have the opportunity to focus on the
advances in molecular biology to help us to identify tumors
with indolent behavior and to adjust our local therapy
accordingly.3,4
Decision making for patients with low-risk disease is often
more complicated than for those with high-risk disease. The
reason is that the benefit from an intervention and its side

effects may not outweigh the risk of the disease. Thus, a
postmenopausal woman with favorable-risk hormone
receptorpositive breast cancer has several options after
breast conservation. In addition to what has been the gold
standard, standard dose external beam radiation therapy
(EBRT), there are several alternatives including hypofractionated EBRT, intraoperative radiation therapy (IORT),
or no radiation, which are all supported by the literature as
at least equivalent, if not superior, strategies to EBRT.5,6 And
yet, trial results have not significantly influenced practice in
the United States.
Over time, the local recurrence rates have been decreasing
from the original reported risk of 10% EBRT. Recent studies
show rates largely less than 5% and even lower when endocrine therapy is used. Data from four trials with populations of similar biology (postmenopausal women with
node-negative, hormone receptorpositive, early-stage breast
cancer) are shown in Table 1. Local recurrence rates are very
low in the EBRT arms or in the complete absence of radiation.
Fyles et al found that women over age 50 in the low-risk
groupimmunohistochemistry subtype luminal Ahad a
local recurrence of 4.9% at 10 years with or without EBRT.7
Intraoperative radiation therapy results, which also show a
very low locoregional recurrence, are consistent with other
modern trial results. It should be noted that there was no
From the University of California, San Francisco, San Francisco, CA.

Corresponding author: Laura Esserman, MD, MBA, University of California, San Francisco, 1600 Divisadero St., 2nd Floor, Box 1710, San Francisco, CA; email: laura.esserman@ucsf.edu.
e5
ESSERMAN, GALLANT, AND ALVARADO
difference in metastatic rates or overall survival for any of

the trial arms.
The question of how to approach patients with earlystage, favorable-risk breast cancer illustrates the complexity
of forces influencing decision making in regard to adoption
of new approaches. Whole-breast EBRT remains the standard of care following breast-conserving surgery. However,
multidose partial breast radiation is increasingly offered as
an alternative for eligible women. Emerging technologies
have provided impetus for shifts in radiation approaches
despite lack of randomized clinical trial data for these devices. The frequency of brachytherapy use increased from
approximately 1% in 2001 to 10% in 2006, despite concerns
about long-term efficacy.8 In fact, over 30,000 women had
been treated with accelerated partial breast irradiation
(APBI) by the time the American Society for Radiation Oncology published its first consensus statement.9
The finding from a randomized trial (CALGB 9343) that
older women with hormone receptorpositive breast cancer
could be effectively treated with tamoxifen without radiation therapy has yet to be adopted into clinical practice.10
The results demonstrated that with or without radiation,
distant recurrence, breast cancer mortality, and mastectomy
rates are the same and very low in the two arms.11 Despite
these results with more than 10 years of follow-up and
corroborating evidence from a similar Canadian trial in all
postmenopausal women,12 as well as the PRIME 2 trial,13
radiation is rarely omitted and postlumpectomy radiation is
considered a quality measure by the American College of
Surgeons for older women. Fear of omitting therapy and
being less aggressive often makes both physicians and patients uncomfortable, even in the face of supporting evidence to the contrary. The cultural bias that more aggressive
treatment of cancer is better, fear of malpractice, and financial rewards all conspire to encourage intervention.
KEY POINTS
e6
Surgery for breast cancer has dramatically changed in

recent years, moving away from extensive surgical
resections for all patients to more individualized, limited
surgery for early stages of disease.
Treatment decisions for patients with early-stage
disease are complex because decision making typically
involves multiple options that likely have little effect on
overall survival.
The widespread use of screening has resulted in
identification of a higher fraction of low-risk cancers.
Women undergoing breast-conserving therapy have
numerous options that include surgery alone or surgery
and limited fractionated radiation and IORT.
Molecular classifiers of indolent tumors can guide a less
aggressive initial strategy and should inform treatment
of both invasive disease and help to reclassify some
types of DCIS.
When the CALGB 9343 results were presented in 2002,

they met a similar reaction to the results of the TARGIT-A
results in 2010. Many demanded 10 years of data prior to
adoption. However, even with 10 years of data demonstrating that mortality is not impacted by radiation in women
over age 70, or even with the publication of the PRIME 2 trial
results,13 the standard of care after breast-conserving surgery has remained EBRT.
Partial breast radiation as an alternative to EBRT has been
on the market for 10 years with variable uptake. There was
a large international randomized trial comparing single
intraoperative radiation using low-energy photons delivered
by the TARGIT device. The results of the TARGIT A trial were
first published in 2010, followed by a second publication in
2014.14 The results showed noninferiority, specifically for
the prepathology stratum (TARGIT given at time of lumpectomy). There was a great deal of criticism and arguments
that it was too early to adopt the findings,15 with bias from
many radiation oncologists not expecting that a treatment
with low-energy photons would work. Indeed, early adoption of technology that turns out to be inferior to the status
quo can be harmful, but late adoption of technology that
turns out to be equivalent or superior to the status quo can
be a missed opportunity that also harms individuals and
society. Using a framework to evaluate the risks and benefits
of early versus late adoption, this technology is one where
early adoption would be preferable to EBRT for all eligible
cases.16
Breast cancer is now recognized as being comprised of
several distinct diseases. One of the reasons that the
locoregional recurrence appears to be so much lower is likely
due to the impact of mammography screening and the
identification of tumors that are biologically more
indolent.17-19 The results from older trials in which there
was a 40% locoregional recurrence in the absence of radiation applied to younger women with tumors with more
aggressive biology.20 Thus, reframing risk and options must
occur as we have the ability to better characterize the biology of newly diagnosed breast cancers21 and offer better
options for women with low-risk disease. This is even more
important where local recurrence is not life threatening.
The risk for distant breast cancer recurrence for women
with hormone receptorpositive disease does extend over a
20-year period.22 This well-known fact leads to an erroneous
conclusion that long-term follow-up for 10 or more years is
needed to assess outcomes. However, the peak hazard rate
for local recurrence is early, and then is low and stable, thus,
the overall conclusions are extremely unlikely to change.
This has been demonstrated by the overview analyses as
well as many trials,23,24 the Canadian trial, the CALGB trial,
and the TARGIT A trial, which showed that early results
predicted the later results.12,25-27
The demonstration that less-aggressive interventions are
equally effective for women with lower-risk tumors is a
critical breakthrough for women and a major advance in our
ability to tailor treatments for women according to the
biology of their tumors. If we can safely accomplish the same
TABLE 1. Recent Studies of Breast Conservation, Radiation, and Local Therapy Outcomes
Study
Accrual Dates
No. of Patients
Study Arms
5-Year LRR
10-Year LRR
12.6-Year LRR
TARGIT-A14
20002012
2,298 prepathology
IORT
2.1%
NA
NA
EBRT
1.1%
4%
7%
10%
Studies Comparing XRT With No XRT

CALGB C934311
19941999
636
Tam
Tam + RT
1%
1%
2%
Luminal A7
19922000
611 (all T1 patients)
Tam
5.5%
13.8%
NA
Tam + RT
0.4%
5.3%
114 (subset of G1/2,

luminal A patients)
Tam
2%
4.9%
1,326
No RT
4.1%
RT
1.3%
3-week EBRT
5-week EBRT
PRIME II13
20032009
Tam + RT
NA
5.5%
NA
NA
2.33%
6.2%
NA
2.17%
6.7%
3-week EBRT
1.9%
5.2%
5-week EBRT
3.3%
3.8%
Studies Comparing Hypofractionated 3-Week EBRT With 5-Week EBRT

Hypofractionated
Radiation5
19931996
START-B6
19992001
1,234
2,215
NA
Abbreviations: LRR, locoregional recurrence; NA, not available; IORT, intraoperative radiation therapy; EBRT, external beam radiation therapy; Tam, tamoxifen; RT, radiation therapy.
goal (preventing cancer recurrence) in a much more efficient, less invasive, and less personally time-consuming
manner for women, the physician community should be
the first to embrace this therapy. The complaint about IORT
seems to be that the data are immature; however, clinical
practitioners can switch to IORT and watch the data mature.
The chance that the results will change has a very small
probability, and practitioners could switch back to EBRT or
switch to another treatment that has been shown to be
more effective than EBRT in the interim. As our analysis
indicates, little will have been lost for any patient who receives IORT, as many are also eligible for hormone therapy
alone.
Daily radiation treatments over many weeks are burdensome to patients, especially those that live far from an
adequate radiation treatment center. It is reasonable to
assume that patients would prefer no radiation, IORT, or
shorter courses of EBRT, even simply on the basis of convenience. Women concerned only about distant recurrence
should be advised about the options of no radiation for
node-negative disease. If a 5% to 10% difference in local
recurrence is important to avoid, IORT is an excellent alternative to no radiation.
The difficulty in adoption of less-aggressive therapy may
best be approached by using biomarkers that characterize
tumors as indolent. One biomarker, specifically adapted for
the express purpose of identifying indolent disease in the
absence of all treatment, could have the potential to provide
reassurance to providers and patients alike that a lessaggressive initial treatment is appropriate.
The widespread prevalence of screening has resulted in a
shift in the kinds of cancers detected. The objective of
screening is to identify cancers at a lower stage, however, a
consequence of screening is the identification of a higher
fraction of tumors with low-risk biology.28 An important way

to mitigate that risk is to recognize that lower-risk biology
tumors are more likely to be detected with screening and to
have tools to identify them at the time of diagnosis and
enable less-aggressive treatment. Using a data set from a
randomized trial of no treatment compared with treatment
with tamoxifen for 2 years, we sought to leverage a validated
molecular classifier to determine if we could define a group
of women whose prognosis was excellent even 20 years after
diagnosis.29 We used a test that was originally designed to
identify a population of patients with breast cancer, in the
absence of systemic therapy, with a good prognosis at
5 years. The 70-gene test (MammaPrint) has been used to
identify women who do not benefit from adjuvant chemotherapy and is currently being evaluated in the MINDACT
clinical trial (6,600 patients; results expected late spring
2016). We created a new threshold with the purpose of
specifically evaluating long-term (. 20 years) breast cancerspecific mortality, which we have designated indolent
threshold.
The women in the STO1 study presented with palpable
tumors, so by definition, they had tumors that had come to
clinical attention. These are not cancers detected by
screening. When the indolent threshold is applied to the STO
study in which women presented with palpable, clinically
detected tumors, we were able to identify women whose
tumors, after surgery and a short course (2 to 5 years) of
tamoxifen, posed no risk for breast cancerspecific death for
15 years.29
An indolent tumor should signal that a more minimal
treatment approach would be sufficient. Although there
may be a small fraction of patients who progress, that
progression is many years later, and early treatments are
unlikely to impact that progression. The critical distinction is
e7
FIGURE 1. Indolent Threshold of the 70-Gene Assay
An indolent threshold of the 70-gene assay has been validated in a randomized

trial of postmenopausal women with node-negative palpable tumors.29 These
tumors are essentially curable at the time of presentation. Thus, there would be no
value to identifying the precursors of such tumors. Such precursors would therefore
fit the definition of IDLE conditions18 and should not be called cancer nor be the target
of screening. The most likely candidates for these precursors are low-grade ductal
carcinoma in situ lesions. Studies are ongoing to investigate the relationship of
low-grade ductal carcinoma in situ and indolent invasive breast cancer.
Abbreviations: IDLE, indolent lesions of epithelial origin; DCIS, ductal carcinoma in
situ; Dx, diagnosis; TAM, tamoxifen.
that additional treatment should be reserved for when and

if that disease progresses, and additional treatment should
be administered at the time of progression (Fig. 1).
We can reflect on the three randomized studies that
demonstrate that women with relatively low-risk biology
have a low risk of recurrence in the setting of hormone
therapy alone and without radiation.11-13 The risk of local
recurrence, as stated earlier, has dropped over the years
likely because of the dramatic increase in the number of
women diagnosed who have indolent cancers.16 In the
CALGB 9343 trial of women treated with lumpectomy alone
and 5 years of tamoxifen, local recurrence was uncommon
and successfully treated at the time of diagnosis without an
impact on mortality or the mastectomy rate. Recurrence
rates in the absence of radiation are low,11-13 and, in particular, the event of recurrence appears to be salvageable
without an adverse impact on mastectomy rates or mortality. In particular, the Canadian trial of radiation compared
with no radiation demonstrates that luminal A tumors have
only a 5% risk of local therapy with 5 years of tamoxifen
therapy and absence of radiation. The indolent threshold
described here is more stringent than the luminal A category, with only 31% of luminal A tumors meeting the indolent designation. Importantly, in all of the radiation versus
no radiation studies, the chance of dying of other causes is
considerably higher than the chance of dying of breast
cancer. The analysis of the Stockholm 1 study confirms the
existence of a tumor type that has an indolent course, which
can be treated less aggressively at the time of diagnosis. In
this population of women, the indolent threshold did not
identify women who had zero risk absent therapy, but almost no risk with a minimal course of tamoxifen.
e8
The indolent threshold could therefore be used to identify

women for whom lumpectomy alone and a short course
of adjuvant endocrine therapy is sufficient. Many women
are unable to tolerate 5 years of endocrine therapy, and as
many as 60% have discontinued aromatase inhibitors by
3 years.30,31 The ability to identify a population of women
with little benefit of extended endocrine risk-reducing
therapy would be of huge benefit. The indolent threshold
does not allow us to say that women would never experience
recurrence over their lifetime (we have not found the
classification for truly IDLE [indolent lesions of epithelial
origin] conditions), but we can say that the chance of any
recurrence is low and that more aggressive therapy can
be administered at the time of recurrence, sparing many
women treatments that will not be of benefit.
It is estimated that over a 10-year period, five to seven
cancers per 1,000 might be diagnosed that might not otherwise come to clinical attention.32 These represent approximately 17% of the patients diagnosed with breast
cancer. Interestingly, approximately 15% of the STO1 trial
population had tumors that met the indolent threshold. In
the MINDACT study, a modern cohort of patients in which
over 80% of women are screened, the fraction of women
with screen-detected tumors with indolent cancers may be
in the range of 37%, or about 20% more than what was
observed in the STO trial, which was conducted prior to the
widespread use of screening. This is consistent with the
additional low-risk cancers diagnosed with screening. Over a
10-year period in the United States, where 50 million women
are screened for 10 years, this might amount to 300,000
cancers with extremely low-risk of recurrence. Characterization of all screen-detected cancers could provide critical
information about how to further personalize treatment.
Even in the setting of presentation of palpable disease,
about 10% of tumors fit the molecular definition of indolent
FIGURE 2. Identification and Therapy of the Indolent

Spectrum of Breast Cancers
Identification of the indolent spectrum of breast cancers and tailoring therapy

accordingly (local and systemic) should be considered part of the evolution
toward personalized breast cancer treatment.
disease. In the setting of screening, that number is more

likely to be in the range of 35% or higher33 in keeping with
observations from screening trials.34 If the biology of these
tumors can be recognized at the time of diagnosis, and
women can be reassured that their prognosis is excellent,
simpler interventions can be used such as excision alone,
sentinel node, and hormone therapy for 2 to 5 years.
Registry studies for using less intervention for postmenopausal women with lower-risk tumors are under
development. The University of Michigan is leading a multiinstitutional study of surgical excision alone for women age
50 to 69 with T1NO tumors with Oncotype Dx results lower
than 18. LUMINA is a Canadian Ontario Cancer Group study
of lumpectomy alone for women older than age 55 with
T1NO tumors that are luminal A. There are several ductal
carcinoma in situ (DCIS) trials starting, two in Europe, including the LORD trial and the LORIS trial, and three in the
United States (a registry trial through the ATHENA network,
the COMET trial recently funded by PCORI, and the ECOGACRIN MRI and Oncotype prospective study) exploring options of using less intervention.
Overdiagnosis is much less of a problem if we accept and
recognize that we will identify indolent disease with
screening and if we have tools to characterize it and be less
aggressive with our interventions. To do so, we have to have
confidence in the tools that we use to identify such indolent
conditions.
These data should also make us reflect on the dilemma of
DCIS. If there are indolent cancers that can be successfully
treated upon diagnosis, then certainly there would not be
any value to finding the precursors of these indolent cancers.
Such precursors would better be reclassified as IDLE, instead

of carcinoma in situ.18 If the indolent invasive tumors can be
treated without radiation, then certainly women with risk
factors or precursors of IDLE conditions such as low and
intermediate DCIS should not be routinely offered radiation.35 Evidence to support this concept was recently
published by Sagara et al.36 The investigators looked at
surgery or no surgery for DCIS in a large SEER database of
57,000 patients. For low-grade DCIS, the 10-year breast
cancerspecific survival for no surgery was 98.8% whereas
the survival for the surgery group was 98.6%. Surely, we can
start advocating for less treatment for these women.
Learning how to safely do less should be considered part of
the evolution of personalized breast cancer treatment
(Fig. 2).
All treatments have side effects. Mastectomy, even with
reconstruction, can be disfiguring and debilitating to
women. They can experience postmastectomy pain. Radiation also has side effects.37,38 Extended endocrine therapy
can also be challenging to take for women, and it has been
reported that over 40% of women stop taking their medicine
after 3 years.30,31 The STO data suggest that 2 years of tamoxifen should be sufficient for women with tumors that
meet the molecular definition of indolent disease. Although
it would be difficult to test, the persistence of a small risk for
recurrence after 15 years over so many years suggests that
perhaps 2 years of hormonal therapy could be repeated 7 to
10 years after diagnosis if we are able to identify those
women with late risk for recurrence. The opportunity to
know when such treatments are not necessary would be a
great boon to patients.
References
1. Wang F, Peled AW, Garwood E, et al. Total skin-sparing mastectomy and
immediate breast reconstruction: an evolution of technique and assessment of outcomes. Ann Surg Oncol. 2014;21:3223-3230.
2. Peled AW, Wang F, Foster RD, et al. Expanding the indications for total
skin-sparing mastectomy: is it safe for patients with locally advanced
disease? Ann Surg Oncol. 2016;23:87-91.
3. Boughey JC, Suman VJ, Mittendorf EA, et al; Alliance for Clinical Trials in
Oncology. Sentinel lymph node surgery after neoadjuvant chemotherapy in patients with node-positive breast cancer: the ACOSOG
Z1071 (Alliance) clinical trial. JAMA. 2013;310:1455-1461.
4. Caudle AS, Yang WT, Krishnamurthy S, et al. Improved axillary
evaluation following neoadjuvant therapy for patients with nodepositive breast cancer using selective evaluation of clipped nodes:
implementation of targeted axillary dissection. J Clin Oncol. Epub
2016 Jan 25.
5. Whelan TJ, Pignol JP, Levine MN, et al. Long-term results of hypofractionated radiation therapy for breast cancer. N Engl J Med. 2010;
362:513-520.
6. Haviland JS, Owen JR, Dewar JA, et al; START Trialists Group. The UK
Standardisation of Breast Radiotherapy (START) trials of radiotherapy
hypofractionation for treatment of early breast cancer: 10-year followup results of two randomised controlled trials. Lancet Oncol. 2013;14:
1086-1094.
7. Fyles A, McCready D, Pintilie M. Luminal A subtype predicts radiation

response in patients with T1N0 breast cancer enrolled in a randomized
trial of tamoxifen with or without breast radiation. Cancer Res. 2011;71:
S2-2.
irradiation consensus statement from the American Society for Radiation Oncology (ASTRO). Int J Radiat Oncol Biol Phys. 2009;74:
987-1001.
irradiation consensus statement from the American Society for Radiation Oncology (ASTRO). J Am Coll Surg. 2009;209:269-277.
10. Soulos PR, Yu JB, Roberts KB, et al. Assessing the impact of a cooperative
group trial on breast cancer care in the medicare population. J Clin
Oncol. 2012;30:1601-1607.
11. Hughes KS, Schnaper LA, Bellon JR, et al. Lumpectomy plus tamoxifen
with or without irradiation in women age 70 years or older with early
breast cancer: long-term follow-up of CALGB 9343. J Clin Oncol. 2013;
31:2382-2387.
12. Fyles AW, McCready DR, Manchul LA, et al. Tamoxifen with or without
breast irradiation in women 50 years of age or older with early breast
cancer. N Engl J Med. 2004;351:963-970.
13. Kunkler IH, Williams LJ, Jack WJ, et al; PRIME II investigators. Breastconserving surgery with or without irradiation in women aged 65 years
e9
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
e10
or older with early breast cancer (PRIME II): a randomised controlled

trial. Lancet Oncol. 2015;16:266-273.
Vaidya J, Wenz F, Bulsara M, et al; TARGIT trialists group. Riskadapted targeted intraoperative radiotherapy versus whole-breast
radiotherapy for breast cancer: 5-year results for local control and
overall survival from the TARGIT-A randomised trial. Lancet. 2014;383:
603-613.
Woloshin S, Schwartz LM. Whats the rush? The dissemination and adoption
of preliminary research results. J Natl Cancer Inst. 2006;98:372-373.
Esserman LJ, Alvarado MD, Howe RJ, et al. Application of a decision
analytic framework for adoption of clinical trial results: are the data
regarding TARGIT-A IORT ready for prime time? Breast Cancer Res Treat.
2014;144:371-378.
Esserman LJ, Shieh Y, Rutgers EJ, et al. Impact of mammographic
screening on the detection of good and poor prognosis breast cancers.
Breast Cancer Res Treat. 2011;130:725-734.
Esserman L, Shieh Y, Thompson I. Rethinking screening for breast cancer
and prostate cancer. JAMA. 2009;302:1685-1692.
Bleyer A, Welch HG. Effect of three decades of screening mammography
on breast-cancer incidence. N Engl J Med. 2012;367:1998-2005.
Fisher B, Montague E, Redmond C, et al. Findings from NSABP Protocol
No. B-04-comparison of radical mastectomy with alternative treatments for primary breast cancer. I. Radiation compliance and its relation
to treatment outcome. Cancer. 1980;46:1-13.
Buyse M, Loi S, vant Veer L, et al; TRANSBIG Consortium. Validation and
clinical utility of a 70-gene prognostic signature for women with nodenegative breast cancer. J Natl Cancer Inst. 2006;98:1183-1192.
Esserman LJ, Moore DH, Tsing PJ, et al. Biologic markers determine both
the risk and the timing of recurrence in breast cancer. Breast Cancer Res
Treat. 2011;129:607-616.
Early Breast Cancer Trialists Collaborative Group (EBCTCG). Effects of
chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials.
Lancet. 2005;365:1687-1717.
Clarke M, Collins R, Darby S, et al; Early Breast Cancer Trialists Collaborative Group (EBCTCG). Effects of radiotherapy and of differences in
the extent of surgery for early breast cancer on local recurrence and
15-year survival: an overview of the randomised trials. Lancet. 2005;
366:2087-2106.
Liu F, Shi W, Done SJ, et al. Identification of a low-risk luminal A breast
cancer cohort that may not benefit from breast radiotherapy. J Clin
Oncol. 2015;33:2035-2040.
26. Vaidya JS, Wenz F, Bulsara M, et al. Targeted intraoperative radiotherapy for early breast cancer: TARGIT-A trial- updated analysis of local
recurrence and first analysis of survival. Cancer Res. 2012;72:S4-2.
27. Hughes KS, Schnaper LA, Berry D, et al; Cancer and Leukemia Group B;
Radiation Therapy Oncology Group; Eastern Cooperative Oncology
Group. Lumpectomy plus tamoxifen with or without irradiation in
women 70 years of age or older with early breast cancer. N Engl J Med.
2004;351:971-977.
28. Esserman LJ, Thompson IM, Reid B, et al. Addressing overdiagnosis and
overtreatment in cancer: a prescription for change. Lancet Oncol. 2014;
15:e234-e242.
29. Esserman LJ, Thompson CK, Yau C, et al. Identification of tumors with an
indolent disease course: MammaPrint ultralow signature validation in a
retrospective analysis of a Swedish randomized tamoxifen trial. Cancer
Res. 2016;76:P6-09-01.
30. Cuzick JF, Sestak I, Howell A, et al. Anastrozole versus tamoxifen for the
prevention of loco-regional and contralateral breast cancer in postmenopausal women with locally excised ductal carcinoma in-situ (IBIS-II
DCIS). Lancet. 2016;387:866-873.
31. Ganz PA, Cecchini RS, Julian TB, et al. Patient-reported outcome (PRO)
results, NRG Oncology/NSABP B-35: a clinical trial of anastrozole (A) vs
tamoxifen (tam) in postmenopausal patients with DCIS undergoing
lumpectomy plus radiotherapy. Cancer Res. 2016;76:S6-04.
32. Marmot MG, Altman DG, Cameron DA, et al. The benefits and harms of
breast cancer screening: an independent review. Br J Cancer. 2013;108:
2205-2240.
33. Drukker CA, Schmidt MK, Rutgers EJ, et al. Mammographic screening
detects low-risk tumor biology breast cancers. Breast Cancer Res Treat.
2014;144:103-111.
34. Miller AB, Wall C, Baines CJ, et al. Twenty five year follow-up for breast
cancer incidence and mortality of the Canadian National Breast
Screening Study: randomised screening trial. BMJ. 2014;348:g366.
35. Narod SA, Iqbal J, Giannakeas V, et al. Breast cancer mortality after a
diagnosis of ductal carcinoma in situ. JAMA Oncol. 2015;1:888-896.
36. Sagara Y, Mallory MA, Wong S, et al. Survival benefit of breast surgery
for low-grade ductal carcinoma in situ: a population-based cohort
study. JAMA Surg. 2015;150:739-745.
37. Darby SC, Ewertz M, McGale P, et al. Risk of ischemic heart disease in
women after radiotherapy for breast cancer. N Engl J Med. 2013;368:
987-998.
38. Esserman L, Yau C. Rethinking the standard for ductal carcinoma in situ
treatment. JAMA Oncol. 2015;1:881-883.
STRATEGIES FOR SUSTAINABLE CANCER CARE
Strategies for Sustainable Cancer Care

David J. Kerr, MD, Anant Jani, PhD, and Sir Muir Gray, CBE, FRCPSGlas, FCLIP
Editors Note: The following article is based on the 2016 ASCO Annual Meeting Education Session NICE or Not? Evaluating
Global Strategies for Sustainable Cancer Care. The authors review strategies for sustainable cancer care and the National
Institute for Health and Care Excellences role in the assessment of clinical and cost-effectiveness of new pharmaceutical and
biopharmaceutical products in the United Kingdom.
OVERVIEW
There is an increasing focus on the relative cost-effectiveness and sustainability of delivering high-quality cancer care,1 with
most emphasis, debatably, given to cost control of innovative treatments. It is difficult to calculate all the direct and indirect
contributors to the total cost of cancer treatment, but it is estimated that cancer drugs constitute 10% to 30% of the total cost
of cancer care. A 2007 study in France2 showed the contribution of drug costs was less than 20%, with approximately 70% of
the total expenditure on cancer accounted for by health care resource use, such as hospitalization. The U.K. government
established the National Institute for Health and Care Excellence (NICE)the dominant function of which is technology
appraisalto assess the clinical and cost-effectiveness of new pharmaceutical and biopharmaceutical products. This is to
ensure that all National Health Service (NHS) patients have equitable access to the most clinically effective and cost-effective
treatments that are viable. NICE has developed a transparent, public process to judge incremental cost-effectiveness using
the quality-adjusted life year (QALY), which allows comparisons of cost-effectiveness across medical specialties. NICE has
been both lauded and criticizedespecially when it passes judgment on marginally effective but expensive anticancer
drugsbut it provides a route to rational rationing and, therefore, may contribute to sustainable cancer care by
highlighting the issue of affordable medicine. This implies a challenge to the wider oncology community as to how we might
cooperate to introduce the concept of value-driven cancer care.
ichael Porter, a leading professor at the Harvard

Business Institute, has developed the following definition of value:
Value in any field must be defined around the customer, not
the supplier. Value must also be measured by outputs, not
inputs. Hence it is patient health results that matter, not the
volume of services delivered. But results are achieved at some
cost. Therefore, the proper objective is patient health
outcomes relative to the total cost (inputs). Efficiency, then,
is subsumed in the concept of value.3(p. 12)
Porter and Teisberg put this another way: Achieving high

value for patients must become the overarching goal of
health care delivery, with value defined as the health outcomes achieved per dollar spent.4 Value, therefore, depends on results, not input, and so value in health care is
measured by the outcomes achieved, not the volumes delivered, and the relative costs of achieving those outcomes.5
Put simply, value is the relationship between the outcome
and the resources used. In turn, the outcome is determined

by the quality and safety of the intervention or service.
STRATEGIES FOR SUSTAINABLE CANCER CARE:

VALUE
Value Versus Quality
Services become high quality for several reasons, many of
which are factors that have been shown to reduce the incidence of errors, notably: (1) good leadership, (2) development of systems, and (3) the creation of a strong culture.
To maximize the probability of benefit and minimize the
probability of harm and errors, professionals need standard
operating procedures (SOPs), commonly defined as a procedure, or set of procedures, to perform a given operation
or evolution in reaction to a specific event. To improve
quality and safety, it is necessary to have a style of general management that will increase effectiveness and reduce harmfor example, promotion of safety culture; use
of systems, including clinical guidelines; clear lines of
From the Nuffield Division of Clinical and Laboratory Science, Nuffield Department of Primary Care Health Sciences, University of Oxford, Oxford, United Kingdom.
Corresponding author: David J. Kerr, MD, Nuffield Division of Clinical and Laboratory Sciences, Level 4, Academic Block, John Radcliffe Infirmary, Headington, Oxford, OX3 9DU
United Kingdom; email: david.kerr@ndcls.ox.ac.uk.
e11
KERR, JANI, AND GRAY
accountability for risk management and safety; and data

collection about risk factors, errors, and near misses.
Certainly there is concern about low-value, low-quality
cancer care, and the Institute of Medicine report, Crossing
the Quality Chasm, on improving the quality of health care
has been welcome, but high-quality health care need not be
synonymous with high-value health care. Low-quality health
care is certainly of lower value, but there are many examples
of health care providing high quality that is of low value. The
semantics of this are important because consumers often
associate value (e.g., in supermarkets) at the lower end of
the market (e.g., Attention Kmart shoppers, as Beetlejuice
once said!).
The assessment of cost-effectiveness by organizations
such as NICE is of vital importance, but cost-effectiveness
refers to an intervention, whereas value refers to a populations services. Perhaps populations will have to make
decisions between different degrees of cost-effectiveness,
worrying not about the financial cost of one particular intervention but about its opportunity costnamely what else
can be done with resources within a finite health budget. If
there is an investment in cancer drugs, there may have to be
disinvestment elsewhere within the cancer program or the
wider health service.
Choosing Wisely
There is increasing recognition of unwarranted variation not
only in investment in cancer care but also in quality, cost, and
outcome. This variation highlights two other problems that
should be addressed not only by policy makers but also by
every clinician providing cancer services.
The first of these is overuse, highlighted in the Choosing
Wisely Campaign in the United States and in the Too Much
Medicine campaign of the British Medical Journal. Overuse
always wastes resources and sometimes does harm (e.g.,
prescribing chemotherapy to patients when palliative care
would be more appropriate, where the concern is not to save
costs, but that the prescription of chemotherapy can be
classified as inappropriate or futile).
KEY POINTS
e12
There is a need to balance cost-effectiveness and

sustainability of delivering high-quality cancer care.
It is estimated that cancer drugs constitute 10% to 30%
of the total cost of cancer care.
The U.K. government established NICE to assess the
clinical effectiveness and cost-effectiveness of new
pharmaceutical and biopharmaceutical products.
NICE has been praised as well as criticized, but it
contributes to sustainable cancer care by highlighting
the issue of affordable medicine.
The worldwide oncology community must cooperate to
introduce the concept of value-driven cancer care for
patients.
The second problem is underuse of effective and costeffective interventions for cancer, principally because of poor
organization in the management of cancer services. One other
aspect of underuse that should be considered is inequity, where
certain subgroups of the populationfor example, distinguished by age, ethnic background, or classreceive lower
levels of high-value treatment than the general population.
Therefore, a focus on value is needed. Three aspects of
value are:
allocative value, determined by how the assets are
distributed to different subgroups in the population;
technical value, determined by how well resources are
used for all the people in need in the population; and
personalized value, determined by how well the decisions relate to the values of each individual.
Triple-Value Health Care

Practitioners providing cancer care should continue to campaign for more resources for cancer, but they should also be
prepared to defend the allocative value of everything they are
currently doing. There are ways in which reallocation can be
facilitated. For example, in Spain, there was a bid for immunotherapy resources for malignant melanoma, the total cost
being about V2 million for the population served. The payers
said they were unable to find this resource, but actually, in that
same population, the annual expenditure on skin disease was
approximately V170 million per year. Practitioners in cancer
care should learn to argue that consideration should be given
to shifting V2 million from lower-value care for other skin
diseases to malignant melanoma.
The main challenge for people providing cancer services is
to consider the allocation of resources between different
cancer groups or between different modalities. The balance
of investment in different patient groups is often a matter of
historical accident, with one particular clinical group having
been more successful than others, but this should be
reviewed to see if the balance is optimal between the different types of common cancer, not excluding rare cancers.
As the path becomes more difficult, providers of cancer care
and patient representatives should consider whether the
balance is right; for example, should the interval between
breast cancer screening be increased to free up resources
that can be switched to chemotherapy for women with
breast cancer?
Technical value is different from efficiency. Technical value
has to take into account not only the use of resources for
the patients seen, but also the possibility of overuse and
underuse. Every cancer service should conduct a review of
inappropriate or futile care, which relates to the third type of
value, personalized value
In making decisions about personalized value, the model
of decision making at the Centre of Evidence-Based Medicine is useful (Fig. 1). Cancer services should ensure that
every patient is fully informed about not only the benefits of
treatment, but also the risks, and help them reflect on how
these odds reflect on values.
FIGURE 1. Center of Evidence-Based Medicines

Personalized Value Model of Decision Making
Managing Clinical Demand

There is a fashion in operations, as there is in sleeves and
skirts: the triumph of some surgeon who has at last found out
how to make a once desperate operation fairly safe is usually
followed by a rage for that operation not only among the
doctors, but actually among their patients.
George Bernard Shaw, preface to The Doctors Dilemma, 1906
Enthusiastic medical professionals, altruistic as we are and

keen to do good, are armed with information provided by
enthusiastic promoters of new technology, expressed in terms
of relative benefit rather than absolute benefit, and are, not
surprisingly, the people who may fuel the growth in demand.
Clinical demand may be overtly expressedfor example by
clinicians leading a campaign for a new drug or a new facility
such as a PET scanneror demand can arise through what
is called creep (i.e., an inexorable increase in testing and
treatment). Innovation should decrease cost as well as increase
it, and there is now a drive to innovate for better value,
harnessing the power of new technology and improved ways
of organizing services. A classic study by Eddy6 in the United
States showed that, in a health care system in which expenditure is not finite, changes in the volume and intensity
of clinical practice are the main factors driving increases
in the cost of care that can be controlled by those who
pay for or manage health care. The other causes of increasing costsaging population, medical and general price
inflationare beyond the power of health service managers
to control (Fig. 2).
Managing innovation usually focuses on a small number of
potentially high-cost interventions; however, the conclusion
from the work of Eddy,6 which is as true today as when it was
published, is that every innovation, no matter how small, should
have its introduction to, or removal from, the health care system
carefully managed, particularly if it is for a common disease.
Managing demand is not easy, but some steps can be
taken; for example:
Be very clear and explicit about the subgroups of cancer
patients most likely to benefit from an intervention and,
as a corollary, those who are least likely to benefit. This
of course plays to the current model of precision medicine, selecting the right drug for the right patient.
Payers also should be explicit about health care conditions that will or will not be given treatment from
publicly/insurance-funded health services.
In addition, services must make explicit decisions and
publish the decisions to not provide interventions they
deem to be of low value, such as expensive but marginally effective cancer therapeutics.
Management of demand may not be possible if left
solely to individual clinicians faced with the distress of
the individual patient. Funders should make decisions
at a population level. This is the implicit tension between the practice of population and clinical medicine
should practicing clinicians always stand outside of
decisions on drug rationing to maintain the integrity of
their relationship with the patients entrusted to their
care?
National Institute for Health and Care Excellence

Since 1999, NICE has provided the United Kingdoms NHS
and those who rely on it for their care, with an increasing
range of advice on effective, good-value health care, and
it has gained a reputation for rigor, independence, and
objectivity (Fig. 3). This has been described by some as
rational rationing. Available guidance takes several forms:
NICE guidelines make evidence-based recommendations that aim to promote integrated care.
Technology appraisal guidance to assess the clinical and
cost-effectiveness of health technologies, such as new
pharmaceutical and biopharmaceutical products, but
FIGURE 2. Causes of Cost Inflation
e13
KERR, JANI, AND GRAY
FIGURE 3. Kerr Led a Review of Scotlands Health

ServiceThe Resemblance Is Evident!
care. These are derived from the best available evidence, particularly NICEs own guidance.
Technology Appraisals
NICE recommendations review clinical and economic evidence, balancing the clinical data in relation to how much it
costs the NHSdoes it represent value for money? This advice
ends the uncertainty and helps to standardize access to health
care across the country. This is reflected in the NHS Constitution, which states that patients have the right to drugs and
treatments that have been recommended by NICE for use in
the NHS, if their doctor believes they are clinically appropriate.
Sir Andrew Dillon, Chief Executive of NICE, said:
also include procedures, devices, and diagnostic agents.

This is to ensure that all NHS patients have equitable
access to the most clinically effective and cost-effective
treatments that are viable.
Quality standards are concise sets of statements with
accompanying metrics designed to drive and measure
priority quality improvements within a particular area of
FIGURE 4. Estimating Cost and Cost-effectiveness
e14
We support the general principle that the NHS should pay a

price which reflects the additional therapeutic benefit of new
drugs. We also share the Governments ambition to ensure
that the option exists for all new licensed drugs to be offered
to those patients who can benefit from them, provided the
price is a fair reflection of their value. We are confident that
the Government will want to take advantage of NICEs
expertise and experience as it develops value-based pricing.
The dominant drivers of NICEs approach are incremental

cost-effectiveness and the QALY, which allows comparisons
of cost-effectiveness across all of medicine. How else would
we compare the relative benefits of cataract or hip surgery
with cancer therapy? The following hypothetical example
gives a clear idea of how the arithmetic is worked out.
As is described in Fig. 4, the model is dependent on inputs.
If drug A had an incremental QALY of only 0.2 with an incremental cost of 6,500, then the incremental costeffectiveness would be 32,500 and possibly outside the
value range.
Operationalizing Strategies for Sustainable Cancer

Care: Affordable Drugs
Almost every time that NICE makes a negative pronouncement on a new cancer drug, there is a volubly
negative press reaction, often driven, understandably, by
disease-oriented interest groups. The NICE leadership has
pushed back on this somewhat and has set a series of
challenges for the wider oncology community,7 namely: (1)
the industry must operate in a much more efficient manner,
(2) the costs of drug development should be slashed, and (3)
oncologists and patient advocacy groups must ask tough
questions of regulators and the pharmaceutical industry.
They also made suggestions for practical implementation
of effective solutions:
They suggest that clinical trial costs could be decreased
by 40%60% without detriment to their quality using
measures such as electronic data capture, reduction in
the length of case management forms, and modified
site management practices.
The use of information technology to seamlessly integrate clinical and translational research and patient
care.
Greater use of adaptive (Bayesian) design techniques in
the design and analysis of randomized controlled trials
could reduce trial duration and the number of patients
needed.
Oncologists and patient advocacy organizations should
challenge regulatory authority data requirements.
Rather than criticize organizations such as NICE for declining reimbursement on grounds of cost-effectiveness,
clinicians and patient advocates should start challenging
pharmaceutical companies about the high prices they
seek for products with modest benefits.
Address the difficulties facing low- and middle-income
countries in accessing affordable cancer care, rather
than only focusing on problems facing high-income
countries.
CONCLUSION
Research and technologic innovation continue to produce
novel therapies, but these are often of marginal clinical
value, while managing to be very costly. Similarly, there is
increasing dependence on expensive new imaging modalities to assess tumor burden throughout treatment, despite
the seeming paradox that patients in the United Kingdom, at
least, present with later-stage disease. There is no doubt
that a programmatic approach to budgeting the totality of
cancer care for a specific population will highlight areas that
represent low value and that, therefore, could be targets for
disinvestment to make way for more effective treatment.
References
1. Sullivan R, Peppercorn J, Sikora K, et al. Delivering affordable cancer care
in high-income countries. Lancet Oncol. 2011;12:933-980.
2. Amalric F. Analyse e conomique des Couts de Cancer en France. Impact Sur
la Qualite de Vie, Prevention, Depistage, Soin, Recherche. Paris: Institut
National Du Cancer; 2007.
3. Porter ME. Defining and introducing value in health care. In Evidencebased Medicine and the Changing Nature of Health Care: 2007 Institute
of Medicine (IOM) Annual Meeting Summary. Washington, DC: Institute
of Medicine; 2008.
4. Porter ME, Teisberg EO. Redefining Health Care. Creating value-based

competition on results. Boston, MA: Harvard Business School Press;
2006.
5. Porter ME. What is value in health care? N Engl J Med. 2010;363:
2477-2481.
6. Eddy DM. Clinical decision making: from theory to practice. Three battles
to watch in the 1990s. JAMA. 1993;270:520-526.
7. Rawlins MD, Chalkidou K. The opportunity cost of cancer care: a statement from NICE. Lancet Oncol. 2011;12:931-932.
e15
ULKA VAISHAMPAYAN
The Role of Nephrectomy for Kidney Cancer in the Era of

Targeted and Immune Therapies
Editors Note: The following article is based on the 2016 ASCO Annual Meeting Education Session Cytoreductive Nephrectomy in
Renal Cell Carcinoma: A Debate. The author reviews the pros and cons of cytoreductive nephrectomy and whether recent
advances in systemic therapy warrant physicians to proceed directly to systemic therapy as in other metastatic solid tumors,
without the integral step of cytoreductive nephrectomy.
OVERVIEW
Although two phase III trials support the recommendation of nephrectomy followed by interferon alpha in metastatic renal
cell carcinoma (RCC), this procedure cannot be applied to every patient with this condition. Systemic therapy has changed
from interferon alpha to antiangiogenic-targeted therapy, and the clinical impact of nephrectomy in the era of targeted
therapy has not been proven. The SEER database shows that only 35% of patients with advanced RCC undergo nephrectomy
as their initial treatment. Retrospective studies showed improved overall survival (OS) outcomes with nephrectomy and
interleukin-2 (IL-2) therapy; however, the inherent selection bias of younger and healthier patients receiving IL-2 likely
accounts for this finding. Neoadjuvant therapy has demonstrated only modest efficacy in unresectable disease, and if
remission is obtained with systemic therapy, it is unclear whether nephrectomy has any incremental benefit. In the absence
of proven benefit of nephrectomy in the setting of targeted therapy, it seems advisable for patients with RCC with severely
symptomatic disease, competing comorbidities, poor performance status, or unresectable disease to avoid nephrectomy
and proceed directly to systemic therapy. The clinical implications of deferred cytoreductive nephrectomy for patients with
metastatic RCC are poorly understood, and patient cohorts that do not undergo this procedure are likely to be comprised of
patients with unfavorable disease characteristics. Unfortunately, the completed trials of targeted therapy were 90%
comprised of patients with prior nephrectomy (the majority of trials incorporate prior nephrectomy as an eligibility requirement) and hence may not reflect the outcomes of the majority of the patients with advanced RCC who have not
undergone nephrectomy. Newer therapies such as nivolumab and cabozantinib have also been evaluated for a population in
which 90% of the patients underwent nephrectomy. Future clinical trials and registry studies must focus on the therapeutic
treatment and overall outcome of patients without nephrectomy and treated with contemporary systemic therapy.
idney cancer is one of the few malignancies in which

surgical resection of the primary disease, or cytoreductive nephrectomy, has held tremendous importance.
Historically, systemic therapy was underdeveloped and ineffective, and surgical resection was considered the only
hope of remission. However, an analysis of the Surveillance,
Epidemiology, and End Results Program (SEER) database
revealed that only 539 of the 1,537 (35%) cases diagnosed
with advanced RCC between 2000 and 2013 had nephrectomy as their initial therapy.1
The major advances in systemic therapy of RCC in recent
years beg the fundamental question whether a large enough
impact has been made to consider proceeding directly to
systemic therapy as in other metastatic solid tumors,
without the integral step of cytoreductive nephrectomy.

The pros and cons of this procedure are summarized in
Table 1.
NEPHRECTOMY IN THE CONTEXT OF IMMUNE

THERAPY
Renal cell carcinoma is unique in the fact that randomized
trials have actually been conducted to establish the role of
nephrectomy in metastatic disease.24 The results revealed
that patients subjected to nephrectomy followed by interferon alpha in the presence of metastatic disease experienced OS benefit (Table 2). The reasons for this
benefit remain unclear. It does not appear that nephrectomy
From the Karmanos Cancer Institute, Wayne State University, Detroit, MI.
Disclosures of potential conflicts of interest provided by the author are available with the online article at asco.org/edbook.
Corresponding author: Ulka N. Vaishampayan, MD, Karmanos Cancer Institute 4100 John R St., 4233 HWRC, Detroit, MI 48201; email: vaishamu@karmanos.org.
e16
ROLE OF NEPHRECTOMY IN THE TARGETED THERAPY ERA
TABLE 1. Pros and Cons of Nephrectomy in Metastatic

Renal Cell Carcinoma
Pros of Nephrectomy in
Metastatic RCC
Cons of Nephrectomy in
Metastatic RCC
Control of symptoms from

primary tumor
Complications of surgery
Phase III trials showing OS benefit No benefit proven with targeted

therapy
No change in response to systemic
therapy
Resection of resistant clones
Delay of systemic therapy
Long-term remissions noted with Impaired creatinine clearance,

surgical resection of metastatic
increased risk of hypertension
RCC
Nephrectomy patients comprise
90% of clinical trial patient
population
Unlikely to benefit patients with

comorbidities and impaired
performance status
Abbreviations: RCC, renal cell carcinoma; OS, overall survival
boosted the response rates to interferon therapy, as the

response rate remained at a dismal 3.3% and 3.6% in the
cytoreductive nephrectomy and interferon-alone arms,
respectively, regardless of whether the patients were
randomly selected to undergo nephrectomy or not. Retrospective case series have been reported favoring cytoreductive nephrectomy prior to immune therapy, such as IL-2,
because of OS benefit.5 However, the inherent biases of a
retrospective study are prominently prevalent in the IL-2
population, as this is a stringently selected group of patients
with excellent cardiopulmonary status and no brain metastases, which predicts a better OS regardless of choice or
sequence of therapy. In fact, even the patients with a
performance status of 1 within the IL-2treated patient cohort had a minimal benefit in OS favoring cytoreductive
KEY POINTS
Two phase III trials support the recommendation of

nephrectomy in metastatic renal cancer when
compared with interferon therapy alone; however, the
clinical applicability of this finding is not clear in the era
of targeted therapy.
A delay in systemic therapy (because of nephrectomy)
can have deleterious effects on the survival of patients
with metastatic renal cancer.
Patients with poor-risk disease or predicted survival less
than 12 months are unlikely to benefit from
nephrectomy. The clinical implications of deferred
nephrectomy for patients with metastatic renal cancer
are not well understood.
Surveillance, Epidemiology, and End Results Program
database analysis revealed that 65% of patients with
advanced renal cancer do not undergo nephrectomy as
their initial treatment. This patient population warrants
further study, especially as it is not included in current
clinical trials.
nephrectomy; median OS was 6.9 months in the cytoreductive nephrectomy arm and 4.8 months in the interferonalone arm.5
There are numerous reports indicating that cytoreductive
nephrectomy improves host immune reaction to the
metastases and is likely to decrease the levels of immunosuppressive factors. It also normalized the nuclear factorkappa B and various other immune defects.6,7 Clinically
spontaneous regressions have been observed. A perioperative (neoadjuvant) checkpoint inhibitor trial has
been proposed based on the finding that peripheral
blood PD-1 expression is reduced significantly after cytoreductive nephrectomy.8 Expression of PD-1 in tumorinfiltrating lymphocytes was associated with a more
aggressive phenotype of RCC (larger tumors, higher nuclear grade, and sarcomatoid differentiation) and increased
risk of cancer-specific death, as reported in a study led
by the Mayo Clinic (risk ratio of 2.24; p = .004).9 However,
a recent report evaluating tumor PD-L1 expression in
417 cases of clear cell RCC from The Cancer Genome
Atlas database reported a lack of association between
PD-L1 expression and unfavorable tumor characteristics
and an improved OS outcome (hazard ratio [HR] 0.59;
p = .006).10
Unfortunately, modern immunotherapy trials cannot inform the benefits of treatment for patients who have not
undergone nephrectomy. For immune checkpoint inhibitors,
the recently reported randomized trial of nivolumab versus
everolimus mainly included patients who had undergone
nephrectomy.11 Similarly for patients on antiangiogenic or
mTOR inhibitor therapies, approximately 90% of the patients
included had nephrectomy prior to study enrollment.1216
Studies are ongoing to evaluate the impact of PD-1 inhibition
on biomarkers of immune response pre- and post-nephrectomy.
Phase II trials such as ADAPTeR (NCT02446860) are assessing
the response and changes in immune-related markers after
the neoadjuvant administration of nivolumab for 8 weeks
followed by nephrectomy.
NEPHRECTOMY IN THE CONTEXT OF VEGF

INHIBITOR THERAPY
The clinical implications of deferred nephrectomy for patients with metastatic renal cancer are not well understood.
As in the case of immunotherapy trials, the percentage of
patients who had undergone nephrectomy in a majority of
clinical trials evaluating VEGF inhibitors is greater than 90%.
The report by Choueiri et al,17 which suggested that patients
treated with targeted therapy had a Karnofsky performance
status less than 80% or poor-risk disease as gauged by the
Memorial Sloan Kettering Cancer Center criteria, experienced no benefit from cytoreductive nephrectomy (p = .08
and .06, respectively; Table 2). A separate report used data
from the International Metastatic Renal Cell Carcinoma
Database Consortium database, which included 982 patients
treated with nephrectomy and targeted therapy and 676
patients treated with targeted therapy alone.18 Although
e17
ULKA VAISHAMPAYAN
TABLE 2. Summary of Study Data on Nephrectomy and Systemic Therapy

First Author, Reference
Type of Study
No. of Patients
Results
Flanigan and Yonover2
Phase III
241
Median OS: CN + IFN: 11.1 months; IFN: 8.1 months; PFS 0; p = .05;
Median OS: CN + IFN: 17.4 months; IFN: 11.7 months; PFS 1; p = .08;
Median OS: CN + IFN: 6.9 months; IFN: 4.8 months
Mickisch et al3
Phase III
84
Median OS: CN + IFN: 17 months; IFN: 7 months (HR 0.54; 95% CI, 0.310.94)
Flanigan et al
Meta-analysis
331
Median OS: CN + IFN: 13.6 months; IFN: 7.8 months (HR 0.69; p = .002)
Pantuck et al5
Retrospective
89
CN + IL-2: median OS, 16.7 months; 5-year OS, 19.6%
Choueiri et al17
Retrospective
targeted-therapy era
314 (CN: 201;

no CN: 113)
Median OS: 19.8 months (CN) vs. 9.4 months (no CN); HR 0.44; p , .01;
multivariate analysis: no benefit; p = .08; poor risk: no benefit; p = .06
Heng et al18
Retrospective
targeted-therapy era
1,658 (CN: 982;

no CN: 676)
Median OS: 20.6 months (CN) vs. 9.5 months (no CN); p , .0001
Abbreviation: CN, cytoreductive nephrectomy; OS, overall survival; IFN, interferon; PFS, progression-free survival; N/A, not applicable.
patients treated with cytoreductive nephrectomy had significantly better OS compared with those treated with
targeted agents (median OS, 20.6 vs. 9.5 months, respectively; HR 0.60; 95% CI, 0.520.69), it is important to
recognize that selection bias is at play; patients with poorrisk disease comprised only 28% of those treated with
cytoreductive nephrectomy versus 54% of the group who did
not undergo the procedure. In addition, only 1% of patients
who did not undergo cytoreductive nephrectomy had favorable risk characteristics.
Several trials are seeking to prospectively evaluate the role
of cytoreductive nephrectomy for patients being treated
with targeted therapy. The CARMENA trial studying nephrectomy followed by sunitinib versus sunitinib alone
will likely shed light on the question of whether cytoreductive nephrectomy remains relevant in the context
of anti-VEGF therapy for metastatic renal cancer. The
primary endpoint is OS, and secondary endpoints are
progression-free survival and operative complication rates.
The study has completed accrual, and results are awaited. The SURTIME study is randomly selecting patients
with metastatic RCC to receive sunitinib followed by
nephrectomy versus sunitinib alone. The study will address whether patients with adequate response to systemic therapy should receive surgical consolidation with
nephrectomy.
SYNERGY BETWEEN NEPHRECTOMY AND

SYSTEMIC THERAPY: REALITY OR PERCEPTION?
Despite a hypothesized synergy between systemic therapy
and nephrectomy, it has not been borne out in clinical
practice. For example, the randomized trials of nephrectomy
and interferon versus interferon alone reported identical
response rates of approximately 3% in both arms. An adjuvant trial of sunitnib versus sorafenib versus placebo
(ASSURE/ECOG 2805) reported lack of benefit for either of
the agents postnephrectomy.18 The only data that suggest
potential synergy come from retrospective trials of patients
treated with IL-2 showing an improved outcome achieved
for patients who underwent nephrectomy as compared with
e18
those without nephrectomy.5 Neoadjuvant therapy has also

shown only modest benefit in advanced RCC.19 In summary,
until the results of randomized trials are available, no
synergy has been proven to date between targeted
antiangiogenic therapy and cytoreductive nephrectomy
in RCC.
THE IMPORTANCE OF MULTIDISCIPLINARY

TEAM INPUT: COLLECTIVE WISDOM AND THE
FUTURE OF PATIENT-CENTERED CARE AND
RESEARCH
Typically, patients who are not offered cytoreductive nephrectomy have poor performance status, major comorbidities, and limited access to care. Racial disparity in rates of
nephrectomy has also been shown to have an impact on
patient outcomes. In a SEER database analysis of advanced
RCC, patients of African-American origin were noted to
have a worse OS outcome as compared with white patients
with metastatic RCC, and an interaction effect was noted
with the disparity in rates of nephrectomy.20 However,
whether this finding reflects other inherent biases noted
above is unknown. This observation highlights the need for
data collection regarding the factors commonly considered
toward the decision-making process of nephrectomy.
In addition, cytoreductive nephrectomy carries a certain
risk that has to be factored into the decision. The major
complication rates of cytoreductive nephrectomy have been
reported to be about 5%, and 11% of the patients did not
start systemic therapy within 60 days after nephrectomy
because of surgical complications. Overall, 61% of patients
did not start systemic therapy in a timely fashion. The
presence of liver metastases, intraoperative transfusion, and
pathologic nodal involvement were noted to be predictors of
delay of systemic therapy more than 60 days after cytoreductive nephrectomy.21
It is not advisable to reflexively send the patient for nephrectomy upon diagnosis of advanced kidney cancer.
Multidisciplinary evaluation to carefully weigh the risk and
benefit ratio factoring in whether the bulk of the patients
symptoms are related to the cancer or to other comorbidities
ROLE OF NEPHRECTOMY IN THE TARGETED THERAPY ERA
would be important. Renal cancer is a heterogeneous disease,22 and to date, the resection of resistant clones is still
considered an ideal way to manage the problem. Stage IV
disease within RCC spans a wide spectrum of outcomes, and
risk prognostication is critically important before nephrectomy is considered.
The reported efficacy of targeted therapy for patients with
metastatic renal cancer is mainly known only in the setting of
cytoreductive nephrectomy. This means that current clinical
trialbased data regarding efficacy of systemic therapy in
RCC do not apply to a large magnitude of the patient
population treated. As systemic therapy options and efficacies evolve, the role of nephrectomy requires re-evaluation.
It does not seem appropriate to subject every patient with
metastatic renal cancer to the morbidity, adverse events,
cost, and potential delay of systemic therapy that result
from the nephrectomy procedure. In addition, special
efforts must be made to collect information regarding the
efficacy of specific systemic therapies for the patients who
are not candidates for nephrectomy. It is likely that the
presence of nephrectomy is an objective measure that
reflects a subgroup within RCC that has a distinctly favorable
outcome.
FACTORS TO BE CONSIDERED PRIOR TO

CYTOREDUCTIVE NEPHRECTOMY
Histology (Clear Cell or Non-clear Cell)
Non-clear cell histology has suboptimal systemic therapy
options, and hence cytoreductive nephrectomy maybe a
more important component of the management. However,
poor prognostic characteristics such as liver metastases,
impaired performance status, and the unresectable nature
of the disease should be considered as relative contraindications to cytoreductive nephrectomy.
Patient Performance Status

Retrospective studies show that patients with impaired
performance status failed to benefit from cytoreductive
nephrectomy.
Current/Impending Symptoms From Disease

Delay in systemic therapy because of cytoreductive nephrectomy is a concern for symptomatic patients with RCC.
Burden of Metastatic Disease

Patients with liver metastases or nodal involvement were
likely to have a higher complication rate with cytoreductive
nephrectomy that resulted in more than a 60-day delay in
systemic therapy.
Memorial Sloan Kettering Cancer Center Risk Criteria

Multivariate analysis revealed no benefit from cytoreductive nephrectomy for patients with poor risk characteristics
per the Memorial Sloan Kettering Cancer Centers risk
classification.
Risk of Surgical Complications

Patients with impaired performance status and comorbidities are likely to have a higher complication rate from
cytoreductive nephrectomy.
CASE ILLUSTRATIONS
Deferral of Nephrectomy as a Result of Disease
Symptoms and Burden
Case 1. A 68-year-old man presented with malena, anemia
(hemoglobin of 4.0 g/dL), and hip pain. Imaging revealed
a large renal mass, hip metastases, and lung metastases.
Colonoscopy showed large mucosal lesions, the biopsy of
which revealed clear cell RCC. This patient, given the lifethreatening nature of his metastatic disease, would not be a
candidate for nephrectomy. The patient was started on an
oral anti-VEGF tyrosine kinase inhibitor. His hemoglobin
improved within 2 weeks of starting therapy, and his lesions
were clinically responding. Clearly, this case needed to
proceed to systemic therapy rapidly, and attempting cytoreductive nephrectomy for this patient was likely to have
a deleterious effect. This case also raises the dilemma of
whether to consider a nephrectomy at a later date after initial
response to targeted therapy. There is no evidence to support
that cytoreductive nephrectomy would be beneficial at this
time in this case. The concerns comprise discontinuing VEGF
inhibitor therapy, the regrowth potential of his disease, and
patient safety during surgery. In a clinical trial of preoperative
sunitinib followed by cytoreductive nephrectomy, about 36%
(17 of 47) of patients had disease progression during the
break from systemic therapy, confirming the concerns regarding this approach.19 The patient continues to be treated
without nephrectomy and with systemic therapy alone.
Case 2. A 58-year-old man presented with severe dyspnea,
noted to be related to pleural effusion. A 6-cm left-sided
renal mass was noted. Thoracentesis was performed, and
lung and pleural metastases were noted. Cytology revealed
malignant cells, and biopsy of kidney revealed clear cell
cancer. The patient was treated with targeted therapy and
demonstrated a response with complete resolution of the
pleural effusion and stable renal mass. In this case, nephrectomy was considered advisable to render the patient in
clinical complete remission. However, there were valid
concerns of holding systemic therapy and risking progression of disease during cytoreductive nephrectomy and the
healing period after the procedure.
Case 3. A 52-year-old man was hospitalized with hypercalcemia; his corrected calcium was 15 mg/dL, and workup
revealed a renal mass and bilateral lung and bone lesions,
including one in the sacrum. Biopsy of bone showed clear cell
cancer consistent with renal primary. This patient needs to
proceed directly to systemic therapy, as attempting a nephrectomy in the presence of uncontrolled metastatic disease would result in worsening patient condition and
compromise survival outcome.
e19
ULKA VAISHAMPAYAN
Deferral of Nephrectomy as a Result of Comorbid

Conditions
Case. An 85-year-old woman with coronary artery disease and angioplasty with stent placement 4 months
ago presented with microscopic hematuria. Workup
revealed a 7-cm right renal mass and bilateral lung
masses. Biopsy of the lung lesion demonstrated clear
cell carcinoma consistent with renal primary. Cytoreductive nephrectomy was considered risky because of
the patients cardiac condition, and, hence, systemic
therapy was started with clinical response. Currently,
the patient remains free of progression for the last
10 months.
CONCLUSION
Absence of cytoreductive nephrectomy is a very common occurrence in advanced RCC (65% of cases) and warrants further
clinical and research investigation. The efficacy and toxicity
reports from clinical trials of targeted and immune therapy
predominantly represent the patient population that has had
cytoreductive nephrectomy and need to be interpreted and
applied in the appropriate context. The complication rates of
nephrectomy range from 7% to 21%19 and may have implications in the delay of or inability to administer systemic therapy.
Outcome-based research and interventional trials focusing on
the clinical treatment of patients with metastatic RCC not undergoing nephrectomy represent a critical unmet need.
References
1. National Cancer Institute. SEER*Stat Databases: November 2014 Submission.
http://seer.cancer.gov/data/seerstat/nov2014/. Accessed March 21, 2016.
2. Flanigan RC, Yonover PM. The role of radical nephrectomy in metastatic
renal cell carcinoma. Semin Urol Oncol. 2001;19:98-102.
3. Mickisch GH, Garin A, van Poppel H, et al; European Organisation for
Research and Treatment of Cancer (EORTC) Genitourinary Group. Radical
nephrectomy plus interferon-alfa-based immunotherapy compared with
interferon alfa alone in metastatic renal-cell carcinoma: a randomised
trial. Lancet. 2001;358:966-970.
4. Flanigan RC, Mickisch G, Sylvester R, et al. Cytoreductive nephrectomy
in patients with metastatic renal cancer: a combined analysis. J Urol.
2004;171:1071-1076.
5. Pantuck AJ, Belldegrun AS, Figlin RA. Nephrectomy and interleukin-2 for
metastatic renal-cell carcinoma. N Engl J Med. 2001;345:1711-1712.
6. Dadian G, Riches PG, Henderson DC, et al. Immunological parameters in
peripheral blood of patients with renal cell carcinoma before and after
nephrectomy. Br J Urol. 1994;74:15-22.
7. Uzzo RG, Clark PE, Rayman P, et al. Alterations in NFkappaB activation in
T lymphocytes of patients with renal cell carcinoma. J Natl Cancer Inst.
1999;91:718-721.
8. MacFarlane AW IV, Jillab M, Plimack ER, et al. PD-1 expression on
peripheral blood cells increases with stage in renal cell carcinoma
patients and is rapidly reduced after surgical tumor resection. Cancer
Immunol Res. 2014;2:320-331.
9. Thompson RH, Dong H, Lohse CM, et al. PD-1 is expressed by tumorinfiltrating immune cells and is associated with poor outcome for
patients with renal cell carcinoma. Clin Cancer Res. 2007;13:1757-1761.
10. Peters I, Tezval H, Kramer MW, et al. Implications of TCGA network data
on 2nd generation immunotherapy concepts based on PD-L1 and PD-1
target structures. Aktuelle Urol. 2015;46:481-485.
11. Motzer RJ, Escudier B, McDermott DF, et al; CheckMate 025 Investigators. Nivolumab versus everolimus in advanced renal-cell carcinoma. N Engl J Med. 2015;373:1803-1813.
12. Escudier B, Eisen T, Stadler WM, et al; TARGET Study Group. Sorafenib in
advanced clear-cell renal-cell carcinoma. N Engl J Med. 2007;356:125-134.
e20
13. Motzer RJ, Hutson TE, Tomczak P, et al. Sunitinib versus interferon
alfa in metastatic renal-cell carcinoma. N Engl J Med. 2007;356:
115-124.
14. Hudes G, Carducci M, Tomczak P, et al; Global ARCC Trial. Temsirolimus,
interferon alfa, or both for advanced renal-cell carcinoma. N Engl J Med.
2007;356:2271-2281.
15. Sternberg CN, Davis ID, Mardiak J, et al. Pazopanib in locally advanced or
metastatic renal cell carcinoma: results of a randomized phase III trial.
J Clin Oncol. 2010;28:1061-1068.
16. Rini BI, Halabi S, Rosenberg JE, et al. Phase III trial of bevacizumab plus
interferon alfa versus interferon alfa monotherapy in patients with
metastatic renal cell carcinoma: final results of CALGB 90206. J Clin
Oncol. 2010;28:2137-2143.
17. Choueiri TK, Xie W, Kollmannsberger C, et al. The impact of cytoreductive nephrectomy on survival of patients with metastatic renal cell
carcinoma receiving vascular endothelial growth factor targeted
therapy. J Urol. 2011;185:60-66.
18. Heng DY, Wells JC, Rini BI, et al. Cytoreductive nephrectomy in patients
with synchronous metastases from renal cell carcinoma: results from
the International Metastatic Renal Cell Carcinoma Database Consortium. Eur Urol. 2014;66:704-710.
19. Haas NB, Manola J, Uzzo R, et al. Initial results from ASSURE (E2805):
adjuvant sorafenib or sunitinib for unfavorable renal carcinoma, an
ECOG-ACRIN-led, NCTN phase III trial. J Clin Oncol. 2015;33;(suppl; abstr
403).
20. Vaishampayan U, Vankayala H, Vigneau FD, et al. The effect of targeted
therapy on overall survival in advanced renal cancer: a study of the
national surveillance epidemiology and end results registry database.
Clin Genitourin Cancer. 2014;12:124-129.
21. Gershman B, Moreira DM, Boorjian SA, et al. Comprehensive characterization of the perioperative morbidity of cytoreductive nephrectomy. Eur Urol. 2016;69:84-91.
22. Gerlinger M, Rowan AJ, Horswell S, et al. Intratumor heterogeneity and
branched evolution revealed by multiregion sequencing. N Engl J Med.
2012;366:883-892.
BREAST CANCER
Breast Cancer Survivorship:

Strategies for Optimal Care
CHAIR
Beverly Moy, MD, MPH
Massachusetts General Hospital Cancer Center
Boston, MA
SPEAKERS
Debra L. Barton, PhD, AOCN, RN
Ann Arbor, MI
William J. Gradishar, MD
Robert H. Lurie Comprehensive Cancer Center of Northwestern University
Chicago, IL
Michelle E. Melisko, MD
UCSF Helen Diller Family Comprehensive Cancer Center
San Francisco, CA
MELISKO, GRADISHAR, AND MOY
Issues in Breast Cancer Survivorship: Optimal Care, Bone

Health, and Lifestyle Modifications
Michelle E. Melisko, MD, William J. Gradishar, MD, and Beverly Moy, MD, MPH
OVERVIEW
There are an estimated 3.1 million survivors of breast cancer in the United States. The predominant reasons for this
substantially large population are that breast cancer is the most common noncutaneous malignancy among women and that
5-year survival rates after breast cancer treatment are approximately 90%. These patients have many medical considerations, including the need to monitor for disease recurrence and to manage complications of their previous cancer
treatments. Most patients remain at risk indefinitely for local and systemic recurrences of their breast cancers and have an
increased risk of developing contralateral new primary breast cancers. Therefore, optimizing care for this patient population
is critical to the overall health care landscape in the United States. Here, we summarize survivorship care delivery and its
challenges, the optimization of bone health in breast cancer survivors, and opportunities for risk reduction through lifestyle
modifications.
here are an estimated 3.1 million survivors of breast

cancer in the United States.1 The predominant reasons
for this substantially large population are that breast cancer
is the most common noncutaneous malignancy among
women and that 5-year survival rates after breast cancer
treatment are approximately 90%.2 In recognition of the
importance of breast cancer survivorship care, multiple
medical organizations, including the Institute of Medicine,
American Society of Clinical Oncology (ASCO), American
Cancer Society, and the Commission on Cancer, have highlighted the importance of addressing the many issues and
concerns of breast cancer survivors following treatment. The
medical community recognizes the multiple challenges for
breast cancer survivors that follow-up care creates. In addition to the significant logistics required to care for a large
population of breast cancer survivors, reasons for these
challenges include the long course of disease, a relentless risk
of recurrence even 15 to 20 years after the original diagnosis,
the need for endocrine therapy for 5 to 10 years in most
patients, and the need to continue screening because of a 1%
risk per year of new contralateral primary breast cancers.
Clinicians must address the physical, emotional, social, and
long-term effects faced by many breast cancer survivors after
they complete active treatment. These long-term effects
include anxiety and depression, vasomotor symptoms, cognitive dysfunction, fatigue, pain, sexual dysfunction, loss of
bone density, infertility, neuropathy, and cardiac toxicity.
From an economic perspective, breast cancer is the

costliest of all cancers, with an estimated $18.1 billion of
national expenditures in the United States in 2014. Almost
$8 billion of these expenditures are for the costs of survivorship care.3 This proportional cost of breast cancer care
exceeds the annual expenditures for the entire treatment of
other types of cancer, such as leukemia and ovarian, brain,
head and neck, or pancreatic cancer.
Despite the calls for action from multiple prestigious
medical societies and the reality of the various medical and
economic challenges, to date, there is no universally agreed
upon standardized follow-up model for patients with earlystage breast cancer who have completed active cancer therapy.
Barriers to survivorship care implementation are well documented and include challenges with staffing and resources.4,5
Many societies have recommended that every patient receive a
survivorship care plan that details follow-up care plans, a surveillance plan, and general health recommendations.6 However,
survivorship care plan adoption in oncology clinics is highly
variable, and there is insufficient evidence that the plans
improve outcomes for patients.4,7
There are three models of comprehensive survivorship
care delivery found in academic medical centers.8 The first
model consists of consultative clinics that offer a one-time
survivorship visit for patients when active therapy is complete. The second model consists of advanced care
practitionerled survivorship clinics, during which patients
From the UCSF Helen Diller Family Comprehensive Cancer Center, San Francisco, CA; Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago, IL; Massachusetts General Hospital Cancer Center, Boston, MA.
Corresponding author: Beverly Moy, MD, Massachusetts General Hospital Cancer Center, 55 Fruit St., Boston, MA 02114; email: bmoy@partners.org.
e22
ISSUES IN BREAST CANCER SURVIVORSHIP
meet with nurse practitioners or physician assistants either

in a stand-alone clinic or embedded within the oncology
practice. The patient is then transitioned back to primary care
as appropriate. The third model is a multidisciplinary survivor
program that relies on collaboration between oncologists,
advanced care practitioners, social workers, and psychologists
to provide comprehensive care.
Some policymakers would argue that follow-up care of these
patients can be transitioned to primary care physicians on the
basis of randomized trial results that show equivalent outcomes with follow-up by either the oncologist or primary care
physician.9-12 The most compelling data are from a study in
which 968 women with early-stage breast cancer were randomly assigned to follow-up care with either a family physician
or an oncologist.9 Recurrence rates, mortality rates, and rates
of serious clinical events, such as spinal cord compression,
pathologic fractures, brachial plexopathy, or hypercalcemia,
were equivalent in both the family physician and the oncologist groups. Health-related quality of life was also similar in
both groups. Lending additional support to simplifying followup for breast cancer survivors, a British study revealed that
twice as many patients preferred simpler, less frequent followup by telephone.13 Finally, patients are less likely to demand
medically inappropriate testing, such as imaging or tumor
markers, if they are educated about the specificity, sensitivity,
and usefulness of the available tests.14
These studies beg the question: Are we providing excessively redundant care to breast cancer survivors? In the current
health care climate with finite resources, are oncology providers obligated to be more discriminating about the
amount of care we provide?15,16 There are myriad reasons
why oncology providers are not prepared to relinquish the
care of breast cancer survivors. Providers and patients often
feel emotionally attached to each other. It is not uncommon
for patients to want to see their oncologists more often for
KEY POINTS
Care for the estimated 3.1 million survivors of breast

cancer is logistically and medically complex, given the
long course of disease, continued risk of recurrence, and
treatment-related side effects.
Additional research to provide evidence about the
optimal way to deliver survivorship care to this large and
growing population is needed.
To date, there is no uniform consensus about the role of
bone-modifying agents in the treatment of early-stage
breast cancer.
Data on breast cancer risk reduction are based on
multiple epidemiologic studies.
Independent of the potential for reducing risk of breast
cancer recurrence, strong evidence exists that lifestyle
factors, such as modification of diet, increasing physical
activity, weight management, and smoking cessation,
influence other medical comorbidities that ultimately
impact overall survival.
reassurance, and, oncologists, driven by a sense of duty to

their patients, may be unwilling to relinquish control of their
patients medical care after developing close relationships
during active cancer treatment. Furthermore, primary care
physicians may be uncomfortable taking the lead in caring for
these patients because they may feel ill-prepared to manage
the many issues faced by breast cancer survivors. Regarding
specific guidance for medical care, we lack evidence about
whether certain types of breast cancer, such as locally advanced or hormone receptornegative disease, would benefit
from more intensive follow-up and imaging. Finally, with the
lack of interinstitutional standards, providers may worry about
the risk of being sued if they decrease their follow-up visits or
imaging studies.
There have been multiple efforts to provide clinical
guidance on how best to care for breast cancer survivors.
The National Comprehensive Cancer Network (NCCN) issued
guidelines for the treatment of breast cancer, which included information on recommended surveillance for cancer
recurrence or new cancers.17 Both ASCO and NCCN have
symptom-specific survivorship care guidelines to address
issues such as cardiac toxicity and neuropathy.18,19 Recently,
ASCO and the American Cancer Society jointly issued a
breast cancer survivorship care guideline to provide concrete recommendations to assist clinicians, both oncologists
and primary care physicians, in the care of these patients.20
Among the concrete recommendations, there is a recommendation patients receive a history and physical every 3 to
6 months for the first 3 years after primary therapy, every 6
to 12 months for the next 2 years, and annually thereafter.
Supported by the recommendations made by the ASCO
Choosing Wisely campaign, clinicians should not offer routine laboratory tests or imaging, except mammography if
indicated, for the detection of disease recurrence in the
absence of symptoms.21 The guideline also recommends that
the cancer treatment team provides a treatment summary
and survivorship care plan to primary care clinicians.
These and other similar guidelines are helpful. However,
they still provide ambiguity about the ideal models for care
and the appropriate intervals between follow-up examinations. A history and physical every 3 to 6 months amounts
to a difference of two to four clinician visits per year. Can
these visits be shared between primary care physicians,
medical oncologists, surgical oncologists, and radiation
oncologists? Would patients benefit more from increased
care coordination? From a purely economical point of view,
one could argue that biannual visits by any oncology provider or by a primary care physician in an otherwise healthy
asymptomatic patient would be supported by the evidence
and may provide the most value in terms of containing
medical costs. However, certain higher-risk patient populations may benefit from more intensive follow-up care.
Unfortunately, we lack clear evidence to be able to identify
these patient subgroups. There is a pressing need for additional research to provide evidence about the optimal way
to deliver care to a large and growing breast cancer survivorship population.
e23
BONE HEALTH AGENTS IN BREAST CANCER

The role of bone agents in the treatment of patients with
early-stage breast cancer has fallen under two broad categories: the established role of preventing loss of bone density
and the more contentious role of reducing the risk of recurrence of breast cancer.22-25 The two classes of bone agents
used are bisphosphonates, of which several agents currently
are in use, and RANK ligand inhibitors, of which there is one,
denosumab. Cancer treatmentinduced bone loss is an iatrogenic disease, a condition created by breast cancer treatments. It is well known that various cancer therapies (e.g.,
aromatase inhibitors, chemotherapy, and surgery) decrease
bone mineral density and increase the risk of fracture.26
Treatment-related fractures lead to decreased quality of life
and shorter survival times, which translate into clinical, social,
and economic consequences.27,28
The ABCSG-12,26,29,30 Z-FAST,31,32 and ZO-FAST33,34 trials
have shown that bone-targeted agents can prevent cancer
treatmentinduced bone loss in early breast cancer. In the
ZO-FAST trial,29,33,34 up-front zoledronic acid increased bone
mineral density in both the lumbar spine and hip. Gnant
et al26,29,30 reported that the addition of zoledronic acid to
both tamoxifen and anastrozole yielded no treatmentinduced bone loss compared with treatment without the
bone-strengthening agent. Additionally, denosumab produced similar results, with increases in bone mineral density
seen over 24 months in trabecular and cortical bone in women
with nonmetastatic breast cancer and low bone mass who
were receiving adjuvant aromatase inhibitor therapy.35
The issue of adjuvant bisphosphonate therapy in breast
cancer has been an area controversy. The results from several
long-term, large clinical trials have added to the confusion.
Improvements in disease-free survival (DFS) were reported
in postmenopausal women who were treated with letrozole
and zoledronic acid compared with letrozole alone (ZOFAST).29,33,35 In contrast, the findings of the AZURE trial35 did
not support the routine use of zoledronic acid in the adjuvant management of breast cancer; no significant benefits
in DFS and overall survival (OS) were reported. In both the
ABCSG-1226,30 and AZURE35 trials, adjuvant bisphosphonates
offered an advantage to postmenopausal women, regardless of
whether their postmenopausal status occurred naturally or was
drug induced. In the results of the 9-year update of the ABCSG12 trial, there was a persistent benefit in DFS with zoledronic
acid in a low-estrogen environment (e.g., a postmenopausal
state). The meta-analysis of 36 trials by Coleman et al,36 which
evaluated nearly 23,000 women, focused on the use of
bisphosphonates in the adjuvant setting. Although no effects on
disease outcomes in premenopausal women were observed,
adjuvant bisphosphonates reduced bone metastases and improved survival in women who had low levels of reproductive
hormones (which included women older than age 55 if menopausal status was unknown). There was a 34% reduction in the
risk of bone recurrence (p = .00001) and a 17% reduction in the
risk of breast cancer death (p = .004). No significant reduction
was reported in first distant tumor recurrence outside of bone.
e24
The phase III ABSCG-18 trial was presented by Gnant et al at

the 2015 San Antonio Breast Cancer Symposium, in which
postmenopausal women with nonmetastatic breast cancer
were randomly assigned to receive the RANK ligand inhibitor,
denosumab, or placebo in addition to aromatase inhibitor
therapy. 37 Previous results from the ABCSG-18 trial demonstrated a reduced fracture rate with adjuvant denosumab
compared with placebo in postmenopausal women with
hormone receptorpositive breast cancer who were treated
with aromatase inhibitors.16 In the most recent presentation,
the primary outcomes were related to bone health, and the
secondary outcomes included DFS. In an intention-to-treat
analysis, the impact of denosumab on DFS reached borderline
significance (hazard ratio [HR] 0.82; 95% CI, 0.661.00). The
benefit was approximately 1% after 3 years, 2% after 5 years,
and 3% after 7 years of follow-up. Exploratory subgroup analyses suggested that the benefit increases when denosumab
is started early, along with aromatase inhibitor therapy, and
that the benefit is greater in patients who have larger tumors
and ductal histology. Overall, the DFS benefit seen with
denosumab was comparable to that reported in the metaanalysis of bisphosphonates.16
The debate about these data largely revolves around how
to translate the apparent benefit into day-to-day management decisions. Different bone agents were used in the
trials, the trial designs were meant to evaluate primarily
bone health endpoints with DFS as a secondary endpoint,
duration of therapy is an open question, and which postmenopausal patients should be considered for this type of
adjuvant therapy remains unclear. To date, though the data
are compelling, the NCCN guideline panel could not reach
consensus for supporting these bone agents as a component of adjuvant therapy of breast cancer. That said, individual patients who have any issues with bone density or
increased risks of fracture may reap the benefits on bone
health as well as some additional reduction in the risk of
disease recurrence.
RISK REDUCTION THROUGH LIFESTYLE

MODIFICATIONS
After a breast cancer diagnosis, most patients have great
interest in adoption of lifestyle changes that might reduce
their risk of cancer recurrence. Despite significant advances
in breast cancer therapies that are supported by randomized
clinical trials demonstrating benefits in DFS and OS, the data
on risk reduction through lifestyle modifications are more
limited and are primarily based on epidemiologic studies.
Patients, the lay press, and physicians alike often confuse the
data about which lifestyle factors may impact risk of developing breast cancer compared with those factors that
have an influence on recurrence after diagnosis. Strong
evidence exists that, independent of the potential for reducing risk of breast cancer recurrence, lifestyle factors that
include modification of diet, increasing physical activity,
weight management, and smoking cessation influence other
medical comorbidities that ultimately impact OS. Similarly,
lifestyle and treatment factors also contribute to an increased risk of second primary malignancies seen among
breast survivors.38
Management of Comorbidities
Cancer survivors face many health challenges, including
morbidity from residual treatment toxicity and increased
mortality from cardiovascular and respiratory diseases. The
impact of comorbidities on survival after a diagnosis of
breast cancer cannot be underestimated. Among 63,566
women diagnosed with breast cancer who were older than
age 66 and identified through the Surveillance, Epidemiology, and End Results database, age and comorbidities at the
time of diagnosis had the largest effects on mortality from
causes other than breast cancer.39 The presence of cardiovascular disease, chronic obstructive pulmonary disease,
or diabetes increased breast cancerspecific mortality by
between 10% and 24%. Among this study population, cardiovascular disease was the primary cause of death in 15.9%
of women, followed closely by breast cancer in 15.1%. Although this study focused on an older population with breast
cancer, exposure to radiation, anthracyclines, and trastuzumab are all associated with a small risk of cardiac toxicity
that may subsequently result in excess morbidity and, in
some cases, mortality.39 In addition, the long-term impacts
of inducing premature menopause and of treatment with
an aromatase inhibitor on cardiovascular health in younger
breast cancer survivors are still unknown. Studies to investigate beta blockers, angiotensin-converting enzyme
inhibitors, and statins to reduce the risk of cardiac toxicity
during and after treatment with anthracyclines and trastuzumab have shown some promise, but none of these
pharmacologic interventions have become standard of
care.40-42 Attention to reducing the risk of subsequent
cardiovascular disease should be a priority after the diagnosis and treatment of breast cancer.
Weight Gain
Weight gain is common during and after treatment of breast
cancer and may be even more pronounced in patients who
are overweight or obese prior to diagnosis.43 Factors contributing to weight gain include use of corticosteroid premedications with chemotherapy, decreased physical activity
during treatment, transition into menopause, and ongoing
hormonal therapy.44 Weight gain after diagnosis has been
associated with a higher rate of breast cancer recurrences
and with worse OS. In an analysis of a cohort of 3,993 women
identified through state registries who were diagnosed with
stage I to III breast cancer, each 5-kg gain was associated
with a 12% increase in all-cause mortality, a 13% increase in
breast cancerspecific mortality, and a 19% increase in
cardiovascular disease mortality.45 A systematic review and
meta-analysis that included a total of 12 cohort studies and
clinical trials with a total of 23,832 patients measured weight
change after breast cancer diagnosis and investigated breast
cancerspecific mortality, all-cause mortality, and recurrence outcomes.46 Weight gain ($ 5.0%) compared with
maintenance (# 5.0% weight change) was associated with

increased all-cause mortality (HR 1.12). In a secondary
analysis to evaluate effects of the amount of weight gain,
moderate-level weight gain (5%10% compared with maintenance of baseline weight) was not associated with a hazard
of overall mortality (HR 0.97; 95% CI, 0.861.11; p = .70), but
higher weight gain (. 10% from baseline compared with
maintenance) was associated with increased mortality (HR
1.23; 95% CI, 1.091.39; p , .001). Among studies for which
data were available, weight gain among patients who were
already overweight at baseline (body mass index [BMI] .
25.0 kg/m2) did not result in worse mortality. In terms of breast
cancer outcomes, moderate weight gain (5%10%) was not
associated with breast cancerspecific mortality, whereas
there was a suggestion that weight gain of greater than 10%
was associated with breast cancerspecific mortality (HR 1.17;
95% CI, 1.001.38; p = .05). Interestingly, in the studies within
this meta-analysis that included data, no amount of weight
gain was associated with an increase in breast cancer recurrences.47,48 The reason for this may be that a larger number of
patients would be required to demonstrate a relatively small
contribution of weight gain on breast cancer recurrences.
Alternatively, the negative impact of weight gain on mortality
and possibly breast cancerspecific mortality after diagnosis of
breast cancer may not be due to increased rate of recurrence
but, rather, to the overall health, comorbidities, treatment
patterns, and responses to treatment after recurrence among
patients with significant weight gain.
Weight Loss Interventions

No matter what the underlying cause, avoidance of significant weight gain after a diagnosis of breast cancer is a desirable goal to improve OS. The challenge of avoiding weight
gain is substantial for breast cancer survivors, many of whom
enter menopause abruptly as a result of chemotherapy and
some of whom remain on hormonal therapy as long as 10
years. Additionally, many patients suffer ongoing toxicities
from their treatment, including fatigue, sleep disturbance,
neuropathy, and arthralgias, all which may interfere with the
ability to initiate and maintain an exercise program supportive
of weight control.
There have been numerous weight loss intervention
studies among breast cancer survivors. A recent systematic
review included 15 studies, among which were eight randomized controlled trials, four single-cohort experimental
intervention studies, two randomized parallel-intervention
trials, one randomized crossover trial, and one controlled
nonrandomized trial.49 The primary objective of each of
these studies was weight loss intervention with a focus on
changing body weight (measured as change in weight, BMI,
percentage body fat, or percent overweight). Studies designs included multicomponent interventions that incorporated diet, physical activity, and behavior modification,
whereas other studies combined personalized lifestyle
telephone counseling with face-to-face group-based education. Several studies used a commercial weight loss
programs, such as Weight Watchers and/or Curves,50,51 and
e25
one study incorporated group teleconferencing for breast

cancer survivors in rural locations.52
Of the 15 studies included in this systematic review, 14
resulted in significant weight loss; weight loss of 5% or greater
from baseline was reported in 10 of the studies (p , .001).
Unfortunately, weight regain was observed in some
studies that included longer follow-up. For example, in the
Exercise and Nutrition to Enhance Recovery and Good
Health for You (ENERGY) study, 692 overweight or obese
breast cancer survivors were randomly assigned to either a
group-based behavioral intervention, supplemented with
telephone counseling and tailored newsletters, or to a lessintensive control intervention to encourage weight loss.
Greater weight loss was observed in the intervention than in
the control group at 12 months (6% vs. 1.5%) and was still
maintained but was less pronounced at 24 months (3.7% vs.
1.3%). Improvements in blood pressure control and physical
activity were also observed in the intervention arm.53
Dietary Interventions
The evidence to support specific dietary recommendations
after a diagnosis of breast cancer is surprisingly limited and is
backed by only a handful of randomized clinical trials, few of
which achieved the desired endpoints of either reduction in
breast cancer recurrences or improvement in DFS or OS.
The Womens Intervention Nutrition Study (WINS) was a
randomized prospective multicenter clinical trial that
included a total of 2,437 patients with early-stage breast
cancer to test the effect of a dietary intervention designed to
reduce fat intake.54 The goal of the dietary intervention was
to reduce percentage of calories from fat to 15%, such that
there would actually be a sustained reduction in fat intake to
approximately 20% of calories in the intervention arm. The
low-fat eating plan included self-monitoring (fat gram
counting and recording), goal setting, modeling, and relapse
support and involved multiple in-person counseling sessions
initially and then subsequent contact with a dietician every
3 months. The patients in the control group had a baseline
dietician visit and then visits every 3 months, during which
general dietary guidelines were reviewed. At a median
follow-up time of 60 months, an interim analysis demonstrated that dietary fat intake was significantly lower in the
intervention than in the control group (p , .001), and the
effect was maintained out to 60 months. Additionally, there
was a 6-pound lower mean body weight in the intervention
group. The hazard of relapse events (local, regional, distant,
or ipsilateral breast cancer recurrence or new contralateral
breast cancer) in the intervention group compared with the
control group was 0.76 (95% CI, 0.600.98). Subgroup analyses were conducted on the basis of BMI, hormone receptor status, and nodal status, and the dietary intervention
had a greater effect on relapse-free survival in women with
estrogen receptornegative cancer (HR 0.58; 95% CI,
0.370.91) than in women with estrogen receptorpositive
disease (HR 0.85; 95% CI, 0.631.14). In this analysis, with a
60-month median follow-up time, there was no difference in
OS between the dietary intervention and control groups.
e26
The Womens Healthy Eating and Living (WHEL) randomized

trial tested whether increasing vegetable, fruit, and fiber intake
and decreasing dietary fat intake could reduce the risk of recurrence or new primary breast cancer and impact mortality in
patients with early-stage breast cancer.55 A total of 3,088
women were randomly assigned to an intervention arm to
receive telephone counseling, cooking classes, and newsletters
promoting high fruit and vegetable intake, increasing fiber, and
reducing energy intake from fat, or to a comparison group that
was provided only with print materials. Analysis validated by
plasma carotenoid concentrations indicated that the intervention group maintained significant differences compared
with the control group for increased fruit, vegetable, and fiber
intake and reduction in fat through 4 years of follow-up. After a
mean follow-up time of 7.3 years, there were no differences
between the intervention and control groups for recurrence of
invasive or noninvasive breast cancer or death. Explanations of
this lack of treatment effect included the fact that many WHEL
participants had already changed their dietary patterns after
diagnosis of breast cancer, so 75% of patients were already
consuming at least five servings of vegetables and fruit a day at
baseline; results suggest that intake of fruits and vegetables
beyond that amount may not be beneficial. Multiple subsequent endocrinologic and behavioral findings and subset
analyses have been reported from the data collected from this
very well-conducted trial, which demonstrates the importance
of this study despite its negative results.
Exercise
Physical activity has been shown to improve quality of life
after a breast cancer diagnosis.56 Multiple cohort studies
have investigated the impact of physical activity after diagnosis on breast cancerspecific and overall mortalities.
Most of these studies describe exercise frequency and intensity in metabolic equivalent (MET) hours, for which 3 MET
hours is equivalent to walking at average pace of 2 to 2.9 mph
for 1 hour. Among 2,987 patients with early-stage breast cancer
in the Nurses Health Study, the relative risk of death from
breast cancer, compared with women who engaged in fewer
than 3 MET hours per week of physical activity, for patients with
9 to 14.9 MET hours per week of exercise was 0.50 (95% CI,
0.310.82); for patients with 15 to 23.9 MET hours per week
was 0.56 (95% CI, 0.380.84); and for patients with 24 or more
MET hours per week was 0.60 (95% CI, 0.400.89). Absolute
mortality risk was also reduced by 6% at 10 years for women
who engaged in 9 or more MET hours per week.57
In the Collaborative Womens Longevity Study (CWLS),
4,482 women completed questionnaires on postdiagnosis
physical activity and other lifestyle factors. Compared
with women engaging in fewer than 2.8 MET hours per
week of physical activity, women who engaged in greater
levels of activity ($ 21 MET hours per week) had a significantly lower risk of dying as a result of breast cancer
(HR 0.51; 95% CI, 0.290.89; p = .05) and had improved OS
(HR 0.44; 95% CI, 0.320.60; p , .001). These results were
consistent independent of age, stage of disease, and
BMI.58
In two smaller-sized cohorts (1,970 women in The Life After

Cancer Epidemiology [LACE] study and 933 women in the
Health, Eating, Activity and Lifestyle [HEAL] study), greater
physical activity after diagnosis was associated with a decrease in overall mortality but not in breast cancerspecific
mortality.59,60
A longitudinal study of 4,643 women diagnosed with invasive breast cancer after entry into the Womens Health
Initiative study evaluated the association between change in
physical activity from before diagnosis to after diagnosis with
all-cause and breast cancerspecific mortalities. Physical activity was reported at baseline (before breast cancer diagnosis)
and after diagnosis (3 or 6 years after baseline visit). Women
who engaged in moderate levels of physical activity (fast walking
for 3 hours/week) before diagnosis had lower all-cause mortality
(HR 0.61; 95% CI, 0.440.87; p = .01) than inactive women.
Women engaging in similar moderate physical activity levels
after diagnosis had both lower breast cancer mortality (HR 0.61;
95% CI, 0.350.99; p = .049) and lower all-cause mortality (HR
0.54; 95% CI, 0.380.79; p , .01). Women who increased or
maintained physical activity of 9 or more MET hours per week
after diagnosis had lower all-cause mortality (HR 0.67; 95% CI,
0.460.96) even if they were inactive before diagnosis.59
With the abundance of epidemiologic data demonstrating
the benefit of exercise after diagnosis of breast cancer, numerous randomized prospective trials are ongoing to better
understand which interventions are most effective to increase
physical activity, what levels and types of physical activity are
optimal, and if associated lifestyle factors have an interaction
with the benefit of exercise.
Alcohol Consumption
There is reasonably strong evidence that moderate alcohol
consumption increases the risk of developing breast cancer,
but the effect of ongoing alcohol intake after diagnosis is
less clear. In the LACE cohort consisting of 1,897 patients
with early-stage breast cancer who enrolled on average 2
years after diagnosis, self-reported drinking of 6 g/day or
more of alcohol (approximately 2.5 ounces of wine, 6
ounces of beer, or 1.5 ounces of hard liquor) compared
with no drinking was associated with an increased risk of

breast cancer recurrence (HR 1.35; 95% CI, 1.001.83) and
death as a result of breast cancer (HR 1.51; 95% CI,
1.002.29). The increased risk of recurrence was most
pronounced in postmenopausal and overweight/obese
women. Alcohol intake was not associated with overall
mortality, possibly because of a cardioprotective effect
and a reduction in noncancer deaths.61
In contrast to the findings in the LACE cohort, alcohol
intake was not significantly associated with breast cancer
recurrence in 3,088 patients participating in the WHEL
trial.62 Among this group of patients, moderate alcohol
intake (. 300 g/month or seven drinks a week) was associated with improved all-cause mortality (HR 0.69; 95%
CI, 0.490.97) when analysis was adjusted for obesity. One
explanation for this may be that obese women were more
likely to be nondrinkers, and patients with moderate alcohol consumption may engage in protective behaviors,
such as increased physical activity.
In the After Breast Cancer Pooling Project, which included
9,329 patients with breast cancer (more than half of whom
were part of the LACE and WHEL cohorts), regular alcohol
intake ($ 6.0 g/day) had no effect on breast cancer recurrence
or overall mortality among the entire population. However,
among women who were postmenopausal, alcohol consumption of 6.0 or more g/day was associated with increased
breast cancer recurrence (HR 1.19; 95% CI, 1.011.40).61
CONCLUSION
Breast cancer survivorship care has emerged as a highpriority issue in the oncology community, given its large
and growing population. Breast cancer survivors face
multiple complex issues, including optimization of bone
health and lifestyle considerations. Clinicians caring for
this substantial population must be knowledgeable on
many medical issues and must be prepared to counsel
their patients on the best health promotion strategies to
minimize risks of recurrence and long-term side effects of
their treatments.
References
1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2015. CA Cancer J Clin.
2015;65:5-29.
2. Howlader N, Noone AM, Krapcho M, et al (eds). SEER Cancer Statistics
Review, 1975-2012. http://seer.cancer.gov/csr.1975-2012/. Bethesda,
MD: National Cancer Institute; 2015.
3. National Institutes of Health. Cancer trends progress report. http://
progressreport.cancer.gov. Accessed January 12, 2016.
4. Mayer DK, Birken SA, Check DK, et al. Summing it up: an integrative
review of studies of cancer survivorship care plans (2006-2013). Cancer.
2015;121:978-996.
5. Birken SA, Mayer DK, Weiner BJ. Survivorship care plans: prevalence
and barriers to use. J Cancer Educ. 2013;28:290-296.
6. Hewitt M, Greenfield S, and Stovall E (eds). From Cancer Patient to

Cancer Survivor: Lost in Transition. Washington, DC: The National
Academies Press; 2005.
7. Salz T, McCabe MS, Onstad EE, et al. Survivorship care plans: is there buy-in
from community oncology providers? Cancer. 2014;120:722-730.
8. Oeffinger KC, McCabe MS. Models for delivering survivorship care. J Clin
Oncol. 2006;24:5117-5124.
9. Grunfeld E, Levine MN, Julian JA, et al. Randomized trial of long-term
follow-up for early-stage breast cancer: a comparison of family physician versus specialist care. J Clin Oncol. 2006;24:848-855.
10. Grunfeld E, Mant D, Yudkin P, et al. Routine follow up of breast cancer in
primary care: randomised trial. BMJ. 1996;313:665-669.
e27
11. Grunfeld E, Fitzpatrick R, Mant D, et al. Comparison of breast cancer patient

satisfaction with follow-up in primary care versus specialist care: results
from a randomized controlled trial. Br J Gen Pract. 1999;49:705-710.
12. Kerrigan D, Waters P, Ryan M, et al. Follow-up arrangements for breast
cancer patients: is it appropriate to transfer surveillance to general
practioners? Ir Med J. 2014;107:273-275.
13. Gulliford T, Opomu M, Wilson E, et al. Popularity of less frequent follow
up for breast cancer in randomised study: initial findings from the
hotline study. BMJ. 1997;314:174-177.
14. Loprinzi CL, Hayes D, Smith T. Doc, shouldnt we be getting some tests?
J Clin Oncol. 2000;18:2345-2348.
15. Jagsi R. Debating the oncologists role in defining the value of cancer
care: we have a duty to society. J Clin Oncol. 2014;32:4035-4038.
16. Sulmasy D, Moy B. Debating the oncologists role in defining the value of
cancer care: our duty is to our patients. J Clin Oncol. 2014;32:4039-4041.
17. Gradishar WJ, Anderson BO, Balassanian R, et al. Breast Cancer, Version
1.2016. J Natl Compr Canc Network. 2015;13:1475-1485.
18. Denlinger CS, Ligibel JA, Are M, et al. Survivorship: screening for cancer
and treatment effects, version 2.2014. J Natl Compr Canc Netw. 2014;
12:1526-1531.
19. Khatcheressian JL, Hurley P, Bantug E, et al; American Society of Clinical
Oncology. Breast cancer follow-up and management after primary
treatment: American Society of Clinical Oncology clinical practice
guideline update. J Clin Oncol. 2013;31:961-965.
20. Runowicz CD, Leach CR, Henry NL, et al. American Cancer Society/
American Society of Clinical Oncology breast cancer survivorship care
guideline. J Clin Oncol. 2016; 34:611-635.
21. Schnipper LE, Smith TJ, Raghavan D, et al. American Society of Clinical
costs: the top five list for oncology. J Clin Oncol. 2012;30:1715-1724.
22. Ben-Aharon I, Vidal L, Rizel S, et al. Bisphosphonates in the adjuvant
setting of breast cancer therapyeffect on survival: a systematic review
and meta-analysis. PLoS One. 2013;8:e70044.
23. Hadji P, Coleman RE, Wilson C, et al. Adjuvant bisphosphonates in early
breast cancer: consensus guidance for clinical practice from a European
Panel. Ann Oncol. 2016;27:379-390.
24. Strobl S, Korkmaz B, Devyatko Y, et al. Adjuvant bisphosphonates and
breast cancer survival. Annu Rev Med. 2016;67:1-10.
25. Gnant M, Mlineritsch B, Luschin-Ebengreuth G, et al; Austrian Breast
and Colorectal Cancer Study Group (ABCSG). Adjuvant endocrine
therapy plus zoledronic acid in premenopausal women with early-stage
breast cancer: 5-year follow-up of the ABCSG-12 bone-mineral density
substudy. Lancet Oncol. 2008;9:840-849.
26. Ezat SW, Syed Junid SM, Noraziani K, et al. Skeletal-related events
among breast and prostate cancer patients: towards new treatment
initiation in Malaysias hospital setting. Asian Pac J Cancer Prev. 2013;
14:3357-3362.
27. Hagiwara M, Delea TE, Chung K. Healthcare costs associated with
skeletal-related events in breast cancer patients with bone metastases.
J Med Econ. 2014;17:223-230.
28. Gnant M, Eidtmann H. The anti-tumor effect of bisphosphonates ABCSG12, ZO-FAST and more. Crit Rev Oncol Hematol. 2010;74:S2-S6 (suppl 1).
29. Gnant M, Mlineritsch B, Stoeger H, et al; Austrian Breast and Colorectal
Cancer Study Group, Vienna, Austria. Adjuvant endocrine therapy plus
zoledronic acid in premenopausal women with early-stage breast
cancer: 62-month follow-up from the ABCSG-12 randomised trial.
Lancet Oncol. 2011;12:631-641.
30. Brufsky AM, Bosserman LD, Caradonna RR, et al. Zoledronic acid effectively prevents aromatase inhibitor-associated bone loss in postmenopausal women with early breast cancer receiving adjuvant
letrozole: Z-FAST study 36-month follow-up results. Clin Breast Cancer.
2009;9:77-85.
e28
31. Brufsky AM, Harker WG, Beck JT, et al. Final 5-year results of Z-FAST trial:
adjuvant zoledronic acid maintains bone mass in postmenopausal
breast cancer patients receiving letrozole. Cancer. 2012;118:
1192-1201.
32. Coleman R, de Boer R, Eidtmann H, et al. Zoledronic acid (zoledronate)
for postmenopausal women with early breast cancer receiving adjuvant
letrozole (ZO-FAST study): final 60-month results. Ann Oncol. 2013;24:
398-405.
33. Eidtmann H, de Boer R, Bundred N, et al. Efficacy of zoledronic acid in
postmenopausal women with early breast cancer receiving adjuvant
letrozole: 36-month results of the ZO-FAST Study. Ann Oncol. 2010;21:
2188-2194.
34. Ellis GK, Bone HG, Chlebowski R, et al. Randomized trial of denosumab in
patients receiving adjuvant aromatase inhibitors for nonmetastatic
breast cancer. J Clin Oncol. 2008;26:4875-4882.
35. Coleman R, Cameron D, Dodwell D, et al; AZURE investigators. Adjuvant
zoledronic acid in patients with early breast cancer: final efficacy
analysis of the AZURE (BIG 01/04) randomised open-label phase 3 trial.
Lancet Oncol. 2014;15:997-1006.
36. Coleman R, Powles T, Paterson A, et al; Early Breast Cancer Trialists
Collaborative Group (EBCTCG). Adjuvant bisphosphonate treatment in
early breast cancer: meta-analyses of individual patient data from
randomised trials. Lancet. 2015;386:1353-1361.
37. Gnant M, Pfeiler G, Dubsky PC, et al; Austrian Breast and Colorectal
Cancer Study Group. Adjuvant denosumab in breast cancer (ABCSG18): a multicentre, randomised, double-blind, placebo-controlled trial.
Lancet. 2015;386:433-443.
38. Curtis RE, Freedman DM, Ron E, et al. New Malignancies among
Cancer Survivors: Seer Cancer Registries, 1973-2000. NIH Publication
No. 05-5302 181-186. Washington, DC: National Cancer Institute;
2006.
39. Zagar TM, Cardinale DM, Marks LB. Breast cancer therapy-associated
cardiovascular disease. Nat Rev Clin Oncol. 2016;13:172-184.
40. Cardinale D, Colombo A, Sandri MT, et al. Prevention of high-dose
chemotherapy-induced cardiotoxicity in high-risk patients by
angiotensin-converting enzyme inhibition. Circulation. 2006;114:
2474-2481.
41. Seicean S, Seicean A, Plana JC, et al. Effect of statin therapy on the risk
for incident heart failure in patients with breast cancer receiving
anthracycline chemotherapy: an observational clinical cohort study.
J Am Coll Cardiol. 2012;60:2384-2390.
42. Pituskin E, Mackey JR, Koshman S, et al. Prophylactic beta blockade
preserves left ventricular ejection fraction in HER2-overexpressing
breast cancer patients receiving trastuzumab: primary results of the
MANTICORE randomized controlled trial. Presented at: San Antonio
Breast Cancer Symposium; December 8-12, 2015; San Antonio, TX.
Abstract S1-05.
43. Chlebowski RT, Aiello E, McTiernan A. Weight loss in breast cancer
patient management. J Clin Oncol. 2002;20:1128-1143.
44. Goodwin PJ, Ennis M, Pritchard KI, et al. Adjuvant treatment and onset
of menopause predict weight gain after breast cancer diagnosis. J Clin
Oncol. 1999;17:120-129.
45. Nichols HB, Trentham-Dietz A, Egan KM, et al. Body mass index before
and after breast cancer diagnosis: associations with all-cause, breast
cancer, and cardiovascular disease mortality. Cancer Epidemiol Biomarkers Prev. 2009;18:1403-1409.
46. Playdon MC, Bracken MB, Sanft TB, et al. Weight gain after breast
cancer diagnosis and all-cause mortality: systematic review and metaanalysis. J Natl Cancer Inst. 2015;107:djv275.
47. Caan BJ, Emond JA, Natarajan L, et al. Post-diagnosis weight gain and
breast cancer recurrence in women with early-stage breast cancer.
48. Jeon YW, Lim ST, Choi HJ, et al. Weight change and its impact on prognosis
after adjuvant TAC (docetaxel-doxorubicin-cyclophosphamide) chemotherapy in Korean women with node-positive breast cancer. Med Oncol.
2014;31:849.
49. Playdon M, Thomas G, Sanft T, et al. Weight loss intervention for breast
cancer survivors: a systematic review. Curr Breast Cancer Rep. 2013;5:
222-246.
50. Djuric Z, DiLaura NM, Jenkins I, et al. Combining weight-loss counseling
with the weight watchers plan for obese breast cancer survivors. Obes
Res. 2002;10:657-665.
51. Greenlee HA, Crew KD, Mata JM, et al. A pilot randomized controlled
trial of a commercial diet and exercise weight loss program in minority
breast cancer survivors. Obesity (Silver Spring). 2013;21:65-76.
52. Befort CA, Klemp JR, Austin HL, et al. Outcomes of a weight loss intervention among rural breast cancer survivors. Breast Cancer Res
Treat. 2012;132:631-639.
53. Rock CL, Flatt SW, Byers TE, et al. Results of the Exercise and Nutrition to
Enhance Recovery and Good Health for You (ENERGY) trial: a behavioral
weight loss intervention in overweight or obese breast cancer survivors.
J Clin Oncol. 2015;33:3169-3176.
54. Chlebowski RT, Blackburn GL, Thomson CA, et al. Dietary fat reduction
and breast cancer outcome: interim efficacy results from the Womens
Intervention Nutrition Study. J Natl Cancer Inst. 2006;98:1767-1776.
55. Pierce JP, Natarajan L, Caan BJ, et al. Influence of a diet very high in
vegetables, fruit, and fiber and low in fat on prognosis following
treatment for breast cancer: the Womens Healthy Eating and Living
(WHEL) randomized trial. JAMA. 2007;298:289-298.
56. Daley AJ, Crank H, Saxton JM, et al. Randomized trial of exercise therapy
in women treated for breast cancer. J Clin Oncol. 2007;25:1713-1721.
57. Holmes MD, Chen WY, Feskanich D, et al. Physical activity and survival
after breast cancer diagnosis. JAMA. 2005;293:2479-2486.
58. Holick CN, Newcomb PA, Trentham-Dietz A, et al. Physical activity and
survival after diagnosis of invasive breast cancer. Cancer Epidemiol
Biomarkers Prev. 2008;17:379-386.
59. Irwin ML, McTiernan A, Manson JE, et al. Physical activity and survival in
postmenopausal women with breast cancer: results from the womens
health initiative. Cancer Prev Res (Phila). 2011;4:522-529.
60. Sternfeld B, Weltzien E, Quesenberry CP Jr, et al. Physical activity and
risk of recurrence and mortality in breast cancer survivors: findings from
the LACE study. Cancer Epidemiol Biomarkers Prev. 2009;18:87-95.
61. Kwan ML, Chen WY, Flatt SW, et al. Postdiagnosis alcohol consumption
and breast cancer prognosis in the after breast cancer pooling project.
Cancer Epidemiol Biomarkers Prev. 2013;22:32-41.
62. Flatt SW, Thomson CA, Gold EB, et al. Low to moderate alcohol intake is
not associated with increased mortality after breast cancer. Cancer
Epidemiol Biomarkers Prev. 2010;19:681-688.
e29
BREAST CANCER
Less Is More: A Multidisciplinary

Conversation on Treatment
Options
CHAIR
Fatima Cardoso, MD
Champalimaud Cancer Centre
Lisbon, Portugal
SPEAKERS
Eun-Sil Shelley Hwang, MD, MPH
Duke University
Durham, NC
Laura Esserman, MD, MBA
San Francisco, CA
TAILORING CHEMOTHERAPY IN EARLY-STAGE BREAST CANCER
Tailoring Chemotherapy in Early-Stage Breast Cancer: Based

on Tumor Biology or Tumor Burden?
Domen Ribnikar, MD, and Fatima Cardoso, MD
OVERVIEW
The question of whether to offer adjuvant chemotherapy to patients with early-stage breast cancer has always been
challenging to answer. It is well known that a substantial proportion of patients with early-stage breast cancer are over
treated, especially when staging and hormonal and HER2 receptors are solely taken into consideration. The advances in our
knowledge of breast cancer biology and its clinical implications were the basis for the discovery of additional reliable
prognostic markers to aid decision making for adjuvant treatment. Gene expression profiling is a molecular tool that more
precisely defines the intrinsic characteristics of each individual tumor. The application of this technology has led to the
development of gene signatures/profiles with relevant prognosticand some predictivevalue that have become important tools in defining which patients with early-stage breast cancer can be safely spared from chemotherapy. However,
the exact clinical utility of these tools will only be determined after the results of two large prospective randomized trials,
MINDACT and TailorX, evaluating their role become available. Notwithstanding the existence of these genomic tools, tumor
burden (defined as tumor size and nodal status) still has independent prognostic value and must be incorporated in decision
making. In addition, these gene signatures have limited predictive value, and new biomarkers and new targets are needed.
Therefore close collaboration between clinicians and scientists is crucial. Lastly, issues of cost-effectiveness, reimbursement,
and availability are crucial and widely variable around the globe.
istorically, breast cancer was viewed as one disease with

different histopathological characteristics and responses
to systemic treatment. The choice of treatment was solely
based on clinicopathologic parameters that are prognostic,
such as tumor size, nodal status, and histologic grade, and the
three predictive markers of response to either endocrine
therapy and/or trastuzumab therapy (estrogen and progesterone receptor expression for endocrine therapy and
HER2 for trastuzumab). These factors are combined with
different algorithms for treatment decision making, such as
those used by Adjuvant! Online1 and the Nottingham
Prognostic Index, and represent the basis of international
treatment guidelines, including those from the National
Comprehensive Cancer Network,2 National Cancer Institute,
and European Society for Medical Oncology 3 as well as
St. Gallens consensus statements.4 The major disadvantage of
this approach is that nearly 60% of all patients with early-stage
breast cancer receive adjuvant chemotherapy, but only 2% to
15% of these patients derive an important benefit and others
only experience its toxic effects.5
Fortunately, breast cancer was one of the first cancers with
advances in molecular profiling technologies, leading to a
deeper understanding of its biology and molecular subtypes.
The advent of genomic signatures allowed us to better

understand the heterogeneity of breast cancer and led to
more individualized systemic treatment approaches. In this
article, we describe traditional clinicopathologic factors and
new molecular tools that may be used to more accurately
tailor adjuvant treatment decisions for patients with earlystage breast cancer. Additionally, we discuss some studies
showing that response to a specific type of treatment is
mainly determined by the intrinsic molecular characteristics
of each individual tumor rather than by anatomic prognostic
parameters. However, these anatomic parameters have
important risk implications and effects on the relative
magnitude of benefit of treatment.
TRADITIONAL FACTORS FOR ADJUVANT

CHEMOTHERAPY DECISION MAKING
The Early Breast Cancer Trialists Collaborative Group (EBCTCG)
meta-analyses (also called Oxford Overview) have demonstrated substantial evidence of the efficacy of adjuvant
chemotherapy in early-stage breast cancer.6,7 The last published results in 2005 report on data from 60 randomized
trials initiated before 1995 comparing polychemotherapy to
no chemotherapy (29,000 patients of whom 10,000 died).7
From the Department of Medical Oncology, Institute of Oncology, Ljubljana, Slovenia; Breast Unit, Champalimaud Cancer Center, Champalimaud Foundation, Lisbon, Portugal.
Corresponding author: Fatima Cardoso, MD, Champalimaud Cancer Center, Av. De Brasilia, Doca De Pedroucos, 1400-048, Lisbon, Portugal; email: fatimacardoso@fundacaochampalimaud.pt.
e31
RIBNIKAR AND CARDOSO
At 15-year follow-up, there was a substantial reduction in

the absolute risk of recurrence and death from breast cancer
for polychemotherapy compared with no chemotherapy.
Among women younger than age 50, polychemotherapy
reduced the annual risk of relapse and death by 37% and
30%, respectively, and the absolute gain in survival was twice
as great at 15 years as it was at 5 years (10% vs. 4.7%). Among
older women (age 5069), at 15-year follow-up, the annual
risk of relapse and breast cancer mortality were reduced by
19% and 12%, respectively, and translated into an absolute
gain of 4.1% and 3%, respectively.7,8 The proportional reductions in both recurrence and breast cancer mortality
were similar in node-negative and node-positive patients,
although the absolute benefit was greater in those with
node-positive disease. Furthermore, it was demonstrated
that the greatest effect of adjuvant chemotherapy occurs in
the initial few years after therapy.
Additional subgroup analyses showed that adjuvant chemotherapy was effective in both estrogen receptornegative
and estrogen receptorpositive disease, albeit the magnitude
of the absolute benefit was larger in estrogen receptor
negative disease. Data from multiple Cancer and Leukemia
Group B (CALGB) trials among women with node-positive
breast cancer reveal greater chemotherapy benefit in estrogen receptornegative disease compared with estrogen
receptorpositive disease.7,9 Substantial reductions were
observed in the absolute risk of recurrence and cancerspecific mortality of 12.3% and 9.2% for patients younger
than age 50 and 8.6% and 6.1% for patients age 5069,
respectively.
KEY POINTS
e32
In the past, breast cancer was perceived to be one

disease with different anatomical extents and different
expressions of estrogen and progesterone receptors and
HER2.
Gene expression analysis has revealed distinct intrinsic
subtypes of breast cancer with different natural
behaviors and responses to treatment.
First-generation gene signatures have the common
ability to further subdivide estrogen receptorpositive
HER2-negative breast cancers into subgroups that differ
in regards to their expression of proliferation-related
genes. Additional biomarkers, particularly those with
predictive value, are urgently needed.
Classic clinicopathologic tools and new molecular
tools should be integrated in and complementary to
adjuvant treatment decision making for individual
patients with breast cancer, in particular for cases where
the benefit of chemotherapy is uncertain.
Treatment of patients with low-risk tumor biology
and high tumor burden is challenging since, despite
being at considerable high risk for distant relapse, the
expected benefit of chemotherapy might not be enough
in terms of the low proliferative biology.
Generally, decisions regarding adjuvant chemotherapy are

based on prognosis, that is, risk of disease recurrence, and on
the likelihood of benefit from the treatment. The threshold
of 10% risk for distant relapse is often used to make recommendation for adjuvant chemotherapy.10
Clinicopathological prognostic factors for survival of earlystage breast cancer include tumor size, nodal status, histologic grade, and estrogen receptor, progesterone receptor, and HER2 expression as well as the presence of
lymphovascular invasion. Several consensus and evidencebased guidelines3,4,11,12 provide recommendations on the
use of adjuvant chemotherapy in early-stage breast cancer.
The St. Gallen guidelines,4 for instance, suggest that patients
with early-stage breast cancer at low enough risk to avoid
adjuvant chemotherapy are characterized by node-negative
disease with all of the following features: pathologic tumor
size of 2 cm or smaller, grade 1, estrogen receptor and progesterone receptor positive, HER2 negative, age older than
35, and no vascular invasion. All other patients who do not
meet these criteria are considered to derive sufficient benefit
from adjuvant chemotherapy because their risk of distant
relapse is considered higher than 10%.
Adjuvant! Online is an independently validated, widely
used web-based tool for assessment of the risk of recurrence
and mortality. It combines multiple clinical and pathologic
factors and produces estimates for recurrence and death
from early-stage breast cancer. It incorporates the effect of
comorbidities in the determination of prognosis and benefit
from different therapeutic strategies.13 However, there are
data suggesting that it generally overestimates prognosis
and it lacks some crucial elements such as HER2 status.
The guidelines described above and Adjuvant! Online cannot reveal the substantial heterogeneity that exists between
patients with similar stages and grades of disease. Several
independent groups have conducted comprehensive gene
expression profiling studies using microarray technology to
develop new tools that are more effective at an individual
level and that can assess risk of relapse to improve decision
making for adjuvant treatment.
INTRINSIC MOLECULAR SUBTYPING

Over 10 years ago, Perou et al14 and Sorlie et al15 demonstrated that estrogen receptorpositive and estrogen
receptornegative breast cancers are completely distinct
diseases on a molecular level. In addition, hierarchical
cluster analyses of genes that vary more between tumors
than between repeated samples of the same tumor, the socalled intrinsic genes, revealed at least four molecular
subtypes of breast cancer, namely, luminal, HER2-enriched,
basal-like, and normal breastlike.14 Multiple independent
datasets have shown the prognostic effect of intrinsic subtyping of breast cancer with luminal A cases (high expression of
estrogen receptor and estrogen receptorregulated genes) to
be at low risk of early recurrence. Apart from its prognostic
implication, the intrinsic molecular subclassification seems to
predict responsiveness to chemotherapy.16 Studies evaluating
the association between molecular subtype and pathologic complete response (pCR) rate to preoperative
(neoadjuvant) chemotherapy have reported the highest
rate of pCR in basal-like (30%45%) and HER2-enriched
(33%55%) subtypes, while pCR rates for luminal B
(1%15%) and luminal A disease (0%7%) were substantially
lower.
It is important to highlight that molecular classification is not without its inherent limitations because up
to 30% of breast cancers do not fit into any of the four
molecular categories. 17 The exact number of true molecular subclasses of breast cancer is currently unknown,
and it is expected that the molecular classification will
further evolve with new technological platforms and
improved understanding of tumor biology. In particular,
the immunohistochemistry-defined triple-negative subtype is currently being subdivided into several molecularly
distinct subtypes with potential future clinical and therapeutic implications.18
PROGNOSTIC GENE EXPRESSION SIGNATURES

Concomitant to the discovery of heterogeneity of breast
cancers, microarray-based gene expression profiling was
used to predict the outcomes of individual patients with
breast cancer and aimed to identify those patients who
could be safely omitted from adjuvant chemotherapy.19 First
attempts were empirical, whereby tumors from patients
with good and poor outcomes were profiled, and collections
of candidate genes that could discriminate between patients
with disease of good and poor prognosis were identified.19,20
Later, studies in which a multigene predictor was generated
on the basis of a hypothesis derived from in vivo or in vitro
experiments and then applied to breast cancer samples were
conducted.21
There are many gene-expression prognostic signatures
that were used during the past decade.22 The so-called firstgeneration signatures have several common features. Despite the different gene sets that compose each of the
signatures, they characterize a unique population with a high
expression of proliferation-related genes associated with
poor prognosis.23 Because the levels of expression of
proliferation-related genes are usually high in estrogen
receptornegative disease, these signatures almost always
classify estrogen receptornegative cancers as being a poor
prognosis disease. In estrogen receptorpositive breast cancer, the strongest prognostic factor is the degree of expression of proliferation-related genes. Based on these
features, it seems that most of these signatures generally
correlate with response to conventional chemotherapy
regimens,24 that is, high proliferative tumors respond better
to chemotherapy; however, they are not predictive of one
type of chemotherapy agent over another and cannot be
used to define the best chemotherapy agent/regimen for
each individual patient.
In the early years of their development, some defended
that these signatures would be a more objective replacement
of common clinicopathologic features traditionally used

for determining prognosis of early-stage breast cancer;
important meta-analyses have shown that they complement but do not replace traditional factors since both
tumor size and nodal status provide independent prognostic information in multivariate analyses.25 In addition,
the accuracy of the outcome prediction of all firstgeneration prognostic signatures is time-dependent,
with more accurate prediction during the first 5 years
after diagnosis and less accurate predictions for late
relapses.26 Finally, there is some evidence that the
prognostic information provided by first-generation
signatures may be comparable with that offered by
semiquantitative and centralized (high-quality) assessment of estrogen and progesterone receptors, HER2, and
Ki-67.27
70-Gene Profile
The 70-gene profile, also known as MammaPrint, was the
first microarray-based prognostic signature to be approved
by the U.S. Food and Drug Administration. The 70-gene
profile may be used for determining the prognosis of patients with stage I and/or II, node-negative, invasive breast
cancer28 as well as for nodes 1 to 3 positive disease.29 This
gene profile was first established using RNA from fresh,
frozen tumor tissues; however, from 2012 onward, it is also
evaluable in formalin-fixed, paraffin-embedded tumor
tissue.28 The assay was based on an empirical microarray
analysis of 78 patients who did not receive adjuvant systemic
therapy for their breast cancer. They were younger than age
55, had tumors up to 5 cm, and were node-negative. If these
patients developed distant metastases within the first 5
years after primary diagnosis, they were classified as poor
prognosis or good prognosis in the event there were no
distant metastases during this timeframe. A list of 70 genes
that could accurately predict poor versus good prognosis for
these patients was identified by a supervised analysis of
25,000 genes included in the microarrays. Subsequent studies
in breast cancer cohorts that were retrospectively accrued
demonstrated the potential of the MammaPrint assay to
determine prognosis of both node-negative and nodepositive disease29 and also for patients with HER2-positive
disease.30 The prognostic subgroups, which were identified
by the MammaPrint assay, correlate sensitivity to chemotherapy in general: patients with high-risk signatures
derive the greatest absolute benefit from adjuvant chemotherapy.31 However, it has no ability to differentiate sensitivity to different types of chemotherapy regimens. One of
main disadvantages of the MammaPrint assay is its small
discriminatory power for patients with estrogen receptor
negative disease (only up to 5% of patients with estrogen
receptornegative disease are classified as having good
prognosis disease).32 The true clinical utility of MammaPrint is
being tested in a randomized prospective phase III trial EORTC
10041/BIG 3-04 MINDACT,33 the results of which will be
presented in April 2016.
e33
Recurrence Score or Oncotype DX
76-Gene Signature
In parallel with the development of microarray-based prognostic signatures, Paik and et al34 developed Oncotype DX, a
quantitative reverse transcription polymerase chain reaction
based signature that measures the expression of 21 genes
(16 cancer-related genes and five reference genes). It can
be performed with RNA extracted from formalin-fixed,
paraffin-embedded tissue samples. Oncotype DX calculates the recurrence score (RS) to predict the risk of distant
relapse within 10 years for patients with estrogen receptor
positive, lymph nodenegative breast cancers. It was developed on the basis of analysis of clinical samples from the
National Surgical Adjuvant Breast and Bowel Project (NSABP)
B-20 clinical trial. Later, it was validated with material from
the B-14 trial. The RS is a continuous variable, ranging
from 0 to 100, and is an independent prognostic factor for
patients with estrogen receptorpositive, node-negative
breast cancer treated with adjuvant tamoxifen. Patients
are classified into three categories, including low risk
(RS , 18), intermediate risk (RS 18-31), and high risk (RS . 31),
which correlate with 10-year relapse rates of 7%, 14%,
and 30%, respectively. The optimal management of the
intermediate-risk group is uncertain and is being studied
in the TailorX trial, in which patients with estrogen
receptorpositive, node-negative breast cancer were assessed
for risk of distant relapse after surgery and assigned
to low-risk (RS , 11), intermediate-risk (RS 11-25), and
high-risk groups (RS . 25). Patients with intermediate-risk
disease were randomly assigned to either endocrine
therapy alone or in combination with chemotherapy. In
2015, the results of the low-risk patients (RS , 11) were
presented and demonstrated excellent outcome at
5 years with endocrine therapy alone, irrespective of age,
tumor size, and grade. The authors concluded that these
patients can be safely treated without chemotherapy.35
The main results of the intermediate-risk score group are
eagerly awaited.
Further analyses also demonstrated that RS generally
correlates with benefit from chemotherapy in estrogen
receptorpositive disease.36 However, it does not provide
predictive value to differentiate between different chemotherapy agents/regimens. It is also not useful to discriminate between subgroups of HER2-positive breast
cancer because HER2 is one of the 21 genes that constitute
the test.
During the last few years, the Oncotype DX assay was
validated among patients who had up to three positive
lymph nodes37 and for those who are supposed to be treated
with an aromatase inhibitors.38
Although the prognostic effect of the Oncotype DX assay
has been extensively validated, there is some evidence
that RS correlates with some clinicopathologic parameters, and, some of them, such as tumor size, nodal status,
and even histologic grade, remain independent of RS.39,40
Hence, a model that combines RS with traditional anatomic pathologic factors may be more prognostic than
RS alone.41
The 76-gene signature, also known as the Veridex signature,

was developed on the basis of a supervised analysis of
microarray data in a set of 115 breast cancers, of which 80%
were estrogen receptorpositive. Contrary to MammaPrint,
estrogen receptorpositive and estrogen receptornegative
cancers were analyzed separately, and this enabled the
identification of 60 genes that could predict the development of distant metastases within 5 years among patients
with estrogen receptorpositive disease and 16 genes that
could predict distant dissemination in estrogen receptor
negative disease.42 All 76 genes together were applied to an
independent test set of 171 patients with node-negative
disease only. The 76-gene predictor was a strong prognostic
factor for development of metastases within the first 5 years
after diagnosis. The main disadvantages of this signature are
time dependency, requirement of fresh or frozen samples,
prognostic power supported by level III evidence only, and,
last but not the least, that the 16-gene signature developed
for estrogen receptornegative disease might not predict
the outcome of patients with triple-negative breast
cancer.23
e34
Genomic Grade Index

In 2006, Sotiriou et al developed the Genomic Grade Index
(GGI). It was based on the expression of 97 genes that were
selected by comparing the gene expression of histologic
grade 2 tumors with that of grade 1 tumors by means of a
DNA microarray.43 Importantly, this signature was able to
divide estrogen receptorpositive grade 2 cancers into grade
1-like with a low frequency of distant relapse and grade
3-like with an aggressive clinical behavior that is similar to
that of histologic grade 3 tumors. Similar to MammaPrint,
GGI also correlates with the benefit from chemotherapy in
general: GGI grade 3 cancers seem to derive higher absolute
benefit from conventional chemotherapy versus GGI grade 1
tumors.44 The main common limitations of first generation
prognostic signatures also apply to GGI.22
PAM 50
PAM 50, also known as Prosigna, was originally developed
for intrinsic subtyping of breast cancer,45 but later it started
to be used to predict recurrence.46 It was developed for
patients receiving adjuvant tamoxifen. Its score, PAM 50
ROR (PAM 50 risk of recurrence), is calculated by using the
expression profile of 50 selected genes from four intrinsic subtypes, a proliferation score (18-gene subset), and
pathologic tumor size47; it classifies patients into those at
low, intermediate, and high risk of recurrence, but it also
provides the score as a continuous variable. It can be used
with formalin-fixed, paraffin-embedded tissue samples and
was approved by the U.S. Food and Drug Administration in
2013. An important finding was the ability of the PAM 50
ROR score to effectively divide patients into the three risk
groups according to risk for recurrence between 5 and 15
years after the primary diagnosis, using the samples from the
Austrian Breast and Colorectal Cancer Study Group 8 trial

(ABCSG-8). It is, therefore, a more useful signature to
predict late distant relapses compared with Oncotype DX
or MammaPrint.
Breast Cancer Index

The Breast Cancer Index (BCI) is another second-generation
gene signature that was found to be a prognostic marker for
patients with early-stage estrogen receptorpositive breast
cancer who have or who have not received tamoxifen. It
is a quantitative reverse transcription polymerase chain
reactionbased method that measures the expression of
two genes, HOXB13 and IL17BR (H/I), and classifies patients
into three recurrence risk categories (low, intermediate, and
high).48 The BCI was found to be useful for the assessment of
both early and late distant recurrence, according to the
Stockholm study. This study included 317 patients with
estrogen receptorpositive node-negative disease, treated
with adjuvant tamoxifen only, and showed that recurrence
rates for the low-risk group based on the BCI were very low,
not only in the first 5 years after the surgery, but also in years
5 to 10.
EndoPredict
The most recently developed second-generation gene signature is EndoPredict. Just like the majority of first- and
second-generation signatures, it is also used for estrogen
receptorpositive and HER2-negative breast cancers. It was
developed based on the analysis of 964 breast cancer
samples, from which eight genes and four control genes
were selected. The expression of these 12 genes is analyzed
with the quantitative reverse transcription polymerase chain
reactionbased method for the classification of patients into
two recurrence risk groups.49 The assay was validated in the
ABCSG-6 and ABCSG-8 trials that included patients with
estrogen receptorpositive, HER2-negative, lymph node
positive or negative breast cancers who were treated with
adjuvant tamoxifen only.50 Additionally, EndoPredictClin,
which combines the EndoPredict score with tumor size and
nodal status in a linear model, identified a subgroup of
patients with an excellent long-term prognosis after a
standard 5 years of endocrine therapy,51 thus making it
useful for evaluating risk of late relapse.
CONCORDANCE AMONG GENE

EXPRESSIONBASED PREDICTORS FOR BREAST
CANCER
Fan et al analyzed a single data set of 295 tumor samples and
applied five gene expressionbased models including intrinsic subtypes, MammaPrint assay, wound response,
Oncotype DX, and the two-gene ratio (BCI) for patients who
had been treated with tamoxifen.52 Four of these models
were similar in prediction for particular risk groups. For
instance, tumors classified as HER2-enriched, basal-like, and
luminal Blike by intrinsic subtyping were almost all classified as having a poor outcome by MammaPrint, Oncotype
DX, and wound response models. Of the five models that Fan
et al analyzed in their study, only the two-gene ratio (H/I)
failed to identify substantial differences in outcome within
that data set.
There are two additional reports evaluating the accuracy of prediction of BCI among patients with estrogen
receptorpositive, node-negative breast cancer. One report
by Reid et al showed that the two-gene model failed to
detect differences in outcome,53 and the other study,
conducted by Goetz et al, showed that the H/I ratio was a
substantial predictor of relapse-free survival and diseasefree survival.54 The possible explanation for these contradictory findings is that a model based on the analysis of only
two genes is much more likely to be sensitive to technical
differences in analysis platforms than one based on many
genes.
Findings from multiple analyses incorporating first- and
second-generation gene signatures among patients with
early-stage breast cancer suggest that even though there is
very little gene overlap and different algorithms are used,
the prediction of outcome for the majority of patients is
similar. One may therefore conclude that, even though
different gene sets are being used as predictors of outcome,
the signatures identify a common set of biologic characteristics that allow for a good distinction between different
subgroups of patients with breast cancer who have distinct
prognoses.
GENE EXPRESSION SIGNATURES AND RESPONSE

TO CHEMOTHERAPY
The predictive value of genomic signatures has been evaluated mostly in the neoadjuvant setting. Several small
studies suggested that the gene expression profiles of
cancers that are highly sensitive to chemotherapy differ
from those of less responsive tumors. For example, Ayers
et al conducted a prospectively designed study in which
needle-biopsy samples from 133 patients with stage I, II, and
III breast cancers were collected before systemic therapy.55
Patients were then treated with preoperative chemotherapy
with weekly paclitaxel and a combination of fluorouracil,
doxorubicin, and cyclophosphamide. A 30-gene predictor
was developed from the data from the first 82 patients, and
it was shown to have a higher sensitivity for prediction of
pCR than a clinical predictor that included age, nuclear
grade, and estrogen receptor status (92% vs. 61%). These
results need further validation and confirmation in independent and larger studies.
Currently no multigene predictor is ready to be used in
clinical practice with the aim of predicting response to
specific chemotherapy agents or regimens.
CONCLUSION
Undoubtedly, the era of molecular oncology has brought a
deeper insight into the complex biology of breast cancer
and its crucial clinical and therapeutic implications. A new
molecular-based classification of breast cancer exists, going
e35
TABLE 1. Most Common Commercially Available Prognostic Gene Signatures for Breast Cancer
MammaPrint
Oncotype DX
Breast Cancer Index
Mapquant DX
PAM 50 ROR
EndoPredict
Provider
Agendia
Genomic Health
Biotheranostics
Ipsogen
NanoString
Sividon
Type of
Assay
70-gene assay
21-gene
recurrence score
2-gene ratio (H/I)

and molecular
grade index
Genomic grade
50-gene assay
12-gene assay
Type of
Sample
Fresh or frozen
or FFPE
FFPE
FFPE
Fresh or frozen
or FFPE
FFPE
FFPE
Technique
DNA microarray
or qRT-PCR
qRT-PCR
qRT-PCR
DNA microarray
or qRT-PCR
qRT-PCR
qRT-PCR
Clinical
Application
Prognosis of
N0, , 5 cm,
stage I/II,
age , 61
Prediction of
recurrence risk
in ER+ and N0
treated with TAM
Prognostic in ER+,
prediction of
response to TAM
Molecular
grading for
ER+, histologic
grade II disease
Originally for
intrinsic subtyping,
recurrence prediction
Recurrence
prediction for
ER+ HER2
Results
Presentation
Dichotomous,
good or poor
prognosis
Continuous variable
Continuous variable
Dichotomous,
GGI I or GGI III
Continuous variable
Dichotomous, low
or high risk
Level of
Evidence
II
III
III
FDA Approval
YES
NO
NO
NO
YES
NO
Abbreviations: ER+, estrogen receptorpositive; FDA, U.S. Food and Drug Administration; FFPE, formalin-fixed, paraffin-embedded; GGI, Genomic Grade Index; qRT-PCR, quantitative
reverse transcription polymerase chain reaction; TAM, tamoxifen.
beyond traditional morphologic oncology. It is, however,

indispensable to use the new molecular tools in conjunction
with classic clinicopathologic parameters because the latter
retain their independent prognostic value for early-stage
breast cancer in multivariate models.
All available tools have advantages and limitations, both of
which must be discussed with patients. First-generation
prognostic gene signatures consistently demonstrate that
about half of the patients with estrogen receptorpositive,
HER2-negative disease are at low risk for early distant relapse and are likely to obtain minimal, if any, benefit from
adjuvant chemotherapy, but do not provide information
about late relapses. Novel second-generation multigene
prognosticators have some additional advantages, such as a
better prediction of late distant relapses, a common situation
in estrogen receptorpositive, HER2-negative breast cancer.
Combination tools, such as EndoPredictClin and Prosigna, may
be even more accurate in determining short- and long-term
prognoses for this subgroup of breast cancer. Therefore,
tumor biology and tumor burden must be complementary
and not mutually exclusive, so that more individualized adjuvant treatment decision making can occur (Table 1).
While we wait for the results of the large prospective randomized trials, MINDACT, TailorX, RxPONDER, and OPTIMA,
all major international guidelines recommend the use of
genomic tools in cases for which the potential benefit of
adjuvant chemotherapy is difficult to estimate based only
using on clinicopathologic factors.
Currently, one of the greatest challenges for the medical
oncologist is determining the treatment strategy for patients
with favorable tumor biology and high tumor burden. These
patients are at substantial risk for distant relapse with endocrine therapy alone.37,56 However, they do not seem to
derive a substantial benefit from adjuvant chemotherapy
and are at risk for the acute and long-term toxicities of this
therapy.
In summary, adjuvant treatment decision making in
breast cancer involves an integration of the available
clinicopathologic factors and new genomic tools, whenever appropriate, as well as a detailed and comprehensible discussion with each individual patient regarding
risk of recurrence, risk of adverse events, comorbidities,
performance status, and, very importantly, patient
preferences.
References
1. Ravdin PM, Siminoff LA, Davis GJ, et al. Computer program to assist in
making decisions about adjuvant therapy for women with early breast
cancer. J Clin Oncol. 2001;19:980-991.
2. Carlson RW, Anderson BO, Burstein HJ, et al. Invasive breast cancer.
J Natl Compr Canc Netw. 2007;5:246-312.
3. Senkus E, Kyriakides S, Ohno S, et al; ESMO Guidelines Committee.
Primary breast cancer: ESMO Clinical Practice Guidelines for diagnosis,
treatment and follow-up. Ann Oncol. 2015;26(Suppl 5):v8-v30.
e36
4. Jackisch C, Harbeck N, Huober J, et al. 14th St. Gallen International

Breast Cancer Conference 2015: evidence, controversies, consensus primary therapy of early breast cancer: opinions expressed by German
experts. Breast Care (Basel). 2015;10:211-219.
5. Early Breast Cancer Trialists Collaborative Group. Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence
and 15-year survival: an overview of the randomized trials. Lancet.
2005;365:1687-1717.
6. Early Breast Cancer Trialists Collaborative Group. Polychemotherapy for

early breast cancer: an overview of the randomised trials. Lancet. 1998;
352:930-942.
7. Early Breast Cancer Trialists Collaborative Group. Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence
and 15-year survival: an overview of the randomised trials. Lancet.
2005;365:1687-1717.
8. Clarke M. Meta-analyses of adjuvant therapies for women with early
breast cancer: the Early Breast Cancer Trialists Collaborative Group
overview. Ann Oncol. 2006;17(Suppl 10):x59-x62.
9. Berry DA, Cirrincione C, Henderson IC, et al. Estrogen-receptor status
and outcomes of modern chemotherapy for patients with node-positive
breast cancer. JAMA. 2006;295:1658-1667.
10. Kelly CM, Hortobagyi GN. Adjuvant chemotherapy in early-stage breast
cancer: what, when, and for whom? Surg Oncol Clin N Am. 2010;19:
649-668.
11. National Comprehensive Cancer Network. NCCN Clinical Practice
Guidelines in Oncology. Breast Cancer, V.I. 2010. http://www.nccn.org/
professionals/physician_gls/PDF/breast.pdf. Accessed January 19, 2010.
12. National Institutes of Health Consensus Development Panel. Adjuvant
therapy for breast cancer. NIH Consens Statement. 2000;17:1-35.
13. Olivotto IA, Bajdik CD, Ravdin PM, et al. Population-based validation of
the prognostic model ADJUVANT! for early breast cancer. J Clin Oncol.
2005;23:2716-2725.
14. Perou CM, Srlie T, Eisen MB, et al. Molecular portraits of human breast
tumours. Nature. 2000;406:747-752.
15. Srlie T, Perou CM, Tibshirani R, et al. Gene expression patterns of
breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci USA. 2001;98:10869-10874.
16. Hu Z, Fan C, Oh DS, et al. The molecular portraits of breast tumors are
conserved across microarray platforms. BMC Genomics. 2006;7:96.
17. Pusztai L, Mazouni C, Anderson K, et al. Molecular classification of
breast cancer: limitations and potential. Oncologist. 2006;11:868-877.
18. Curtis C, Shah PS, Chin SF, et al. The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups. Nature.
2012;486:346-352.
19. Weigelt B, Baehner FL, Reis-Filho JS. The contribution of gene expression profiling to breast cancer classification, prognostication and
prediction: a retrospective of the last decade. J Pathol. 2010;220:
263-280.
20. Van;t Veer LJ, Dai H, van de Vijver MJ, et al. Gene expression profiling
predicts clinical outcome of breast cancer. Nature. 2002;415:530536.
21. Kim C, Paik S. Gene-expression-based prognostic assays for breast
cancer. Nat Rev Clin Oncol. 2010;7:340-347.
22. Sotiriou C, Pusztai L. Gene-expression signatures in breast cancer.
N Engl J Med. 2009;360:790-800.
23. Wirapati P, Sotiriou C, Kunkel S, et al. Meta-analysis of gene expression
profiles in breast cancer: toward a unified understanding of breast
cancer subtyping and prognosis signatures. Breast Cancer Res. 2008;10:
R65.
24. Iwamoto T, Bianchini G, Booser D, et al. Gene pathways associated with
prognosis and chemotherapy sensitivity in molecular subtypes of breast
cancer. J Natl Cancer Inst. 2011;103:264-272.
25. Desmedt C, Haibe-Kains B, Wirapati P, et al. Biological processes associated with breast cancer clinical outcome depend on the molecular
subtypes. Clin Cancer Res. 2008;14:5158-5165.
26. Desmedt C, Piette F, Loi S, et al; TRANSBIG Consortium. Strong time
dependence of the 76-gene prognostic signature for node-negative
breast cancer patients in the TRANSBIG multicenter independent
validation series. Clin Cancer Res. 2007;13:3207-3214.
27. Cuzick J, Dowsett M, Pineda S, et al. Prognostic value of a combined

estrogen receptor, progesterone receptor, Ki-67 and human epidermal
growth factor receptor 2 immunohistochemical score and comparison
with the genomic health recurrence score in early breast cancer. J Clin
Oncol. 2011;29:4273-4278.
28. Sapino A, Roepman P, Linn SC, et al. MammaPrint molecular diagnostics
on formalin-fixed, paraffin-embedded tissue. J Mol Diagn. 2014;16:
190-197.
29. Mook S, Schmidt MK, Viale G, et al; TRANSBIG Consortium. The 70-gene
prognosis-signature predicts disease outcome in breast cancer patients
with 1-3 positive lymph nodes in an independent validation study.
30. Knauer M, Cardoso F, Wesseling J, et al. Identification of a low-risk
subgroup of HER-2-positive breast cancer by the 70-gene prognosis
signature. Br J Cancer. 2010;103:1788-1793.
31. Knauer M, Mook S, Rutgers EJ, et al. The predictive value of the 70-gene
signature for adjuvant chemotherapy in early breast cancer. Breast
32. Buyse M, Loi S, vant Veer L, et al; TRANSBIG Consortium. Validation and
clinical utility of a 70-gene prognostic signature for women with nodenegative breast cancer. J Natl Cancer Inst. 2006;98:1183-1192.
33. Bogaerts J, Cardoso F, Buyse M, et al; TRANSBIG consortium. Gene
signature evaluation as a prognostic tool: challenges in the design of the
MINDACT trial. Nat Clin Pract Oncol. 2006;3:540-551.
34. Paik S, Shak S, Tang G, et al. A multigene assay to predict recurrence of
tamoxifen-treated, node-negative breast cancer. N Engl J Med. 2004;
351:2817-2826.
35. Sparano JA, Gray RJ, Makower DF, et al. Prospective validation of a 21gene expression assay in breast cancer. N Engl J Med. 2015;373:20052014.
36. Paik S, Tang G, Shak S, et al. Gene expression and benefit of chemotherapy in women with node-negative, estrogen receptor-positive
37. Albain KS, Barlow WE, Shak S, et al; Breast Cancer Intergroup of North
America. Prognostic and predictive value of the 21-gene recurrence
score assay in postmenopausal women with node-positive, oestrogenreceptor-positive breast cancer on chemotherapy: a retrospective
analysis of a randomised trial. Lancet Oncol. 2010;11:55-65.
38. Dowsett M, Cuzick J, Wale C, et al. Prediction of risk of distant recurrence using the 21-gene recurrence score in node-negative and
node-positive postmenopausal patients with breast cancer treated with
anastrozole or tamoxifen: a TransATAC study. J Clin Oncol. 2010;28:
1829-1834.
39. Habel LA, Shak S, Jacobs MK, et al. A population-based study of tumor
gene expression and risk of breast cancer death among lymph nodenegative patients. Breast Cancer Res. 2006;8:R25.
40. Tang G, Shak S, Paik S, et al. Comparison of the prognostic and predictive
utilities of the 21-gene Recurrence Score assay and Adjuvant! for
women with node-negative, ER-positive breast cancer: results from
NSABP B-14 and NSABP B-20. Breast Cancer Res Treat. 2011;127:
133-142.
41. Tang G, Cuzick J, Wale C, et al. Recurrence risk of node-negative and ERpositive early-stage breast cancer patients by combining recurrence
score, pathologic and clinical information: a meta-analysis approach. J
Clin Oncol. 2010;28 (suppl; abstr 509).
42. Wang Y, Klijn JG, Zhang Y, et al. Gene-expression profiles to predict
distant metastasis of lymph-node-negative primary breast cancer.
Lancet. 2005;365:671-679.
43. Sotiriou C, Wirapati P, Loi S, et al. Gene expression profiling in breast
cancer: understanding the molecular basis of histologic grade to improve prognosis. J Natl Cancer Inst. 2006;98:262-272.
e37
44. Liedtke C, Hatzis C, Symmans WF, et al. Genomic grade index is associated with response to chemotherapy in patients with breast cancer.
J Clin Oncol. 2009;27:3185-3191.
45. Parker JS, Mullins M, Cheang MC, et al. Supervised risk predictor of breast
cancer based on intrinsic subtypes. J Clin Oncol. 2009;27:1160-1167.
46. Gnant M, Filipits M, Greil R, et al; Austrian Breast and Colorectal Cancer
Study Group. Predicting distant recurrence in receptor-positive breast
cancer patients with limited clinicopathological risk: using the PAM50
Risk of Recurrence score in 1478 postmenopausal patients of the
ABCSG-8 trial treated with adjuvant endocrine therapy alone. Ann
Oncol. 2014;25:339-345.
47. Filipits M, Nielsen TO, Rudas M, et al; Austrian Breast and Colorectal
Cancer Study Group. The PAM50 risk-of-recurrence score predicts risk
for late distant recurrence after endocrine therapy in postmenopausal
women with endocrine-responsive early breast cancer. Clin Cancer Res.
2014;20:1298-1305.
48. Zhang Y, Schnabel CA, Schroeder BE, et al. Breast cancer index identifies
early-stage estrogen receptor-positive breast cancer patients at risk for
early- and late-distant recurrence. Clin Cancer Res. 2013;19:4196-4205.
49. Filipits M, Rudas M, Jakesz R, et al; EP Investigators. A new molecular
predictor of distant recurrence in ER-positive, HER2-negative breast
cancer adds independent information to conventional clinical risk
factors. Clin Cancer Res. 2011;17:6012-6020.
50. Dubsky PC, Jakesz R, Mlineritsch B, et al. Tamoxifen and anastrozole as a
sequencing strategy: a randomized controlled trial in postmenopausal
e38
51.
52.
53.
54.
55.
56.
patients with endocrine-responsive early breast cancer from the

Austrian Breast and Colorectal Cancer Study Group. J Clin Oncol. 2012;
30:722-728.
Dubsky P, Brase JC, Jakesz R, et al; Austrian Breast and Colorectal Cancer
Study Group (ABCSG). The EndoPredict score provides prognostic information on late distant metastases in ER+/HER2- breast cancer patients. Br J Cancer. 2013;109:2959-2964.
Fan C, Oh DS, Wessels L, et al. Concordance among gene-expressionbased predictors for breast cancer. N Engl J Med. 2006;355:560-569.
Reid JF, Lusa L, De Cecco L, et al. Limits of predictive models using
microarray data for breast cancer clinical treatment outcome. J Natl
Cancer Inst. 2005;97:927-930.
Goetz MP, Suman VJ, Ingle JN, et al. A two-gene expression ratio of
homeobox 13 and interleukin-17B receptor for prediction of recurrence
and survival in women receiving adjuvant tamoxifen. Clin Cancer Res.
2006;12:2080-2087.
Ayers M, Symmans WF, Stec J, et al. Gene expression profiles predict
complete pathologic response to neoadjuvant paclitaxel and fluorouracil, doxorubicin, and cyclophosphamide chemotherapy in breast
Saghatchian M, Mook S, Pruneri G, et al. Combining genomic profiling
(70-gene MammaPrint) with nodal status allows to classify patients
with primary breast cancer and positive lymph nodes (1-9) into very
distinct prognostic subgroups that could help tailor treatment strategies. Cancer Res. 2009;69:a102.
BREAST CANCER
Targeted Therapies in Hormone

ReceptorPositive Breast Cancer
CHAIR
Hope S. Rugo, MD
San Francisco, CA
SPEAKERS
Cynthia Ma, MD, PhD
Washington University School of Medicine
St. Louis, MO
Dejan Juric, MD
Boston, MA
RUGO, VIDULA, AND MA
Improving Response to Hormone Therapy in Breast Cancer:

New Targets, New Therapeutic Options
Hope S. Rugo, MD, Neelima Vidula, MD, and Cynthia Ma, MD, PhD
OVERVIEW
The majority of breast cancer expresses the estrogen and or progesterone receptors (ER and PR). In tumors without
concomitant HER2 amplification, hormone therapy is a major treatment option for all disease stages. Resistance to
hormonal therapy is associated with disease recurrence and progression. Recent studies have identified a number of
resistance mechanisms leading to estrogen-independent growth of hormone receptorpositive (HR+) breast cancer as a
result of genetic and epigenetic alterations, which could be exploited as novel therapeutic targets. These include acquired
mutations in ER-alpha (ESR1) in response to endocrine deprivation; constitutive activation of cyclin-dependent kinases (CDK)
4 and 6; cross talk between ER and growth factor receptor signaling such as HER family members, fibroblast growth factor
receptor (FGFR) pathways, intracellular growth, and survival signals PI3K/Akt/mTOR; and epigenetic modifications by
histone deacetylase (HDAC) as well as interactions with tumor microenvironment and host immune response. Inhibitors of
these pathways are being developed to improve efficacy of hormonal therapy for treatment of both metastatic and earlystage disease. Two agents are currently approved in the United States for the treatment of metastatic HR+ breast cancer,
including the mTOR inhibitor everolimus and the CDK4/6 inhibitor palbociclib. Management of toxicity is a critical aspect of
treatment; the primary toxicity of everolimus is stomatitis (treated with topical steroids) and of palbociclib is neutropenia
(treated with dose reduction/delay). Many agents are in clinical trials, primarily in combination with hormone therapy; novel
combinations are under active investigation.
he estrogen-dependent nature of breast cancer has been

well established since the historical observation that
removal of the ovaries was effective in treating breast tumors.1
A number of pharmacological agents, referred to as endocrine
therapy, are now available in the clinic. These include aromatase inhibitors (AI) and gonadotropin-releasing hormone
(GnRH) agonists that reduce estrogen biosynthesis, the
selective estrogen receptor modulators (SERM) tamoxifen,
and the selective ER down regulator (SERD) fulvestrant. The
clinical application of these agents has led to substantial
improvements in survival outcomes for patients with ER+
breast cancer.2 However, despite standard therapy, over
20% of patients with early-stage disease experience relapse,
and, essentially, all patients with metastatic disease succumb to their illness.2,3 Although the mechanisms of endocrine resistance are not fully understood, significant
progress has been made in recent years in uncovering
several key molecular pathways that promote ligandindependent activation of ER and tumor growth. In this
article, we will focus on evidence for ER-alpha (ESR1) mutation, growth factor receptor signaling, PI3K/Akt/mTOR,
CDK4/6, and epigenetic and immune checkpoints as resistance mechanisms and therapeutic targets in ER+ breast
cancer (Fig. 1).
UNDERSTANDING THE BIOLOGY OF HORMONE

RECEPTORS AND POTENTIAL MECHANISMS OF
RESISTANCE
Mechanisms of Action of Estrogen Receptors
Estrogen receptors belong to the family of steroid hormone
receptors and its main mechanism of action is a transcription
factor (Fig. 2). There are two different forms of ER, ER-alpha
and ER-beta, encoded by ESR1 and ESR2, respectively. There
is substantial sequence homology between ER-alpha and
ER-beta, although the function of ER-beta is less known.
Upon activation by estrogen, the cytosolic ER forms dimers
and translocates to the nucleus where ER binds directly to
the estrogen response elements (ERE) or is tethered to
the promoter regions of target genes via its interaction
with SP1. 4 Additional coregulators are then recruited
to the complex to regulate the transcription levels of
target genes, including cyclin D, to promote cell cycle
From the UCSF Helen Diller Family Comprehensive Cancer Center, San Francisco, CA; University of San Francisco School of Medicine, San Francisco, CA; Department of Medicine,
Washington University School of Medicine in St. Louis, St. Louis, MO.
Corresponding author: Hope S. Rugo, MD, UCSF Helen Diller Family Comprehensive Cancer Center, 1600 Divisadero St., San Francisco, CA 94115; email: hope.rugo@ucsf.edu.
e40
IMPROVING RESPONSE TO HORMONE THERAPY
FIGURE 1. Key Pathways of Endocrine Resistance
Pink and blue color code activating and suppression signals, respectively, for pathway activation.
Abbreviations: HDAC, histone deacetylase; RTK, receptor tyrosine kinase.
progression.4,5 In addition to this classic genomic action of

ER, growth factor receptor tyrosine kinases and a variety of
proximal signaling molecules, such as G protein, Ras, Src,
PI3K, and Shc, could activate ER in the absence of estrogen,
leading to estrogen-independent cell proliferation and resistance

to endocrine therapy.6-10
ESR1 Mutation
KEY POINTS
Resistance to endocrine therapy in HR+ breast cancer is

associated with diverse molecular mechanisms,
including acquired mutations in ESR1 in response to
endocrine deprivation, constitutive activation of CDK4/6,
cross talk between the ER and growth factor receptor
signaling, epigenetic modifications by HDAC, and
interactions with tumor microenvironment and host
immune response.
Increased understanding of endocrine resistance
mechanisms has led to the development of targeted
agents, including two agents which are approved for
clinical use in the United States, that have improved
progression-free survival in combination with standard
hormone agents.
PI3K is the most commonly altered pathway in HR+
breast cancer; multiple agents that target this pathway
or related growth factor receptors are being studied. To
date, markers of PI3K pathway activation have not
identified a specific population most likely to benefit
from this therapy.
Toxicities of targeted therapies and impact on quality of
life must be taken into account and managed
expectantly when treating patients; common toxicities
are stomatitis (everolimus) and neutropenia
(palbociclib).
Ongoing trials are evaluating inhibitors of the PI3K
pathway, CDK4/6, and HDAC, primarily in combination
with hormone therapy in the metastatic, adjuvant, and
neoadjuvant settings; combination therapy targeting
two or more pathways is also under investigation.
Although initially discovered in the 1990s, the ESR1 mutation

was largely overlooked until recent years when recurrent,
rather than primary, breast cancers were studied. In contrast
to the extremely low incidence in treatment-naive primary
breast cancers, mutations in ESR1 have been reported in
11%55% of recurrent or metastatic cancers that progressed
after long-term endocrine therapy (Fig. 3).11-14 The hot-spot
mutations cluster in the ligand-binding-domain of ER, which
induce a constitutive agonist conformation. Compared with
the wild-type ER, mutant ER is resistant to estrogen deprivation and much less responsive to tamoxifen or fulvestrant. Successful treatment of ESR1-mutant breast cancer
therefore requires the development of newer generation
SERMs or SERDs, or other strategies, that are effective in
targeting the mutated ER.
HER2 Gene Amplification

HER family members, including HER1 (EGFR), HER2, HER3,
and HER4, are tyrosine kinase receptors important in
promoting cell growth and survival.15 Cross talk between
ER and HER family members has been well recognized in
preclinical and clinical studies. HER2 amplification occurs
in about 10% of ER+ breast cancers and is an established
mechanism of endocrine resistance.16 Compared with ER+
HER2- disease, ER+ HER2+ breast cancer is associated with a
higher risk of relapse on adjuvant endocrine therapy17 and
incomplete cell cycle arrest to neoadjuvant endocrine
therapy.18 Adjuvant trastuzumab reduces relapse in ER+
HER2+ breast cancer and is the standard of care.19 In the
metastatic setting, hormonal therapy in combination with
HER2-targeted agents provides an alternative to chemotherapy regimens.20-22
e41
FIGURE 2. Estrogen Receptor Pathway and Cross Talk With Growth Factor Receptor, PI3K/AKT/mTOR, and
CDK4/6 Pathways
Estrogen-bound estrogen receptor (ER), in complex with coactivators (CoA) or corepressors (CoR), regulates target gene expression via the estrogen response element (ERE)
or the AP1 binding sites via its interaction with other transcription factors. Additionally, ER can interact with receptor tyrosine kinases (FGFR, EGFR/HER), which promote
estrogen-independent ER phosphorylation through pathways such as PI3K/Akt/mTOR and MAPK (a and b). The G1 to S phase transition is controlled by CDK4/6, which is
activated by the ER downstream target cyclin D. Constitutive activation of CDK4/6 is associated with endocrine resistance.
HER2 Mutation
Somatic mutations in HER2 in otherwise HER2-negative
breast cancer have attracted much attention in recent years.
Although the overall HER2 mutation rate is approximately
2% in primary breast cancers,23 it reaches over 20% in invasive lobular cancers.24 These mutations cluster in the
kinase domain and the extracellular domain important for its
e42
interaction with other HER family members. In preclinical

models, many of these mutations have shown to be oncogenic and are sensitive to treatment with the irreversible
tyrosine kinase inhibitor of HER1/HER2, neratinib.23 Results
of the breast cancer cohort of the SUMMIT study, a multicenter, open-label, multihistology phase II basket trial of
neratinib for patients with HER2-mutant, nonamplified,
FIGURE 3. ESR1 Mutation as an Acquired Mechanism

of Endocrine Resistance
The incidence of ESR1 mutations increases dramatically after treatment with longterm endocrine therapy for recurrent or metastatic ER+ disease. The hot-spot ligandbinding domain mutations induce a constitutive agonist conformation of ER, leading
to estrogen independence.
Abbreviations: AF1, activation function 1; DBD, DNA-binding domain; ER, estrogen
receptor; LBD, ligand-binding domain.
advanced solid tumors (NCT01953926), was recently reported in an abstract form.25 Among the 19 evaluable patients, objective response was observed in six (overall
response rate 32%). As a majority of breast cancers with
HER2 mutation are ER+, the combination of fulvestrant and
neratinib is being studied among this patient population
(NCT01670877).
Fibroblast Growth Factor Receptors and Ligands

Amplification
The fibroblast growth factor (FGF) signaling is composed
of 18 ligands that exert their function through four highly
conserved transmembrane tyrosine kinase receptors
(FGFR1, FGFR2, FGFR3, and FGFR4) to control a wide range
of biologic functions and regulating cellular proliferation,
survival, migration, and differentiation.26 FGFR1 amplification has been identified in 16%27% of luminal B ER+
breast cancers and associated with increased Ki67, early
relapse, and poor survival.27 Less frequently, the pathway
could be activated by amplification of FGFR2 or different
ligands including FGF3 and FGF4.26 In preclinical studies,
FGFR1 amplification enhanced PI3K and MAPK pathway
signaling and rendered cancer cells resistance to endocrine
therapy, which was reversed by RNAi silencing of FGFR1.27
Dovitinib (TKI258), a first-generation oral tyrosine kinase inhibitor of FGFR1-3, VEGFR, and PDGFR, inhibited proliferation
of FGFR1/2-amplifiedbut not FGFRnormalbreast cancer
cell lines, and demonstrated promising antitumor activity for
patients with FGFR1/2/3-amplified breast cancers in a phase II
trial.28 Several FGFR inhibitors are currently being investigated in advanced HR+ breast cancer to overcome
endocrine resistance.
PI3K/Akt/mTOR Pathway
The PI3K/Akt/mTOR pathway is a cardinal nodal point in the
transduction of extracellular and intracellular growth and
survival signals, and acquired endocrine resistance is often

accompanied by upregulation of PI3K pathway signaling.29-32
Gain of function mutations in the alpha catalytic subunit of PI3
kinase (PIK3CA) occurs at a frequency of 30%40% and is the
most frequent genetic abnormality in ER+ breast cancer.33,34
A majority of the mutations, including the three hot-spot
mutations E542K, E545K, and H1047R, are missense activating mutations that cluster in the evolutionarily conserved
accessary domain and the kinase domain.35 In preclinical
models, ER+ breast cancer carrying PIK3CA mutations was
highly dependent on p110 alpha for cell survival.31,36
Rapalogs are among the first agents introduced in the
clinic that indirectly inhibit the activity of mTOR complex 1
(mTOC1) through its interaction with FKBP12.37 Everolimus
is now approved for AI-resistant metastatic breast cancer.
However, inhibition of mTORC1 has the potential to upregulate Akt activity due to the negative feedback loop between the downstream S6K and the upstream PI3K.38,39
Direct kinase inhibitors against both mTORC1 and mTORC2,
as well as inhibitors against Akt and PI3K, are in clinical trials
for more effective pathway inhibition. Promising activity has
been observed in some studies.40-43 However, it has become
evident that the efficacy of pan-PI3K inhibitors is limited by
dose-limiting toxicities that include rash, diarrhea, and elevated transaminases.43,44 Alpha-specific inhibitors are
likely advantageous in improving the therapeutic window,
but resistance mechanisms through acquiring mutations in
PTEN, leading to growth dependence on PI3K-beta, have
been reported in clinical trials of alpelisib, an alpha-specific
inhibitor.45 Preclinical modeling also indicated that effective
inhibition of mTOR remains important for upstream inhibitors,46 but combined PI3K and mTOR inhibition may not
be feasible due to overlapping toxicities. There has been
significant interest in developing biomarkers of treatment
efficacy, in particular PIK3CA mutation status. However, no
clear association has been identified even with the direct
PI3K inhibitors.40-43,47
Cyclin D/CDK4/6/Rb Pathway

The G1 to S phase transition is controlled by CDK4/6, which
are activated upon binding to D-type cyclins,48-51 leading to
phosphorylation of retinoblastoma susceptibility (RB1) gene
product (Rb) and the release of the E2F transcription factors.51 There is a strong link between the action of estrogen
and CDK4/6 activity.52-56 Persistent cyclin D1 expression and
Rb phosphorylation has been associated with resistance to
endocrine therapy in ER+ breast cancer.57 The poorer prognosis, luminal B ER+ diseases are preferentially enriched with
gains of CCND1 (cyclin D1; 58% in luminal B vs. 29% in luminal
A) and CDK4 (25% in luminal B vs. 14% in luminal A) and loss of
CDKN2A (p16) and CDKN2C (p18), which are negative regulators of CDK4/6.34
In preclinical studies, the CDK4/6 inhibitor palbociclib was
preferentially effective in inhibiting the cell proliferation of
ER+ cancer cells, including those resistant to anti-estrogens.58 Synergism was observed when combining palbociclib
e43
with various anti-estrogens,58,59 and efficacy was demonstrated in patient-derived xenograft models of ER+ breast
cancer carrying ESR1 mutations.59 Cells lacking Rb (and
hence not dependent on cyclin D1/CDK4/6 for proliferation)
were resistant to palbociclib. Therefore lack of Rb may serve
as a biomarker of resistance to this class of agents. Palbociclib in combination with letrozole is now approved as firstline treatment of postmenopausal women with metastatic
ER+ HER2 breast cancer. Several other CDK4/6 inhibitors
are in various stages of clinical development.
Combining PI3K and CDK4/6 Inhibitors in PIK3CAMutant Breast Cancer

A drug screen of 42 agents identified the CDK4/6 inhibitor
ribociclib as a strong sensitizer to PI3K inhibition in three ER+
breast cancer cell lines with acquired resistance to PI3K
inhibitors.60 In this study, acquired resistance to PI3K inhibition was associated with maintained phosphorylation of
S6 and Rb. Addition of ribociclib reduced Rb phosphorylation
and induced synergist antitumor effect of the PI3K inhibitor,
particularly in those with PIK3CA mutation and intact Rb. The
synergism was also observed in the parental cell lines and those
with de novo resistance to PI3K inhibitors. The impressive antitumor activity and the nonoverlapping toxicities of these two
classes of agents has led to clinical evaluation of triplet regimens
composed of letrozole, ribociclib, and alpelisib.
complicated because of its functional redundancy with the

insulin receptor.75 In addition to mediating endocrine therapy
resistance, data also support the importance of IGF1R in
mediating treatment resistance to inhibitors against mTOR or
Akt in endocrine-resistant ER+ breast cancer because of the
negative feedback loop between the downstream target S6K
and the upstream IGF1R/IRSs/PI3K signaling.38,76 Preclinical
data support the clinical investigation for the combined inhibition of IGF1R/IR and Akt, along with ER targeting.76
Vascular Endothelial Growth Factor

In addition to contributing to neoangiogenesis, VEGF has
been shown to induce breast cancer cell proliferation and
resistance to endocrine therapy via an autocrine mechanism.77,78 VEGF and the VEGF receptor have shown to be
overexpressed in breast cancers, and high levels of VEGF
have been linked to early recurrence and resistance to
hormonal therapy.79,80 These provided the preclinical rational for investigating the addition of VEGF inhibitors to
hormonal therapy for the treatment of ER+ breast cancer.
The challenge so far has been the difficulty in identifying a
target patient population who would derive the most benefit
from these agents.
Microenvironment and Immune Checkpoint Pathway
Preclinical studies indicated that epigenetic silencing of

ER and deregulation of growth factor receptorpathway
components play an important role in the development of
endocrine resistance.61-64 In addition, an array of mutations
have been identified in genes that regulate histone and DNA
modification.34 An example is the MLL3 (myeloid/lymphoid
or mixed-lineage leukemia gene) mutation, which occurred
in 8% of luminal A and 6% of luminal B breast cancers.34
The immune checkpoint pathway, PD-1, and the PD-L1 pathway has been implicated in tumor immune invasion.81 Inhibitors against PD-1 and PD-L1, which disrupt the interaction
between PD-1 on T cells and its ligands PD-L1, have shown
antitumor activity in a variety of tumor types. Approximately
4%20% of ER+ breast cancers, perhaps more frequently in the
more proliferative luminal B versus luminal A subtypes, have
been reported to overexpress PD-L1 in cancer and stroma
cells.82-84 PD-1 and PD-L1 antibodies are being tested in HR+
breast cancer, with preliminary results available.
Insulin-Like Growth Factor/Type 1 Insulin-Like

Growth Factor Signaling
REVERSING HORMONE RESISTANCE IN THE

CLINIC
The insulin-like growth factor (IGF)/type 1 insulin-like

growth factor receptor (IGF1R) signaling has a critical role
in cell growth, survival, and migration and is required for
mammary gland development.65 The IGF ligands, IGF1 and
IGF2, stimulate cell growth and survival signaling primarily via binding to IGF1R and the subsequent activation of
the PI3K/Akt and RAS/MAPK pathways.65,66 Extensive bidirectional regulation exists between ER and the IGF1R
pathway.67 ER-alpha is a major regulator of IGF signaling as
IGF1R and other IGF signaling components are its direct
transcription target,68-70 while IGF1R upregulates the transcription level of ER-alpha and increases ER phosphorylation via
activation of mTOR/S6K signaling.71 Hyperactivation of IGF1R
and downstream signaling has been associated with the development of endocrine resistance in ER+ breast cancer
preclinical models and patient specimens.65,72-74 However
targeting IGF1R in the endocrine-resistant setting has been
With a greater understanding of the mechanisms of resistance to hormone therapy, new therapies have emerged
that hold the potential to overcome hormone resistance and
improve response, duration of response, and, hopefully,
survival. Targeted therapies that are being actively investigated in the clinic include mTOR inhibitors (everolimus), PI3K inhibitors (buparlisib, alpelisib, and taselisib),
CDK4/6 inhibitors (palbociclib, ribociclib, and abemaciclib),
HDAC inhibitors (entinostat), FGFR inhibitors (dovitinib and
lucitinib), and IGFR inhibitors. Altering the host immune response with checkpoint inhibition is also being investigated in
advanced disease. Table 1 summarizes the major ongoing
phase II and III trials with the most-promising novel targeted
agents. Table 2 provides a summary of these agents, their
mechanisms of action, developmental status, and toxicity
profile. This section provides a more detailed review of these
emerging therapies, focusing on non-HER2 amplified disease.
Epigenetic Pathways
e44
Of note, two randomized trials have demonstrated improved progression-free survival (PFS) with the addition of
HER2-targeted agents to AIs in the treatment of metastatic,
HR, and HER2+ disease.22,85 However, given the demonstrated
improvement in overall survival (OS) with chemotherapy,
trastuzumab, and pertuzumab in the first-line metastatic setting, this is generally the preferred treatment approach, with
hormone therapy reserved for maintenance in combination
with HER2-targeted therapy following response.
Inhibition of mTOR
Several randomized trials have demonstrated that inhibition
of the mTOR pathway can improve duration of response to
hormone therapy, despite an initial rocky start. The first
randomized trial evaluated the efficacy of temsirolimus in
combination with the AI letrozole as first-line therapy for
metastatic HR+ breast cancer.86 There was no difference in PFS
between the two arms. It may be that the dose and schedule of
temsirolimus were not optimal, as stomatitis (a marker of drug
exposure) was seen at a lower rate than expected.
Greater success leading to the first approval of a targeted
therapy in combination with hormone therapy in HER2
disease was seen with everolimus. A phase II neoadjuvant
trial was conducted to evaluate efficacy and biomarkers,
with the primary endpoint of clinical response.87 Two
hundred-seventy postmenopausal women with HR+ earlystage breast cancer were randomly selected to receive
letrozole plus everolimus or placebo for 4 months prior to
surgery. Clinical response was higher in the combination arm
compared with placebo (68.1% vs. 59.1%; p = .062). More
importantly, a significant decrease in cell proliferation on
repeat biopsy performed 2 weeks after treatment was noted
in the combination therapy arm compared with placebo
(57% vs. 30% decrease; p , .01). These data encouraged
further investigation of everolimus in combination with
hormone therapy.
Two subsequent trials tested everolimus in the metastatic
setting. The TAMRAD trial was a phase II open-label trial that
randomly selected 111 postmenopausal women whose
disease had progressed on an AI to receive tamoxifen and
everolimus or tamoxifen alone.88 The primary endpoint,
clinical benefit rate at 6 months, was improved with combination therapy compared with tamoxifen alone (61% vs.
42%; p = .045). Progression-free survival in the combination
therapy arm was significantly longer at 8.6 months versus
4.5 months (p = .002).
Finally, regulatory approval of everolimus in combination
with exemestane for the treatment of metastatic breast
cancer progressing on nonsteroidal AIs was based on data
from the BOLERO-2 trial, a double-blind phase III trial that
randomly selected 724 postmenopausal women in a 2:1 ratio
to receive exemestane with everolimus at 10 mg a day or
placebo.89 At a median follow-up of 18 months, investigatorassessed PFS (the primary endpoint) was more than doubled
with the addition of everolimus (7.8 months for everolimus
compared with 3.2 months for placebo; hazard ratio [HR]
0.45; 95% CI, 0.380.54). Median OS was not significantly

improved at 31 months with everolimus compared with
26.6 months with placebo (HR 0.89; 95% CI, 0.731.10;
p = .14).90 A subset analysis of BOLERO-2 demonstrated
efficacy of combination therapy among patients with visceral
metastases and those with nonvisceral disease.91 Based on
the PFS data, everolimus is now being used clinically in
combination with exemestane in the metastatic setting.
Toxicity is a substantial issue with everolimus, but data
from subsequent trials suggest that physician and patient
education and expectant management reduces common
toxicity and increases tolerability.
The most common toxicity in BOLERO-2 was stomatitis
(56% all grade, aphthous-type ulcers), although the incidence of grade 3 stomatitis was low (8%), and no grade 4
events were oberved.92 Stomatitis occurs early, with 80% of
cases occurring in the first 6 weeks of therapy.93 A metaanalysis of seven phase III studies of everolimus in different
cancer types evaluated whether the incidence of stomatitis
correlated with clinical efficacy.94 There was commonality of
everolimus-induced stomatitis across solid tumors (66.9% all
grades of all patients), with 8.6% grade 3/4 stomatitis, and
the incidence of stomatitis correlated with at least as good if
not better PFS compared with the control population. This
suggests that stomatitis may be a marker of drug exposure
for everolimus and that dose reductions and delays for this
toxicity do not impact efficacy. A phase II trial evaluated a
commercially available steroid-based mouthwash among
patients taking prophylactic exemestane and everolimus to
determine if the incidence and severity of stomatitis might
be reduced with this approach.95 Data will be presented at
the 2016 American Society of Clinical Oncology (ASCO)
Annual Meeting.
Other commonly noted side effects with everolimus include fatigue, diarrhea, rash, hyperglycemia, pneumonitis,
weight loss, anemia, and thrombocytopenia. The risk of
pneumonitis persists over the course of therapy at a low
rate, with a 3% rate of grade 3 pneumonitis observed in
BOLERO-2. Additional clinical trials with everolimus are ongoing, summarized in Table 1. These include studies in the
neoadjuvant, adjuvant, and metastatic settings. Results from
BOLERO-6, comparing everolimus plus exemestane with
capecitabine, are expected this year.
Inhibition of PI3K
Two classes of PI3K inhibitors are being studied in HR+ breast
cancer, the pan-class I inhibitors (buparlisib [BKM120] and
pictilisib [GDC0941]) and the alpha-specific inhibitors
(alpelisib [BYL719] and taselisib [GDC0032]).
Results with the pan-PI3K inhibitors have been complicated by toxicity and disappointing efficacy. The FERGI trial
was a phase II trial that randomly selected 168 women with
HR+ postmenopausal metastatic breast cancer previously
treated with an AI to receive fulvestrant with pictilisib or
placebo.44 Toxicity was increased in the pictilisib arm; 17% of
patients had at least grade 3 rash and 7% had at least grade 3
e45
TABLE 1. Phase II/III Trials of the Novel Targeted Agents Under Development for Hormone ReceptorPositive
Breast Cancer
Trial
Novel Agent
Primary Endpoint
No.
Patients
Trial Status
Everolimus: mTOR Inhibitor

Baselga et al87
Everolimus
Clinical examination response
270
Completed; response rate of 68.1% in

everolimus/letrozole arm compared with
59.1% in placebo/letrozole arm (p = .062)
GINECO88
Everolimus
Clinical benefit rate at 6 months
111
Completed; clinical benefit rate of 61% in

everolimus/tamoxifen arm compared with
42% in tamoxifen arm (p = .045)
BOLERO-289
Everolimus
PFS
724
Completed; median PFS of 7.8 months in

everolimus/exemestane arm compared
with 3.2 months in exemestane/placebo
(HR 0.45; 95% CI, 0.380.54)
UNIRAD127
Everolimus
Disease-free survival
1,984
Ongoing
SWOG 1207128
Everolimus
Invasive disease-free survival using

a stratified log-rank test, assessed
up to 10 years
1,900
Ongoing, recruiting patients
NCT02088684130
Everolimus
PFS
46
BOLERO-4143
Everolimus
Percentage of patients progressionfree after completion of first-line

treatment
202
BOLERO-6144
Everolimus
PFS
297
BRE-43145
Everolimus
Time to progression
LEO146
Everolimus
PFS
137
Ongoing, active recruitment
FEVEX
Everolimus
PFS
745
Not yet open
PrE0102148
Everolimus
PFS
130
Ongoing
NCT02236572113
Everolimus
Achievement of a PEPI score of 0

following neoadjuvant treatment
with everolimus and an aromatase
inhibitor at 26 weeks
NCT02123823149
BI 836845 (monoclonal antibody

to insulin like growth factor)
and everolimus
147
33
Ongoing
Ongoing, recruitment completed

Completed
66
PFS
174
Taselisib (GDC0032): Alpha-Specific PI3K Inhibitor

LORELEI150
Taselisib
Objective response rate and

pathologic complete response
rate at 16 weeks of treatment
330
SANDPIPER98
Taselisib
PFS
600
Buparlisib (BKM120) Pan-Class I PI3K Inhibitor

BELLE-296
Buparlisib
PFS
1147
BELLE-3151
Buparlisib
PFS
420
PFS
168
PFS of 6.6 months for pictilisib/fulvestrant

arm compared with 5.1 months in
fulvestrant/placebo arm (HR 0.7; 95% CI,
0.521.06)
Completed; PFS was slightly improved with

buparlisib vs. placebo (6.9 vs. 5 months; HR
0.78; 95% CI, 0.670.89)
Pictlisib (GDC0941): Pan-Class I PI3K Inhibitor

FERGI44
Pictlisib
Palbociclib: CDK4/6 Inhibitor

PALOMA-1/TRIO1899
Palbociclib
PFS
165
Completed; PFS for palbociclib/letrozole was

20.2 months compared with 10.2 months
in letrozole arm (HR 0.49; 95% CI,
0.3190.748)
PALOMA-2100
Palbociclib
PFS
650
Ongoing, closed to accrual

Continued
e46
TABLE 1. Phase II/III Trials of the Novel Targeted Agents Under Development for Hormone ReceptorPositive
Breast Cancer (Contd)
No.
Patients
Trial
Novel Agent
Primary Endpoint
PALOMA-3101
Palbociclib
PFS
PEARL104
Palbociclib
PFS
348
PENELOPEB103
Palbociclib
Invasive disease-free survival
800
PALOMA-4152
Palbociclib
PFS
330
PALOMA-2139
Palbociclib
Treatment discontinuation rate
160
Ongoing, not recruiting patients
PALLET153
Palbociclib
Change in proliferation rate (Ki67)

at 2 weeks
306
FLIPPER154
Palbociclib
PFS
190
Ongoing, recruiting participants
PARSIFAL155
Palbociclib
PFS
304
Ongoing, currently recruiting patients
NCT02384239156
Palbociclib
Tumor Progression as measured by

RECIST 1.1
70
Ongoing, currently recruiting patients
NCT02592746157
Palbociclib
PFS
122
Ongoing, not yet recruiting patients
220
417
Trial Status
Completed; PFS of 9.2 months in palbociclib/
fulvestrant arm vs. 3.8 months in fulvestrant/placebo arm (HR 0.42; 95% CI,
0.320.56)
Abemaciclib: CDK4/6 Inhibitor

neoMONARCH134
Abemaciclib
Change in proliferation rate (Ki67)

at 2 weeks
monarcHER158
Abemaciclib
PFS
225
Ongoing, not yet recruiting patients
MONARCH-1110
Abemaciclib
Objective response rate
128
Closed to accrual
MONARCH-2112
Abemaciclib
PFS
550
Ongoing
MONARCH-3113
Abemaciclib
PFS
450
Ongoing
NCT02308020114
Abemaciclib
Percentage of participants achieving

complete response or partial
response: objective intracranial
response rate
247
Recruiting
Ribociclib: CDK4/6 Inhibitor

MONALEESA-2141
Ribociclib
PFS
667
Ongoing
159
Ribociclib
PFS
660
Ongoing, recruiting participants
MONALEESA-3
Entinostat: HDAC Inhibitor

Yardley et al117
Entinostat
PFS
130
Completed; PFS was 4.3 months in

exemestane/entinostat arm compared
with PFS of 2.3 months with exemestane
alone arm (p = .055); overall survival was
28.1 months in exemestane/entinostat
arm compared with 19.8 months in
exemestane alone arm (HR 0.59; 95% CI,
0.360.97)
E2112118
Entinostat
PFS
600
Ongoing
Azacitidine: DNA Methylation Inhibitor

NCT01349959120
Azacitidine and entinostat
Confirmed response rate
40
Ongoing
NCT02374099119
Azacitidine
Time period of PFS
92
Abbreviations: HR, hazard ratio; PFS, progression-free survival.
diarrhea, and 18% of patients discontinued due to adverse

events. Progression-free survival was similar between the
two treatment arms at 6.6 months for pictilisib and
5.1 months for placebo (HR 0.7; 95% CI, 0.521.06; p = .096),
and among patients with PI3KCA-mutant tumors. In an
exploratory analysis, there was a suggestion of benefit for
patients with PR+ disease.
The pan-PI3K inhibitor buparlisib was evaluated in the
phase III BELLE-2 trial.96 Patients with prior AI therapy were
randomly selected to fulvestrant with either buparlisib or

placebo, with a primary endpoint of PFS in the full population and in those with PI3K pathwayactivated tumors. In
the 1,147 women who received study therapy, PFS was
slightly improved with buparlisib compared with placebo
(6.9 vs. 5 months; HR 0.78; 95% CI, 0.670.89; p , .001).
Progression-free survival was not significantly improved for
patients with PI3K activation (6.8 vs. 4 months). In an exploratory analysis of PIK3CA mutations in circulating tumor
e47
TABLE 2. Summary of the Most-Promising Targeted Agents Under Development in the Treatment of Hormone
ReceptorPositive Breast Cancer
Agent
Mechanism of Action
Developmental Status
Everolimus
mTOR inhibition
FDA approval for everolimus plus exemestane as at Main toxicities include stomatitis, hyperglycemia,
least second-line therapy for HR+ advanced
fatigue, pneumonitis
breast cancer
Palbociclib, Ribociclib, CDK4/6 inhibition

Abemaciclib
Toxicity Profile
FDA approval for palbociclib plus letrozole as first- Main toxicities include neutropenia without
line therapy, and palbociclib plus fulvestrant as
an increase in febrile neutropenia, QTc
at least second-line therapy for advanced breast
prolongation, and fatigue; diarrhea most
cancer; ongoing trials with all 3 agents in the
common with abemaciclib
metastatic, adjuvant, and neoadjuvant settings
Alpha-Specific
Alpelisib, Taselisib
PI3K inhibition
Ongoing trials
Main toxicities include hyperglycemia, rash, and

liver dysfunction
Buparlisib, Pictilisib
PI3K inhibition
Ongoing trials
Main toxicities include hyperglycemia, rash, and

liver dysfunction
Entinostat
HDAC inhibition
FDA breakthrough status; ongoing phase III trial

with exemestane
Main toxicities include fatigue, myelosuppression,

diarrhea, nausea, and vomiting
Dovitinib, Lucitinib
Pre-clinical and preliminary clinical efficacy with

FGFR inhibition
dovitinib and lucitinib
FGFR/VEGF inhibition
(lucitinib)
Main toxicities include diarrhea, nausea, and

hypertension, proteinuria, and liver dysfunction
Pan-PI3K
Abbreviation: FDA, U.S. Food and Drug Administration.
DNA (ctDNA; 87 patients for buparlisib and 113 for placebo),

PFS was improved for those with PIK3CA mutations receiving
buparlisib compared with placebo (7.0 vs. 3.2 months; HR
0.56), with no difference in the wild-type population.
Toxicity was increased in the buparlisib arm, with grade
3/4 transaminitis (. 20%; approximately 0% grade 4), hyperglycemia (15.4%), rash (7.9%), anxiety (5.4%), and depression (4.4%); 13% of patients discontinued treatment due
to adverse events.
The data from BELLE-2, although disappointing in the
overall population, provide the first data suggesting that
ctDNA could be a potential marker to select patients who
might benefit from specific targeted therapies. Ongoing
trials are evaluating the use of PIK3CA mutations in ctDNA as
predictive markers for response to alpha-specific PI3K inhibitors, such as alpelisib and taselisib.
The alpha-specific PI3K inhibitors alpelisib and taselisib
have demonstrated encouraging efficacy in early-phase
trials, with modest toxicity. Two ongoing placebo-controlled
phase III trials are evaluating the potential improvement in
PFS with the addition of alpelisib (SOLAR-197) or taselisib
(SANDPIPER98) to fulvestrant among patients with HR+
metastatic breast cancer, and neoadjuvant trials in combination with AIs are also ongoing (the buparlisib arm of the
NeoORB study was recently closed based on results from
BELLE-2). More details regarding ongoing studies are provided in Table 1. Results are eagerly awaited.
Inhibition of CDK4/6
Three CDK4/6 inhibitors (palbociclib, ribociclib, and abemaciclib) are in advanced clinical testing, and palbociclib has
regulatory approval for treatment of metastatic breast
e48
cancer. Palbociclib and ribociclib are given on a 3-week-on,

1-week-off schedule, and abemaciclib is given continuously.
A number of studies are evaluating CDK4/6 inhibitors in the
neoadjuvant, adjuvant, and metastatic settings, as summarized in Table 1. The results of pivotal trials and key
planned trials are summarized here.
PALOMA-1 was a phase II trial that randomly selected 165
postmenopausal women with untreated HR metastatic
breast cancer to receive letrozole with or without palbociclib.99
Progression-free survival with combination therapy was
20.2 months vs. 10.2 months in the control arm (HR 0.49;
95% CI 0.3190.748; p = .0004). The trial was not powered
to evaluate OS, which was similar between the two arms.
The primary toxicity was at least grade 3 neutropenia,
occurring in 54% of patients, with no cases of febrile
neutropenia; 13% of patients in the study arm discontinued therapy due to adverse events.
Based on this data, in 2015, the U.S. Food and Drug Administration (FDA) granted accelerated approval for palbociclib and letrozole for the first-line treatment of
postmenopausal patients with HR+ metastatic breast cancer. PALOMA-2 is a phase III trial designed to validate these
findings; data are expected to be presented at the 2016
ASCO Annual Meeting.100
PALOMA-3, a phase III trial, randomly selected 521 women
with advanced pretreated breast cancer 2:1 to treatment
with palbociclib and fulvestrant versus fulvestrant and
placebo.101 Progression-free survival was significantly improved in the combination arm versus placebo (9.2 months vs.
3.8 months; HR 0.42; 95% CI, 0.320.56; p , .001). Premenopausal women with ovarian suppression and patients with one line of prior chemotherapy appeared to
have similar benefit. Toxicity was similar to PALOMA-1, although only 2.6% of patients discontinued treatment due to
adverse events. Fulvestrant and palbociclib were approved
by the FDA in early 2016.
An international adjuvant trial of 4,600 patients is comparing 2 years of treatment with palbociclib versus placebo
(PALLAS),102 and PENELOPE-B103 is studying 1 year of palbociclib as postneoadjuvant therapy for high-risk patients. Numerous other trials are evaluating palbociclib as neoadjuvant
therapy compared with capecitabine in metastatic disease
(PEARL)104 and following chemotherapy, among others.
Ribociclib is a similar CDK4/6 inhibitor that is being studied
in the metastatic and neoadjuvant settings in the ongoing
MONALEESA trials.105,106 In addition, ongoing studies are
evaluating ribociclib combined with exemestane and everolimus,107 and letrozole and the PI3K inhibitor alpelisib.108
Abemaciclib is a potent inhibitor of CDK4, more than CDK6,
and is the only agent in this class that can be given continuously. A phase IB trial including 47 patients demonstrated single-agent activity with a clinical benefit rate of
61.1%,109 leading to the single agent MONARCH-1 trial of
patients with heavily pretreated HR+ metastatic breast
cancer; data are expected to be presented at the 2016 ASCO
Annual Meeting.110 An expansion cohort of the phase IB trial
evaluated abemaciclib plus fulvestrant; the clinical benefit
rate was 72.2%.111 The most common toxicity from abemaciclib is diarrhea, with a lower incidence of neutropenia.
Additional ongoing MONARCH trials are evaluating abemaciclib in the neoadjuvant setting,105 in metastatic disease
in combination with fulvestrant,112 or nonsteroidal AIs113 in
brain metastases114 and in HER2+ disease.112
Inhibition of FGFR
Dovitinib, an oral tyrosine kinase inhibitor of FGFR, has been
shown to have both preclinical and clinical efficacy.28 In a
study of 81 patients, five patients with tumor FGFR amplification experienced clinical benefit at 6 months. These
results prompted further trials with dovitinib and hormonal
therapy combinations, although toxicity is a limiting factor.115 Lucitanib is a FGFR/VEGFR tyrosine kinase inhibitor
with encouraging phase IB response data under evaluation
in a phase II single-agent trial.116 Toxicity is again limiting,
including hypertension, proteinuria, and transaminitis.
Inhibition of HDAC
A randomized phase II trial evaluated the addition of entinostat or placebo to exemestane for 130 women with advanced breast cancer.117 Progression-free survival was
slightly improved in those receiving entinostat compared
with placebo (4.3 vs. 2.3 months; p = .055), but the exploratory endpoint of OS was significantly improved (28.1 vs.
19.8 months; HR 0.59; 95% CI, 0.360.97; p = .036), leading
to breakthrough drug status by the FDA. Interestingly,
protein lysine hyperacetylation in the entinostat biomarker
subset was associated with prolonged PFS. Fatigue and
neutropenia were the most common grade 3 or worse
toxicities; 11% of patients discontinued due to adverse

events. A phase III trial, E2112, with a similar design is
ongoing (Table 1).118
Azacitidine, a DNA methylation inhibitor, is also being
investigated in breast cancer with hormonal therapy119 and
entinostat.120
Inhibition of IGFR
A number of IGFR inhibitors have been studied in the clinic, with
generally disappointing results to date. A randomized phase II
trial evaluated the benefit of adding ganitumab to either fulvestrant or exemestane as second-line therapy for 156 patients
with metastatic HR+ breast cancer.121 There was no difference
in PFS (3.9 months for the ganitumab arm vs. 5.7 months for
placebo). New agents that do not cause hyperglycemia may
offer advantages; the monoclonal antibody BI 836845 is being
studied in combination with everolimus and exemestane in a
randomized phase II trial based on promising phase I results.107
Inhibition of Angiogenesis
Although not currently in the clinical arena, a discussion of
this approach is important. Two randomized phase III trials
evaluated the potential efficacy from adding bevacizumab,
the VEGF antibody, to first-line hormone therapy with
letrozole or fulvestrant.122,123 Both trials demonstrated improvement in PFS comparing letrozole or fulvestrant plus
bevacizumab with hormone therapy alone, although only the
CALGB trial found this difference to be statistically significant
(LEA trial: 19.3 vs. 14.4 months; HR 0.83; 95% CI, 0.651.06;
p = .126; CALGB 40503: 20 vs. 16 months; HR 0.75; 95% CI,
0.590.96; p = .016); no difference was seen in OS. Adverse
events included at least grade 3 hypertension and proteinuria.
Further combined biomarkers analysis is ongoing.
Immune Checkpoint Inhibition

Based on the expression of PD-L1 in HR+ breast cancer, there
has been interest in evaluating the efficacy of this approach
in the advanced disease setting. Pembrolizumab, a PD-1
antibody, was tested in a number of solid malignancies in the
phase IB KEYNOTE-28 trial.124 Out of 248 patients who had
tissue available for screening, 48 (19.4%) had at least 1%
staining for PD-L1. The overall response rate among 25
treated patients was 12%, with a clinical benefit rate of 20%.
The median time to response was 8 weeks, and the median
duration of response had not yet been reached.
The phase IB Javelin trial evaluated the efficacy of the antiPD-L1 antibody avelumab for patients with metastatic solid
tumors, including breast cancer.125 Out of the 72 patients
with HR+ disease, an objective response was seen in only
2.8%, but 54% were found to have PD-L1 expression. This
brings into question the antibodies used for PD-L1 testing as
well as potential differences in efficacy between different
checkpoint inhibitors.
Overall, therapy was well tolerated in both trials, with
the primary toxicity reversible and immune related (e.g.,
thyroid disorders, hepatitis). Further studies will evaluate
e49
combination therapy with agents that could increase

expression of neo-antigens (such as HDAC inhibitors), and
therefore enhance immunogenicity.
CONCLUSION
Progress in the understanding of endocrine resistance
mechanisms has led to the development of a number of
agentssome approved and many in clinical trialsto
target growth factor receptor signaling, PI3K/Akt/mTOR,
CDK4/6, HDAC, and immune checkpoints, among others.
The development of ESR1 mutation following endocrine

therapy provides further evidence that ER remains relevant
in resistant tumors. Combination strategies that target both
ER and resistant pathways are therefore rational approaches
to overcome endocrine resistance. Toxicity and cost are
increased with the addition of targeted agents. Awareness of
both expected toxicity and appropriate management is
important. There is a critical need to develop predictive
biomarkers; recent data using ctDNA are encouraging.
Further studies are ongoing.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
e50
Beatson GT. On the treatment of inoperable carcinoma of the

mamma: suggestions for a new method of treatment with illustrative
cases. Lancet. 1896;2:104-107.
Davies C, Godwin J, Gray R, et al; Early Breast Cancer Trialists Collaborative Group (EBCTCG). Relevance of breast cancer hormone receptors and other factors to the efficacy of adjuvant tamoxifen: patientlevel meta-analysis of randomised trials. Lancet. 2011;378:771-784.
Mauri D, Pavlidis N, Polyzos NP, et al. Survival with aromatase inhibitors and inactivators versus standard hormonal therapy in advanced breast cancer: meta-analysis. J Natl Cancer Inst. 2006;98:
1285-1291.
Osborne CK, Schiff R. Estrogen-receptor biology: continuing progress
and therapeutic implications. J Clin Oncol. 2005;23:1616-1622.
Egeland NG, Lunde S, Jonsdottir K, et al. The role of MicroRNAs as
predictors of response to tamoxifen treatment in breast cancer patients. Int J Mol Sci. 2015;16:24243-24275.
Losel R, Wehling M. Nongenomic actions of steroid hormones. Nat Rev
Mol Cell Biol. 2003;4:46-56.
Levin ER. Integration of the extranuclear and nuclear actions of estrogen. Mol Endocrinol. 2005;19:1951-1959.
Bjornstrom L, Sjoberg M. Mechanisms of estrogen receptor signaling:
convergence of genomic and nongenomic actions on target genes.
Mol Endocrinol. 2005;19:833-842.
Campbell RA, Bhat-Nakshatri P, Patel NM, et al. Phosphatidylinositol
3-kinase/AKT-mediated activation of estrogen receptor alpha: a new
model for anti-estrogen resistance. J Biol Chem. 2001;276:9817-9824.
Kato S, Endoh H, Masuhiro Y, et al. Activation of the estrogen receptor
through phosphorylation by mitogen-activated protein kinase. Science. 1995;270:1491-1494.
Li S, Shen D, Shao J, et al. Endocrine-therapy-resistant ESR1 variants
revealed by genomic characterization of breast-cancer-derived xenografts. Cell Reports. 2013;4:1116-1130.
Robinson DR, Wu YM, Vats P, et al. Activating ESR1 mutations in
hormone-resistant metastatic breast cancer. Nat Genet. 2013;45:
1446-1451.
Toy W, Shen Y, Won H, et al. ESR1 ligand-binding domain mutations in
hormone-resistant breast cancer. Nat Genet. 2013;45:1439-1445.
Roodi N, Bailey LR, Kao WY, et al. Estrogen receptor gene analysis in
estrogen receptor-positive and receptor-negative primary breast
cancer. J Natl Cancer Inst. 1995;87:446-451.
Saxena R, Dwivedi A. ErbB family receptor inhibitors as therapeutic
agents in breast cancer: Current status and future clinical perspective.
Med Res Rev. 2012;32:166-215.
Dowsett M. Overexpression of HER-2 as a resistance mechanism to
hormonal therapy for breast cancer. Endocr Relat Cancer. 2001;8:
191-195.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
Ejlertsen B, Aldridge J, Nielsen KV, et al. Prognostic and predictive role

of ESR1 status for postmenopausal patients with endocrineresponsive early breast cancer in the Danish cohort of the BIG 1-98
trial. Ann Oncol. 2011;23:1138-1144.
Ellis MJ, Tao Y, Young O, et al. Estrogen-independent proliferation is
present in estrogen-receptor HER2-positive primary breast cancer
after neoadjuvant letrozole. J Clin Oncol. 2006;24:3019-3025.
Perez EA, Romond EH, Suman VJ, et al. Four-year follow-up of trastuzumab plus adjuvant chemotherapy for operable human epidermal
growth factor receptor 2-positive breast cancer: joint analysis of data
from NCCTG N9831 and NSABP B-31. J Clin Oncol. 2011;29:3366-3373.
Schwartzberg LS, Franco SX, Florance A, et al. Lapatinib plus letrozole
as first-line therapy for HER-2+ hormone receptor-positive metastatic
breast cancer. Oncologist. 2010;15:122-129.
Marcom PK, Isaacs C, Harris L, et al. The combination of letrozole and
trastuzumab as first or second-line biological therapy produces durable responses in a subset of HER2 positive and ER positive advanced
breast cancers. Breast Cancer Res Treat. 2007;102:43-49.
Kaufman B, Mackey JR, Clemens MR, et al. Trastuzumab plus anastrozole versus anastrozole alone for the treatment of postmenopausal
women with human epidermal growth factor receptor 2-positive,
hormone receptor-positive metastatic breast cancer: results from the
randomized phase III TAnDEM study. J Clin Oncol. 2009;27:5529-5537.
Bose R, Kavuri SM, Searleman AC, et al. Activating HER2 mutations in
HER2 gene amplification negative breast cancer. Cancer Discov. 2013;
3:224-237.
Ross JS, Wang K, Sheehan CE, et al. Relapsed classic E-cadherin (CDH1)mutated invasive lobular breast cancer shows a high frequency of HER2
(ERBB2) gene mutations. Clin Cancer Res. 2013;19:2668-2676.
Hyman DM, Piha-Paul SA, Rodon
J, et al. Neratinib for ERBB2 mutant,
HER2 non-amplified, metastatic breast cancer: preliminary analysis
from a multicenter, open-label, multi-histology phase II basket trial.
Cancer Res. 2016;76 (suppl 4; abstr PD5-05).
Turner N, Grose R. Fibroblast growth factor signalling: from development to cancer. Nat Rev Cancer. 2010;10:116-129.
Turner N, Pearson A, Sharpe R, et al. FGFR1 amplification drives
endocrine therapy resistance and is a therapeutic target in breast
cancer. Cancer Res. 2010;70:2085-2094.
Andre F, Bachelot T, Campone M, et al. Targeting FGFR with dovitinib
(TKI258): preclinical and clinical data in breast cancer. Clin Cancer Res.
2013;19:3693-3702.
deGraffenried LA, Friedrichs WE, Russell DH, et al. Inhibition of mTOR
activity restores tamoxifen response in breast cancer cells with aberrant Akt Activity. Clin Cancer Res. 2004;10:8059-8067.
Miller TW, Hennessy BT, Gonzalez-Angulo AM, et al. Hyperactivation
of phosphatidylinositol-3 kinase promotes escape from hormone
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
dependence in estrogen receptor-positive human breast cancer. J Clin

Invest. 2010;120:2406-2413.
Sanchez CG, Ma CX, Crowder RJ, et al. Preclinical modeling of combined phosphatidylinositol-3-kinase inhibition with endocrine therapy
for estrogen receptor-positive breast cancer. Breast Cancer Res. 2011;
13:R21.
Vivanco I, Sawyers CL. The phosphatidylinositol 3-Kinase AKT pathway
in human cancer. Nat Rev Cancer. 2002;2:489-501.
Ellis MJ, Ding L, Shen D, et al. Whole-genome analysis informs breast
cancer response to aromatase inhibition. Nature. 2012;486:353-360.
Cancer Genome Atlas Network.; Cancer Genome Atlas Network.
Comprehensive molecular portraits of human breast tumours. Nature. 2012;490:61-70.
Saal LH, Holm K, Maurer M, et al. PIK3CA mutations correlate with
hormone receptors, node metastasis, and ERBB2, and are mutually
exclusive with PTEN loss in human breast carcinoma. Cancer Res.
2005;65:2554-2559.
Crowder RJ, Phommaly C, Tao Y, et al. PIK3CA and PIK3CB inhibition
produce synthetic lethality when combined with estrogen deprivation
in estrogen receptor-positive breast cancer. Cancer Res. 2009;69:
3955-3962.
Guertin DA, Sabatini DM. The pharmacology of mTOR inhibition. Sci
Signal. 2009;2:pe24.
OReilly KE, Rojo F, She QB, et al. mTOR inhibition induces upstream
receptor tyrosine kinase signaling and activates Akt. Cancer Res. 2006;
66:1500-1508.
Wan X, Harkavy B, Shen N, et al. Rapamycin induces feedback activation of Akt signaling through an IGF-1R-dependent mechanism.
Oncogene. 2007;26:1932-1940.
Mayer IA, Abramson VG, Isakoff SJ, et al. Stand up to cancer phase Ib
study of pan-phosphoinositide-3-kinase inhibitor buparlisib with
letrozole in estrogen receptor-positive/human epidermal growth
factor receptor 2-negative metastatic breast cancer. J Clin Oncol.
2014;32:1202-1209.
Ma CX, Sanchez C, Gao F, et al. A phase I study of the AKT inhibitor MK2206 in combination with hormonal therapy in postmenopausal
women with estrogen receptor positive metastatic breast cancer. Clin
Cancer Res. Epub 2016 Jan 18.
Ma CX, Luo J, Naughton M, et al. A phase 1 trial of BKM120 (buparlisib)
in combination with fulvestrant in postmenopausal women with
estrogen receptor positive metastatic breast cancer. Clin Cancer Res.
2016;22:1583-1591.
Baselga J, Im S-A, Iwata H, et al. PIK3CA status in circulating tumor DNA
(ctDNA) predicts efficacy of buparlisib (BUP) plus fulvestrant (FULV) in
postmenopausal women with endocrine-resistant HR+/HER2 advanced breast cancer (BC): First results from the randomized, phase III
BELLE-2 trial. Presented at: San Antonio Breast Cancer Symposium;
2015; San Antonio, Texas.
Krop I, Johnston S, Mayer I, et al. The FERGI phase II study of the PI3K
inhibitor pictilisib (GDC-0941) plus fulvestrant vs fulvestrant plus
placebo in patients with ER+, aromatase inhibitor (AI)-resistant advanced or metastatic breast cancer Part I results. Cancer Res. 2015;
75:S2-02.
Juric D, Castel P, Griffith M, et al. Convergent loss of PTEN leads to
clinical resistance to a PI(3)Ka inhibitor. Nature. 2015;518:240-244.
Elkabets M, Vora S, Juric D, et al. mTORC1 inhibition is required for
sensitivity to PI3K p110a inhibitors in PIK3CA-mutant breast cancer.
Sci Transl Med. 2013;5:196ra99.
Hortobagyi GN, Piccart-Gebhart MJ, Rugo HS, et al. Correlation of
molecular alterations with efficacy of everolimus in hormone receptor
positive, HER2-negative advanced breast cancer: Results from BOLERO-2.
J Clin Oncol. 2013 (suppl; abstr LBA509).
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
Sherr CJ. Cancer cell cycles. Science. 1996;274:1672-1677.

Sherr CJ, Roberts JM. CDK inhibitors: positive and negative regulators
of G1-phase progression. Genes Dev. 1999;13:1501-1512.
van den Heuvel S, Harlow E. Distinct roles for cyclin-dependent kinases
in cell cycle control. Science. 1993;262:2050-2054.
Weinberg RA. The retinoblastoma protein and cell cycle control. Cell.
1995;81:323-330.
Butt AJ, McNeil CM, Musgrove EA, et al. Downstream targets of
growth factor and oestrogen signalling and endocrine resistance: the
potential roles of c-Myc, cyclin D1 and cyclin E. Endocr Relat Cancer.
2005;12:S47-S59.
Altucci L, Addeo R, Cicatiello L, et al. Estrogen induces early and timed
activation of cyclin-dependent kinases 4, 5, and 6 and increases cyclin
messenger ribonucleic acid expression in rat uterus. Endocrinology.
1997;138:978-984.
Geum D, Sun W, Paik SK, et al. Estrogen-induced cyclin D1 and D3 gene
expressions during mouse uterine cell proliferation in vivo: differential
induction mechanism of cyclin D1 and D3. Mol Reprod Dev. 1997;46:
450-458.
Said TK, Conneely OM, Medina D, et al. Progesterone, in addition to
estrogen, induces cyclin D1 expression in the murine mammary epithelial cell, in vivo. Endocrinology. 1997;138:3933-3939.
Tong W, Pollard JW. Progesterone inhibits estrogen-induced cyclin D1
and cdk4 nuclear translocation, cyclin E- and cyclin A-cdk2 kinase
activation, and cell proliferation in uterine epithelial cells in mice. Mol
Cell Biol. 1999;19:2251-2264.
Thangavel C, Dean JL, Ertel A, et al. Therapeutically activating RB:
reestablishing cell cycle control in endocrine therapy-resistant breast
cancer. Endocr Relat Cancer. 2011;18:333-345.
Finn RS, Dering J, Conklin D, et al. PD 0332991, a selective cyclin D
kinase 4/6 inhibitor, preferentially inhibits proliferation of luminal
estrogen receptor-positive human breast cancer cell lines in vitro.
Breast Cancer Res. 2009;11:R77.
Wardell SE, Ellis MJ, Alley HM, et al. Efficacy of SERD/SERM HybridCDK4/6 inhibitor combinations in models of endocrine therapy resistant breast cancer. Clin Cancer Res. 2015;21:5121-5130.
Vora SR, Juric D, Kim N, et al. CDK 4/6 inhibitors sensitize PIK3CA
mutant breast cancer to PI3K inhibitors. Cancer Cell. 2014;26:136-149.
Pathiraja TN, Stearns V, Oesterreich S. Epigenetic regulation in estrogen receptor positive breast cancerrole in treatment response.
J Mammary Gland Biol Neoplasia. 2010;15:35-47.
Yi X, Wei W, Wang SY, et al. Histone deacetylase inhibitor SAHA induces ERalpha degradation in breast cancer MCF-7 cells by CHIPmediated ubiquitin pathway and inhibits survival signaling. Biochem
Pharmacol. 2008;75:1697-1705.
Thomas S, Thurn KT, Biaku E, et al. Addition of a histone deacetylase
inhibitor redirects tamoxifen-treated breast cancer cells into apoptosis, which is opposed by the induction of autophagy. Breast Cancer
Res Treat. 2011;130:437-447.
Biaku E, Marchion DC, Schmitt ML, et al. Selective inhibition of
histone deacetylase 2 silences progesterone receptor-mediated signaling. Cancer Res. 2008;68:1513-1519.
Farabaugh SM, Boone DN, Lee AV. Role of IGF1R in breast cancer
subtypes, stemness, and lineage differentiation. Front Endocrinol
(Lausanne). 2015;6:59.
Shelton JG, Steelman LS, White ER, et al. Synergy between PI3K/Akt
and Raf/MEK/ERK pathways in IGF-1R mediated cell cycle progression
and prevention of apoptosis in hematopoietic cells. Cell Cycle. 2004;3:
372-379.
Fagan DH, Yee D. Crosstalk between IGF1R and estrogen receptor
signaling in breast cancer. J Mammary Gland Biol Neoplasia. 2008;13:
423-429.
e51
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
e52
Lee AV, Jackson JG, Gooch JL, et al. Enhancement of insulin-like growth
factor signaling in human breast cancer: estrogen regulation of insulin
receptor substrate-1 expression in vitro and in vivo. Mol Endocrinol.
1999;13:787-796.
Maor S, Mayer D, Yarden RI, et al. Estrogen receptor regulates insulinlike growth factor-I receptor gene expression in breast tumor cells:
involvement of transcription factor Sp1. J Endocrinol. 2006;191:
605-612.
Mauro L, Salerno M, Panno ML, et al. Estradiol increases IRS-1 gene
expression and insulin signaling in breast cancer cells. Biochem Biophys Res Commun. 2001;288:685-689.
Becker MA, Ibrahim YH, Cui X, et al. The IGF pathway regulates ERa
through a S6K1-dependent mechanism in breast cancer cells. Mol
Endocrinol. 2011;25:516-528.
Knowlden JM, Hutcheson IR, Barrow D, et al. Insulin-like growth
factor-I receptor signaling in tamoxifen-resistant breast cancer:
a supporting role to the epidermal growth factor receptor. Endocrinology. 2005;146:4609-4618.
Arnedos M, Drury S, Afentakis M, et al. Biomarker changes associated
with the development of resistance to aromatase inhibitors (AIs) in
estrogen receptor-positive breast cancer. Ann Oncol. 2014;25:
605-610.
Fox EM, Miller TW, Balko JM, et al. A kinome-wide screen identifies the
insulin/IGF-I receptor pathway as a mechanism of escape from
hormone dependence in breast cancer. Cancer Res. 2011;71:
6773-6784.
Fagan DH, Uselman RR, Sachdev D, et al. Acquired resistance to tamoxifen is associated with loss of the type I insulin-like growth factor
receptor: implications for breast cancer treatment. Cancer Res. 2012;
72:3372-3380.
Fox EM, Kuba MG, Miller TW, et al. Autocrine IGF-I/insulin receptor
axis compensates for inhibition of AKT in ER-positive breast cancer
cells with resistance to estrogen deprivation. Breast Cancer Res. 2013;
15:R55.
Liang Y, Hyder SM. Proliferation of endothelial and tumor epithelial
cells by progestin-induced vascular endothelial growth factor from
human breast cancer cells: paracrine and autocrine effects. Endocrinology. 2005;146:3632-3641.
Schoeffner DJ, Matheny SL, Akahane T, et al. VEGF contributes to
mammary tumor growth in transgenic mice through paracrine and
autocrine mechanisms. Lab Invest. 2005;85:608-623.
Foekens JA, Peters HA, Grebenchtchikov N, et al. High tumor levels of
vascular endothelial growth factor predict poor response to systemic
therapy in advanced breast cancer. Cancer Res. 2001;61:5407-5414.
Manders P, Beex LV, Tjan-Heijnen VC, et al. Vascular endothelial
growth factor is associated with the efficacy of endocrine therapy in
patients with advanced breast carcinoma. Cancer. 2003;98:
2125-2132.
Pardoll DM. The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer. 2012;12:252-264.
Ghebeh H, Mohammed S, Al-Omair A, et al. The B7-H1 (PD-L1)
T lymphocyte-inhibitory molecule is expressed in breast cancer patients with infiltrating ductal carcinoma: correlation with important
high-risk prognostic factors. Neoplasia. 2006;8:190-198.
Muenst S, Schaerli AR, Gao F, et al. Expression of programmed death
ligand 1 (PD-L1) is associated with poor prognosis in human breast
Sabatier R, Finetti P, Mamessier E, et al. Prognostic and predictive
value of PDL1 expression in breast cancer. Oncotarget. 2015;6:
5449-5464.
Johnston S, Pippen J Jr, Pivot X, et al. Lapatinib combined with
letrozole versus letrozole and placebo as first-line therapy for
postmenopausal hormone receptor-positive metastatic breast cancer. J Clin Oncol. 2009;27:5538-5546.

86. Wolff AC, Lazar AA, Bondarenko I, et al. Randomized phase III placebocontrolled trial of letrozole plus oral temsirolimus as first-line endocrine therapy in postmenopausal women with locally advanced or
metastatic breast cancer. J Clin Oncol. 2013;31:195-202.
87. Baselga J, Semiglazov V, van Dam P, et al. Phase II randomized study of
neoadjuvant everolimus plus letrozole compared with placebo plus
letrozole in patients with estrogen receptor-positive breast cancer.
J Clin Oncol. 2009;27:2630-2637.
88. Bachelot T, Bourgier C, Cropet C, et al. Randomized phase II trial of
everolimus in combination with tamoxifen in patients with hormone
receptor-positive, human epidermal growth factor receptor 2-negative metastatic breast cancer with prior exposure to aromatase inhibitors: a GINECO study. J Clin Oncol. 2012;30:2718-2724.
89. Yardley DA, Noguchi S, Pritchard KI, et al. Everolimus plus exemestane
in postmenopausal patients with HR(+) breast cancer: BOLERO-2 final
progression-free survival analysis. Adv Ther. 2013;30:870-884.
90. Piccart M, Hortobagyi GN, Campone M, et al. Everolimus plus
exemestane for hormone-receptor-positive, human epidermal
growth factor receptor-2-negative advanced breast cancer: overall
survival results from BOLERO-2. Ann Oncol. 2014;25:2357-2362.
91. Campone M, Bachelot T, Gnant M, et al. Effect of visceral metastases
on the efficacy and safety of everolimus in postmenopausal women
with advanced breast cancer: subgroup analysis from the BOLERO-2
study. Eur J Cancer. 2013;49:2621-2632.
92. Baselga J, Campone M, Piccart M, et al. Everolimus in postmenopausal
hormone-receptor-positive advanced breast cancer. N Engl J Med.
2012;366:520-529.
93. Rugo HS, Pritchard KI, Gnant M, et al. Incidence and time course of
everolimus-related adverse events in postmenopausal women with
hormone receptor-positive advanced breast cancer: insights from
BOLERO-2. Ann Oncol. 2014;25:808-815.
94. Dean JL, Thangavel C, McClendon AK, et al. Therapeutic CDK4/6 inhibition in breast cancer: key mechanisms of response and failure.
Oncogene. 2010;29:4018-4032.
95. NCT02069093. Phase II study of stomatitis prevention with a steroidbased mouthwash in post-menopausal women with ER+, HER2metastatic or locally advanced breast cancer. https://clinicaltrials.
gov/ct2/show/NCT02069093. Accessed January 30,2016.
96. Baselga J, Im SA, Iwata H, et al. PIK3CA status in circulating tumor DNA
(ctDNA) predicts efficacy of buparlisib (BUP) plus fulvestrant (FULV) in
postmenopausal women with endocrine-resistant HR+/HER2 advanced breast cancer (BC): First results from the randomized, phase III
BELLE-2 trial. Cancer Res. 2015;76 (4 suppl; abstr S6-01).
97. NCT02437318. Study assessing the efficacy and safety of alpelisib
plus fulvestrant in men and postmenopausal women with advanced
breast cancer which progressed on or after aromatase inhibitor
treatment. (SOLAR-1). https://clinicaltrials.gov/ct2/show/NCT02437318?
term=alpelisib&rank=1. Accessed March 3, 2016.
98. NCT02340221. SANDPIPER Study: A study of taselisib + fulvestrant
versus placebo + fulvestrant in patients with advanced or metastatic
breast cancer who have disease recurrence or progression during or
after aromatase inhibitor therapy. https://clinicaltrials.gov/ct2/
show/NCT02340221. Accessed March 8, 2016.
99. Finn RS, Crown JP, Lang I, et al. The cyclin-dependent kinase 4/6
inhibitor palbociclib in combination with letrozole versus letrozole
alone as first-line treatment of oestrogen receptor-positive, HER2negative, advanced breast cancer (PALOMA-1/TRIO-18): a randomised phase 2 study. Lancet Oncol. 2015;16:25-35.
100. NCT01740427. A study of palbociclib (PD-0332991) + letrozole vs. letrozole
for 1st line treatment of postmenopausal women with ER+/HER2- advanced
101.
102.
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
breast cancer (PALOMA-2). http://www.cancer.gov/clinicaltrials/search/

view?cdrid=743730&version=HealthProfessional&protocolsearchid=
6378902. Accessed May 16, 2014.
Turner NC, Ro J, Andre F, et al. Palbociclib in hormone-receptorpositive advanced breast cancer. N Engl J Med. 2015;373:209-219.
NCT02513394. PALbociclib CoLlaborative Adjuvant Study: a randomized
phase III trial of palbociclib with standard adjuvant endocrine therapy versus standard adjuvant endocrine therapy alone for hormone
receptor positive (HR+) / human epidermal growth factor receptor 2
(HER2)-negative early breast cancer (PALLAS). https://clinicaltrials.
gov/show/NCT02513394undefined. Accessed March 3, 2016.
NCT01864746. A study of palbociclib in addition to standard endocrine treatment in hormone receptor positive Her2 normal patients
with residual disease after neoadjuvant chemotherapy and surgery
(PENELOPE-B). http://clinicaltrials.gov/show/NCT01864746. Accessed
May 16, 2014.
NCT02028507. Phase III study of palbociclib (PD-0332991) in combination with exemestane versus chemotherapy (capecitabine) in
hormonal receptor (HR) positive/HER2 negative metastatic breast
cancer (MBC) patients with resistance to non-steroidal aromatase
inhibitors (PEARL). http://clinicaltrials.gov/ct2/show/NCT02028507.
Accessed May 16, 2014.
NCT01919229. A pharmacodynamics pre-surgical study of LEE011 in
early breast cancer patients (MONALEESA-1). http://clinicaltrials.gov/
show/NCT01919229. Accessed May 27, 2014.
NCT01958021. Study of efficacy and safety of LEE011 in postmenopausal women with advanced breast cancer: (MONALEESA-2).
http://clinicaltrials.gov/show/NCT01958021. Accessed May 27, 2014.
Bardia A, Modi S, Chavez-Mac M, et al. Phase Ib/II study of LEE011,
everolimus, and exemestane in postmenopausal women with ER
+/HER2- metastatic breast cancer. J Clin Oncol. 2014;32 (suppl;
abstr 535).
Juric D, Ismail-Khan R, Campone M, et al. Phase Ib/II study of ribociclib
and alpelisib and letrozole in ER+, HER2 breast cancer: Safety,
preliminary efficacy and molecular analysis. Cancer Res. 2016;76
(4 suppl; abstr P3-14-01).
Patnaik A, Rosen LS, Tolaney SM, et al. Clinical activity of
LY2835219, a novel cell cycle inhibitor selective for CDK4 and
CDK6, in patients with metastatic breast cancer. Cancer Res. 2014;
74 (suppl; abstr CT232).
NCT02102490. A study of abemaciclib (LY2835219) in participants
with previously treated breast cancer that has spread (MONARCH 1).
https://clinicaltrials.gov/ct2/show/NCT02102490?term=ABEMACICLIB&
rank=16. Accessed March 3, 2016.
Patnaik A, Rosen LS, Tolaney SM, et al. LY2835219, a novel cell cycle
inhibitor selective for CDK 4/6, in combination with fulvestrant for
patients with hormone receptor positive (HR+) metastatic breast
cancer. J Clin Oncol. 2014;32 (suppl; abstr 534).
NCT02107703. A study of abemaciclib (LY2835219) combined with
fulvestrant in women with hormone receptor positive HER2 negative
breast cancer (MONARCH 2). https://clinicaltrials.gov/ct2/show/
NCT02107703. Accessed March 26, 2016.
NCT02236572. Neoadj ph 2 AI plus everolimus in postmenopausal
women w/ ER pos/HER2 neg, low risk score. https://clinicaltrials.gov/
ct2/show/NCT02236572. Accessed March 26, 2016.
NCT02308020. A study of abemaciclib (LY2835219) in participants with
breast cancer, non-small cell lung cancer, or melanoma that has spread to
the brain. https://clinicaltrials.gov/ct2/show/NCT02308020?term=
abemaciclib+brain&rank=1. Accessed March 3, 2016.
NCT01484041. Dovitinib plus an aromatase inhibitor for metastatic
breast cancer. https://clinicaltrials.gov/show/NCT01484041. Accessed
January 31, 2016.
116. NCT01723774. PD 0332991 and anastrozole for stage 2 or 3 estrogen

receptor positive and HER2 negative breast cancer. http://www.
cancer.gov/clinicaltrials/search/view?cdrid=742660&version=
HealthProfessional&protocolsearchid=12773356. Accessed May
27, 2015.
117. Yardley DA, Ismail-Khan RR, Melichar B, et al. Randomized phase II,
double-blind, placebo-controlled study of exemestane with or
without entinostat in postmenopausal women with locally recurrent
or metastatic estrogen receptor-positive breast cancer progressing on
treatment with a nonsteroidal aromatase inhibitor. J Clin Oncol. 2013;
31:2128-2135.
118. Connolly RM, Zhao F, Miller K, et al. E2112: a randomized phase III trial
of endocrine therapy plus entinostat/placebo in patients with hormone receptor-positive advanced breast cancer. J Clin Oncol. 2015;33
(suppl; abstr TPS636).
119. NCT02374099. Study to assess the efficacy and safety of the epigenetic modifying effects of CC-486 (oral azacitidine) in combination
with Fulvestrant. https://clinicaltrials.gov/ct2/show/NCT02374099?
term=vidaza+and+breast+cancer&rank=5. Accessed January 31, 2016.
120. NCT01349959. Azacitidine and entinostat in treating patients with
advanced breast cancer. https://clinicaltrials.gov/ct2/show/NCT01349959?
term=vidaza+and+breast+cancer&rank=1. Accessed January 31, 2016.
121. Robertson JF, Ferrero JM, Bourgeois H, et al. Ganitumab with either
exemestane or fulvestrant for postmenopausal women with advanced, hormone-receptor-positive breast cancer: a randomised,
controlled, double-blind, phase 2 trial. Lancet Oncol. 2013;14:
228-235.
122. Dickler MN, Barry WT, Cirrincione CT, et al. Phase III trial evaluating
the addition of bevacizumab to letrozole as first-line endocrine
therapy for treatment of hormone-receptor positive advanced breast
cancer: CALGB 40503 (Alliance). J Clin Oncol. 2015;33 (suppl; abstr
501).
123. Martn M, Loibl S, von Minckwitz G, et al. Phase III trial evaluating the
addition of bevacizumab to endocrine therapy as first-line treatment
for advanced breast cancer: the letrozole/fulvestrant and avastin
(LEA) study. J Clin Oncol. 2015;33:1045-1052.
124. Rugo HS, Delord JP, Im SA, et al. Preliminary efficacy and safety of
pembrolizumab (MK-3475) in patients with PD-L1positive, estrogen
receptor-positive (ER+)/HER2-negative advanced breast cancer enrolled in KEYNOTE-028. Cancer Res. 2016;76 (4 suppl; abstr S5-07).
125. Dirix LY, Takacs I, Nikolinakos P, et al. Avelumab (MSB0010718C), an
anti-PD-L1 antibody, in patients with locally advanced or metastatic
breast cancer: a phase Ib JAVELIN solid tumor trial. Cancer Res. 2016;
76 (suppl 4; abstr S1-04).
126. Munster PN, Hamilton EP, Franklin C, et al. Phase Ib study of LEE011
and BYL719 in combination with letrozole in estrogen receptorpositive, HER2-negative breast cancer (ER+, HER2- BC). J Clin Oncol.
2014;32:5s (suppl; abstr 533).
127. Andre F. UNIRAD. Randomised, double blind, multicentre phase III
trial evaluating the safety and benefit of adding everolimus to adjuvant hormone therapy in women with high risk of relapse, ER+ and
HER2- primary breast cancer who remain free of disease after receiving at least 1 year of adjuvant hormone therapy. http://www.icr.
ac.uk/our-research/our-research-centres/clinical-trials-andstatistics-unit/clinical-trials/unirad. Accessed January 30, 2016.
128. NCT01674140. SWOG 1207. Phase III randomized, placebo-controlled
clinical trial evaluating the use of adjuvant endocrine therapy +/2 one
year of everolimus in patients with high-risk, hormone receptorpositive and HER2/neu negative breast cancer. e3 breast cancer
study- evaluating everolimus with endocrine therapy. http://swog.org/
Visitors/ViewProtocolDetails.asp?ProtocolID=2248. Accessed January 30,
2016.
e53
129. NCT02308020. A study of abemaciclib (LY2835219) in participants

with breast cancer that has spread to the brain. https://clinicaltrials.
gov/ct2/show/record/NCT02308020?term=abemaciclib&rank=6.
Accessed January 30, 2016.
130. NCT02088684. Study of LEE011 with fulvestrant and BYL719 or
BKM120 in advanced breast cancer. http://www.cancer.gov/
clinicaltrials/search/view?cdrid=759365&version=HealthProfessional&
protocolsearchid=12773476. Accessed May 27, 2014.
131. NCT01872260. Study of LEE011. BYL719 and letrozole in advanced ER+
breast cancer. http://www.cancer.gov/clinicaltrials/search/view?cdrid=
750319&version=HealthProfessional&protocolsearchid=
12773476. Accessed May 27, 2014.
132. NCT02057133. A study of LY2835219 in combination with therapies
for breast cancer that has spread. https://clinicaltrials.gov/ct2/show/
record/NCT02057133. Accessed May 27, 2014.
133. NCT02117648. A study of LY2835219 in participants with cancer.
https://clinicaltrials.gov/ct2/show/NCT02117648. Accessed March
26, 2016.
134. NCT02441946. A neoadjuvant study of abemaciclib (LY2835219) in
postmenopausal women with hormone receptor positive, HER2
negative breast cancer (neoMONARCH). https://clinicaltrials.gov/ct2/
show/NCT02441946?term=abemaciclib&rank=3. Accessed May 15,
2015.
135. Caldon CE, Sergio CM, Kang J, et al. Cyclin E2 overexpression is associated with endocrine resistance but not insensitivity to CDK2 inhibition in human breast cancer cells. Mol Cancer Ther. 2012;11:
1488-1499.
136. Casimiro MC, Velasco-Velazquez M, Aguirre-Alvarado C, et al. Overview of cyclins D1 function in cancer and the CDK inhibitor landscape:
past and present. Expert Opin Investig Drugs. 2014;23:295-304.
137. Finn RS, Crown JP, Lang I, et al. Final results of a randomized phase II
study of PD 0332991, a cyclin-dependent kinase (CDK-4/6 inhibitor), in
combination with letrozole vs letrozole alone for first-line treatment
of ER+/HER2- advanced breast cancer (PALOMA-1; TRIO-18). Cancer
Res. 2014;74 (suppl; abstr CT101).
138. NCT01320592. PD0332991 and paclitaxel in treating patients with
metastatic breast cancer. http://www.cancer.gov/clinicaltrials/
search/view?cdrid=697909&version=HealthProfessional&protocolsearchid=
12773356. Accessed May 27, 2014.
139. NCT02040857. Palbociclib and endocrine therapy in treating patients
with hormone receptor-positive stage II-III breast cancer. http://
www.cancer.gov/clinicaltrials/search/view?cdrid=757030&version=
HealthProfessional&protocolsearchid=12773356. Accessed May 27,
2014.
140. Byrd JC, Peterson BL, Gabrilove J, et al; Cancer and Leukemia Group B.
Treatment of relapsed chronic lymphocytic leukemia by 72-hour
continuous infusion or 1-hour bolus infusion of flavopiridol: results
from Cancer and Leukemia Group B study 19805. Clin Cancer Res.
2005;11:4176-4181.
141. NCT01709370. Letrozole and CDK 4/6 inhibitor for ER positive, HER2
negative breast cancer in postmenopausal women. https://clinicaltrials.
gov/ct2/show/NCT01709370. Accessed March 26, 2016.
142. NCT02102490. A study of abemaciclib (LY2835219) in participants
with previously treated breast cancer that has spread (MONARCH 1).
26, 2016.
143. NCT01698918. Open-label, phase II, study of everolimus plus letrozole
in postmenopausal women with ER1, HER2- metastatic or locally
advanced breast cancer (Bolero-4). https://clinicaltrials.gov/ct2/
144. NCT01783444. A phase II study of everolimus in combination with
exemestane versus everolimus alone versus capecitabine in advance
e54
145.
146.
147.
148.
149.
150.
151.
152.
153.
154.
155.
156.
157.
158.
159.
breast cancer. (BOLERO-6). https://clinicaltrials.gov/ct2/show/

NCT00570921. Study of combined fulvestrant and everolimus in
advanced/metastatic breast cancer after aromatase inhibitor failure
(BRE-43). https://clinicaltrials.gov/ct2/show/NCT00570921. Accessed
March 26, 2016.
NCT02344550. Study of GnRH plus leterozole +/2 everolimus for
premenopausal women with metastatic breast cancer (LEO). https://
clinicaltrials.gov/ct2/show/NCT02344550. Accessed March 26, 2016.
NCT02404051. Fulvestrant and EVerolimus Plus EXemestane in
metastatic breast cancer (FEVEX). https://clinicaltrials.gov/ct2/show/
NCT01797120. Study of fulvestrant +/2 everolimus in post-menopausal,
hormone-receptor 1 metastatic breast ca resistant to AI (PrE0102). https://
NCT02123823. A phase Ib/II randomized study of BI 836845 in combination with exemestane and everolimus versus exemestane and
everolimus alone in women with locally advanced or metastatic breast
cancer. https://clinicaltrials.gov/ct2/show/record/NCT02123823?term5
BI1836845&rank53. Accessed March 26, 2016.
NCT02273973. A study of neoadjuvant letrozole 1 GDC-0032 versus
letrozole 1 placebo in post-menopausal women with breast cancer
(LORELEI). https://clinicaltrials.gov/ct2/show/NCT02273973. Accessed
March 26, 2016.
NCT01633060. A phase III study of BKM120 with fulvestrant in
patients with HR1, HER2-, AI treated, locally advanced or metastatic breast cancer who progressed on or after mTORi (BELLE-3).
26, 2016.
NCT02297438. A study of palbociclib (PD-0332991) 1 letrozole vs.
placebo1 letrozole for 1st line treatment of Asian postmenopausal women with ER1/HER2- advanced breast cancer [PALOMA-4].
26, 2016.
NCT02296801. A phase II randomized study evaluating the biological
and clinical effects of the combination of palbociclib with letrozole
as neoadjuvant therapy in post-menopausal women with estrogenreceptor positive primary breast cancer (PALLET). https://clinicaltrials.
NCT02690480. Comparison of efficacy and tolerability of fulvestrant1
placebo vs fulvestrant1palbociclib as first line therapy for postmenopausal women with HR1 metastatic BC treated with 5 years of
hormonal therapy remaining disease free more than 12 months after
completion or have de novo metastatic disease (FLIPPER). https://
NCT02491983. Study to evaluate the efficacy and safety of palbociclib
in combination with fulvestrant or letrozole in patients with ER1,
HER2- locally advanced or metastatic breast cancer (PARSIFAL). https://
NCT02384239. A study of palbociclib in combination with fulvestrant
or tamoxifen as treatment for metastatic breast cancer. https://
NCT02592746. A study of palbociclib with exemestane plus goserelin
versus capecitabine in premenopausal women with HR1 MBC. https://
NCT02675231. A study of abemaciclib (LY2835219) in women with
HR1, HER21 locally advanced or metastatic breast cancer (monarcHER).
26, 2016.
NCT02422615. Study of efficacy and safety of LEE011 in postmenopausal
women with advanced breast cancer (MONALEESA-3). https://clinicaltrials.
BREAST CANCER
Treatment of Premenopausal
Women With Endocrine-Sensitive
Breast Cancer
CHAIR
Nancy E. Davidson, MD
University of Pittsburgh Cancer Institute
Pittsburgh, PA
SPEAKERS
Hatem A. Azim Jr., MD, PhD
Institut Jules Bordet
Brussels, Belgium
Kathryn J. Ruddy, MD, MPH
Mayo Clinic
Rochester, MN
TREATING PREMENOPAUSAL ENDOCRINE-SENSITIVE BREAST CANCER
Challenges in Treating Premenopausal Women with

Endocrine-Sensitive Breast Cancer
Hatem A. Azim Jr., MD, PhD, Nancy E. Davidson, MD, and Kathryn J. Ruddy, MD, MPH
OVERVIEW
For the hundreds of thousands of premenopausal women who are diagnosed annually with endocrine-sensitive breast
cancer, treatment strategies are complex. For many, chemotherapy may not be necessary, and endocrine therapy decision
making is paramount. Options for adjuvant endocrine regimens include tamoxifen for 5 years, tamoxifen for 10 years,
ovarian function suppression (OFS) plus tamoxifen for 5 years, and OFS plus an aromatase inhibitor for 5 years. There are
modest differences in efficacy between these regimens, with a benefit from OFS most obvious among patients with higher-risk
disease; therefore, choosing which should be used for a given patient requires consideration of expected toxicities and
patient preferences. An aromatase inhibitor cannot be safely prescribed without OFS in this setting. Additional research is
needed to determine whether genomic tests such as Prosigna and Endopredict can help with decision making about
optimal duration of endocrine therapy for premenopausal patients. Endocrine therapy side effects can include hot flashes,
sexual dysfunction, osteoporosis, and infertility, all of which may impair quality of life and can encourage nonadherence
with treatment. Ovarian function suppression worsens menopausal side effects. Hot flashes tend to be worse with
tamoxifen/OFS, whereas sexual dysfunction and osteoporosis tend to be worse with aromatase inhibitors/OFS. Pregnancy is safe after endocrine therapy, and some survivors can conceive naturally. Still, embryo or oocyte cryopreservation
should be considered at the time of diagnosis for patients with endocrine-sensitive disease who desire future childbearing,
particularly if they will undergo chemotherapy.
n the United States, approximately 20% of patients with

breast cancer are diagnosed before age 50.1 This translates
into almost 50,000 new patients per year, which well exceeds the total number of women diagnosed with acute
leukemia and brain and stomach cancers in all ages combined.1 A higher proportion of patients with breast cancer
are diagnosed at a younger age in other parts of the world,
such as Africa and the Middle East, where the average age of
developing breast cancer hardly exceeds 50.2
Over the past decade, several studies have indicated
that diagnosis at a younger age is associated with poorer
prognosis,3-5 especially for women with endocrine-sensitive
disease.3,5 Compared with patients with breast cancer who
are postmenopausal, younger patients are more commonly
diagnosed with highly proliferative estrogen receptor
positive tumors (i.e., luminal-B breast cancer).6 Furthermore, these tumors are enriched for molecular aberrations
that could render them more aggressive, including high
expression of stem cell markers and high prevalence of
GATA3 mutations,3,7 the latter of which has been implicated
in endocrine resistance. In addition, younger patients often
have unique challenges, most notably fertility-related issues,

poorer perceived quality of life, and difficulties adhering
to prescribed endocrine therapy. Therefore, treating premenopausal women with early breast cancer, particularly
those with endocrine-sensitive disease, is rather complex.
HOW TO CHOOSE ENDOCRINE THERAPY FOR

PREMENOPAUSAL PATIENTS?
In recent years, several trials have evaluated different
strategies of adjuvant endocrine therapy for premenopausal
patients. These include tamoxifen for 5 years, tamoxifen for
10 years, OFS plus tamoxifen for 5 years, and OFS plus an
aromatase inhibitor for 5 years. As reviewed below, modest
differences between these options in terms of long-term
outcome have been reported. These approaches have different
toxicity profiles. Deciding which approach should be used for
a given patient requires consideration of risk of recurrence,
fertility considerations, and patient preferences regarding
expected toxicities.8 Unfortunately, premenopausal patients
with breast cancer have been found to have poor persistence
From the Department of Medicine, Institut Jules Bordet, Universite Libre de Bruxelles, Brussels, Belgium; University of Pittsburgh Cancer Institute and UPMC Cancer Center, Pittsburgh,
PA; Department of Oncology, Mayo Clinic, Rochester, MN.
Corresponding author: Hatem A. Azim Jr., MD, PhD, Institut Jules Bordet, Boulevard de Waterloo, 121, 1000 Brussels, Belgium; email: hatemazim@icloud.com.
23
AZIM, DAVIDSON, AND RUDDY
with endocrine therapy.9,10 Some studies have shown that only

50% of women complete the 5 years of scheduled tamoxifen.9
Therefore, it is very important for clinicians to engage young
patients in therapeutic decision making and to discuss toxicities
and expectations. Salient points pertaining to the use of the
different endocrine therapy options are discussed in this article.
Figure 1 summarizes a schema to help optimize treatment
decisions in daily practice.
Value of Ovarian Function Syndrome

Induction of amenorrhea for premenopausal patients with
endocrine-sensitive breast cancer has consistently been
shown to reduce risk of recurrence.11,12 This is true for OFS
by luteinizing hormone-releasing hormones (LHRH) agonists
or surgical or radiation ablation. However, because adjuvant
tamoxifen was not routinely administered to patients who
were premenopausal until the early 2000s, uncertainty
remained regarding the benefit of amenorrhea induction in
the presence of tamoxifen. Within the last several years, two
studies have provided important insights on this issue. The
first is the E-3193, INT-0142 trial, which randomly assigned
node-negative, estrogen receptorpositive patients to tamoxifen or OFS plus tamoxifen, both given for 5 years.13 This
KEY POINTS
24
Several adjuvant endocrine therapy options are

currently available for premenopausal patients with
endocrine-responsive early-stage breast cancer, and it
can be challenging to decide which is best for a given
patient.
Recent evidence from randomized trials indicates that a
substantial proportion of premenopausal women with
early-stage breast cancer can achieve excellent
outcomes with endocrine therapy only, in the absence
of chemotherapy, challenging the notion that adjuvant
chemotherapy should be considered in all young
patients with breast cancer.
Careful attention must be paid to the toxicities of
endocrine therapies including hot flushes, sexual
dysfunction, osteoporosis, and infertility, which could all
reduce adherence to endocrine therapy and
detrimentally impact patients quality of life.
Pregnancy following endocrine-responsive breast
cancer is safe, although challenges exist in the era of
extended adjuvant therapy. A prospective trial currently
is addressing the feasibility of temporary interruption of
endocrine treatment to allow pregnancy.
Embryo or oocyte cryopreservation prior to
chemotherapy is the best option to preserve fertility for
premenopausal patients with breast cancer, even those
who have endocrine-sensitive disease. Evolving data
suggest that LHRH agonists could reduce risk of
chemotherapy-induced amenorrhea and, thus, may
serve as a valid alternative to embryo or oocyte
cryopreservation if these are not feasible.
study did not accrue as planned, and it closed early with only
345 patients. Notably, 85% of patients had T1 disease and,
per protocol, none were treated with adjuvant chemotherapy. This underpowered study showed no difference in
10-year disease-free survival (DFS; 85.9% vs. 87.1%; p = .62)
or overall survival (OS; 92.5% vs. 92.4%; p = .67) between
both arms. The Suppression of Ovarian Function Trial (SOFT)
was a larger, properly powered trial that included 2,033
patients who were randomly assigned to either tamoxifen
with OFS or tamoxifen alone.14 In this trial, the main analysis
indicated no significant difference in 5-year DFS between the
OFS plus tamoxifen and tamoxifen alone arms (84.7% vs.
86.6%; p = .10). Subgroup analyses suggested a benefit with
the addition of OFS in women who had received chemotherapy (who were more likely to have node-positive disease) and those younger than age 35 at diagnosis. Of those
diagnosed with breast cancer younger than age 35 (94% of
whom received chemotherapy), the proportion who were
free of breast cancer for at least 5 years was 67.7% with
tamoxifen alone and 78.9% with tamoxifen with OFS, suggesting that OFS cut the risk of recurrence by approximately
one-third in this group.
These results underscore several important points. First,
endocrine therapy alone, without chemotherapy, is a very
effective option for selected premenopausal women with
early-stage breast cancer. Second, in older premenopausal
women with low-risk early-stage breast cancer, OFS adds
little if any benefit to 5 years of tamoxifen. Finally, induction
of amenorrhea is therapeutic for younger premenopausal
patients with higher-risk disease.
Tamoxifen Versus Aromatase Inhibitors

In the postmenopausal setting, aromatase inhibitors have
largely replaced tamoxifen as the standard endocrine adjuvant therapy.15 Among premenopausal patients, the final
results of the ABCSG-12 trial did not show a difference in DFS
between tamoxifen and anastrozole when administered for
3 years in combination with LHRH agonists, but OS was inferior
in the aromatase inhibitor arm.16 This raised concerns about
the use of aromatase inhibitors with OFS among premenopausal patients. Recently, the results of a joint analysis
of the long-awaited SOFT and TEXT trials were published.17 In
total, 4,690 women were randomly assigned to receive either
tamoxifen or exemestane in addition to OFS for 5 years.
Approximately 40% of patients had node-positive disease and
nearly 50% received adjuvant chemotherapy. The results
demonstrated a 28% reduction in DFS hazard ratio (HR) at
5 years with exemestane compared to tamoxifen (91.1% vs.
87.3%, p , .001), with no difference in OS.
Although in line with the hypothesis and results observed
in the postmenopausal setting, the results of this pooled
analysis contrasted with those observed in the ABCSG-12
trial. However, it is important to note that the ABCSG-12 trial
was smaller in size and administered therapy for only
3 years, which does not reflect todays standard practice. An
unplanned analysis of ABCSG-12 suggested that the worse
FIGURE 1. Schema to Guide Decision Making of Primary Systemic Therapy of Premenopausal Women With
Endocrine ReceptorPositive (HER-Negative) Breast Cancer
Abbreviations: AI, aromatase inhibitor; ER, estrogen receptor; LHRH, luteinizing hormone-releasing hormone; OFS, ovarian function suppression; PR, progesterone receptor; y, years.
*No evidence on extended adjuvant therapy in women treated with adjuvant OFS/AI during first 5 years.
**Clinical utility of genomic tests to guide extended endocrine therapy is yet to be proven.
outcome observed in the aromatase inhibitor arm was restricted to patients who were overweight or obese18; these
patients constituted nearly one-third of the study population.
This was hypothesized to reflect incomplete estrogen level
suppression by aromatase inhibitor for these women, which
has been seen among postmenopausal patients who were
obese,19 or incomplete ovarian suppression by LHRH agonists
in larger patients. Research is limited regarding the impact of
body weight on the efficacy of OFS, but it is possible that
surgical or radiation ablation would be more effective than
LHRH agonists in women who are overweight and obese.20
Alternatively, cross talk between estrogen signaling and insulin growth factor signaling may play a role in the impact of
weight on the efficacy of OFS and AI.21 A randomized prospective study from Japan in the neoadjuvant setting showed
superiority of aromatase inhibitor plus OFS over tamoxifen
plus OFS with a near doubling of clinical response rate on
MRI.22 However, in this study, only 17% of patients were
overweight or obese. Correlation between body mass index
(BMI) and outcome in the joint analysis of SOFT and TEXT is yet
to be published, but high BMI was shown to be associated
with high on-treatment estradiol levels in SOFT patients
randomly assigned to exemestane and OFS.23 Although most
SOFT patients maintained a low estradiol level on treatment,
34% of patients on OFS and aromatase inhibitor had high
estradiol levels at some point during the 12 months after
study initiation. No correlation with long-term outcome was
seen, perhaps because of the small number of patients in this
analysis (86 patients on LHRH agonist plus exemestane).
Thus, although the combination of OFS and exemestane

may be the most effective 5-year endocrine therapy regimen, adequate suppression of estradiol levels appears to be
more challenging for patients who are overweight or obese,
in whom ABCSG-12 suggests that OFS with tamoxifen may
produce equal if not better outcomes. Yet it is unclear
whether measurement of estradiol levels during LHRH agonists therapy will be clinically useful in guiding the choice of
adjuvant endocrine therapy in daily practice.24
One final point that should be underscored is the very
favorable outcome of the 50% of women in the joint analysis
who did not receive adjuvant chemotherapy. Although these
patients were mostly low risk by virtue of stage, their 5-year
DFS was greater than 93%. Because the benefit of adjuvant
chemotherapy is largely a reduction in risk of recurrence
during the first few years after diagnosis, it is likely that
chemotherapy would not have been beneficial in the majority of these patients. These results strongly challenge the
notion that young premenopausal women should be offered
more chemotherapy, and they indicate that endocrine
therapy alone remains a very effective option for a large
fraction of young patients with endocrine-sensitive breast
cancer.
10-Year Versus 5-Year Schedule

Unlike estrogen receptornegative tumors, endocrinesensitive cancers have a propensity for late relapse. 25
Therefore, extending treatment beyond 5 years seems
25
logical to improve the outcomes of these patients. In the

ATLAS and aTTom trials,26,27 taking tamoxifen for 10 years
was shown to be superior to the standard 5-year schedule,
and benefit was observed independent of menopausal
status. Given the established role of aromatase inhibitors in
treating postmenopausal women during the first 5 years,
these results appear to be most relevant for the treatment
of premenopausal women for whom tamoxifen is still
considered a mainstay treatment. Another study, MA17,
which previously showed the added benefit of extended
therapy with letrozole after 5 years of tamoxifen, demonstrated that this benefit is mostly observed in women who
were premenopausal at the time they initiated tamoxifen
but became postmenopausal during its administration.28
These results in aggregate underscore the value of considering extending endocrine therapy to 10 years for premenopausal women. However, the question remains as to
whether all women need this approach. This question is
relevant as prolonged treatment increases toxicity and could
impair quality of life. Acknowledging the adherence issues
highlighted earlier, it is important to define tools that can
identify the subsets of patients who will benefit the most
from this approach and spare others the burden of long-term
treatment.
Recently, several genomic tests, particularly Prosigna and
Endopredict, have been evaluated for their ability to predict
long-term recurrence in women who were treated with
5 years of adjuvant endocrine therapy in the context of
randomized phase III trials.29-31 Notably, these trials were
conducted among postmenopausal women and genomic
testing was not part of the initial study plan. In the ABCSG-8
trial, which randomly assigned postmenopausal patients to
receive either tamoxifen or anastrozole, approximately
1,300 patients were free of relapse at completion of 5 years
of endocrine therapy and were analyzed by Prosigna testing.29 None of these patients received adjuvant chemotherapy and 30% were node-positive. Thirty-seven percent
of patients were classified as low risk by the genomic test, of
whom only 2.4% developed distant recurrence. In contrast,
17.5% of the 30% of patients classified as high risk developed
distant recurrence between years 5 and 15. Similar results
were observed in a combined analysis of the ABCSG-8 and
ATAC datasets.30 Endopredict showed comparable findings
when tested on the ABCSG-6 and ABCSG-8 trial. In 1,702
patients free of relapse after 5 years of endocrine therapy,
Endopredict identified 64% as low risk, with an absolute risk
of recurrence of 1.8% at 10 years.31
Thus, it appears that extending endocrine therapy beyond
5 years for women in the low-risk group category is very
unlikely to add any benefit because their risk of recurrence is
negligible. On the other hand, extended therapy could be more
useful for patients classified as high risk. Of note, none of these
studies included premenopausal patients, yet data with other
genomic tests such as Oncotype Dx32 and an in silico analysis3
indicated that the prognostic performance of genomic tests
is largely comparable irrespective of age. Currently, several
studies investigating extended endocrine therapy have yet to
26
report, and thus validation of the role of genomic tests to

inform on the benefit of this approach is warranted. Until then,
no solid statements could be made regarding their role in
guiding the use of extended endocrine therapy.
TOXICITIES OF ENDOCRINE THERAPY FOR

PREMENOPAUSAL WOMEN
Premenopausal women with breast cancer who require
endocrine therapy are vulnerable to a variety of treatment
toxicities. Although excess tamoxifen-associated uterine
cancer diagnoses and thromboembolic events have not been
observed among patients younger than age 50, some side
effects of endocrine therapy are more severe for young
patients, particularly if OFS is used. These side effects are
discussed below.
Hot Flashes
Hot flashes are problematic for the majority of young recipients of adjuvant endocrine therapy,33 especially with the
addition of OFS,34 and perhaps more substantially with
tamoxifen than aromatase inhibitors. These hot flashes
often are accompanied by night sweats, which can interrupt
sleep and impair quality of life. In 345 young women enrolled
in the E-3193, INT-0142 phase III trial, grade 3 hot flashes
occurred in 4.7% of those receiving tamoxifen monotherapy
compared with 16.1% of those receiving tamoxifen with
OFS.13 In SOFT, grade 3 or 4 hot flashes occurred in 7.6% of
patients assigned to receive tamoxifen, compared with
13.2% of those assigned to OFS and tamoxifen.14 However,
the use of an aromatase inhibitor instead of tamoxifen with
OFS was associated with less sweating (reported by 54.5%
vs. 59%, respectively).17
Sexual Dysfunction
In contrast, sexual dysfunction is most severe for patients
who received an aromatase inhibitor with OFS.17 In the TEXT
trial, 52.4% of 2,318 women who received exemestane with
OFS and 47.4% of 2,325 women who received tamoxifen
with OFS had vaginal dryness. Forty-five percent of the
exemestane/OFS group and 40.9% of the tamoxifen/OFS
group had decreased libido, although dyspareunia was reported in 30.5% and 25.8%, respectively. Of note, ovarian
dysfunction secondary to chemotherapy can cause similar
vaginal symptoms and sexual dysfunction.35 Tamoxifen
alone is less problematic,13 and it may even improve vaginal
dryness in some survivors by causing vaginal discharge.34
Sexual dysfunction in survivors of breast cancer can be
compounded by body image issues associated with local
therapies and the weight gain that many experience during
systemic chemotherapy.
Depression and Cognitive Dysfunction

SOFT results showed a nonstatistically significant trend
toward more mood impairment in those who received tamoxifen with OFS than in those who received tamoxifen
monotherapy, with depressive symptoms identified in

51.9% and 46.6%, respectively.14 An aromatase inhibitor
does not appear to cause more depression than tamoxifen.17
Treatment of depression can be complex for patients receiving tamoxifen because of concerns that certain antidepressants (e.g., fluoxetine, paroxetine, and sertraline)
may reduce the efficacy of tamoxifen by inhibiting
CYP2D6.36,37 One study found that survivors of breast cancer
who took paroxetine with tamoxifen had increased breast
cancer mortality, although another study identified no impact of CYP2D6 inhibitor use on recurrence risk for patients
taking tamoxifen. Possibly, the CYP2D6 genotype is particularly relevant for tamoxifen efficacy for premenopausal
patients,38 but definitive studies of this issue are needed.
Cognitive dysfunction related to endocrine therapy is
understudied among premenopausal patients. However, in
the ZIPP trial, patient self-evaluation of memory and concentration was similar for patients with breast cancer who
received chemotherapy with or without goserelin, goserelin
plus tamoxifen, or tamoxifen alone.34 In addition, preliminary results at 1 year from 86 of a planned 321 participants in the cognitive function substudy of SOFT
demonstrated no worse objective cognitive function in
those who received tamoxifen plus OFS compared with
tamoxifen monotherapy.39
Bone Health
Tamoxifen is known to improve bone mineral density for
postmenopausal women, but it can thin bones in premenopausal women.40 This is likely because of its mixed
agonist/antagonist effect on estrogen receptor. Therefore,
bone loss is a concern in most young patients with breast
cancer, not just in those who experience chemotherapyrelated premature menopause or receive OFS. A recent
study of survivors of breast cancer who were premenopausal
at the time of breast cancer diagnoses and had no known
history of osteoporosis revealed a 12% fracture risk during a
median 3.1-year follow-up after the completion of 56 years
of endocrine therapy. This was similar to the 15% fracture
risk this study identified in survivors of postmenopausal
breast cancer.41 However, the fractures in survivors of
premenopausal breast cancer were more commonly toe/
finger fractures, although those in postmenopausal women
were more commonly hip fractures. The median time to
first fracture was shorter in premenopausal than in postmenopausal women (1.4 vs. 2.4 years; p = .01). Zoledronic
acid has been proven an effective preventative agent against
bone loss in this population.42,43 The lifetime fracture risk of
young recipients of endocrine therapy deserves additional
study in the new era of increased use of OFS and prolonged
duration of treatment.
Cardiovascular Toxicity
Although it occurs at low frequency in young patients,
cardiovascular toxicity also must be considered. Hypertension developed more frequently in SOFT participants who
received OFS with tamoxifen compared with those who

received tamoxifen monotherapy,14 identified in 23.3%
versus 17.2% of patients, respectively. But among TEXT
and SOFT participants who received OFS, hypertension
was equally likely with tamoxifen versus exemestane.17
Thrombosis or embolism occurred in 2.2% (95% CI, 1.62.8)
of patients who received OFS plus tamoxifen and 1.0% (95%
CI, 0.71.5) of those who received OFS plus aromatase
inhibitors.17
PREGNANCY AFTER ESTROGEN

RECEPTORPOSITIVE BREAST CANCER: SAFETY
AND FEASIBILITY IN THE ERA OF EXTENDED
ADJUVANT THERAPY
Infertility remains a main concern for young patients with
breast cancer. A recent prospective study including more
than 600 patients demonstrated that up to 50% of young
patients with breast cancer are concerned about infertility,
and, in a substantial fraction, this has an impact on treatment decisions.44
The safety of subsequent pregnancy in women with a
history of estrogen receptorpositive breast cancer continues to be an area of concern. In a recent survey, approximately 40% of oncologists believed that hormonal
changes secondary to pregnancy could increase risk of recurrence particularly during the first 2 years after cancer
diagnosis.45 Although this question has been addressed in
several studies over the past 3 to 4 decades,46-48 definitive
answers were complicated by lack of information on estrogen receptor status in many of these studies and
methodological concerns about patient selection. In 2013, a
large multicenter study was published in which the longterm outcome of 333 women who became pregnant after
breast cancer diagnosis was compared with 874 survivors of
breast cancer who did not become pregnant.49 Both groups
were matched according to relevant tumor characteristics
and use of adjuvant systemic therapy. The primary endpoint
was 5-year DFS of the estrogen receptorpositive cohort. In
this cohort, which comprised 194 pregnant and 492 nonpregnant patients, no difference in 5-year DFS was observed
by pregnancy status (HR 0.91; 95% CI, 0.671.24; p = .55).
Secondary endpoints, which included DFS in estrogen
receptornegative patients and OS, also did not differ between the pregnant and nonpregnant groups. These results
underscore the safety of pregnancy following breast cancer
irrespective of the estrogen receptor status of the primary
tumor. Similar to other studies,47 approximately 30% of
patients underwent abortion in this study. A preplanned
subgroup analysis showed that abortion did not have any
impact on patient prognosis, indicating that abortion should
not be recommended for therapeutic purposes.
However, with the increasing use of long-term endocrine
therapy up to 10 years, the feasibility of becoming pregnant
after completion of 510 years of endocrine therapy is a
major challenge. This raises the question of whether temporary interruption of endocrine therapy to allow pregnancy
27
is safe. This approach is sometimes considered in selected

cases, particularly for patients with low risk of recurrence or
advanced age at the time of diagnosis. Presently, it is not
known whether this approach is detrimental because it was
not investigated in any of the reported studies. As such, the
first international prospective trial (POSITIVE) was recently
launched to address the safety of temporary interruption of
endocrine therapy to allow pregnancy in young women
with estrogen receptorpositive breast cancer who wish to
become pregnant.50 This trial allows patients to receive
1830 months of endocrine therapy. Patients with adequate
ovarian function can temporarily interrupt endocrine therapy
for up to 2 years to allow pregnancy with or without breastfeeding. Afterward, they are encouraged to resume endocrine
therapy for a total course of 510 years according to local
practice. POSITIVE is sponsored by the International Breast
Cancer Study Group and is performed in collaboration with
ALLIANCE in the United States. This study is expected to recruit
approximately 500 patients, which would provide solid evidence for the safety and feasibility of endocrine therapy interruption to allow pregnancy.
ASSISTED REPRODUCTION TO BECOME

PREGNANT AFTER BREAST CANCER: TIMING
AND IMPACT ON OUTCOME
Given the progressive decline in ovarian function with age51
and the detrimental impact of chemotherapy on ovarian
function,52 fertility preservation must be considered and
offered soon after breast cancer diagnosis to improve the
chances of future conception. Ovarian stimulation for oocyte
or embryo cryopreservation is a standard procedure. It is
currently endorsed by several guidelines for patients who
are newly diagnosed with breast cancer and are also considering future pregnancy.53-55 However, several concerns
hinder the wide application of this approach.
Potential Delay of Adjuvant Systemic Therapy

Optimally, ovarian stimulation should be performed before
the initiation of adjuvant therapy to precede the destruction of oocytes by cytotoxic agents. Standard stimulation protocols are classicaly applied during the first days
of the menstrual cycle, which can result in a delay in initiation of chemotherapy as some woman can have to wait
up to 3 weeks for stimulation. Such a delay could be potentially detrimental to patient outcome because the interval between surgery and initiation of chemotherapy is
crucial, particularly in treating young patients with breast
cancer.56 Over the past 5 years, several studies were
published on random stimulation protocols, in which ovarian
stimulation is performed at any time during the menstrual
cycle.57-59 The number of days required for ovarian stimulation is slightly longer using the random protocol, on average 1 to 2 days more, yet the oocyte yield is almost
identical. This approach increases the feasibility of performing ovarian stimulation prior to starting systemic
therapy without substantial delay.
28
Potential Detrimental Impact on Breast Cancer

Outcome Particularly in Patients With Estrogen
ReceptorPositive Disease
This is perhaps the greatest concern when considering assisted
reproduction in young patients with breast cancer, as this
procedure is associated with significant rise in serum estradiol
levels. Despite the fact that estradiol peaks only last for a few
days, defining protocols that would reduce estradiol peaks
without majorly compromising oocyte yield is desirable. Oktay
et al have performed a series of studies investigating the
addition of letrozole to standard ovarian stimulation protocols
to reduce estradiol peaks.60,61 They showed that this strategy
was associated with a comparable oocyte yield and significantly
lower estradiol peaks. In one study, the mean estradiol level at
human chorionic gonadotropin (known as hCG) triggering was
483.4 pg/mL with letrozole compared with 1,464 pg/mL for
the standard protocol (p , .001), with similar numbers of
mature oocytes (8.7 vs. 9.7; p = .43) and fertilization rates
(74.1% vs. 73.2%; p = .71). In addition, there was no difference
in days needed for stimulation, which was approximately
12 days.61 Other studies have confirmed these results.62,63 In
terms of fertility and pregnancy outcome, recent results from
131 patients with breast cancer who underwent ovarian
stimulation with letrozole before initiating adjuvant chemotherapy showed that 40 patients attempted to transfer
embryos back at an average time of 5.2 years following initial
embryo cryopreservation. The overall live birth rate was 45%,
which is comparable to figures observed in the general population of women of the same age. Importantly, no congenital
malformations were observed in 25 children after a mean
follow-up of 40 months.64
To provide more robust evidence regarding the safety of
this approach, a prospective study was initiated to evaluate the association between using ovarian stimulation
with letrozole and long-term breast cancer outcome.62 In
this study, 337 young patients with breast cancer (median
age 35) were included, of whom 120 underwent ovarian
stimulation with letrozole. At a median follow-up of
5 years, there was no difference in DFS between patients
who underwent stimulation and those who did not.65 Notably, in this study, patients who underwent stimulation
initiated adjuvant chemotherapy later compared with the
control group (45 days vs. 33 days; p , .01). This is not
unexpected because this study did not offer random stimulation protocols. A subsequent analysis investigating patient
outcome according to the number of ovarian stimulation
cycles (one vs. two cycles) showed no difference in patient
prognosis with better fertility preservation cycle outcome in
those undergoing two cycles in terms of number of mature
oocytes (10.3 vs. 6.2; p = .004), fertilized oocytes (7.4 vs. 4.4;
p = .04) and embryos (6.4 vs. 3.7; p =.019).66
Although more data on larger series of patients with
breast cancer are still needed to establish the safety and
effectiveness of using ovarian stimulation with letrozole
in newly diagnosed patients, this approach is currently
preferred.
Feasibility and Safety of Ovarian Stimulation for

Patients Already Treated With Chemotherapy
p = .04), although absolute numbers remain low (33 vs. 19

pregnancies) in the combined results from five different
trials.72
Although these data suggest that giving LHRH agonists
before and during chemotherapy could improve chances of
future fertility for young patients with breast cancer,
several important considerations should be taken into
account. First, these trials evaluated chemotherapyinduced amenorrhea, which is an imperfect surrogate for
adequate ovarian function. Previous studies have shown
that even women who resume menses after chemotherapy have reduced ovarian reserve73 and will most
likely develop early menopause.74 Chemotherapy has
been shown to induce structural changes in the ovary
including hyalinization of cortical vessels, collagen deposition, and cortical fibrosis, even in women who
resume menses after chemotherapy.75 Whether the coadministration of LHRH agonists avoids any of these phenomena remains unclear. Second, long-term data are scant.
The PROMISE trial did report that the 5-year cumulative
incidence estimate of resumption of menses was 72.6% in
the LHRH agonist group versus 64% in the control group
(age-adjusted HR 1.48; 95% CI, 1.121.95; p = .006).76
Unfortunately, none of the published studies has adequate evaluation of ovarian function parameters over
time.
An individual-patient meta-analysis including all randomized trials of LHRH agonist in this setting currently is
underway. This will allow subgroup analyses according to
age and estrogen receptor status. Yet based on current data,
LHRH agonists can be considered during chemotherapy to
potentially preserve the fertility of women in whom ovarian
stimulation was not possible.
The postchemotherapy setting is suboptimal for ovarian

stimulation because of the detrimental effects of adjuvant
systemic therapy on ovarian function.52 Nevertheless, often
we are confronted by situations in which a patient did not
freeze embryos or oocytes before chemotherapy initiation
and wants to become pregnant but requires assisted reproduction to conceive. In the one study reported in this
setting, 25 women underwent ovarian stimulation some
years after primary chemotherapy to conceive. Their
outcome was compared with more than 170 women
who had spontaneous pregnancy after breast cancer
diagnosis.67 Approximately 50% of patients had endocrinesensitive disease. At a median follow-up of 4.5 years after
conception, no difference in breast cancer outcome was
observed between the two groups. This small study suggests that this approach may be considered in selected
patients who did not have fertility preservation at diagnosis, completed their adjuvant therapy, and cannot
conceive spontaneously.
USE OF CONCOMITANT ADMINISTRATION OF

LHRH AGONISTS WITH CHEMOTHERAPY TO
PRESERVE FERTILITY
Several randomized studies have addressed the role of LHRH
agonists as a mean of preserving fertility with conflicting
results.68-71 Table 1 summarizes results from the main trials,
the largest of which, PROMISE and POEMS, showed reduced
rates of chemotherapy-induced amenorrhea with the concurrent administration of LHRH agonists.70,71 A recent metaanalysis of published data showed higher odds of achieving
pregnancy as well (odds ratio 1.83; 95% CI, 1.023.28;
TABLE 1. Main Randomized Trials That Investigated the Role of LHRH Agonist to Reduce Chemotherapy-Induced
Amenorrhea
PROMISE71,76
GBG 37 ZORO68
NCT0009084469
POEMS70
Country
Italy
Germany
United States
International
Number of Patients
281
61
49
135
Median Age
39
36
39
38
Chemotherapy
Any chemotherapy
AC-T
AC-T or FEC
Any chemotherapy
LHRH Agonists/28d
Triptorelin
Goserelin
Triptorelin
Goserelin
ER Status
Any
Negative
Any
Negative
Median Follow-up
7.3 years
NR
1.5 years
4.1 years
Primary Endpoint (POF defined as)
Amenorrhea + postmenopausal
FSH and E2 at 12 months
Amenorrhea at 6
months
Amenorrhea at
12 months
Amenorrhea + postmenopausal
FSH at 24 months
Absence of POF (%)
LHRH: 91%
LHRH: 56%
LHRH: 88%
LHRH: 92%
No LHRH: 74%
No LHRH: 70%
No LHRH: 90%
No LHRH: 78%
Positive
Negative
Negative
Positive
LHRH: 8
LHRH: 1
LHRH: 0
LHRH: 22
No LHRH: 3
No LHRH: 1
No LHRH: 2
No LHRH: 12
Pregnancies Achieved (number)
Abbreviations: AC, doxorubicin, cyclophosphamide; E2, estradiol; ER, estrogen receptor; FEC, 5-fluorouracil, epirubicin, cyclophosphamide; FSH, follicle-stimulating hormone; LHRH,
luteinizing hormone-releasing hormone; POF, premature ovarian failure; T, taxane.
29
References
1. DeSantis C, Ma J, Bryan L, et al. Breast cancer statistics, 2013. CA Cancer
J Clin. 2014;64:52-62.
2. DeSantis CE, Bray F, Ferlay J, et al. International variation in female
breast cancer incidence and mortality rates. Cancer Epidemiol Biomarkers Prev. 2015;24:1495-1506.
3. Azim HA, Jr, Michiels S, Bedard PL, et al. Elucidating prognosis and
biology of breast cancer arising in young women using gene expression
profiling. Clin Cancer Res. 2012;18:1341-1351.
4. Kim EK, Noh WC, Han W, et al. Prognostic significance of young age (,35
years) by subtype based on ER, PR, and HER2 status in breast cancer:
a nationwide registry-based study. World J Surg. 2011;35:1244-1253.
5. Cancello G, Maisonneuve P, Rotmensz N, et al. Prognosis and adjuvant
treatment effects in selected breast cancer subtypes of very young
women (,35 years) with operable breast cancer. Ann Oncol. 2010;21:
1974-1981.
6. Azim HA, Jr, Partridge AH. Biology of breast cancer in young women.
Breast Cancer Res. 2014;16:427.
7. Azim HA, Jr, Nguyen B, Brohee S, et al. Genomic aberrations in young
and elderly breast cancer patients. BMC Med. 2015;13:266.
8. Mathew A, Davidson NE. Adjuvant endocrine therapy for premenopausal women with hormone-responsive breast cancer. Breast.
2015;24(Suppl 2):S120-S125.
9. Huiart L, Bouhnik AD, Rey D, et al. Early discontinuation of tamoxifen
intake in younger women with breast cancer: is it time to rethink the
way it is prescribed? Eur J Cancer. 2012;48:1939-1946.
10. Ruddy K, Mayer E, Partridge A. Patient adherence and persistence with
oral anticancer treatment. CA Cancer J Clin. 2009;59:56-66.
11. Cuzick J, Ambroisine L, Davidson N, et al; LHRH-agonists in Early Breast
Cancer Overview group. Use of luteinising-hormone-releasing hormone
agonists as adjuvant treatment in premenopausal patients with hormonereceptor-positive breast cancer: a meta-analysis of individual patient data
from randomised adjuvant trials. Lancet. 2007;369:1711-1723.
12. Swain SM, Jeong JH, Geyer CE, Jr, et al. Longer therapy, iatrogenic
amenorrhea, and survival in early breast cancer. N Engl J Med. 2010;
362:2053-2065.
13. Tevaarwerk AJ, Wang M, Zhao F, et al. Phase III comparison of tamoxifen
versus tamoxifen plus ovarian function suppression in premenopausal
women with node-negative, hormone receptor-positive breast cancer
(E-3193, INT-0142): a trial of the Eastern Cooperative Oncology Group.
J Clin Oncol. 2014;32:3948-3958.
14. Francis PA, Regan MM, Fleming GF, et al; SOFT Investigators; International Breast Cancer Study Group. Adjuvant ovarian suppression in
premenopausal breast cancer. N Engl J Med. 2015;372:436-446.
15. Dowsett M, Cuzick J, Ingle J, et al. Meta-analysis of breast cancer
outcomes in adjuvant trials of aromatase inhibitors versus tamoxifen.
J Clin Oncol. 2010;28:509-518.
16. Gnant M, Mlineritsch B, Stoeger H, et al. Zoledronic acid combined with
adjuvant endocrine therapy of tamoxifen versus anastrozol plus ovarian
function suppression in premenopausal early breast cancer: final
analysis of the Austrian Breast and Colorectal Cancer Study Group Trial
12. Annal Oncol. 2015;26:313-320.
17. Pagani O, Regan MM, Walley BA, et al; TEXT and SOFT Investigators;
International Breast Cancer Study Group. Adjuvant exemestane with
ovarian suppression in premenopausal breast cancer. N Engl J Med.
2014;371:107-118.
18. Pfeiler G, Konigsberg R, Fesl C, et al. Impact of body mass index on the
efficacy of endocrine therapy in premenopausal patients with breast
cancer: an analysis of the prospective ABCSG-12 trial. J Clin Oncol. 2011;
29:2653-2659.
30
19. Ligibel JA, Winer EP. Aromatase inhibition in obese women: how much
is enough? J Clin Oncol. 2012;30:2940-2942.
20. Dowsett M, Lnning PE, Davidson NE. Incomplete estrogen suppression
with gonadotropin-releasing hormone agonists may reduce clinical
efficacy in premenopausal women with early breast cancer. J Clin Oncol.
Epub 2016 Jan 4.
21. Hamelers IH, Steenbergh PH. Interactions between estrogen and
insulin-like growth factor signaling pathways in human breast tumor
cells. Endocr Relat Cancer. 2003;10:331-345.
22. Masuda N, Sagara Y, Kinoshita T, et al. Neoadjuvant anastrozole versus
tamoxifen in patients receiving goserelin for premenopausal breast
cancer (STAGE): a double-blind, randomised phase 3 trial. Lancet Oncol.
2012;13:345-352.
23. Bellet M, Gray KP, Francis PA, et al. Twelve-month estrogen levels in
premenopausal women with hormone receptor-positive breast cancer
receiving adjuvant triptorelin plus exemestane or tamoxifen in the
Suppression of Ovarian Function Trial (SOFT): the SOFT-EST substudy.
J Clin Oncol. Epub 2016 Jan 4.
24. Burstein HJ, Lacchetti C, Anderson H, et al. Adjuvant endocrine therapy
for women With hormone receptor-positive breast cancer: American
Society of Clinical Oncology clinical practice guideline update on ovarian
suppression. J Clin Oncol. Epub 2016 Feb 16.
25. Copson E, Eccles B, Maishman T, et al; POSH Study Steering Group.
Prospective observational study of breast cancer treatment outcomes
for UK women aged 18-40 years at diagnosis: the POSH study. J Natl
Cancer Inst. 2013;105:978-988.
26. Davies C, Pan H, Godwin J, et al; Adjuvant Tamoxifen: Longer Against
Shorter (ATLAS) Collaborative Group. Long-term effects of continuing
adjuvant tamoxifen to 10 years versus stopping at 5 years after diagnosis of oestrogen receptor-positive breast cancer: ATLAS, a randomised trial. Lancet. 2013;381:805-816.
27. Gray RG, Rea D, Handley K, et al. aTTom: Long-term effects of continuing
adjuvant tamoxifen to 10 years versus stopping at 5 years in 6,953
women with early breast cancer. J Clin Oncol. 2013;31s (suppl; abstr 5).
28. Goss PE, Ingle JN, Martino S, et al. Impact of premenopausal status at breast
cancer diagnosis in women entered on the placebo-controlled NCIC CTG
MA17 trial of extended adjuvant letrozole. Ann Oncol. 2013;24:355-361.
29. Filipits M, Nielsen TO, Rudas M, et al. The PAM50 risk-of-recurrence
score predicts risk for late distant recurrence after endocrine therapy in
postmenopausal women with endocrine-responsive early breast cancer. Clin Cancer Res. 2014;20:1298-1305.
30. Sestak I, Cuzick J, Dowsett M, et al. Prediction of late distant recurrence
after 5 years of endocrine treatment: a combined analysis of patients
from the Austrian breast and colorectal cancer study group 8 and
arimidex, tamoxifen alone or in combination randomized trials using the
PAM50 risk of recurrence score. J Clin Oncol. 2015;33:916-922.
31. Dubsky P, Brase JC, Jakesz R, et al; Austrian Breast and Colorectal Cancer
Study Group (ABCSG). The EndoPredict score provides prognostic information on late distant metastases in ER+/HER2- breast cancer patients. Br J Cancer. 2013;109:2959-2964.
351:2817-2826.
33. Ruddy KJ, Gelber S, Ginsburg ES, et al. Menopausal symptoms and fertility
concerns in premenopausal breast cancer survivors: a comparison to ageand gravidity-matched controls. Menopause. 2011;18:105-108.
34. Nystedt M, Berglund G, Bolund C, et al. Side effects of adjuvant endocrine treatment in premenopausal breast cancer patients: a prospective randomized study. J Clin Oncol. 2003;21:1836-1844.
35. Azim HA Jr, de Azambuja E, Colozza M, et al. Long-term toxic effects of

adjuvant chemotherapy in breast cancer. Ann Oncol. 2011;22:
1939-1947.
36. Jin Y, Desta Z, Stearns V, et al. CYP2D6 genotype, antidepressant use,
and tamoxifen metabolism during adjuvant breast cancer treatment.
J Natl Cancer Inst. 2005;97:30-39.
37. Binkhorst L, Mathijssen RH, van Herk-Sukel MP, et al. Unjustified
prescribing of CYP2D6 inhibiting SSRIs in women treated with tamoxifen. Breast Cancer Res Treat. 2013;139:923-929.
38. Margolin S, Lindh JD, Thoren L, et al. CYP2D6 and adjuvant tamoxifen:
possible differences of outcome in pre- and post-menopausal patients.
Pharmacogenomics. 2013;14:613-622.
39. Phillips KA, Feng Y, Ribi K, et al. Co-SOFT: the cognitive function sub-study of
the suppression of ovarian function trial (SOFT). Paper presented at: 37th
Annual CTRC-AACR San Antonio Breast Cancer Symposium; December
2014; San Antonio, TX.
40. Powles TJ, Hickish T, Kanis JA, et al. Effect of tamoxifen on bone mineral
density measured by dual-energy x-ray absorptiometry in healthy
premenopausal and postmenopausal women. J Clin Oncol. 1996;14:
78-84.
41. Koopal C, Janssen-Heijnen ML, van de Wouw AJ, et al. Fracture
incidence in pre- and postmenopausal women after completion of
adjuvant hormonal therapy for breast cancer. Breast. 2015;24:
153-158.
42. Hershman DL, McMahon DJ, Crew KD, et al. Zoledronic acid prevents
bone loss in premenopausal women undergoing adjuvant chemotherapy for early-stage breast cancer. J Clin Oncol. 2008;26:4739-4745.
43. Gnant M, Mlineritsch B, Luschin-Ebengreuth G, et al; Austrian Breast
and Colorectal Cancer Study Group (ABCSG). Adjuvant endocrine
therapy plus zoledronic acid in premenopausal women with early-stage
breast cancer: 5-year follow-up of the ABCSG-12 bone-mineral density
substudy. Lancet Oncol. 2008;9:840-849.
44. Ruddy KJ, Gelber SI, Tamimi RM, et al. Prospective study of fertility
concerns and preservation strategies in young women with breast
45. Biglia N, Torrisi R, DAlonzo M, et al. Attitudes on fertility issues in breast
cancer patients: an Italian survey. Gynecol Endocrinol. 2015;31:
458-464.
46. Kroman N, Jensen MB, Wohlfahrt J, et al; Danish Breast Cancer Cooperative Group. Pregnancy after treatment of breast cancera
population-based study on behalf of Danish Breast Cancer Cooperative
Group. Acta Oncol. 2008;47:545-549.
47. Ives A, Saunders C, Bulsara M, et al. Pregnancy after breast cancer:
population based study. BMJ. 2007;334:194.
48. Azim HA, Jr, Santoro L, Pavlidis N, et al. Safety of pregnancy following
breast cancer diagnosis: a meta-analysis of 14 studies. Eur J Cancer.
2011;47:74-83.
49. Azim HA, Jr, Kroman N, Paesmans M, et al. Prognostic impact of
pregnancy after breast cancer according to estrogen receptor status:
a multicenter retrospective study. J Clin Oncol. 2013;31:73-79.
50. NCT02308085. Pregnancy Outcome and Safety of Interrupting Therapy
for Women with Endocrine Responsive Breast Cancer (POSITVE). http://
51. Lobo RA. Potential options for preservation of fertility in women. N Engl
J Med. 2005;353:64-73.
52. Torino F, Barnabei A, De Vecchis L, et al. Chemotherapy-induced ovarian
toxicity in patients affected by endocrine-responsive early breast
cancer. Crit Rev Oncol Hematol. 2014;89:27-42.
53. Loren AW, Mangu PB, Beck LN, et al; American Society of Clinical
Oncology. Fertility preservation for patients with cancer: American
Society of Clinical Oncology clinical practice guideline update. J Clin
Oncol. 2013;31:2500-2510.
54. Lambertini M, Del Mastro L, Pescio MC, et al. Cancer and fertility
preservation: international recommendations from an expert meeting.
BMC Med. 2016;14:1.
55. Peccatori FA, Azim HA, Jr, Orecchia R, et al; ESMO Guidelines Working
Group. Cancer, pregnancy and fertility: ESMO clinical practice guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2013;24(suppl
6):vi160-vi170.
56. Balduzzi A, Leonardi MC, Cardillo A, et al. Timing of adjuvant systemic
therapy and radiotherapy after breast-conserving surgery and mastectomy. Cancer Treat Rev. 2010;36:443-450.
57. Buendgen NK, Schultze-Mosgau A, Cordes T, et al. Initiation of ovarian
stimulation independent of the menstrual cycle: a case-control study.
Arch Gynecol Obstet. 2013;288:901-904.
58. Cakmak H, Katz A, Cedars MI, et al. Effective method for emergency
fertility preservation: random-start controlled ovarian stimulation.
Fertil Steril. 2013;100:1673-1680.
59. Cakmak H, Rosen MP. Ovarian stimulation in cancer patients. Fertil
Steril. 2013;99:1476-1484.
60. Oktay K, Buyuk E, Libertella N, et al. Fertility preservation in breast
cancer patients: a prospective controlled comparison of ovarian
stimulation with tamoxifen and letrozole for embryo cryopreservation.
J Clin Oncol. 2005;23:4347-4353.
61. Oktay K, Hourvitz A, Sahin G, et al. Letrozole reduces estrogen and
gonadotropin exposure in women with breast cancer undergoing
ovarian stimulation before chemotherapy. J Clin Endocrinol Metab.
2006;91:3885-3890.
62. Azim AA, Costantini-Ferrando M, Oktay K. Safety of fertility preservation
by ovarian stimulation with letrozole and gonadotropins in patients
with breast cancer: a prospective controlled study. J Clin Oncol. 2008;
26:2630-2635.
63. Goldrat O, Gervy C, Englert Y, et al. Progesterone levels in letrozole
associated controlled ovarian stimulation for fertility preservation in
breast cancer patients. Hum Reprod. 2015;30:2184-2189.
64. Oktay K, Turan V, Bedoschi G, et al. Fertility preservation success
subsequent to concurrent aromatase inhibitor treatment and ovarian
stimulation in women with breast cancer. J Clin Oncol. 2015;33:
2424-2429.
65. Kim J, Turan V, Oktay K. Long-term safety of letrozone and gonadotropin
stimulation for fertility preservation in women with breast cancer. J Clin
Endocrinol Metab. 2016;101:1364-1371.
66. Turan V, Bedoschi G, Moy F, et al. Safety and feasibility of performing
two consecutive ovarian stimulation cycles with the use of letrozolegonadotropin protocol for fertility preservation in breast cancer patients. Fertil Steril. 2013;100:1681-1685.e1.
67. Goldrat O, Kroman N, Peccatori FA, et al. Pregnancy following breast
cancer using assisted reproduction and its effect on long-term outcome.
Eur J Cancer. 2015;51:1490-1496.
68. Gerber B, von Minckwitz G, Stehle H, et al; German Breast Group Investigators. Effect of luteinizing hormone-releasing hormone agonist on
ovarian function after modern adjuvant breast cancer chemotherapy:
the GBG 37 ZORO study. J Clin Oncol. 2011;29:2334-2341.
69. Munster PN, Moore AP, Ismail-Khan R, et al. Randomized trial using
gonadotropin-releasing hormone agonist triptorelin for the preservation of ovarian function during (neo)adjuvant chemotherapy for breast
70. Moore HC, Unger JM, Phillips KA, et al; POEMS/S0230 Investigators.
Goserelin for ovarian protection during breast-cancer adjuvant chemotherapy. N Engl J Med. 2015;372:923-932.
71. Del Mastro L, Boni L, Michelotti A, et al. Effect of the gonadotropinreleasing hormone analogue triptorelin on the occurrence of
chemotherapy-induced early menopause in premenopausal women
with breast cancer: a randomized trial. JAMA. 2011;306:269-276.
31
72. Lambertini M, Ceppi M, Poggio F, et al. Ovarian suppression using

luteinizing hormone-releasing hormone agonists during chemotherapy
to preserve ovarian function and fertility of breast cancer patients:
a meta-analysis of randomized studies. Ann Oncol. 2015;26:2408-2419.
73. Partridge AH, Ruddy KJ, Gelber S, et al. Ovarian reserve in women who
remain premenopausal after chemotherapy for early stage breast
cancer. Fertil Steril. 2010;94:638-644.
74. Partridge A, Gelber S, Gelber RD, et al. Age of menopause among
women who remain premenopausal following treatment for early
32
breast cancer: long-term results from International Breast Cancer Study

Group Trials V and VI. Eur J Cancer. 2007;43:1646-1653.
75. Meirow D, Dor J, Kaufman B, et al. Cortical fibrosis and blood-vessels
damage in human ovaries exposed to chemotherapy. Potential
mechanisms of ovarian injury. Hum Reprod. 2007;22:1626-1633.
76. Lambertini M, Boni L, Michelotti A, et al; GIM Study Group. Ovarian
suppression with triptorelin during adjuvant breast cancer chemotherapy and long-term ovarian function, pregnancies, and disease-free
survival: a randomized clinical trial. JAMA. 2015;314:2632-2640.
BREAST CANCER
Triple-Negative Breast Cancer: Is

Change on the Horizon?
CHAIR
Carey K. Anders, MD
The University of North Carolina at Chapel Hill
Chapel Hill, NC
SPEAKERS
Rebecca Dent, MD
National Cancer Center Singapore
Singapore
Vandana Abramson, MD
Vanderbilt University Medical Center
Nashville, TN
ANDERS ET AL
The Evolution of Triple-Negative Breast Cancer: From Biology

to Novel Therapeutics
Carey K. Anders, MD, Vandana Abramson, MD, Tira Tan, MBBS, and Rebecca Dent, MD
OVERVIEW
Triple-negative breast cancer (TNBC) is clinically defined as lacking expression of the estrogen receptor (ER), progesterone receptor
(ER), and HER2. Historically, TNBC has been characterized by an aggressive natural history and worse disease-specific outcomes
compared with other breast cancer subtypes. The advent of next-generation sequencing (NGS) has allowed for the dissection of
TNBC into molecular subtypes (i.e., basal-like, claudin-low). Within TNBC, several subtypes have emerged as immune-activated,
consistently illustrating better disease outcome. In addition, NGS has revealed a host of molecular features characteristic
of TNBC, including high rates of TP53 mutations, PI3K and MEK pathway activation, and genetic similarities to serous ovarian
cancers, including inactivation of the BRCA pathway. Identified genetic vulnerabilities of TNBC have led to promising therapeutic
approaches, including DNA-damaging agents (i.e., platinum salts and PARP inhibitors), as well as immunotherapy. Platinum salts are
routinely incorporated into the treatment of metastatic TNBC; however, best outcomes are observed among those with deficiencies
in the BRCA pathway. Although the incorporation of platinum in the neoadjuvant care of patients with TNBC yields higher pathologic
complete response (pCR) rates, the impact on longer-term outcome is less clear. The presence of immune infiltrate in TNBC has
shown both a predictive and prognostic role. Checkpoint inhibitors, including PD-1 and PD-L1 inhibitors, are under investigation in
the setting of metastatic TNBC and have shown responses in initial clinical trials. Finally, matching emerging therapeutic strategies to
optimal subtype of TNBC is of utmost importance as we design future research strategies to improve patient outcome.
riple-negative breast cancer is a unique subset of breast

cancer that accounts for approximately 15%20% of all
breast cancer diagnoses.1 Clinically defined as lacking expression of ER/PR and HER2, TNBC is characterized by an
aggressive natural history and worse disease-specific outcomes compared with other breast cancer subtypes.2 Despite
initial responses to chemotherapy, early and higher rates of
distant, typically visceral, recurrences are observed for TNBC.3
Anthracycline and taxane-based chemotherapy traditionally
has been the mainstay of therapy for TNBC, but with the
advent of NGS, molecular dissection of TNBC is revealing novel
targets currently under investigation.4-6 Herein, we will review
the natural history of TNBC, our current understanding of the
molecular characteristics of this unique subset of breast
cancer, and emerging clinical strategies in the modern era,
with a highlight on immunotherapy and DNA damaging agents.
DEFINITION OF TRIPLE-NEGATIVE BREAST

CANCER
The term triple-negative breast cancer is a pathologic
definition based on the protein expression of the three most
commonly targeted biomarkers in the treatment of breast

cancer: ER, PR, and HER2. The definition of TNBC has varied
throughout the decades, namely around the definition of
ER and/or PR positivity (i.e., 1%10% vs. $ 1% by immunohistochemistry [IHC]). More recently, ER/PR positivity
has been more strictly defined as greater than or equal to 1% by
the American Society of Clinical Oncology guidelines.7 Endocrine therapy is generally recommended for breast cancers
exhibiting greater than or equal to 1% ER and/or PR positivity
across all stages of breast cancer. HER2-postivity also has been
strictly defined as protein expression of 3+, and/or HER2/neu
gene amplification greater than or equal to 2.0 by fluorescence
in situ hybridization (FISH).8 Thus, the pathologic definition of
TNBC is currently defined as ER and/or PR IHC expression of
0 and HER2 negativity defined as either IHC expression of 01+
or lack of gene amplification (FISH , 2.0). This universal
definition is being incorporated into the design of TNBC clinical
trials. Such an approach will help interpret results of individual
clinical trials in the context of others because the biology of
breast cancers expressing hormone receptors at 1%10% is
likely distinct from those without any expression.
From the Department of Medicine, Vanderbilt University, Vanderbilt-Ingram Cancer Center, Nashville, TN; Department of Medicine, National Cancer Center Singapore, Singapore;
Department of Medicine, The University of North Carolina at Chapel Hill, Chapel Hill, NC; UNC Lineberger Comprehensive Cancer Center, Chapel Hill, NC.
Corresponding author: Carey K. Anders, MD, Division of Hematology Oncology, University of North Carolina at Chapel Hill, 710 Oxfordshire Ln., Chapel Hill, NC 27517; email:
carey_anders@medunc.edu.
34
TRIPLE-NEGATIVE BREAST CANCER: EVOLUTION OF BIOLOGY AND NOVEL THERAPEUTICS
NATURAL HISTORY AND CLINICAL BEHAVIOR OF

TRIPLE-NEGATIVE BREAST CANCER
In addition to its distinct pathologic definition, TNBC is
characterized by a unique clinical history as compared with
other subsets of breast cancer. Smid and colleagues reported
among 344 primary lymph nodenegative breast cancers
that the site of distant recurrence differed by breast cancer
subtype.9 In this analysis, visceral relapses, including lung
(p = .01) and brain metastases (p = .0035), more commonly were
observed among patients with basal-like tumors. In contrast,
bone relapse was more commonly seen among the luminal
subtype (p = .0031), and those with HER2-enriched breast
cancer more commonly recurred in the liver (p = .17; Fig. 1).
Moreover, Harrell et al evaluated tumor gene expression
using a dataset of approximately 1,000 human breast tumors
and confirmed that HER2-enriched tumors more commonly
spread to the liver, whereas basal-like and claudin-low breast
tumors, both commonly triple-negative, spread to the brain
and lung.10 Among patients with advanced TNBC, brain relapse has been observed in approximately 50% of patients.11
Response to chemotherapy also differs by breast cancer
subtype, with the high-grade TNBC subset typically more
responsive to preoperative chemotherapy (i.e., higher rates
of pCR) but with higher overall rates of relapse.3,12 This
phenomenon has been coined the triple-negative paradox.3 The higher rate of distant recurrence for patients with
TNBC following neoadjuvant chemotherapy is likely attributed to residual disease (i.e., inability to achieve pCR).
Multiple studies have shown superior survival for patients
who experience pCR compared with those who do not
experience pCR in this setting. Efforts to convert patients
with residual disease following neoadjuvant chemotherapy
KEY POINTS
Triple-negative breast cancer (TNBC) is a pathologic

classification defined as lack of estrogen and
progesterone receptors (0% by IHC) expression and
HER2 negativity (01+ by IHC or lack of HER2/neu gene
amplification).
Although TNBC comprises only 15% to 20% of all breast
cancer diagnoses, it is over-represented among
advanced disease as recurrences are early and typically
visceral (e.g., lung and brain).
Next-generation sequencing has provided a powerful
platform to dissect the biology of TNBC to provide novel
therapeutic targets under clinical investigation.
Checkpoint inhibitors, including PD-1 and PD-L1, are
showing responses in TNBC in initial clinical trials, and
several studies are underway to confirm activity and
understand biomarkers of response.
Platinum-based chemotherapy, with a DNA-damaging
mechanism of action, is active in TNBC and is routinely
incorporated into the neoadjuvant and metastatic
settings. Investigation of platinums in the adjuvant
setting is ongoing.
to a pCR with novel chemotherapy combinations (platinums)

and novel targeted therapies (inhibitors of PARP inhibitors)
have shown superior pCR rates on a trial-level basis.13-15 It is
unclear, however, if incremental increases in pCR rates ultimately will translate into improvement in recurrence-free
survival.16
Triple-negative breast cancer is characterized by recurrence
patterns and natural history that are distinct from other
subtypes of breast cancer. Dent et al compared the natural
history and clinical behavior of TNBC to non-TNBC subtypes
among 1,601 patients diagnosed with early-stage breast
cancer in Toronto between 1987 and 1997.2 At a median
follow-up of 8 years, this analysis illustrated that women with
TNBC were more likely to experience a distant recurrence
(hazard ratio [HR] 2.6; p , .0001); survival for patients with
TNBC was inferior compared with others (HR 3.2; p , .001)
within 5 years of diagnosis. Interestingly, and compared with
patients with non-TNBC breast cancer, the risk of recurrence
peaked at the 3-year mark and declined thereafter. This
observation is in contrast to non-TNBC, in which the risk of
recurrence is lower during the initial 3-year follow-up period,
but it remains constant over time. These important distinctions between the clinical behaviors of TNBC versus
non-TNBC must be considered when designing clinical
interventions to prevent recurrence for this subset of
aggressive breast cancer.
TRIPLE-NEGATIVE BREAST CANCER SUBTYPES

Breast cancer is a heterogeneous disease characterized by
distinct intrinsic subtypes, traditionally defined as luminal A
and B, normal-like, HER2-enriched, and basal-like, each with
corresponding unique clinical behavior.1,17,18 Over the past
decade, TNBC also has been recognized as a similarly heterogeneous disease. Several groups have sought to dissect
the biology of TNBC using IHC, gene expression, and sequencing tools. Perhaps the most well-known classification
of TNBC by gene expression, when compared with all breast
cancers, is the basal-like subtype. The basal-like subtype,
which represents 15%20% of all breast cancers, originally
was classified by a basal epithelial cell gene expression
cluster including keratin 5, keratin 17, integrin-b4, and
laminin.1 Basal-like breast cancer is approximately 80%
concordant with the IHC classification of TNBC.19 A more
rare subtype of breast cancer, termed the claudin-low
subtype, is also commonly triple-negative.20 The claudinlow subtype is characterized by low to absent expression of
luminal differentiation markers, high enrichment for
epithelial-to-mesenchymal transition (EMT) markers, immune response genes, and cancer stem celllike features.
This breast cancer subtype, with high frequency of metaplastic and medullary differentiation, illustrates a response
to therapy that is intermediate between the basal-like and
luminal breast cancer subtypes.
A different, yet complementary, approach has been taken
by other investigators in which only TNBCs have been
subtyped using gene expression and sequencing tools to
35
ANDERS ET AL
FIGURE 1. Subtype-Specific Patterns of Distant Recurrence From 344 Primary Breast Cancers
Copyright pending as of February 1, 2016.
reveal distinct biologic subtypes with unique therapeutic

implications.5,6 Lehmann and colleagues analyzed gene
expression from over 500 TNBC tumors and identified six
unique subtypes: basal-like (two subtypes, BL1 and BL2),
immunomodulatory (IM), mesenchymal (M), mesenchymal
stemlike (MSL), and luminal androgen receptor (LAR).5 Cellline models representative of each subtype were identified
and therapeutically targeted. Results indicated that cell lines
of the BL1 and BL2 subtypes, with higher expression of cell
cycle and DNA damage response genes, were sensitive to
cisplatin. Mesenchymal and MSL subtype cell lines, which
were enriched for EMT and growth factor pathways, illustrated sensitivity to dual PI3K/mTOR inhibition and Src inhibition. Finally, LAR subtype cell lines were sensitive to an
androgen receptor (AR) antagonist, a strategy that has shown
early success in the clinical arena.21 In parallel, Burnstein et al
performed DNA and RNA sequencing on approximately 200
TNBC tumors with confirmation in independent datasets and
identified four subtypes: LAR, mesenchymal (MES), basal-like
immunosuppressed (BLIS), and basal-like immune-activated
(BLIA).6 The BLIS subtype conferred the worst prognosis and
BLIA the best prognosis, for both disease-free and diseasespecific survival. In this analysis, subtype-specific targets were
identified (such as AR in the LAR and STAT signal transduction
molecules and cytokines in the BLIA) that are worthy of
continued exploration.
36
MOLECULAR DISSECTION OF TRIPLE-NEGATIVE

BREAST CANCER
As with other breast cancer subtypes, the biology of triplenegative, specifically basal-like, breast cancer was a focus of
The Cancer Genome Atlas (TCGA) project.4 In this large-scale
analysis of primary human breast cancers, genomic DNA
copy number arrays, DNA methylation, exome sequencing,
messenger RNA arrays, microRNA sequencing, and reversephase protein arrays were evaluated. Among the 81 basallike breast cancers, of which 65 were both triple-negative
and basal-like, several hallmark findings were reported: (1) a
high frequency of TP53 mutations (80%); (2) loss of RB1 and
BRCA1; (3) high activation of the PI3K pathway (either
through PIK3CA gene mutation of approximately 9% or loss of
the negative regulators INPP4B and/or PTEN); (4) an enhanced
proliferation signature with hyperactivated FOXM1 as a transcriptional driver of this signature; and (5) genetic similarity
between basal-like breast cancer and serous ovarian cancers.
These similarities included high frequencies of ATM and TP53
mutations, high expression of AKT3 and MYC, inactivation of
BRCA1 and BRCA2, RB1 loss, and cyclin E1 amplification. From a
therapeutic targeting perspective, approximately 20% of basallike tumors harbored a germline or somatic BRCA mutation,
which may confer sensitivity to platinum and/or PARP inhibitors that will be discussed below. Finally, copy number
analysis revealed several amplifications or deletions that may

be targetable in the clinical arena. Amplifications included
PIK3CA (49%), KRAS (32%), BRAF (30%), and EGFR (23%). Other
less common amplifications were seen in FGFR1, FGFR2, IGFR1,
KIT, MET, and PDGFRA. As cited above, deletions were seen in
PTEN and INPP4B, which confer activation of the PI3K pathway.
Results of this valuable analysis are summarized in Sidebar 1
and provide a wealth of knowledge pertaining to our understanding of the biology of TNBC and ways to target this
disease more effectively.
TRIPLE-NEGATIVE BREAST CANCER: IS CHANGE

ON THE HORIZON?
In addition to our evolving understanding of the biology of
TNBC over the decades, therapeutic strategies to treat TNBC
across different stages are maturing. Although many molecularly targeted strategies under investigation exist, for
the purposes of this review, we will focus on the use of
platinums and the incorporation of immunotherapy into the
treatment of TNBC.
Optimal Use of Platinum in Triple-Negative

Breast Cancer
A subset of TNBC has been postulated to be phenotypically
and molecularly similar to familial BRCA1-associated breast
cancers, so called BRCAness.22 It is estimated that more
than 75% of tumors arising in BRCA1 mutation carriers are
triple-negative and/or have a basal-like phenotype.23 These
tumors are characterized by their heightened sensitivity to
damage by DNA cross-linking.24-26 Double-stranded DNA
breaks caused by agents such as platinum salts are considered the most incisive form of DNA damage, and they are
repaired by homologous recombination, an error-free
process, and nonhomologous end-joining, which is error
SIDEBAR 1. The Cancer Genome Atlas Molecular

Characteristics of Basal-like Breast Cancer
Characteristics
High concordance with TNBC (approximately 80%)
High rates of TP53 pathway alterations (TP53 mutations in
84%; gain of MDM2 14%)
High rates of PI3K pathway alterations (PIK3CA mutation
7%; PTEN mutation or loss 35%; INPP4B loss 30%)
High rates of RB pathway alterations (RB1 mutation or loss
20%; cyclin E1 amplification 9%; high expression of
CDKN2A; low expression of RB1)
High genomic instability
Hypomethylation compared with other breast cancer
subtypes
Similarities with serous ovarian carcinomas, including
approximately 20% rate of germline or somatic BRCA
mutations
Abbreviation: TNBC, triple-negative breast cancer.
prone. BRCA1 and BRCA2 are well-described tumor suppressor genes that participate directly in homologous
recombinationmediated repair of double-stranded breaks.27
In humans, mutation in one copy of either of these genes in the
germline results in hereditary breast and ovarian cancer syndrome; tumors that subsequently develop are defective in
homologous recombinationmediated DNA repair.27 Recent
reports have expanded the range of genes implicated in familial
breast cancers, many of which are involved in the homologous
recombination pathway.23,28-30 In addition, defective homologous recombination also can be found in sporadic breast
cancers as shown by comparative genome hybridization
studies.31 Finally, impairment of homologous recombination
deficiency can occur through epigenetic mechanisms, such as
methylation of BRCA1/2, which is also part of the BRCAness
spectrum.22
Incorporation of Platinums in the Neoadjuvant

Treatment of Triple-Negative Breast Cancer
Over the past few decades, there has been considerable
interest in the use of platinum salts in the treatment of TNBC
on the basis that dysfunction of homologous recombination
DNA repair sensitizes tumor cells to these agents and induces cell death. As a proof of concept, a small neoadjuvant
study reported a strikingly high pCR of 90% to four cycles of
neoadjuvant cisplatin in a group of 10 patients from Poland
with BRCA1 mutation, the majority of whom were the TNBC
phenotype.32 Larger phase II studies involving patients with
predominantly sporadic TNBC have described conflicting
results. Alba and colleagues reported a randomized phase II
study of standard neoadjuvant epirubicin plus cyclophosphamide chemotherapy followed by docetaxel with or
without carboplatin AUC 6 given every 21 days.33 The
prespecified primary endpoint of improvement in pCR rates
(defined as pCR in breast) was not met, with pCR rates in
breast and axilla of 30% of patients in both treatment
arms.33 In contrast, the German Breast Cancer Group administered neoadjuvant paclitaxel, liposomal doxorubicin,
and bevacizumab with or without weekly carboplatin AUC 1.5-2
in the GeparSixto trial. Of the 315 patients with TNBC, 53.2%
of those treated with the addition of weekly carboplatin, as
compared with 36.9% treated without the addition of carboplatin, achieved a pCR (p = .005).14 Similarly, CALGB 40603, a
two-by-two factorial randomized phase II trial evaluating
weekly paclitaxel with or without carboplatin (AUC 6) and/or
bevacizumab followed by dose dense doxorubicin plus cyclophosphamide showed an increase in pCR rate (defined as
ypT0/is) from 46% to 60% (odds ratio 1.76; p = .0018).15
Furthermore, an important clinical question revolves
around these data: does the purported improvement in pCR
rates translate to improved survival outcomes? Early survival
analysis of GeparSixto and CALGB 40603 simultaneously
were reported at the San Antonio Breast Cancer Symposium
in December 2015. In the GeparSixto study, the improved
pCR rate translated into a significant increase in 3-year
disease-free survival (DFS) from 76.1% to 85.8% (HR 0.56;
37
ANDERS ET AL
95% CI, 0.330.96; p = .035).34 This contrasts with the CALGB

40603 study, which reported a numerical improvement, but
no statistical difference in 3-year event-free survival nor
overall survival (OS), with the caveat that this trial was
underpowered to determine a survival benefit.13 Of note,
GeparSixto used a nonstandard, more intensive regimen of
concurrent anthracycline-taxane chemotherapy with the
incorporation of bevacizumab in all arms. Pharmacologic
synergy, as well as potentially the weekly schedule of
carboplatin, may have contributed to the better overall
prognosis in the GeparSixto trial.
Incorporation of Platinums in the Metastatic

Treatment of Triple-Negative Breast Cancer
In the metastatic setting, retrospective studies have been
conflicting and largely focused on comparing TNBC with
other subtypes.35-38 In the largest randomized phase III trial,
the Triple Negative Breast Cancer Trial (TNT), 376 patients in
U.K. centers were randomly assigned to receive either
carboplatin or docetaxel monotherapy as first-line treatment with cross-over on progression.39 In the unselected
TNBC population, there was no difference in the primary
outcome of objective response rates in both treatment arms
when treated up front (31.4% vs. 35.6%; p = .44) or following
cross-over (22.8% vs. 25.6%; p = .73).39 Similarly, the secondary endpoint of progression-free survival (PFS) did not
significantly differ in both arms.39 To add to the burgeoning
evidence in favor of a role for platinum-based chemotherapy, two studies from China have been reported: (1) a
randomized phase II trial comparing docetaxel and cisplatin
versus docetaxel and capecitabine and (2) a multicenter
randomized phase III trial of gemcitabine and cisplatin versus
gemcitabine and paclitaxel.40,41 In the phase II study, 53
patients with metastatic TNBC were randomly selected to
receive docetaxel with cisplatin or capecitabine in the firstline setting. Both overall response rates (63.0% vs. 15.4%;
p = .001) and PFS (10.9 vs. 4.8 months; p # .001) were
significantly better in the cisplatin group.40 CBCSG006 was a
phase III study in which 240 patients were randomly assigned
to receive a maximum of eight cycles of gemcitabine and
either cisplatin or paclitaxel dosed every 21 days.41 The trial
used a hybrid design and was powered for noninferiority by
allowing for a prespecified superiority hypothesis to be
tested. Gemcitabine plus cisplatin was both noninferior to and
superior to gemcitabine plus paclitaxel (PFS: HR 0.692; 95% CI,
0.5230.915; p , .0001 [noninferiority] and p , .009
[superiority]).41 Therefore, in both studies, the cisplatincontaining regimen outperformed the comparator.
Challenges in Incorporating Platinums Into Standard

Practice: Toxicities and Dosing Strategies
Given this encouraging clinical data on the role of platinum
agents in TNBC, important clinical questions still remain on
how best to choose the dose and/or schedule of the platinum agent. No trials to date have been powered adequately
to demonstrate an improvement in long-term survival
38
outcome. The addition of carboplatin enhances acute toxicities associated with chemotherapy with higher discontinuation rates, more dose reductions, and higher grade 3/4
adverse events, which could potentially compromise the
current standard of care in an already high-risk population
of patients. It may be reasonable to consider off-trial use of
carboplatin in locally advanced TNBC for rapid control of
locoregional disease and in young, fit patients with a high
risk of relapse as we await definitive results from ongoing
phase III trials powered for survival outcomesPEARLY
(NCT02441933), Gerpar-Octo (NCT021253144), NRGBR003 (NCT02488967), and TPPC (NCT02455141).
BIOMARKER DISCOVERY IN TRIPLE-NEGATIVE

BREAST CANCER
Triple-negative breast cancer is a heterogeneous disease.
Given the trade-off of greater toxicities in the early setting
and mixed results in the metastatic setting, it would be ideal
to be better able to identify patients most likely to benefit
from this treatment strategy. TBCRC009, a phase II biomarker discovery trial, has reported an improved response
rate in patients who harbored deleterious germline BRCA
mutation(s) without durability of response nor improvement in PFS or OS.42 The TNT trial lends support to BRCA
genotyping and gene expression profiling to improved
therapy selection in metastatic TNBC.39 Patients with BRCA1
or BRCA2 mutations treated with carboplatin fared better
with a greater response rate (68% vs. 33.3%; p = .03) and PFS
(6.8 vs. 3.1 months; p = .03) as compared with docetaxel,
whereas nonbasal-like tumors did better with docetaxel
(73.7% vs. 16.7%; p # .01).39 Lastly, DNA-based assays have
been developed based on whole genome tumor loss of
heterozygosity score (LOH),43 telomeric allelic imbalance
score (TAI),44 and large-scale state transitions score (LST).45
These assays measure patterns of genomic instability derived
through comparison of BRCA1/2 mutated and nonmutated
tumors. The unweighted sum of LOH, TAI, and LST together
form the homologous recombination deficiency score. This
has been evaluated in the translational component of the
TBCRC009 and TNT trials. Again, results are conflicting. In
TBCRC009, tumors from patients with platinum-responsive
disease exhibited higher values for LST and LOH assays.42 This
was not so in TNT, in which the homologous recombination
deficiency score did not select for sensitivity to carboplatin
over docetaxel.39 Therefore, further development of an assay
to identify homologous recombination defect in non-BRCA1/2
tumors is needed. Finally, the group from Memorial Sloan
Kettering Cancer Center reported a comparison of homologous
recombination deficiency scoring methodologies and concluded that homologous recombination DNA repair gene
sequencing demonstrated good sensitivity and specificity,
providing a potential biomarker for clinical trials of homologous
recombination deficiencytargeted therapies.46
In summary, for the practicing oncologist, what would be
the current standard of care with respect to the use of
platinum salts? In the neoadjuvant setting, platinum is a
reasonable choice for selected patients, especially if those

who are BRCA positive. There are no data to support platinum
salts in the adjuvant setting; clinical trials are ongoing. In the
metastatic setting, where the goal of therapy is palliative,
platinum is one of the standard chemotherapies oncologists
include in their armamentarium of chemotherapies, although
with the caveat that the most convincing data are in the frontline setting for BRCA-germline mutation carriers.
IMMUNOTHERAPY AND IMMUNOCONJUGATES

FOR TRIPLE-NEGATIVE BREAST CANCER:
PROMISE AND POTENTIAL
Modern immunotherapy, in the form of monoclonal antibodies that act as checkpoint inhibitors, has revolutionized
the landscape of the treatment of metastatic melanoma in a
relatively short period of time. Antibodies to cytotoxic
T-lymphocyte antigen 4 (CTLA-4), PD-1, and PD-L1, all of which
increase the immune response against the tumor by blocking
immune regulating proteins that downregulate the immune
system, have increased response rates and OS.47-49 Perhaps
more interestingly, other solid tumor malignancies, such as lung
cancer, which traditionally are not considered immunogenic,
have shown clinical benefit to checkpoint inhibitors.50 The role
of immunotherapy in breast cancer has yet to be defined, but
increasing evidence points to TNBCs as possibly having unique
characteristics that may make them more responsive to
checkpoint inhibition. Given the lack of targeted treatment
options for TNBC, immunotherapy poses an exciting opportunity for the treatment of this aggressive disease.
OVERVIEW OF TUMOR IMMUNE SURVEILLANCE

IN THE SETTING OF BREAST CANCER
In a process called tumor immune surveillance, tumor immune
cells, including CD8+ T cells, recognize tumor-associated antigens presented by tumor cells and attack the tumor cells. The
PD-1 receptor pathway plays an important role in modulating
immune responses and provides an escape mechanism to
tumor immune surveillance. PD-1 is an inhibitory immune
checkpoint receptor expressed on activated T cells, B cells,
natural killer cells, and other lymphocytes that can remain
active in inflammatory states such as cancer.51,52 PD-L1 is a
ligand of PD-1 and acts to suppress antitumor immunity by
binding PD-1. Ligation of PD-L1 with PD-1 inhibits T-cell proliferation, cytokine production, and cytolytic activity, which
leads to the functional inactivation or exhaustion of T cells.53
Through upregulation of PD-L1 and other adaptive immune
resistance mechanisms, tumors are able to use this pathway to
evade the antitumor immune response.
The balance of cells in the tumor microenvironment, which
includes tumor cells, stromal cells (i.e., fibroblasts and microvasculature), and immune cells (i.e., lymphocytes), appears to influence breast cancer outcome.54 An association
between greater tumor infiltrative lymphocytes (TILs) and
better prognosis in breast cancer has been recognized for
some time, but newer studies have shown the specific
relevance in TNBC, which has been shown to have
substantial infiltration with TILs.55-58 Higher levels of TILs

generally are associated with poor-prognostic clinicopathologic features, including ER negativity, higher grade, higher
proliferative rate, and lymph node positivity.55,59-62 However, despite worse clinical features, higher levels of TILs are
associated with improved DFS and OS, independent of
systemic therapy.59,63,64 This apparent paradox highlights
the role that the immune system may play in a subset of
TNBCs and suggests that TILs may be a surrogate for an
adaptive immune response in these cancers.
ASSOCIATION OF TUMOR INFILITRATIVE

LYMPHOCYTES AND RESPONSE OF TRIPLENEGATIVE BREAST CANCER TO CHEMOTHERAPY
Generally, TILs are divided into intratumoral TILs that have
direct contact with tumor cells, and stromal TILs, which are
found between tumor cells within the tumor stroma, but
they do not have direct contact with tumor cells. A study of
patients with breast cancer receiving neoadjuvant chemotherapy with an anthracycline/taxane combination showed
an independent association between percentage of intratumoral TILs and pCR.64 Those with lymphocyte-predominant
breast cancer, defined as more than 60% of stromal or
tumoral infiltration, had a particularly high rate of pCR (41.7%)
compared with only a 2% pCR rate in those tumors without any
tumoral lymphocyte infiltration. In the GeparSixto study that
evaluated the addition of carboplatin to chemotherapy in TNBC
or HER2-positive cancers, the presence of stromal TILs was
predictive of response to the addition of carboplatin, although
this correlation reached statistical significance only in HER2positive tumors.65 Regardless, this association between response to chemotherapy based on the presence of TILs suggests
that TILs not only are predictive markers of response in TNBC,
but that cell death induced by platinum agents may prime or
generate neoantigens to stimulate nearby lymphocytes.
PROGNOSTIC IMPLICATIONS OF IMMUNE GENE

SIGNATURES IN TRIPLE-NEGATIVE BREAST
CANCER
Several other studies promote the exploration of the PD-1
pathway and immunotherapy in TNBC. Data from TCGA4
have confirmed higher PD-L1 mRNA expression in TNBC
versus non-TNBC samples.66 This and other studies have
shown that PD-L1 is not detected in normal breast tissue but
is expressed in about half of all breast cancers, including
approximately 20% to 30% of TNBCs.67,68 PD-L1 expression
is associated with TILs,69 correlates with higher histologic
grade, greater tumor size, and higher expression of the
proliferation marker Ki-67.70
Further supporting the link between immunotherapy and
TNBC are gene expression profiling studies that demonstrate
an association between expression of IM genes and better
clinical outcomes in TNBC.71 Desmedt et al were among the
first to create gene modules and correlate them with outcome in different subtypes of breast cancer. Of the seven
gene expression modules they described (tumor invasion,
39
ANDERS ET AL
immune response, angiogenesis, apoptosis, proliferation,

and ER and HER2 signaling), only the immune response
module correlated with prognosis in the ER-/HER2- subgroup. Since that time, several groups have described the
prognostic value of immune gene signatures and TNBC.72-74
More recently, analysis of gene expression profiles of 587
TNBC samples identified six distinct subtypes, including an
IM subtype characterized by high expression of immunerelated genes.5 This subtype is composed of immuneactivated and associated signaling components contributed
from both the tumor and the infiltrating lymphocytes, and
it has been associated with improved relapse-free survival
compared with other subtypes.5 RNA sequencing showed
this subtype to have substantially higher expression of PD-L1,
PD-1, and CTLA-4. These and other data provide evidence that
there may be a subset of TNBC in which checkpoint inhibitors
may have particular efficacy.
CLINICAL EXPERIENCE AND OUTCOMES WITH

IMMUNOTHERAPY IN TRIPLE-NEGATIVE BREAST
CANCER
The first reported study of an immune checkpoint inhibitor in
TNBC was a single-arm phase IB study of single-agent
pembrolizumab, a PD-1 antibody (KEYNOTE-012).75 This study
showed that pembrolizumab 10 mg/kg given every 2 weeks was
not only well-tolerated; it also showed activity in heavily pretreated patients with metastatic PD-L1positive TNBC. Interestingly, of all patients with TNBC who were screened, 58%
tested positive for PD-L1 in greater than 1% of tumor stroma or
tumor cells. Over 45% of patients had received more than three
prior treatments in the metastatic setting, and 21.9% had
received five or more treatments. Of the 27 participants
with centrally confirmed measurable disease (32 patients
total enrolled), one participant (3.7%) had a complete
response (CR), four participants (14.8%) had a confirmed
partial response (PR), 25.9% had stable disease, and 44.4%
had progressive disease by central review. The median
time to response was 18 weeks (range, 732 weeks), and
the median duration of response had not been reached
(range, 15more than 40 weeks).
The PD-L1 antagonist atezolizumab (MPDL3280A) also is
being evaluated in TNBC. An ongoing phase I study of atezolizumab has an expansion cohort in heavily pretreated patients
with PD-L1positive and negative TNBC.76 Initial data show
the overall response rate for evaluable patients with metastatic
TNBC is 19%, including 9.5% CR and 9.5% PR, with 75% of
responses ongoing (range, 18 to more than 56 weeks).
FUTURE DIRECTIONS FOR IMMUNOTHERAPY IN

TRIPLE-NEGATIVE BREAST CANCER
A decade ago, very little was known about the make-up of
TNBC, and the only promising treatments on the horizon
involved cytotoxic chemotherapy. Now, with a better
understanding of the heterogeneity of TNBC and the
understanding that immune targets are relevant in this
disease, we are in a new era in which we are exploring
checkpoint inhibitors, vaccines, and immune antagonists.
Two large phase III randomized studies of atezolizumab
and pembrolizumab are now ongoing in metastatic TNBC;
the first is nab-paclitaxel with or without atezolizumab
and the other is pembrolizumab versus single agent
chemotherapy. Studies with both agents are also planned
in the neoadjuvant setting. Finally, two studies evaluating
carboplatin with atezolizumab are to begin accrual
through the TBCRC in upcoming months to capitalize on
data with platinums in TNBC. Tissue samples from these
studies will further clarify the relationship of response and
resistance to PD-1 and PD-L1 expression, and to the
presence of TILs. The reported studies to date that show
responses in heavily pretreated patients with TNBC are
notable, but more impressive is the duration of response
seen in a disease that usually has a PFS of 2 to 3 months
after the first or second line of treatment of metastatic
disease. Immunotherapy rapidly has become a great
contender in the treatment of TNBC. Our next challenges
will be to identify how to use it best, whether it is with
chemotherapy or in combination with other immune
modulators, and to identify biomarkers of response.
CONCLUSION
Triple-negative breast cancer is an established subset of
breast cancer with characteristic clinical behavior and
natural history. Until recently, the mainstay of therapy
against TNBC has been cytotoxic chemotherapy. The
advent of NGS has opened our eyes to the possibility of
targeted therapy in a subset of breast cancer that lacks the
classic biomarkers (ER, PR, and HER2) in breast cancer.
Despite this challenge, investigators spanning the research spectrum from bench-to-bedside are tackling TNBC
head on. We are seeing an explosion of activity in basic,
translational, and clinical research that includes platinums, immunotherapy, and others. Coordinated research
efforts such as this are the way forward to improve the
outcome for our thousands of patients with TNBC.
References
1. Perou CM, Srlie T, Eisen MB, et al. Molecular portraits of human breast
tumours. Nature. 2000;406:747-752.
2. Dent R, Trudeau M, Pritchard KI, et al. Triple-negative breast cancer:
clinical features and patterns of recurrence. Clin Cancer Res. 2007;
13:4429-4434.
40
3. Carey LA, Dees EC, Sawyer L, et al. The triple negative paradox: primary
tumor chemosensitivity of breast cancer subtypes. Clin Cancer Res.
2007;13:2329-2334.
4. The Cancer Genome Atlas Network. Comprehensive molecular portraits
of human breast tumours. Nature. 2012;490:61-70.
5. Lehmann BD, Bauer JA, Chen X, et al. Identification of human triplenegative breast cancer subtypes and preclinical models for selection of
targeted therapies. J Clin Invest. 2011;121:2750-2767.
6. Burstein MD, Tsimelzon A, Poage GM, et al. Comprehensive genomic
analysis identifies novel subtypes and targets of triple-negative breast
cancer. Clin Cancer Res. 2015;21:1688-1698.
7. Hammond ME, Hayes DF, Dowsett M, et al. American Society of Clinical
Oncology/College of American Pathologists guideline recommendations for immunohistochemical testing of estrogen and progesterone
receptors in breast cancer. J Clin Oncol. 2010;28:2784-2795.
8. Wolff AC, Hammond ME, Hicks DG, et al; American Society of Clinical
Oncology; College of American Pathologists. Recommendations for
human epidermal growth factor receptor 2 testing in breast cancer:
American Society of Clinical Oncology/College of American Pathologists
clinical practice guideline update. J Clin Oncol. 2013;31:3997-4013.
9. Smid M, Wang Y, Zhang Y, et al. Subtypes of breast cancer show
preferential site of relapse. Cancer Res. 2008;68:3108-3114.
10. Harrell JC, Prat A, Parker JS, et al. Genomic analysis identifies unique
signatures predictive of brain, lung, and liver relapse. Breast Cancer Res
Treat. 2012;132:523-535.
11. Lin NU, Claus E, Sohl J, et al. Sites of distant recurrence and clinical
outcomes in patients with metastatic triple-negative breast cancer: high
incidence of central nervous system metastases. Cancer. 2008;113:
2638-2645.
12. Liedtke C, Mazouni C, Hess KR, et al. Response to neoadjuvant therapy
and long-term survival in patients with triple-negative breast cancer.
J Clin Oncol. 2008;26:1275-1281.
13. Sikov WM, Berry DA, Peou CM, et al. Event-free and overall survival
following neoadjuvant weekly paclitaxel and dose-dense AC +/2 carboplatin and/or bevacizumab in triple-negative breast cancer: outcomes from CALGB 40603 (Alliance). 2015, San Antonio, TX: 2016 San
Antonio Breast Cancer Symposium.
14. von Minckwitz G, Schneeweiss A, Loibl S, et al. Neoadjuvant carboplatin
in patients with triple-negative and HER2-positive early breast cancer
(GeparSixto; GBG 66): a randomised phase 2 trial. Lancet Oncol. 2014;
15:747-756.
15. Sikov WM, Berry DA, Perou CM, et al. Impact of the addition of carboplatin and/or bevacizumab to neoadjuvant once-per-week paclitaxel
followed by dose-dense doxorubicin and cyclophosphamide on pathologic complete response rates in stage II to III triple-negative breast
cancer: CALGB 40603 (Alliance). J Clin Oncol. 2015;33:13-21.
16. Cortazar P, Zhang L, Untch M, et al. Pathological complete response and
long-term clinical benefit in breast cancer: the CTNeoBC pooled
analysis. Lancet. 2014;384:164-172.
17. Srlie T, Perou CM, Tibshirani R, et al. Gene expression patterns of
breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci USA. 2001;98:10869-10874.
18. Sorlie T, Tibshirani R, Parker J, et al. Repeated observation of breast
tumor subtypes in independent gene expression data sets. Proc Natl
Acad Sci USA. 2003;100:8418-8423.
19. Livasy CA, Karaca G, Nanda R, et al. Phenotypic evaluation of the basal-like
subtype of invasive breast carcinoma. Mod Pathol. 2006;19:264-271.
20. Prat A, Parker JS, Karginova O, et al. Phenotypic and molecular characterization of the claudin-low intrinsic subtype of breast cancer. Breast
Cancer Res. 2010;12:R68.
21. Gucalp A, Tolaney S, Isakoff SJ, et al; Translational Breast Cancer
Research Consortium (TBCRC 011). Phase II trial of bicalutamide in
patients with androgen receptor-positive, estrogen receptor-negative
metastatic breast cancer. Clin Cancer Res. 2013;19:5505-5512.
22. Turner N, Tutt A, Ashworth A. Hallmarks of BRCAness in sporadic
cancers. Nat Rev Cancer. 2004;4:814-819.
23. Foulkes WD, Smith IE, Reis-Filho JS. Triple-negative breast cancer.
N Engl J Med. 2010;363:1938-1948.
24. Bhattacharyya A, Ear US, Koller BH, et al. The breast cancer susceptibility
gene BRCA1 is required for subnuclear assembly of Rad51 and survival
following treatment with the DNA cross-linking agent cisplatin. J Biol
Chem. 2000;275:23899-23903.
25. Moynahan ME, Cui TY, Jasin M. Homology-directed DNA repair,
mitomycin-c resistance, and chromosome stability is restored with
correction of a Brca1 mutation. Cancer Res. 2001;61:4842-4850.
26. Hastak K, Alli E, Ford JM. Synergistic chemosensitivity of triple-negative
breast cancer cell lines to poly(ADP-ribose) polymerase inhibition,
gemcitabine, and cisplatin. Cancer Res. 2010;70:7970-7980.
27. Roy R, Chun J, Powell SN. BRCA1 and BRCA2: different roles in a common
pathway of genome protection. Nat Rev Cancer. 2011;12:68-78.
28. Kurian AW, Hare EE, Mills MA, et al. Clinical evaluation of a multiplegene sequencing panel for hereditary cancer risk assessment. J Clin
Oncol. 2014;32:2001-2009.
29. Castera L, Krieger S, Rousselin A, et al. Next-generation sequencing for
the diagnosis of hereditary breast and ovarian cancer using genomic
capture targeting multiple candidate genes. Eur J Hum Genet. 2014;22:
1305-1313.
30. Couch FJ, Hart SN, Sharma P, et al. Inherited mutations in 17 breast
cancer susceptibility genes among a large triple-negative breast cancer
cohort unselected for family history of breast cancer. J Clin Oncol. 2015;
33:304-311.
31. Stefansson OA, Jonasson JG, Johannsson OT, et al. Genomic profiling of
breast tumours in relation to BRCA abnormalities and phenotypes.
32. Byrski T, Huzarski T, Dent R, et al. Response to neoadjuvant therapy with
cisplatin in BRCA1-positive breast cancer patients. Breast Cancer Res
Treat. 2009;115:359-363.
33. Alba E, Chacon JI, Lluch A, et al. A randomized phase II trial of platinum
salts in basal-like breast cancer patients in the neoadjuvant setting.
Results from the GEICAM/2006-03, multicenter study. Breast Cancer
Res Treat. 2012;136:487-493.
34. von Minckwitz G, Loibl S, Schneeweiss A, et al. Early survival analysis of
the randomized phase II trial investigating the addition of carboplatin to
neoadjuvant therapy for triple-negative and HER2-positive early breast
cancer (GeparSixto). 2015, San Antonio, TX: 2015 San Antonio Breast
Cancer Symposium.
35. Staudacher L, Cottu PH, Dieras V, et al. Platinum-based chemotherapy in
metastatic triple-negative breast cancer: the Institut Curie experience.
Ann Oncol. 2011;22:848-856.
36. Khalaf D, Hilton JF, Clemons M, et al. Investigating the discernible and
distinct effects of platinum-based chemotherapy regimens for metastatic triple-negative breast cancer on time to progression. Oncol Lett.
2014;7:866-870.
37. Uhm JE, Park YH, Yi SY, et al. Treatment outcomes and clinicopathologic
characteristics of triple-negative breast cancer patients who received
platinum-containing chemotherapy. Int J Cancer. 2009;124:1457-1462.
38. Villarreal-Garza C, Khalaf D, Bouganim N, et al. Platinum-based chemotherapy in triple-negative advanced breast cancer. Breast Cancer Res
Treat. 2014;146:567-572.
39. Tutt A, Ellis P, Kilburn L, et al. The TNT trial: A randomized phase III trial
of carboplatin (C) compared with docetaxel (D) for patients with
metastatic or recurrent locally advanced triple negative or BRCA1/2
breast cancer (CRUK/07/012). 2014, San Antonio, TX: 2014 San Antonio
Breast Cancer Symposium.
40. Fan Y, Xu BH, Yuan P, et al. Docetaxel-cisplatin might be superior to
docetaxel-capecitabine in the first-line treatment of metastatic triplenegative breast cancer. Ann Oncol. 2013;24:1219-1225.
41
ANDERS ET AL
41. Hu XC, Zhang J, Xu BH, et al. Cisplatin plus gemcitabine versus paclitaxel
plus gemcitabine as first-line therapy for metastatic triple-negative
breast cancer (CBCSG006): a randomised, open-label, multicentre,
phase 3 trial. Lancet Oncol. 2015;16:436-446.
42. Isakoff SJ, Mayer EL, He L, et al. TBCRC009: a multicenter phase II clinical
trial of platinum monotherapy with biomarker assessment in metastatic
triple-negative breast cancer. J Clin Oncol. 2015;33:1902-1909.
43. Abkevich V, Timms KM, Hennessy BT, et al. Patterns of genomic loss of
heterozygosity predict homologous recombination repair defects in
epithelial ovarian cancer. Br J Cancer. 2012;107:1776-1782.
44. Birkbak NJ, Wang ZC, Kim JY, et al. Telomeric allelic imbalance indicates
defective DNA repair and sensitivity to DNA-damaging agents. Cancer
Discov. 2012;2:366-375.
45. Popova T, Manie E, Rieunier G, et al. Ploidy and large-scale genomic
instability consistently identify basal-like breast carcinomas with
BRCA1/2 inactivation. Cancer Res. 2012;72:5454-5462.
46. Powell SN, Riaz N, Mutter RW, et al. A functional assay for homologous
recombination (HR) DNA repair and whole exome sequencing reveal
that HR-defective sporadic breast cancers are enriched for genetic
alterations in DNA repair genes. 2015, San Antonio, TX: 2015 San
Antonio Breast Cancer Symposium.
47. Hodi FS, ODay SJ, McDermott DF, et al. Improved survival with ipilimumab
in patients with metastatic melanoma. N Engl J Med. 2010;363:711-723.
48. Robert C, Long GV, Brady B, et al. Nivolumab in previously untreated
melanoma without BRAF mutation. N Engl J Med. 2015;372:320-330.
49. Hamid O, Robert C, Daud A, et al. Safety and tumor responses with
lambrolizumab (anti-PD-1) in melanoma. N Engl J Med. 2013;369:134-144.
50. Garon EB, Rizvi NA, Hui R, et al; KEYNOTE-001 Investigators. Pembrolizumab for the treatment of non-small-cell lung cancer. N Engl J
Med. 2015;372:2018-2028.
51. Blank C, Gajewski TF, Mackensen A. Interaction of PD-L1 on tumor cells
with PD-1 on tumor-specific T cells as a mechanism of immune evasion:
implications for tumor immunotherapy. Cancer Immunol Immunother.
2005;54:307-314.
52. Keir ME, Butte MJ, Freeman GJ, et al. PD-1 and its ligands in tolerance
and immunity. Annu Rev Immunol. 2008;26:677-704.
53. Butte MJ, Keir ME, Phamduy TB, et al. Programmed death-1 ligand 1
interacts specifically with the B7-1 costimulatory molecule to inhibit
T cell responses. Immunity. 2007;27:111-122.
54. Dunn GP, Bruce AT, Ikeda H, et al. Cancer immunoediting: from
immunosurveillance to tumor escape. Nat Immunol. 2002;3:991-998.
55. Loi S, Sirtaine N, Piette F, et al. Prognostic and predictive value of tumorinfiltrating lymphocytes in a phase III randomized adjuvant breast
cancer trial in node-positive breast cancer comparing the addition of
docetaxel to doxorubicin with doxorubicin-based chemotherapy: BIG
02-98. J Clin Oncol. 2013;31:860-867.
56. Liu S, Lachapelle J, Leung S, et al. CD8+ lymphocyte infiltration is an
independent favorable prognostic indicator in basal-like breast cancer.
57. Liyanage UK, Moore TT, Joo HG, et al. Prevalence of regulatory T cells is
increased in peripheral blood and tumor microenvironment of patients with
pancreas or breast adenocarcinoma. J Immunol. 2002;169:2756-2761.
58. Bates GJ, Fox SB, Han C, et al. Quantification of regulatory T cells enables
the identification of high-risk breast cancer patients and those at risk of
late relapse. J Clin Oncol. 2006;24:5373-5380.
59. Ono M, Tsuda H, Shimizu C, et al. Tumor-infiltrating lymphocytes are
correlated with response to neoadjuvant chemotherapy in triplenegative breast cancer. Breast Cancer Res Treat. 2012;132:793-805.
42
60. Issa-Nummer Y, Darb-Esfahani S, Loibl S, et al. Prospective validation of

immunological infiltrate for prediction of response to neoadjuvant
chemotherapy in HER2-negative breast cancera substudy of the
neoadjuvant GeparQuinto trial. PLoS One. 2013;8:e79775.
61. Ladoire S, Arnould L, Apetoh L, et al. Pathologic complete response to
neoadjuvant chemotherapy of breast carcinoma is associated with the
disappearance of tumor-infiltrating Foxp3+ regulatory T cells. Clin
Cancer Res. 2008;14:2413-2420.
62. Loi S, Michiels S, Salgado R, et al. Tumor infiltrating lymphocytes are
prognostic in triple negative breast cancer and predictive for trastuzumab benefit in early breast cancer: results from the FinHER trial. Ann
Oncol. 2014;25:1544-1550.
63. Mahmoud SM, Paish EC, Powe DG, et al. Tumor-infiltrating CD8
+ lymphocytes predict clinical outcome in breast cancer. J Clin Oncol.
2011;29:1949-1955.
64. Denkert C, Loibl S, Noske A, et al. Tumor-associated lymphocytes as an
independent predictor of response to neoadjuvant chemotherapy in
65. Denkert C, von Minckwitz G, Brase JC, et al. Tumor-infiltrating lymphocytes and response to neoadjuvant chemotherapy with or without
carboplatin in human epidermal growth factor receptor 2-positive
and triple-negative primary breast cancers. J Clin Oncol. 2015;33:
983-991.
66. Mittendorf EA, Philips AV, Meric-Bernstam F, et al. PD-L1 expression in
triple-negative breast cancer. Cancer Immunol Res. 2014;2:361-370.
67. Ghebeh H, Mohammed S, Al-Omair A, et al. The B7-H1 (PD-L1)
T lymphocyte-inhibitory molecule is expressed in breast cancer patients with infiltrating ductal carcinoma: correlation with important
high-risk prognostic factors. Neoplasia. 2006;8:190-198.
68. Emens LA, Braiteh FS, Cassier P, et al. Inhibition of PD-L1 by MPDL3280A
leads to clinical activity in patients with metastatic triple-negative
breast cancer. 2014, San Antonio, TX: 2014 San Antonio Breast Cancer
Sympsoium.
69. Ali HR, Glont SE, Blows FM, et al. PD-L1 protein expression in breast
cancer is rare, enriched in basal-like tumours and associated with
infiltrating lymphocytes. Ann Oncol. 2015;26:1488-1493.
70. Muenst S, Schaerli AR, Gao F, et al. Expression of programmed death
ligand 1 (PD-L1) is associated with poor prognosis in human breast
71. Desmedt C, Haibe-Kains B, Wirapati P, et al. Biological processes associated with breast cancer clinical outcome depend on the molecular
subtypes. Clin Cancer Res. 2008;14:5158-5165.
72. Nagalla S, Chou JW, Willingham MC, et al. Interactions between immunity, proliferation and molecular subtype in breast cancer prognosis.
Genome Biol. 2013;14:R34.
73. Calabro` A, Beissbarth T, Kuner R, et al. Effects of infiltrating lymphocytes
and estrogen receptor on gene expression and prognosis in breast
74. Rody A, Holtrich U, Pusztai L, et al. T-cell metagene predicts a favorable
prognosis in estrogen receptor-negative and HER2-positive breast
cancers. Breast Cancer Res. 2009;11:R15.
75. Nanda R, Chow LQ, Dees EC, et al. A phase Ib study of pembrolizumab
(MK-3475) in patients with advanced triple-negative breast cancer.
2014, San Antonio, TX: 2014 San Antonio Breast Cancer Symposium.
76. Emens LA, Braiteh FS, Cassier P, et al. Inhibition of PD-L1 by MPDL3280A
leads to clinical activity in patients with metastatic triple-negative
breast cancer (TNBC) 2015, Philadelphia, PA: American Association
for Cancer Research 106th Annual Meeting.
BREAST CANCER
Updates and Controversies in

HER2-Positive Breast Cancer
CHAIR
Sunil Verma, MD, MSEd, FRCPC
Cumming School of Medicine, University of Calgary
Calgary, AB, Canada
SPEAKERS
Kimberly L. Blackwell, MD
Durham, NC
Speaker
Miguel Martin, MD, PhD
~
Hospital General Universitario Gregorio Maranon
Madrid, Spain
NIXON AND VERMA
A Value-Based Approach to Treatment of HER2-Positive

Breast Cancer: Examining the Evidence
Nancy Nixon, MD, FRCPC, and Sunil Verma, MD, MSEd, FRCPC
OVERVIEW
Over the past decade, treatment of HER2-positive breast cancer has been revolutionized with the introduction of targeted
therapies. Survival in both early and advanced HER2-positive breast cancer has improved significantly. With evidence for
major clinical benefit, it is imperative that health systems evaluate new treatments to maximize the value of health
expenditures. Physicians, funding agencies, and policy makers are tasked with analyzing available evidence to ensure that
each individual patient receives the optimal treatment in a resource-challenged environment.
pproximately 15% to 20% of breast cancer patients

overexpress HER2 protein. 1 Over the past decade,
treatment of HER2-positive breast cancer has been revolutionized. Although previously a poor prognostic marker,
HER2-positive breast cancer now gives patients an equivalent, if not superior, survival to those patients that are HER2negative.2 In metastatic disease, the median overall survival
(OS) has dramatically increased over the course of 15 years,
from 20 months, 50 months, and beyond.3 For metastatic
patients, American Society of Clinical Oncology (ASCO)
guidelines4 now recommend HER2-targeted therapies in all
lines of treatment of metastatic disease.
TREATMENT OF METASTATIC HER2-POSITIVE

BREAST CANCER
the EGFR family. The rationale for the addition of pertuzumab to trastuzumab is to overcome resistance to trastuzumab caused specifically by formation of HER2:HER3
heterodimers. It was evaluated for metastatic disease in the
CLEOPATRA trial,6 which studied patients with HER2-positive
disease with no prior treatment of metastatic disease and
randomly assigned them 1:1 to trastuzumab and docetaxel,
plus either pertuzumab or placebo. Ninety percent of the
patients in this trial had not been on prior trastuzumab
therapy in the adjuvant setting, meaning the majority of the
patients were naive to anti-HER2 therapies. The results of
this trial established a new first-line standard of metastatic
HER2-positive breast cancer, with an increase in progressionfree survival (PFS) of 6 months (18.6 vs. 12.4 months; p , .001)
and OS (56.5 vs. 40.8 months; p = .002; Table 1).7
First Line
Trastuzumab, a human monoclonal IgG antibody that selectively targets HER2, was the first targeted drug to be
approved. It has since has been shown to be effective in
combination with multiple chemotherapy drugs including
docetaxel, paclitaxel, vinorelbine, and capecitabine. In the
landmark trial by Slamon et al,5 patients were randomly
assigned 1:1 to standard chemotherapy alone versus
standard chemotherapy plus trastuzumab. In this study,
there was a substantial improvement in time to progression (TTP, 7.4 vs. 4.6 months; p , .001). Median OS
increased from 20.3 to 25.1 months (p = .046), establishing
trastuzumab as a standard of care in combination with
taxanes (Table 1).
Pertuzumab, a recombinant humanized monoclonal antibody, binds to a separate domain on the HER2 receptor,
preventing the dimerization of HER2 with other members of
Second Line
For patients progressing on an anti-HER2 therapy combined
with a cytotoxic or endocrine agent, additional anti-HER2
therapy should be offered based on evidence showing it is
beneficial to continue suppression of the HER2 pathway.8-12
Laptinib is an oral tyrosine kinase inhibitor that functions
downstream of HER2, inhibiting both EGFR and HER2 receptors. It was shown to be an effective option for secondline treatment in 2006 by Geyer et al13 for patients who
had locally advanced or metastatic HER2-positive breast
cancer previously treated with regimens that included an
anthracycline, a taxane, and trastuzumab. The addition of
lapatinib resulted in a significant improvement in median
TTP of 8.4 versus 4.4 months at interim analysis, meeting
specified criteria for early reporting (hazard ratio [HR] 0.49;
p , .001). There was no noticeable improvement in OS
From the University of Calgary, Calgary, Alberta, Canada.

Corresponding author: Sunil Verma, MD, MSEd, FRCPC, University of Calgary, 1331 29 St. NW, Calgary, Alberta, Canada; email: drsunil.verma@albertahealthservices.ca.
e56
HER2-POSITIVE BREAST CANCER: A VALUE-ORIENTED APPROACH TO TREATMENT
TABLE 1. Clinical Benefit of Targeted Therapies for Early HER2-Positive Breast Cancer
Regimen
Overall Survival
Disease-Free Survival
Adjuvant early breast

NSABP-B31 and
cancer
NCCTG N983120
Indication
AC T vs. AC TH
HR 0.63; 75.2% vs. 84.0% at

8.4 years follow-up
HR 0.60; p , .001; 62.3% vs.

73.7% at 8.4 years follow-up
10-Year Follow-up Adjuvant early breast

cancer
BCIRG 00621
AC T vs. AC TH vs. TCH
HR 0.63 (AC-TH p , .0001; 85.9% HR 0.72 (AC-TH p , .001; 74.6%

vs. 78.7%)
vs. 67.9%)
HR 0.76 (TCH p = .0075; 83.8% vs. HR 0.77 (TCH p = .0011; 73.0%
78.7%)
vs. 67.9%)
10-Year Follow-up Adjuvant early breast

HERA
cancer
Standard chemotherapy vs. standard

chemotherapy + 1-y trastuzumab
HR 0.74; p , .001; 79.4% vs.

72.9% at 10 years follow-up
HR 0.75; p , .001; 69.3% vs.

62.5% at 10 years follow-up
Abbreviations: T, taxane; H, trastuzumab, C, cyclophosphamide, c, carboplatin; AC, doxorubicin and cyclophosphamide; HR, hazard ratio.
observed in this trial, likely in part because of early crossover

(Table 1).
In 2012, the EMILIA study14 was published, establishing a
new standard of care for second-line treatment. This trial
looked at T-DM1, an antibody-drug conjugate of trastuzumab
and the cytotoxic agent emtansine (DM1). The antibody
trastuzumab binds to the HER2 on tumor cell surfaces and
upon internalization the emtansine moiety is released,
binding to tubules and disrupting microtubule dynamics.
EMILIA compared T-DM1 versus lapatinib plus capectiabine
in patients with locally advanced or metastatic HER2-positive
breast cancer who had previously received zero to more than
three prior regimens. All the patients on this trial had prior
exposure to trastuzumab and a taxane. T-DM1 showed a
significant improvement in median PFS (9.6 vs. 6.4 months, HR
0.65; p , .001) and median OS (30.9 vs. 25.1 months; HR 0.68;
p , .001; Table 1). In addition, rates of grade 3 or higher
adverse events were higher in the lapatinib plus capecitabine
group. This established T-DM1 as the new standard for secondline treatment.
KEY POINTS
Treatment of HER2-positive breast cancer, and its

prognosis, has been revolutionized since the
introduction of targeted treatment.
Significant improvements in overall survival have been
observed in both metastatic and adjuvant settings over
the past decade.
Despite significant clinical benefit with introduction of
targeted treatments, there is a risk that cost on a
population level may become prohibitive for
widespread use.
Physicians, funding agencies, and policy makers are
tasked with finding a means to ensure that treatments
are cost-effective in an environment of limited
resources.
Careful evaluation of the evidence and consideration of
clinical benefit alongside the cost-effectiveness of novel
therapies will help to guide clinical practice.
Beyond Second Line

After progression on trasuzumab, pertuzumab, and treatment with T-DM1, there remain multiple options of treatment, still including anti-HER2targeted treatments. These
include capecitabine plus lapatinib, trastuzumab plus
lapatinib,15-17 or trastuzumab in combination with another
cytotoxic agent. There is insufficient evidence to recommend
one regimen over another. Alternatively, patients may be
considered for enrollment on clinical trial.
In 2015, the THERESA trial was presented at the San
Antonio Breast Cancer Symposium. Showing OS benefit for
T-DM1 for patients who have had prior therapy with taxane,
trastuzumab, and lapatinib. This confirms the efficacy for
T-DM1 in later lines of therapy for patients who may not
have had received T-DM1 in the second-line setting.
TREATMENT OF EARLY HER2-POSITIVE BREAST

CANCER
Adjuvant
With the success of anti-HER2 therapy in the metastatic
setting, trastuzumab was evaluated in the adjuvant setting
in a series of pivotal trials, first reported in 2005. In the HERA
trial,18 patients were randomly assigned 1:1:1 to observation, versus 1 or 2 years of trastuzumab given every 3 weeks.
When compared with observation, trastuzumab given after
primary therapy reduced the rate of recurrence, in particular
distant recurrence, by approximately 50% after interim efficacy analysis with median follow-up of 12 months. This
benefit was confirmed in a 2012 meta-analysis of eight trials
of chemotherapy plus trastuzumab versus trastuzumab
alone.19 The HR for recurrence was 0.60, also with an improvement in OS and HR for death of 0.66. The benefit in OS
was most strongly associated with concurrent administration of trastuzumab with chemotherapy, as opposed to with
sequential treatment. In 2014, Perez et al published the
results of a planned joint analysis on survival from NSABP-31
and NCCTG N9831, where patients were randomly assigned
to doxorubicin plus cyclophosphamide followed by paclitaxel plus trastuzumab for 1 year versus paclitaxel alone.20
Adding trastuzumab led to 37% relative improvement in OS
(HR 0.63; p , .001). An improvement in disease-free survival
(DFS) of 40% was observed (HR 0.60; p , .001) with increase
in 10-year DFS rate from 62.2% to 73.7% (Table 2).
e57
NIXON AND VERMA
TABLE 2. Clinical Benefit of Targeted Therapies for Early and Advanced HER2-Positive Breast Cancer
Indication
Regimen
Overall Survival
Progression-Free Survival
Slamon et al,
20015
First-line advanced breast

cancer
Standard chemotherapy vs. standard chemotherapy + H
25.1 vs. 20.3 months (p = .046)
7.4 vs. 4.6 months

(p , .001)
CLEOPATRA6
First-line advanced breast

cancer
TH + placebo vs. THP
56.5 months vs. 40.8 months

(HR 0.68; p , .001)
7.7 vs. 6. 3 months

(HR 0.68; p , .001)
EMILIA14
Second-line advanced
breast cancer
T-DM1 vs. capecitbine + lapatinib
20.9 vs. 25.1 months (HR 0.68;

p , .001)
9.6 vs. 6.4 months

(HR 0.65; p , .001)
Geyer et al,
200613
Second-line advanced
breast cancer
Lapatinib + capecitabine vs. capecitabine
TTP 8.4 vs. 4.4 months

(HR 0.49; p , .001)
Abbreviations: T, taxane; H, trastuzumab; P, pertuzumab; DM1, emtansine; HR, hazard ratio; TTP, time to progression.
The benefit of adjuvant trastuzumab has been maintained

with longer follow-up. An updated analysis of the HERA trial
showed continued significant DFS and OS benefit of trastuzumab despite substantial crossover (Table 2). Ten-year
follow-up from the BCIRG 006 trial, which randomly assigned
patients to three arms (doxorubicin plus cyclophosphamide
followed by taxane plus or minus trastuzumab, vs. docetaxel
plus carboplatin plus trastuzumab) was presented in 2015
at the San Antonio Breast Cancer Symposium (Table 2).
With 10.3 years of follow-up, there was significant benefit
in OS for both trastuzumab-containing arms compared
with chemotherapy alone (p , .001 for doxorubicin plus
cyclophosphamide followed by docetaxel plus trastuzumab, and p = .0018 for docetaxel plus carboplatin plus
trastuzumab).
Several studies have been done to look at lapatinib in

neoadjuvant treatment, both in combination with or in place
of trastuzumab. Lapatinib has been consistently inferior to
trastuzumab when compared head to head.25-27 The combination was looked at in the NeoALLTO study, which enrolled women with HER2-positive, early breast cancer and
randomly assigned them 1:1:1 to lapatinib, trastuzumab, or
trastuzumab plus laptinib for 6 weeks followed by the antiHER2 treatment combined with paclitaxel for 12 weeks.
After surgery, women received three cycles of FEC followed
by 34 weeks of the assigned anti-HER2 neoadjuvant therapy.
Although patients in the combination group had improvement in pCR, this did not translate to improved DFS or OS in
the confirmatory phase III adjuvant ALTTO study.
Near Future Directions

Neoadjuvant
There is greater interest in evaluating HER2 therapies in the
neoadjuvant setting, given the correlation between pathologic complete response (pCR) and long-term outcomes
including event-free survival (EFS).22 The U.S. Food and Drug
Administration granted approval to the addition of neoadjuvant pertuzumb to trastuzumab based on two randomized phase II studies, in which higher pCR rates were seen
compared with trastuzumab arms. NeoSphere compared four
regimens preoperatively (docetaxel plus trastuzumab vs.
docetaxel plus trastuzumab plus pertuzumab vs. trastuzumab
plus pertuzumab. vs. docetaxel plus pertuzumab).23 After
surgery, patients in the docetaxel arms received three cycles
of 5-fluorouracil, epirubicin, and cyclophosphamide (FEC)
and completed a year of trastuzumab, whereas patients
who had received only antibody treatment received both
docetaxel and FEC followed by trastuzumab for a full year.
pCR rates were 29%, 46%, 17%, and 24% in each group,
respectively. In the TRYPHA-ENA study, women were randomly assigned to trastuzumab plus pertuzumab and concurrent FEC followed by concurrent docetaxel, FEC alone
followed by concurrent docetaxel, or concurrent docetaxel
and carboplatin.24 After surgery, patients completed 1 year
of trastuzumab. The pCR rate was equivalent across arms. In
both studies, patients with ER-negative disease were more
likely to achieve pCR.
e58
There continues to be new data that will add to our

knowledge of how to optimize treatment of early disease.
The APHINITY trial evaluating adjuvant pertuzumab plus
trastuzumab versus trastuzumab after surgery is yet to be
reported but may provide evidence for adjuvant utilization
of pertuzumab. Similarly, the KATHERINE trial, which recently
completed accrual, is evaluating the addition of T-DM1 as adjuvant treatment of patients treated with neoadjuvant chemotherapy and trastuzumab who do not achieve pCR. T-DM1 is
also being evaluated in direct comparison to paclitaxel and
trastuzumab in the adjuvant setting for patients with low-risk
HER2-positive disease (ATEMPT trial).
VALUE-BASED APPROACH
Over the past decade, there has been rapid growth in the
cost of cancer care as a whole. In the United States, it is
expected to increase from $125 billion in 2010 to $158 billion
in 2020.28 Although drug treatment costs account for as
much as 20% of total cost, it is imperative that health
systems evaluate new treatments strategically to maximize
value of health expenditures. In HER2 breast cancer, the
benefits gained by new treatments have been undeniably
important clinically. Physicians, policy makers, and funding
agencies are tasked with identifying economic trade-offs and
ensuring patients have access to treatments with meaningful health benefits in a resource-limited environment.
Clinical Benefit
Value Assessment
The definition of clinical benefit is variable. Ideally, it would

include all patient-important outcomes. In practice, the evidence used to create treatment guidelines is the highest quality
available, such as the randomized controlled trials described. In
ASCOs 2015 value framework, clinical endpoints used to assess
benefit (OS, PFS, DFS) were selected based on their representation of data collected in clinical trials.29 Based on this
framework, Table 1 summarizes the clinical benefit observed in
landmark trials in early breast cancer. Table 2 summarizes
clinical benefit in advanced disease.
Economists use a variety of methods to assess value.

Commonly used in evaluations of medical intervention are
quality-adjusted life-years (QALYs) and incremental cost
effectiveness ratios (ICERs; Table 3). QALY incorporates both
quality and quantity of life achieved with treatment. A
treatment with a low cost per QALY is more cost effective
than one with high cost per QALY. An ICER is simply the ratio
between the difference in cost and the difference in benefit
of two interventions. In cost-effectiveness analyses, ICER are
commonly expressed as incremental cost per QALY. Defining
TABLE 3. Summary of Key North American Cost-Effectiveness Analyses on Targeted HER2 Therapies in Various
Lines of Treatment
Study
Neoadjuvant Pertuzumab
Attard et al
33
Country
HER2 Test and Treatment
ICER/QALY
Canada
HER2 test: not included in analysis
Neosphere: $25,388 per QALY
NeoSphere: neoadjuvant
docetaxel + trastuzumab 6
pertuzumab surgery
fluorouracil + epirubicin +
cyclophosphamide (FEC)
TRYPHAENA: $46,196 per

QALY
TRYPHAENA: docetaxel + carboplatin + trastuzumab +

pertuzumab
Adjuvant Trastuzumab
Kurian et al32
Garrison et al46
United States
United States
Anthracycline based: $39,98

per QALY
Treatment: anthracycline-based
regimens in NSABP B-31 and
NCCTGN9831 trials; nonanthracycline regimen from
BCIRG 006 + trastuzumab for 52
weeks vs. chemotherapy only
Nonanthracycline is more expensive and less effective,

therefore both with and
without trastuzumab exceed $100,000 per QALY
compared with anthracycline regimens
$26,417 per QALY over

a lifetime
Treatment: anyhtacycline-based
regimens from NSABP B-31 and
NCCTG N9831 trials plus
trastuzumab (at actual dose/
duration received in trial) vs.
chemotherapy alone
Metastatic trastuzumab
Elkin et al34
United States
HER2 test: IHC +3; IHC with FISH

for IHC 2+ and 3+; IHC with FISH
for IHC 2+; FISH only
Initial IHC with FISH confirmation: $125,0000 per

QALY
Treatment: trastuzumab +
paclitaxel vs. paclitaxel only
Initial FISH: $145,000 per

QALY
Other strategies for HER2
testing less cost effective
First-line pertuzumab
Durkee et al36
United States
HER2 testing: not included in

analysis
$713,219 per QALY
Treatment: docetaxel + trastuzumab + pertuzumab vs. docetaxel and trastuzumab alone

Metastatic T-DM1
PCODR
Canada

analysis
$162,839 per QALY
Treatment: T-DM1 vs. lapatinib

capecitabine
Metastatic capecitabine + lapatinib
Le et al35
United States

analysis
$166,113 per QALY
Treatment: capecitabine + lapatinib vs. capecitabine alone

Abbreviations: ICER, incremental cost-effectiveness ratio; QALY, quality-adjusted life-years; IHC, immunohistochemistry; FISH, fluorescence in situ hybridization; DM1, emtansine.
e59
NIXON AND VERMA
the threshold for cost effectiveness is a focus for funding

sources worldwide and is variable depending on the funding
party and the basic structure of the health care system. In
most analysis from the United States using USD, a threshold
level of $100,000 per QALY is considered cost-effective
(range 50,000 to three times the United States per capita
gross domestic product, or $160,000 per QALY).30 This
method of assessment is fraught with variability that is
highly subjective to the approach used to estimate OS in the
control group when crossover is allowed in the trial, inclusion of treatment past progression, and the cost of HER2
testing as demonstrated in a systemic review by Parkinson
et al.31
In the adjuvant setting, trastuzumab was shown to be costeffective by Kurian et al, who estimated an ICER of $39,982
USD per QALY.32 Cost-effectiveness of pertuzumab in the
neoadjuvant setting was investigated by Attard et al based
on pCR data from both NeoSphere and TRYPHAENA.33
Published EFS and OS data for patients achieving and not
achieving pCR were used in combination with percentage
achieving pCR in trials to estimate survival. The incremental
cost per QALY ranged from $25,388 CAD (NeoSphere) to
$46,196 CAD (TRYPHAENA analysis), making the addition of
pertuzumab cost-effective with a threshold of $100,000.
For treatment of advanced disease, the cost of even a
single targeted agent increases significantly per QALY. For
HER2 testing and treatment with trastuzumab in the metastatic setting, Elkin et al determined an ICER of $125,000
USD per QALY.34 A similar method was used to determine
the ICER for lapatinib pus capecitabine to be $166,113 USD
per QALY, not including testing for HER2 positivity.35 The
pan-Canadian Oncology Drug Review (pCODR) calculated a
cost of $162,839 CAD for T-DM1 based on EMILIA trial results. The addition of pertuzumab to docetaxel and trastuzumab according to the CLEOPATRA trial, was found to be,
again, incrementally higher in a recent publication, with an
estimated ICER of $713,219 USD per QALY gained.36 Despite
the high cost, most funding agencies have adopted these
therapies into clinical practice. This brings to question what
practical considerations can be made to maximize access to
highly effective, costly treatments in an environment of
limited resources.
PRACTICAL CONSIDERATIONS
Metastatic Disease
There are several limitations to QALY as a means of evaluation for treatment of advanced breast cancer patients. In
the metastatic setting, successful treatments are victims of
their own success, because longer PFS extrapolates to more
time accruing costs for expensive therapies. Additionally,
QALY does not capture an individual patients priorities. For
patients with metastatic disease, the balance between
quality versus quantity of life is a much more personal
determination than for early breast cancer, where the goal
is to cure. It also does not take into account the benefit of
time-off treatment or symptom improvement. Based on
these limitations, we pose questions for discussion that still
e60
focus on treatment according to best evidence, while finding

ways to minimize costs and avoid overtreatment.
Should dual anti-HER2 maintenance be continued after
maximal response? There is limited evidence to answer
whether maintenance with dual anti-HER2 blockade should
be continued once maximal response is achieved. There is,
however, evidence of continued efficacy of trastuzumab
after progression on a trastuzumab-based regimen. A trial
by Von Minckwitz et al looked at the combination of
capecitabine plus trastuzumab versus trastuzumab alone
as a second-line treatment.10 This showed a significant improvement in TTP (8.2 vs. 5.6 months; p = .03) and a nonsignificant trend in OS (25.5 vs. 20.4 months; p = .257). In a
separate study, the combination of trastuzumab plus laptinib
versus lapatinib alone after progression on a trastuzumabbased regimen was investigated.16 This showed a significant
improvement for PFS in the combination arm (12 weeks vs.
8.1 weeks; p = .008) and OS (14 vs. 9.5 months).
Currently, no trials have been done to address continuation of targeted therapy after maximal response is
achieved. In most trials, HER2-targeted therapy was administered until disease progression or until significant
toxicity leading to discontinuation of therapy. Current
recommendations on duration from ASCO and the costeffectiveness analysis mentioned before are based on the
approach used in relevant clinical trials. Potential gains in
both quality of life and cost-effectiveness may be achieved
if we consider treatment breaks in patients who have
sustained a response with no evidence of progression. This
approach needs to evaluated and studied in the metastatic
setting and could be guided by utilization of functional
imaging/circulating tumor cells.
Do all patients need dual anti-HER2 blockade up front? If
we are able to identify a certain subgroup within the HER2
population that derives the greatest benefit from dualantibody treatment, it allows us to be selective in who receives the treatment, sparing others both toxicities and cost.
Baselga et al looked to answer this question by assessing a
protocol-prespecified panel of biomarkers in tumor and
serum samples from patients on the CLEOPATRA trial.17
Biomarkers were identified based on previous understanding of HER2-signaling pathways, or HER2-targeted
therapy resistance. Pertuzumab was found to have a consistent benefit for PFS, independent of biomarker subgroups. This indicates that to our current knowledge, HER2 is
the only marker that should be used for patient selection for
the combination of trastuzumab plus pertuzumab. There is a
need for further studies that may help to elucidate tumor, or
patient-related factors that may ultimately help us determine which patients are best candidates for dual antibodies up front.
One of the key questions that must be asked when applying evidence to practice is whether the patient population
in the study matches the patient being treated in the realworld setting. In the CLEOPATRA trial, 90% of patients were
nave to any HER2-targeted therapy. In practice, many of the

patients we treat in the first line for metastatic disease are
being treated for recurrence, and have already received
antibody treatment in the adjuvant setting. Swain et al
recently published the results that suggest incremental
improvement for patients who relapsed postadjuvant
trastuzumab; however, although the HR was similar, the
prognosis and overall results for this population were inferior compared with de novo patients.37 Forty-one patients
in the control group and 47 patients in the treatment arm
were enrolled after 1 year of adjuvant trastuzumab. The OS
was 46.6 months and 53.8 months, respectively (HR 0.8;
95% CI, 0.441.47), with a large confidence interval owing
to relatively small numbers. The study authors point out,
however, that since the introduction of trastuzumab in the
adjuvant setting the proportion of patients who had recurrent disease had significantly decreased, as shown in the
SystHERs trial.38 This means that the results of the CLEOPATRA trial may reflect what is seen in practice today, with
increased proportion of patients being treated with HER2positive de novo presentation.
Early Breast Cancer

Treatment of early breast cancer is distinct from advanced
breast cancer because the treatment goals are more clearly
defined and uniform across patients. Treatment with the
goal of cure makes decreasing treatment intensity an uncommon target. However, because outcomes of HER2positive breast cancer exceed survival rates of 95%, it is
important to consider whether a subset of patients exists in
whom we can de-escalate therapy.
Is there a population who do not need HER2-targeted
treatment? For patients who are HER2-negative, chemotherapy is generally not recommended unless they are
lymph nodepositive, or high-risk lymph nodenegative. In
the HER2+ population, however, the threshold for treatment
is lower based on our knowledge that even earlier stage
tumors are more likely to recur. OSullivan et al performed a
meta-analysis of five trastuzumab randomized controlled
trials in early breast cancer and tumors 2 cm or smaller.39
They concluded that patients with HER2-positive and tumors
2 cm or smaller had major DFS and OS benefits from trastuzumab as part of the treatment regimen.
For patients with T1a or T1b tumors that are nodenegative, prognosis is less well-defined, even in the setting of HER2 positivity. Ferenbacher et al did a retrospective
analysis of 234 patients with HER2-positive, lymph node
negative, and T1a or T1b tumors.40 Distant invasive recurrence was the primary endpoint. Five-year distant
relapse-free survival was 98.2% for T1a, 89.4% for T1b, and
84.5% for 1-cm tumors. A majority (74%) of these patients
had no adjuvant treatment, raising the possibility of not
requiring therapy for certain T1a patients. In contrast, a
study by Rodriguez et al41 looked at 276 patients with
node-negative, T1a-b HER2-positive breast cancer. Of these
patients, 47% had been given adjuvant chemotherapy and

trastuzumab. Disease-free survival at 40 months was 93% for
patients who had not received chemotherapy versus 99% for
those who had. In multivariate analysis, tumor size was not
associated with clinical outcomes, leading the authors to
state adjuvant treatment should be discussed with all HER2positive T1a-b tumors. Based on the available data, one has
to weigh the potential benefit:risk of toxicity and cost implications specifically for patients with T1a/T1b tumors.
Does everyone need neoadjuvant pertuzumab? In the
adjuvant setting similarly to advanced disease, there is
question of whether patients who are positive for both HER2
and hormone receptors should be treated the same. In the
NeoSphere study, patients were stratified according to
hormone receptor status at the time of randomization. They
subsequently compared pCR results across these groups.
Patients that were estrogen receptor (ER) and progesterone
receptor (PgR)negative showed a pCR rate of 63% when
pertuzumab was added to docetaxel and trastuzumab, compared with 36.8% with trastuzumab and docetaxel alone. In
patients who were also ER-positive, PgR-positive, or both, the
three-drug regimen had a pCR of 26%, compared with 20% with
docetraxel and trastuzumab. The updated EFS data show a
nonsignificant trend for improvement in EFS. One requires
confirmatory adjuvant data to truly assess the benefit of adding
pertuzumab to standard trastuzumab-based therapy.42
Do patients need 1 year of trastuzumab? Duration of
trastuzumab has been looked at in multiple studies. In the
original HERA study, trastuzumab for 1 year plus chemotherapy was compared with chemotherapy alone, and in a
third arm trastuzumab for 2 years was given. At a median
follow-up of 8 years, there was no difference between 1 and
2 years with respect to DFS (HR 0.99; p = .86) or OS (HR 1.05;
p = .63). The other pivotal trials discussed before used the
duration of 1 year somewhat empirically. Shorter duration of
adjuvant trastuzumab was studied in the FinHER trial.43 This
study tested nine cycles of weekly trastuzumab with adjuvant docetaxel or vinorelbine followed by FEC in a subset of
early breast cancer patients with HER2-positive, high-risk
disease (node-positive or node-negative with tumors at least
2 cm). The primary endpoint of recurrence-free survival
favored the trastuzumab arm (HR 0.41; p = .01), as did OS (HR
0.41; p = .07). Documented clinical improvement was similar
to standard of 1-year therapy seen in previous trials.
In a larger phase III study, PHARE randomly assigned 1,693
patients with HER2-positive high-risk (node-positive) disease
to chemotherapy plus 6 months versus 1 year of trastuzumab.44 This study failed to achieve its primary outcome of
noninferiority at a median follow-up of 3.5 years. Additionally, more patients in the 6 months group had a distant
recurrence as the first DFS event.
There are a number of ongoing studies to evaluate
shorter treatments of adjuvant trastuzumab comparing 6 to
12 months (e.g., the Hellenic group trial, the Persephone
study) and 3 to 12 months (e.g., SOLD study, Short-HER
e61
NIXON AND VERMA
study). These studies will provide insight into possibly reducing

the duration of adjuvant treatment and, subsequently, costs.
While awaiting these results, 1 year continues to be the
standard of care.
Biosimilars
The most intuitive way to receive the benefit of targeted
treatments without the cost is to find a cheaper, equally
effective alternative. For this purpose, there has been great
interest in the production of biosimilar agents. Biosimilars
are distinct from generic formulations in that they are not
biochemically identical to the original. Biologic agents are
complex, with larger size and more intricate structure, such
that exact replication is impossible.45 Biosimilars, therefore,
must be tested to ensure noninferior biologic activity, safety,
and efficacy.46 At least 10 trastuzumab biosimilar monoclonal antibodies are in active development. The furthest
along in testing is CT-P6, produced by Celltrion. CT-P6 has
been through phase III testing and has demonstrated an
equivalent overall response rate (primary endpoint) to
trastuzumab in patients with metastatic disease. Based on
this trial, CT-P6 has already been approved as a biosimilar in
South Korea. As this becomes available, the cost of trastuzumab may drop by as much as 40%.
There are important questions raised that must be considered despite the attraction of reduced cost. These include
questions on what patient population was included to establish comparable clinical efficacy. To date, studies have
been in metastatic patients, which is a varied population
compared with patients with early breast cancer. Primary
endpoints have also been questioned, with the overall response rate being most commonly chosen. Another concern
is that, up to now, biosimilar trials have looked at trastuzumab as a single agent, but the new standard of care in firstline advanced disease is a combination treatment with
pertuzumab. Ongoing efforts are underway to evaluate
biosimilars in a neoadjuvant setting and with more clinically
relevant endpoints to guide us about its use in early-stage
and advanced breast cancer.
Overall, we have made major strides in the management
of HER2-positive early-stage and advanced breast cancer.
These impressive gains are based on inclusion of newer
HER2-targeted therapies and the related financial impact of
integrating these drugs. The future in HER2-positive breast
cancer clinical research needs to focus on how best to
optimize therapy that provides us with a more individualized
approach based on inclusion of biomarkers, imaging, and
patient factors.
References
1. Slamon DJ, Clark GM, Wong SG, et al. Human breast cancer: correlation
of relapse and survival with amplification of the HER-2/neu oncogene.
Science. 1987;235:177-182.
2. Zurawksa U, Baribeau DA, Giilck S, et al. Outcomes of HER 2-positive
early-stage breast cancer in the trastuzumab era: a population based
study of Canadian patients. Curr Oncol. 2013;20:6.
3. Verma S, Joy AA, Rayson D, et al. HER story: the next chapter in HER-2directed therapy for advanced breast cancer. Oncologist. 2013;18:
1153-1166.
4. Giordano SH, Temin S, Kirshner JJ, et al; American Society of Clinical
Oncology. Systemic therapy for patients with advanced human epidermal growth factor receptor 2-positive breast cancer: American
Society of Clinical Oncology clinical practice guideline. J Clin Oncol.
2014;32:2078-2099.
5. Slamon DJ, Leyland-Jones B, Shak S, et al. Use of chemotherapy plus a
monoclonal antibody against HER2 for metastatic breast cancer that
overexpresses HER2. N Engl J Med. 2001;344:783-792.
6. Swain SM, Baselga J, Kim SB, et al; CLEOPATRA Study Group. Pertuzumab, trastuzumab, and docetaxel in HER2-positive metastatic breast
7. Swain SM, Kim SB, et al. Confirmatory overall survival (OS) analysis of
CLEOPATRA: A randomized, double-blind, placebo controlled Phase III
study with peruzumab (P), trastuzumab (T), and docetaxel (D) in patients with HER2-postiive first-line metastatic breast cancer (MBC).
Cancer Res. 2012;72:3s (suppl; abstr P5-18-26).
8. Cameron D, Casey M, Oliva C, et al. Lapatinib plus capecitabine in
women with HER-2-positive advanced breast cancer: final survival
analysis of a phase III randomized trial. Oncologist. 2010;15:924-934.
9. Cameron D, Casey M, Press M, et al. A phase III randomized comparison
of lapatinib plus capecitabine versus capecitabine alone in women with
e62
10.
11.
12.
13.
14.
15.
16.
advanced breast cancer that has progressed on trastuzumab: updated

efficacy and biomarker analyses. Breast Cancer Res Treat. 2008;112:
533-543.
von Minckwitz G, du Bois A, Schmidt M, et al. Trastuzumab beyond
progression in human epidermal growth factor receptor 2-positive
advanced breast cancer: a german breast group 26/breast international
group 03-05 study. J Clin Oncol. 2009;27:1999-2006.
Von Minckwitz G, Vogel P, Schmidt M, et al. Trastuzumab treatment
beyond progression in patients with HER-2 positive metastatic breast
cancer: the TBP study (GBG 26/BIG 3-05). Presented at: 30th San
Antonio Breast Cancer Symposium; December 2007; San Antonio, TX.
Abstract 4056. (abstr 4056).
Von Minckwitz G, Zielinski C, Maarteense P, et al. Capecitabine vs.
capecitabine + trastuzumab in patients with HER2-positive metastatic
breast cancer progressing durgin trastuzumab treatment: the TBP
phase III study (GBG 26/BIG 3-05). 2008. J Clin Oncol. 26:47s (suppl;
abstr 1025).
Geyer CE, Forster J, Lindquist D, et al. Lapatinib plus capecitabine for
HER2-positive advanced breast cancer. N Engl J Med. 2006;355:
2733-2743.
Verma S, Miles D, Gianni L, et al; EMILIA Study Group. Trastuzumab
emtansine for HER2-positive advanced breast cancer. N Engl J Med.
2012;367:1783-1791.
Blackwell KL, Burstein HJ, Storniolo AM, et al. Randomized study of
Lapatinib alone or in combination with trastuzumab in women with
ErbB2-positive, trastuzumab-refractory metastatic breast cancer. J Clin
Oncol. 2010;28:1124-1130.
Blackwell KL, Burstein HJ, Storniolo AM, et al. Overall survival benefit
with lapatinib in combination with trastuzumab for patients with human epidermal growth factor receptor 2-positive metastatic breast
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
cancer: final results from the EGF104900 Study. J Clin Oncol. 2012;30:
2585-2592.
Baselga J, Corte s J, S-A Im, et al. Biomarker analyses in CLEOPATRA:
a phase III, placebo-controlled study of pertuzumab in human epidermal growth factor receptor 2positive, first-line metastatic breast
Piccart-Gebhart MJ, Procter M, Leyland-Jones B, et al; Herceptin Adjuvant (HERA) Trial Study Team. Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer. N Engl J Med. 2005;353:
1659-1672.
Moja L, Tagliabue L, Balduzzi S, et al. Trastuzumab containing regimens for early breast cancer. Cochrane Database Syst Rev. 2012;4:
CD006243.
Perez EA, Romond EH, Suman VJ, et al. Trastuzumab plus adjuvant
chemotherapy for human epidermal growth factor receptor 2-positive
breast cancer: planned joint analysis of overall survival from NSABP B-31
and NCCTG N9831. J Clin Oncol. 2014;32:3744-3752.
Slamon DJ, Eiermann W, Robert NJ, et al. Ten year follow up of the
BCIRG-006 Trial comparing doxorubicin plus cyclophosphamide
followed by docetaxel (ACT) with doxorubicin plus cyclosphosphamide
followed by docetaxel and trastuzumab (ACTH) with docetaxel, carboplatin, and trastuzumab (TCH) in HER2+ early breast cancer. Presented at:
San Antonio Breast Cancer Symposium; December 2015; San Antonio,
TX. Abstract S5-S04.
Cortazar P, Zhang L, Untch M, et al. Pathological complete response and
Gianni L, Pienkowski T, Im YH, et al. Efficacy and safety of neoadjuvant
pertuzumab and trastuzumab in women with locally advanced, inflammatory, or early HER2-positive breast cancer (NeoSphere):
a randomised multicentre, open-label, phase 2 trial. Lancet Oncol. 2012;
13:25-32.
Schneeweiss A, Chia S, Hickish T, et al. Pertuzumab plus trastuzumab in
combination with standard neoadjuvant anthracycline-containing and
anthracycline-free chemotherapy regimens in patients with HER2positive early breast cancer: a randomized phase II cardiac safety
study (TRYPHAENA). Ann Oncol. 2013;24:2278-2284.
Baselga J, Bradbury I, Eidtmann H, et al; NeoALTTO Study Team.
Lapatinib with trastuzumab for HER2-positive early breast cancer
(NeoALTTO): a randomised, open-label, multicentre, phase 3 trial.
Lancet. 2012;379:633-640.
Guarneri V, Frassoldati A, Bottini A, et al. Preoperative chemotherapy plus trastuzumab, lapatinib, or both in human epidermal
growth factor receptor 2-positive operable breast cancer: results of
the randomized phase II CHER-LOB study. J Clin Oncol. 2012;30:
1989-1995.
Robidoux A, Tang G, Rastogi P, et al. Lapatinib as a component of
neoadjuvant therapy for HER2-positive operable breast cancer (NSABP
protocol B-41): an open-label, randomised phase 3 trial. Lancet Oncol.
2013;14:1183-1192.
Mariotto AB, Yabroff KR, Shao Y, et al. Projections of the cost of cancer
care in the United States: 2010-2020. J Natl Cancer Inst. 2011;103:
117-128.
Schnipper LE, Davidson NE, Wollins DS, et al; American Society of
Clinical Oncology. American Society of Clinical Oncology Statement: A
Conceptual Framework to Assess the Value of Cancer Treatment Options. J Clin Oncol. 2015;33:2563-2577.
30. Ubel PA, Loewenstein G, Jepson C. Whose quality of life? A commentary

exploring discrepancies between health state evaluations of patients
and the general public. Qual Life Res. 2003;12:599-607.
31. Parkinson B, Pearson SA, Viney R. Economic evaluations of trastuzumab
in HER2-positive metastatic breast cancer: a systematic review and
critique. Eur J Health Econ. 2014;15:93-112.
32. Kurian AW, Thompson RN, Gaw AF, et al. A cost-effectiveness analysis of
adjuvant trastuzumab regimens in early HER2/neu-positive breast
33. Attard CL, Pepper AN, Brown ST, et al. Cost-effectiveness analysis of
neoadjuvant pertuzumab and trastuzumab therapy for locally advanced, inflammatory, or early HER2-positive breast cancer in Canada.
J Med Econ. 2015;18:173-188.
34. Elkin EB, Weinstein MC, Winer EP, et al. HER-2 testing and trastuzumab
therapy for metastatic breast cancer: a cost-effectiveness analysis. J Clin
Oncol. 2004;22:854-863.
35. Le QA, Hay JW. Cost-effectiveness analysis of lapatinib in HER-2-positive
advanced breast cancer. Cancer. 2009;115:489-498.
36. Durkee BY, Qian Y, Pollom EL, et al. Cost-effectiveness of pertuzumab in
human epidermal growth factor receptor 2-positive metastatic breast
37. Swain S, Baselga, J. Treatment of HER2-positive metastatic breast
cancer. N Engl J Med. Correspondence. 2015;372:1964-1965.
38. Tripathy D, Brufsky A, Cobleigh M, et al. Increasing proportion of de
novo compared with recurrent HER2-positive metastatic breast cancer:
early results from the systemic therapies for HER2-positive metastatic
breast cancer registry study. Presented at the 37th Annual San Antonio
Breast Cancer Symposium; December 2014; San Antonio, TX.
39. OSullivan CC, Bradbury I, Campbell C, et al. Efficacy of adjuvant
trastuzumab for patients with human epidermal growth factor receptor
2-positive early breast cancer and tumors #2cm: A meta-analysis of the
randomized trastuzumab trials. J Clin Oncol. 2015;33:2600-2608.
40. Fehrenbacher L, Capra AM, Quesenberry CP Jr, et al. Distant invasive
breast cancer recurrence risk in human epidermal growth factor receptor 2-positive T1a and T1b node-negative localized breast cancer
diagnosed from 2000 to 2006: a cohort from an integrated health care
delivery system. J Clin Oncol. 2006;32:2151-2158.
41. Rodrigues JM, Wassermann J, Albiges L, et al. Trastuzumab treatment
in T1ab, node-negative human epidermal growth factor receptor
2-overexpressing breast carcinomas. J Clin Oncol. 2010;28:e541-e542.
42. Gianni L, Pienkowski T, Im Y, et al. Five-year analysis of the phase II
NeoSphere trial evaluating four cycles of neoadjuvant docetaxel (D)
and/or trastuzumab (T) and/or pertuzumab (P). J Clin Oncol. 2015;33:
15s (suppl; abstr 505).
43. Joensuu H, Bono P, Kataja V, et al. Fluorouracil, epirubicin, and cyclophosphamide with either docetaxel or vinorelbine, with or without
trastuzumab, as adjuvant treatments of breast cancer: final results of
the FINHER trial. J Clin Oncol. 2009;27:5684-5692.
44. Pivot X, Romieu G, Debled M, et al. 6 months versus 12 months of
adjuvant trastuzumab for patients with HER2-positive early breast cancer
PHARE: a randomized phase 3 trial. Lancet Oncol. 2013;14:741-748.
45. Thill M. New frontiers in oncology: biosimilar monoclonal antibodies for
the treatment of breast cancer. Expert Rev Anticancer Ther. 2015;15:
331-338.
46. Garrison LP Jr, Lubeck D, Lalla D, et al. Cost-effectiveness analysis of
trastuzumab in the adjuvant setting for treatment of HER2-positive
breast cancer. Cancer. 2007;110:489-498.
e63

MARTIN AND LOPEZ-TARRUELLA
Emerging Therapeutic Options for HER2-Positive Breast

Cancer
Miguel Martin, MD, PhD, and Sara Lopez-Tarruella,
MD, PhD
OVERVIEW
The natural history of HER2-positive breast cancer has progressively improved since the introduction of the first anti-HER2
directed therapy (trastuzumab). Trastuzumab has significantly increased survival of patients with HER2-positive metastatic
breast cancer and, after the standardization of the use of this drug in the adjuvant setting in 2005, has also avoided many
disease recurrences and, consequently, saved many lives. Later on, the introduction of lapatinib offered new choices for
patients with advanced HER2-positive breast cancer, although the drug has failed to show a clear efficacy in the adjuvant
setting. New promising drugs have been approved to broaden the horizon of HER2-positive breast cancer such as pertuzumab or T-DM1, but we need new options to further improve the management of these diseases. In this review, we cover
new strategies that are currently under evaluation for the treatment of patients with HER2-positive breast cancer, including
new tyrosine kinase inhibitors (neratinib, ONT-380), new antibody-drug conjugates targeting HER2 (MM-302), and new
indications of already approved drugs (T-DM1), as well as the potential dual combinations of anti-HER2 therapy with
phosphoinositide 3-kinase/mTOR or cell cycle inhibitors (palbociclib, abemaciclib). Last but not least, we briefly review a new
paradigm of emerging approaches that involve the host immune response, HER2 breast cancer vaccines, and other immune
strategies, including immune checkpoint inhibition.
dvances in the treatment of patients with HER2-positive

breast cancer have been made over the past few decades, both in the metastatic setting and in the neo/adjuvant
treatment of earlier stages. In the neo/adjuvant setting, the
combination of chemotherapy, trastuzumab, and perhaps
pertuzumab (for high-risk patients) has significantly improved patient outcomes. In the metastatic setting, the
outcome of patients with HER2-positive tumors has also
significantly improved, owing to the introduction of effective
anti-HER therapies, including trastuzumab, pertuzumab,
lapatinib, and T-DM1. The reported median survival time
using these modern combination therapies is now approximately 5 years compared with approximately 1.5 years in
the pretrastuzumab era.1,2
Despite this, 15%20% of patients with HER2-positive localregional breast cancer still relapse after receiving the currently
available neo/adjuvant therapies. In addition, metastatic HER2positive breast cancer essentially remains an incurable disease,
despite the improvement in survival seen after the introduction
of the new anti-HER2 therapies. Newer therapies and newer
approaches are therefore necessary to further improve the
current results. Here, we review the most promising emerging
therapeutic options currently in development for the treatment of HER2-positive breast cancer (Table 1).
ORAL, SMALL-MOLECULE INHIBITORS

Neratinib
Neratinib is an oral, small-molecule, anti-HER receptor tyrosine kinase inhibitor with a complex history. It was designed
in the early 1990s at Lederle Laboratories (Pearl River, NY) and
was later was developed by Wyeth-Ayerst (Quebec, Canada),
Pfizer (Quebec, Canada), and, finally, Puma Biotechnology
(Los Angeles, CA), the current owner of the drug.3,4 Neratinib
is a covalent drug that, differently from lapatinib, binds irreversibly to the ATP active site of the tyrosine kinase domain
of HER2. A phase I study of neratinib among patients with
solid tumors showed that the maximum tolerated dose and
the recommended dose for phase II trials was 320 mg and
240 mg once daily, respectively. Grade 3 diarrhea was the
dose-limiting toxicity.5 The drug has good oral bioavailability
and, in the presence of food, the half-life elimination on day
1 after a single 240-mg administration is 14 hours, supporting
a once-daily dosing regimen.
Neratinib was initially tested in a phase II trial of patients
with breast cancer with HER2-positive metastatic tumors (66
patients with and 70 without prior trastuzumab therapy,
respectively) at a dose of 240 mg once daily. The study
showed that single-agent neratinib has relevant antitumor
activity among patients with HER2-positive breast cancer
~on,
From the Instituto de Investigacion
Sanitaria Gregorio Maran
Universidad Complutense, Madrid, Spain.
~on,
Corresponding author: Miguel Martin, MD, PhD, Servicio de Oncologa Medica,
Instituto de Investigacion
Sanitaria Gregorio Maran
Universidad Complutense, Dr. Esquerdo 46,
28007 Madrid, Spain; email: mmartin@geicam.org.
e64
EMERGING THERAPEUTIC OPTIONS FOR HER2 BREAST CANCER
who were pretreated. Objective response rates (ORRs) were

24% for patients who received prior trastuzumab treatment
and 56% for trastuzumab-naive patients. Responses were of
long duration (9 months for patients pretreated with trastuzumab and 12 months for trastuzumab-naive patients).
Median progression-free survival (PFS) was approximately
5 months for patients pretreated with trastuzumab and
approximately 9 months for trastuzumab-naive patients.
The downside of the treatment was the high incidence of
adverse events, particularly diarrhea and, to a much lesser
extent, nausea, vomiting, and fatigue. Diarrhea was reported
for 85% of patients pretreated with trastuzumab (grade 3 to
4 for 30% of patients). The incidence of diarrhea was greater
in the first week of therapy (85%) but significantly declined
with loperamide therapy from that time onward (60% at
weeks 2 to 4, less than 40% at month 2, and approximately
15% at month 3).6
Based on these appealing phase II data, single-agent
neratinib was compared with the combination of lapatinib
and capecitabine in a phase II noninferiority study. The
comparator was the standard therapy for trastuzumabpretreated HER2-positive metastatic breast cancer at the
time the study was designed (capecitabine plus lapatinib). Of
the study patients, 117 were allocated to receive 240 mg
of neratinib daily and 116 patients received 1,250 mg of
lapatinib daily plus 2,000 mg/m2 of capecitabine daily on
days 1 to 14 of each 21-day cycle. The study failed to show
noninferiority of neratinib compared with lapatinib plus
KEY POINTS
The currently available anti-HER2 therapies have

changed the natural history of HER2-positive breast
cancer, but new therapeutic options are necessary
because the disease is essentially incurable in the
metastatic setting and a relevant proportion of patients
with early-stage disease still relapse in spite of the use of
currently available therapies.
Neratinib is one of the most promising new drugs for
HER2-positive breast cancer, although the management
of its main side effect, diarrhea, should be addressed
and resolved to allow safe use of the drug.
Other drugs, such as ONT-380 (an oral, small-molecule,
HER2-selective inhibitor) and MM-302 (an anti-HER2
antibody-drug conjugate carrying pegylated liposomal
doxorubicin) have shown initial promising activity and
good tolerance among patients with HER2-positive
breast cancer.
Interest in the inhibition of downstream signaling of the
HER2 pathway with mTOR/phosphoinositide 3-kinase/
Akt and CDK4/6 inhibitors is currently under clinical
evaluation.
Some immunologic approaches (vaccines, alone or in
combination with trastuzumab, inhibitors of
immunocheckpoints, and others) are also being tested
in HER2-positive breast cancer.
TABLE 1. Emerging Therapies for HER2-Positive Breast

Cancer
Class
Drug
Oral, Anti-HER2,
Small
Molecules
Neratinib
Tyrosine kinase inhibitor (HER2

and HER1)
ONT-380
Tyrosine kinase inhibitor (HER2)
Anti-HER2 AntiBody-Drug
Conjugates
T-DM1
Trastuzumab-like effect
Intracellular release of emtansine-cytotoxic toxicity in HER2overexpressing cells
MM-302
mTOR/PI3K/Akt
Inhibitors
Mechanism of Action
Intracellular release of pegylated

liposomal doxorubicin in
HER2-overexpressing cells
Everolimus
mTOR inhibition
Buparlisib
Pan-class I PI3K inhibitor
Pictilisib
Pan-class I PI3K inhibitor
Taselisib
Alpha-specific PI3K inhibitor
Alpelisib
Alpha-specific PI3K inhibitor
CDK4/6
Inhibitors
Palbociclib
CDK4/6 inhibition, cell cycle arrest
Abemaciclib
CDK4/6 inhibition, cell cycle arrest
Vaccines
E75
Peptide-based vaccine
GP2
Peptide-based vaccine
Immune
Checkpoint
Inhibitors
Pembrolizumab
AntiPD-1
Abbreviation: PI3K, phosphoinositide 3-kinase.
capecitabine. However, neratib showed indisputable single-agent

activity in the trial, with a response rate of 29% versus 41%
with the combination of lapatinib and capecitabine (p = .067).
Median PFS and overall survival (OS) times were 4.5 and
19.7 months (neratinib) versus 6.8 and 23.6 months for the
combination, respectively. Eighty-five percent of patients taking
neratinib experienced diarrhea (grade 3 or greater for 28% of
patients). Diarrhea was more frequent in the first cycle of therapy
and was properly managed with loperamide or dose modifications for most patients.7
In view of the lack of demonstration of noninferiority of
neratinib versus the standard lapatinib plus capecitabine and to
prepare the registration trial of neratinib for trastuzumabpretreated HER2-positive metastatic breast cancer, a phase
I/II study combining neratinib and capecitabine was conducted.8 The study was carried out in two parts. The first part
was a dose-escalation trial among 33 patients with advanced
solid tumors; they received oral neratinib once a day continuously plus capecitabine twice a day (days 1 to 14 every
3 weeks) at predefined dose levels, to define the recommended dose/schedule for phase II/III trials. In part two, 72
patients with trastuzumab-pretreated HER2-positive metastatic breast cancer received the combination at the recommended dose for phase II trials (240 mg of neratinib per day
continuously plus 1,500 mg/m2 of capecitabine for 2 weeks
every 21 days).
In part two, the ORR was 64% for patients with no prior
lapatinib exposure and 57% for patients previously treated with
lapatinib. Median PFS was approximately 9 and 8 months,
e65

respectively. The most common side effects of the combination

were diarrhea (88%) and palmar-plantar erythrodysesthesia
syndrome (48%). This study showed that the combination of
neratinib and capecitabine possesses significant antitumor
activity in trastuzumab-pretreated metastatic HER2-positive
breast cancer. A randomized registration trial (NALA;
NCT01808573) comparing neratinib plus capecitabine with
lapatinib plus capecitabine is underway.
A randomized phase II trial (NEFERTT; NCT00915018),
which included 479 patients with HER2-positive breast
cancer without any therapy for metastatic disease, compared paclitaxel plus trastuzumab with paclitaxel plus neratinib. No statistically significant differences in PFS (the
primary endpoint) or ORR (one of the trials secondary
endpoints) were seen.9 Grade 3 diarrhea was reported for
approximately 30% of patients receiving neratinib plus
paclitaxel compared with approximately 4% of patients in
the paclitaxel plus trastuzumab arm. Central nervous system
(CNS) progression occurred in 8.3% of patients in the neratinib arm versus 17.3% in the trastuzumab arm (RR 0.48;
p = .002). However, the proportion of patients with CNS
involvement at study initiation was superior in the paclitaxel
arm (12 patients; 5.1%) compared with the neratinib arm
(6 patients; 2.5%).
The significant activity of neratinib observed in metastatic
HER2-positive breast cancer has prompted the evaluation of
the drug in earlier stages. A phase II trial conducted by the
NSABP group (NSABP FG-7 Trial) compared weekly paclitaxel
plus neratinib or trastuzumab or the combination of neratinib and trastuzumab followed by standard doxorubicin
and cyclophosphamide as neoadjuvant therapy among
126 patients with HER2-positive locally advanced, operable
breast cancer.10 The proportions of patients with inflammatory breast cancer (a bad prognostic feature) were 16.7%
(neratinib arm) and 9.5% (in both the trastuzumab and the
trastuzumab plus neratinib arms). The pCR rates (breast plus
axillary lymph nodes) were 38.1% (trastuzumab arm), 33.3%
(neratinib arm), and 50% (neratinib plus trastuzumab arm).
The apparent superiority of the combination arm was
mainly attributable to the effect on patients with hormone
receptor (HR)negative disease (pCR rates of 57.1%, 46.2%,
and 73.7%, in the trastuzumab, neratinib, and combination
arms, respectively). For patients with HR-positive disease,
the pCR rates were not significantly different among arms
(29.6%, 27.6%, and 30.4%, respectively). Grade 3 diarrhea
was the most relevant side effect of neratinib (31% in both
neratinib arms vs. 0% in the trastuzumab arm). Intensive
primary prophylaxis with loperamide in neratinib-treated
patients implemented at European sites apparently reduced
the incidence of grade 3 diarrhea (17% in the neratinib
arm and 24% in the neratinib plus trastuzumab arm across
all cycles).
Neratinib has also been evaluated in the I-SPY 2 program,
a multicenter neoadjuvant trial among women with operable breast cancer using adaptive randomization within
biomarker subtypes to evaluate novel antitumor drugs added
to conventional therapy.11 I-SPY 2 intends to identify new drugs
e66
that, in combination with standard therapy, meet a high

Bayesian predictive probability of statistical significance in a
future phase III neoadjuvant trial performed among the same
population. Patients with HER2-positive breast cancer were
randomly assigned to receive 12 weekly administrations of
paclitaxel plus trastuzumab or paclitaxel plus neratinib, followed by four cycles of doxorubicin hydrochloride plus cyclophosphamide (AC) before surgery. Neratinib met the
predictive probability criterion in the population with HRnegative/HER2-positive tumors. A phase III confirmatory trial
with neratinib is planned for patients with HER2-positive breast
cancer (I-SPY 3, a phase III registration trial).
Neratinib has been evaluated in an adjuvant phase III trial,
the ExteNET study. Eligible patients had operable-stage
HER2-positive breast cancer with no relapse after surgery,
chemotherapy, and up to 2 years after the completion of
1 year of trastuzumab. Patients were randomly assigned to
receive neratinib or placebo for 1 year (together with
hormonal therapy for patients with HR-positive tumors). In
the initial design, the trial had a 90% power to detect a
hazard ratio of 0.7 for invasive disease-free survival (iDFS),
at a two-sided 5% significance level. The study design suffered from some modifications over time because the drug
owner changed and three different sponsors assumed responsibility. In the initial design, patients who were T$1c
node negative were eligible, a sample size of 3,850 patients
was established, and iDFS was the main endpoint. In view of
the good outcome of node-negative patients with standard
trastuzumab-containing chemotherapy later reported in
other adjuvant trials (particularly the BCIRG 006 study), an
initial amendment restricted recruitment to patients with
node-positive disease with trastuzumab completion 1 year
or earlier before randomization. In October 2011, the new
sponsor introduced two key changes: cessation of enrollment (only 2,842 patients of the initially planned of 3,850
had been enrolled at that time) and shortening of follow-up
to 2 years from randomization. In January 2014, a new
amendment re-established the extension of data collection
for disease events and deaths to 5 years postrandomization
for consenting patients and the primary endpoint of iDFS at
2 years of follow-up was resumed. Treatment allocation
remained blinded to the sponsor prior to the first analysis
and an independent statistical evaluation by a reputed
biostatistician who had access to all data was performed.
The detailed analyses showed that the results remained
consistent throughout the trial and were trustable. The
primary analysis was first reported in 201512 and published
in detail in 2016.13 At 2 years from randomization, the hazard
ratio for iDFS in the neratinib group compared with the
placebo group was 0.67 (p = .009), with 2-year iDFS rates of
93.9% and 91.6%, respectively. A preplanned subgroup
analysis according to HR status showed that patients
with HR-positive tumors (the majority of whom were receiving concurrent hormonal therapy) obtained the greatest
benefit with neratinib (hazard ratio for iDFS of 0.51, p = .001),
whereas patients with HR-negative tumors had little or
no benefit (hazard ratio for iDFS, 0.93; p = .735; interaction
p = .054). Another preplanned analysis of iDFS for 1,463

patients with centrally confirmed HER2-positive disease
showed a hazard ratio of 0.51 (95% CI, 0.330.77; p = .002).
The 3-year update of the ExteNET trial was recently reported
and confirmed the earlier findings.14 OS data were not
mature but will continue to be monitored.
Diarrhea was the most frequent side effect of neratinib in
the ExteNET trial and occurred among most patients (grade 2
in 32.5%, grade 3 in 39.8%, and grade 4 in 0.1%). Ustaris
et al15 recently reviewed the characteristics and treatment
of neratinib-induced diarrhea. This side effect usually starts
during the first month of therapy and resolves or decreases
later among most patients either spontaneously or with
loperamide or dose reductions. Diarrhea is a cause of dose
reduction for 10%15% of patients. Permanent discontinuation of neratinib because of diarrhea occurs in a minority
of patients (0%14% of patients in the reviewed trials) and
seems to be reduced by the introduction of early, intense
loperamide prophylaxis. A prospective clinical trial is evaluating the efficacy of prophylactic loperamide among patients with HER2-positive operable breast cancer receiving
extended adjuvant neratinib (NCT02400476). Loperamide is
administered at a dose of 4 mg three times a day for the first
2 weeks of neratinib and 4 mg twice a day until week 8.
ONT-380
ONT-380 is an oral, small-molecule, HER2-selective inhibitor in
clinical development. It is speculated that the lack of inhibition
of HER1 could translate into a lower frequency of skin and
gastrointestinal toxicity with respect to HER2/HER1 dual inhibitors (i.e., lapatinib, or neratinib). In fact, a phase I trial with
ONT-380 as a single agent showed no treatment-related grade
3 diarrhea and only minimal skin toxicity. In a phase Ib trial of
ONT-380 in combination with T-DM1 among patients who
received prior taxane and trastuzumab treatment, a 41% ORR
was reported.16 A second study reported the CNS activity of
ONT-380 in combination with either T-DM1 or trastuzumab or
capecitabine. Patients with brain metastases assessable for
response were included in the combined analysis. Responses
and clinical benefit in the CNS were reported with the three
combinations tested, supporting future development of the
drug for this particular indication.17
ANTIBODY-DRUG CONJUGATES
T-DM1
T-DM1 is the first antibody-drug conjugate (ADC) approved for
metastatic HER2-positive breast cancer for the indication of
patients pretreated with taxanes and trastuzumab.18 The role
of T-DM1 in earlier stages of the disease is currently under
evaluation. The KAITLIN trial (NCT01966471) is an adjuvant
study in operable HER2-positive breast cancer. After surgery,
patients are randomly assigned to receive four cycles of
anthracycline-containing chemotherapy followed by combination treatment with taxane, trastuzumab, and pertuzumab
or with T-DM1 and pertuzumab. The study enrollment has
stopped and the protocol was amended in view of the results of
the MARIANNE trial, in which T-DM1 plus pertuzumab was not

superior to taxane plus trastuzumab as a first-line treatment for
patients in the metastatic setting.
The ADAPT trial19 is a phase II trial that enrolled 375 patients
with HR-positive, operable, HER2-positive tumors and randomly assigned them to receive neoadjuvant T-DM1, T-DM1
plus endocrine therapy, or trastuzumab plus endocrine therapy
for 12 weeks prior to surgery. The pathologic complete response rates (breast plus axilla) were 41%, 41.5%, and 15.1%,
respectively.
The phase III KATHERINE trial (NCT01772472) compares
adjuvant trastuzumab versus adjuvant T-DM1 for patients with
HER2-positive tumors in which the tumor persisted in the
breast or axilla after neoadjuvant chemotherapy plus trastuzumab. The KRISTINE trial (TRIO-021; NCT02131064) is an
ongoing phase III trial in which patients are randomly assigned
to neoadjuvant standard docetaxel, carboplatin, trastuzumab,
and pertuzumab or to T-DM1 plus pertuzumab, with pCR as the
main endpoint.
MM-302
MM-302 is an ADC composed of a HER2-directed antibody
(lacking anti-HER2 activity) linked to pegylated liposomal
doxorubicin. The antibody acts as a carrier delivering pegylated
liposomal doxorubicin into HER2-positive breast cancer cells.
The drug has been tested in a phase I trial with 69 patients
with HER2-positive metastatic breast cancer after a median of
four prior regimens for metastatic disease. Patients were
treated with either MM-302 alone, MM-302 plus trastuzumab,
or MM-plus trastuzumab and cyclophosphamide. The most
frequent adverse events (. 20% of patients) were constipation, cough, decreased appetite, diarrhea, dyspnea, fatigue, nausea, neutropenia, stomatitis, and vomiting. As a
single agent, a maximum tolerated dose was not reached at
50 mg/m2. Cardiotoxicity was seen only among patients with
extensive prior exposure to regular anthracyclines. Antitumor
activity was observed for 49 patients treated with 30 mg/m2 of
MM-302 or greater, alone or in combination with trastuzumab,
with a response rate of 12% and median PFS of 7.6 months.
Responses were seen for patients pretreated with trastuzumab, T-DM1, and pertuzumab. The median PFS among the 13
patients receiving MM-302 plus trastuzumab and cyclophosphamide was 10.6 months.20 In view of these data, a randomized phase II study (the HERMIONE trial; NCT02213744) is
ongoing. The study includes patients with metastatic HER2positive breast cancer who received prior trastuzumab, T-DM1,
and pertuzumab treatment (but not anthracyclines in any
setting) and randomly assigns patients to receive MM-302 plus
trastuzumab or trastuzumab plus a single chemotherapy agent
at the physicians choice.
INHIBITORS OF DOWNSTREAM SIGNALING

PATHWAYS
mTOR/Phosphoinositide 3-Kinase/Akt Inhibitors
The mTOR/phosphoinositide 3-kinase (PI3K)/Akt pathway
plays an important role in the downstream signaling of
e67

the HER2 pathway and is a potential target for combined

therapies.21 The BOLERO 3 trial studied adding the mTOR
inhibitor everolimus to conventional chemotherapy
(vinorelbine-trastuzumab) among patients with metastatic
HER2-positive breast cancer who were pretreated with
trastuzumab and taxanes. The addition of everolimus significantly improved PFS compared with vinorelbinetrastuzumab alone.22 The BOLERO 3 trial provided the
proof of principle that the blockade of mTOR can play a role
in the management of HER2-positive breast cancer; however, the significant increase in toxicity associated with
everolimus administration makes this combination unattractive. The BOLERO 1 trial, a first-line study comparing
paclitaxel plus trastuzumab with or without everolimus,
failed to show any significant advantage in PFS with everolimus in the overall group of 719 patients. However, the HRnegative subpopulation obtained an apparent benefit with
the addition of everolimus (median PFS of 20.27 months vs.
13.08 months with placebo, p = .0049), although the preplanned protocol-specified significance threshold (p = .0044)
was not crossed.23 Slamon et al24 used next-generation
sequencing (NGS) and immunohistochemistry (IHC) to establish the status of the PI3K pathway in tumor samples
(mainly primary tumors) from patients from the BOLERO
1 and 3 trials. Exons of 282 cancer-related genes were analyzed by NGS, whereas PTEN (phosphatase and tensin
homolog) levels were determined by IHC. Patients with low
PTEN or known PIK3CA or AKT1 E17K mutations were
considered to have PI3K pathway activation. Patients with
PI3K pathway hyperactivity showed a significantly better PFS
with exemestane, but the remaining patients did not.24 The
role of everolimus in the routine therapy of metastatic HER2positive breast cancer is therefore currently undefined.
Clinical trials of PI3K pathway inhibitors (both pan-class
I and alpha-specific inhibitors) for the treatment of metastatic HER2-positive breast cancer are underway. In a phase
Ib trial of buparlisib plus trastuzumab for patients with HER2positive metastatic breast cancer that progressed while they
were taking trastuzumab, the combination was well tolerated and showed some activity; at the recommended dose
for phase II trials, there were two partial responses (17%)
and seven disease stabilizations of 6 weeks or longer
(58%).25 Several studies are testing the activity of other
PI3K inhibitors, including alpelisib (NCT02038010), taselisib
(NCT02390427), and pictilisib (NCT00960960).
CDK4/6 Inhibitors
CDK4/6 controls a key pathway downstream of HER2. The
activated cyclin DCDK4/6 heterodimer promotes cell cycle
progression. Deregulation of cell cycle control is not uncommon in estrogen receptorpositive and HER2-positive
breast cancer; therefore, inhibition of CDK4/6 could be a
way to augment the effectiveness of standard anti-HER2
therapies.
Palbociclib, a CDK4/6 inhibitor, is synergistic with trastuzumab
and T-DM1 in HER2-positive preclinical models.26,27 A phase Ib
e68
study is exploring the combination of palbociclib and T-DM1

in metastatic HER2-positive breast cancer after prior trastuzumab or other HER2-directed therapies (NCT01976169).
The PATRICIA trial (NCT02448420) is a phase II study of
palbociclib and trastuzumab with or without letrozole among
patients with HER2-positive metastatic breast cancer who
have received chemotherapy and treatment with trastuzumab for their metastatic disease.
Abemaciclib, another CDK4/6, has been safely combined
with trastuzumab in a phase Ib study (NCT02057133). Diarrhea was a dose-limiting toxicity. The recommended dose
for further studies is 150 mg/12 hours.28
VACCINES AND OTHER FORMS OF

IMMUNOTHERAPY
Tumor Vaccines
This strategy aims to amplify the patients immune response
against HER2-positive breast cancer. The majority of breast
cancer vaccine trials were initially performed in advanced
stages of the disease and the results were somewhat disappointing.29 The high genomic instability and the loss of
antigens,30 together with defects in the antitumor immune
response (T-cell activation/function and regulation) found in
metastatic breast cancer, can explain these negative results.
Afterward, the cornerstones of clinical development of
cancer vaccines were redefined,31 emphasizing the need for
selecting an appropriate clinical context; herein, early or
minimal residual disease stages were proposed as more
suitable scenarios to be explored. Several vaccine formulations have been tested among patients with HER2-positive
breast cancer.32
Peptide-based vaccines use antigen epitopes of cancer
cells combined with different adjuvants to elicit immune
responses. The main advantages of this approach include
their easy manufacture at low cost, storage stability, safety,
and measurable antigen-specific immune response in the
blood (direct monitoring of T-cell response).33
E75 is an immunogenic peptide derived from the extracellular domain of HER2. Small-size phase I trials tested E75
with different immune-adjuvants in metastatic breast and
ovarian cancer, demonstrating that the peptide was able to
induce specific cytotoxic T-lymphocyte responses with a
favorable safety profile. However, no meaningful antitumor
activity was seen. Because the metastatic studies showed
that this active immunization could last several years after
the vaccination process, the vaccine was later tested in the
adjuvant setting. E75, in combination with granulocytemacrophage colony-stimulating factor (GM-CSF), was evaluated in a small randomized trial of patients with high-risk
breast cancer with any level of HER2 and HLA-A2+, demonstrating its safety and effectiveness in eliciting immune
response in vivo in node-positive breast cancer. Recurrence
rates were lower in the vaccination arm (21% in the observation group vs. 8% in the vaccine group at a median
22 months of follow-up).34 The final results of the phase I/II
clinical trial of E75 among 195 patients with early breast
cancer were recently reported with a 60-month follow-up
time. A nonstatistically significant difference in 5-year

disease-free survival (DFS) in favor of the vaccine group was
reported (89.7% vs. 80.2%, p = .08). A phase III clinical trial
with E75 is currently ongoing (NCT01479244).
GP2 is a peptide from the transmembrane HER2 protein
domain found to be able to stimulate CD8 lymphocytes to
destroy HER2-overexpressing cells. The preliminary results
of a phase II trial among patients with HLA-A2+ with any level
of expression of HER2 were recently reported.35 The combination of GP2 plus GM-CSF was compared with GM-CSF
alone in patients with breast cancer whose tumors had
diverse expression of HER2. The treatment was well tolerated. In the subset of patients with HER2-overexpressing
tumors treated with trastuzumab, the combination of GP2
vaccine plus GM-CSF was associated with a numerically
superior DFS (94% vs. 89% in the GM-CSF group, p = .86).
Other immunogenic approaches are also under clinical
testing, including peptides derived from the HER2 intracellular
domain, DNA-based and whole tumor cell vaccines (either
autologous or from cell-line cultures), and dendritic cell

vaccines.36,37
The synergy of trastuzumab and vaccines was explored in a
phase I/II trial obtaining prolonged and robust T-cell responses with very low toxicity.38 Ongoing trials are even
considering the addition of chemotherapy to vaccine/
antibody combinations (NCT00266110). A further step in
the HER2 vaccination field is the generation of HER2 vaccineprimed autologous T cells for therapeutic infusion reported
to be feasible and well tolerated in early phase I trials.39
Immune Checkpoint Inhibitors

The clinical testing of immune checkpoint inhibitors in breast
cancer, initially focused on triple-negative breast cancer, has
recently been extended to the HER2-positive subtype.
PANACEA (NCT02129556) is a phase IB/II trial evaluating the
efficacy of pembrolizumab and trastuzumab in trastuzumabresistant metastatic breast cancer.
References
1. Swain SM, Baselga J, Kim SB, et al; CLEOPATRA Study Group. Pertuzumab, trastuzumab, and docetaxel in HER2-positive metastatic breast
3. Tsou HR, Overbeek-Klumpers EG, Hallett WA, et al. Optimization of 6,7disubstituted-4-(arylamino)quinoline-3-carbonitriles as orally active,
irreversible inhibitors of human epidermal growth factor receptor-2
kinase activity. J Med Chem. 2005;48:1107-1131.
4. Lopez-Tarruella
S, Jerez Y, Marquez-Rodas I, et al. Neratinib (HKI-272) in

the treatment of breast cancer. Future Oncol. 2012;8:671-681.
5. Wong KK, Fracasso PM, Bukowski RM, et al. A phase I study with
neratinib (HKI-272), an irreversible pan ErbB receptor tyrosine kinase
inhibitor, in patients with solid tumors. Clin Cancer Res. 2009;15:
2552-2558.
6. Burstein HJ, Sun Y, Dirix LY, et al. Neratinib, an irreversible ErbB receptor
tyrosine kinase inhibitor, in patients with advanced ErbB2-positive
7. Martin M, Bonneterre J, Geyer CE Jr, et al. A phase two randomised
trial of neratinib monotherapy versus lapatinib plus capecitabine
combination therapy in patients with HER2+ advanced breast cancer.
Eur J Cancer. 2013;49:3763-3772.
8. Saura C, Garcia-Saenz JA, Xu B, et al. Safety and efficacy of neratinib in
combination with capecitabine in patients with metastatic human
epidermal growth factor receptor 2-positive breast cancer. J Clin Oncol.
2014;32:3626-3633.
9. Awada A, Colomer R, Bondarenko I, et al. Efficacy and CNS progression
analysis from the randomized phase 2 trial of neratinib + paclitaxel vs
trastuzumab + paclitaxel as first-line treatment for HER2+ MBC
(NEfERTT). J Clin Oncol. 2015;33 (suppl; abstr 610).
10. Jacobs SA, Robidoux A, Garcia JMP, et al. NSABP FB-7: a phase II
randomized trial evaluating neoadjuvant therapy with weekly paclitaxel
(P) plus neratinib (N) or trastuzumab (T) or neratinib and trastuzumab
(N+T) followed by doxorubicin and cyclophosphamide (AC) with
postoperative T in women with locally advanced HER2-positive breast
cancer. Presented at: San Antonio Breast Cancer Symposium. San
Antonio, TX; 2015. Abstract PD5-04.
11. Park JW, Liu MC, Yee D, et al. Neratinib plus standard neoadjuvant
therapy for high-risk breast cancer: efficacy results from the I-SPY 2
TRIAL. Cancer Res. 2014;74 (abstr CT227).
12. Chan A, Delaloge S, Holmes FA, et al. Neratinib after adjuvant chemotherapy and trastuzumab in HER2-positive early breast cancer:
primary analysis at 2 years of a phase 3 randomized, placebo-controlled
trial (ExteNET). J Clin Oncol. 2015;33 (suppl; abstr 508).
13. Chan A, Delaloge S, Holmes FA, et al. Neratinib after trastuzumab-based
adjuvant therapy in HER2-positive breast cancer (ExteNET): a randomized, double-blind, phase 3 trial. Lancet Oncol. 2016;17:367-377.
14. Chan A, Delaloge S, Holmes FA, et al. Neratinib after trastuzumab-based
adjuvant therapy in early-stage HER2+ breast cancer: 3-year analysis
from a phase 3 randomized, placebo-controlled double-blind trial
(ExteNET). Presented at: San Antonio Breast Cancer Symposium. San
Antonio, TX; 2015. Abstract S5-02.
15. Ustaris F, Saura C, Di Palma J, et al. Effective management and prevention of neratinib-induced diarrhea. Am J Hematol Oncol. 2015;11:
13-22.
16. Ferrario C, Hamilton E, Aucoin N, et al. A phase 1b study of ONT 380, an
oral HER2-specific inhibitor, combined with ado trastuzumab emtansine
(T-DM1), in HER2+ metastatic breast cancer (MBC). Presented at: San
Antonio Breast Cancer Symposium. San Antonio, TX; 2015. Abstract P414-20.
17. Murthy RK, Hamilton E, Borges VF, et al. ONT-380 in the treatment of
HER2+ breast cancer central nervous system (CNS) metastases (mets).
Presented at: San Antonio Breast Cancer Symposium. San Antonio, TX;
2015. Abstract P4-14-19.
18. Krop IE, Kim SB, Gonza lez-Martn A, et al; TH3RESA Study Collaborators. Trastuzumab emtansine versus treatment of physicians
choice for pretreated HER2-positive advanced breast cancer
(TH3RESA): a randomised, open-label, phase 3 trial. Lancet Oncol.
2014;15:689-699.
19. Harbeck N, Gluz O, Christgen M, et al. Final analysis of WSG-ADAPT HER2
+/HR+ phase II trial: Efficacy, safety, and predictive markers for 12weeks of neoadjuvant TDM1 with or without endocrine therapy versus
trastuzumab+endocrine therapy in HER2-positive hormone-receptorpositive early breast cancer. Presented at: San Antonio Breast Cancer
Symposium. San Antonio, TX; 2015. Abstract S5-03.
e69

20. LoRusso P, Krop I, Miller K, et al. A phase I study of MM-302, a HER2targeted PEGylated liposomal doxorubicin, in patients with HER2+
metastatic breast cancer. Cancer Res. 2015;75 (abstr CT234).
21. Rexer BN, Arteaga CL. Optimal targeting of HER2-PI3K signaling in breast
cancer: mechanistic insights and clinical implications. Cancer Res. 2013;
73:3817-3820.
22. Andre F, ORegan R, Ozguroglu M, et al. Everolimus for women with
trastuzumab-resistant, HER2-positive, advanced breast cancer
(BOLERO-3): a randomised, double-blind, placebo-controlled phase 3
23. Hurvitz SA, Andre F, Jiang Z, et al. Combination of everolimus with
trastuzumab plus paclitaxel as first-line treatment for patients with
HER2-positive advanced breast cancer (BOLERO-1): a phase 3, randomised, double-blind, multicentre trial. Lancet Oncol. 2015;16:816-829.
24. Slamon DJ, Hurvitz SA, Chen D, et al. Predictive biomarkers of everolimus efficacy in HER2+ advanced breast cancer: combined exploratory
analysis from BOLERO-1 and BOLERO-3. J Clin Oncol. 2015;33 (suppl;
abstr 512).
25. Saura C, Bendell J, Jerusalem G, et al. Phase Ib study of buparlisib plus
trastuzumab in patients with HER2-positive advanced or metastatic
breast cancer that has progressed on tsrastuzumab-based therapy. Clin
Cancer Res. 2014;20:1935-1945.
26. Cadoo KA, Gucalp A, Traina TA. Palbociclib: an evidence-based review of
its potential in the treatment of breast cancer. Breast Cancer (Dove Med
Press). 2014;6:123-122.
27. Knudsen E, Cox D, Franco J, et al. Targeting CDK4/6 in HER2 positive
breast cancer: therapeutic effect, markers, and combination strategies.
Ann Oncol. 2014;25(suppl 1):i21-i22.
28. Goetz MP, Beeram M, Beck T, et al. Abemaciclib, an inhibitor of CDK4
and CDK6, combined with endocrine and HER2-targeted therapies for
women with metastatic breast cancer. Presented at: San Antonio Breast
Cancer Symposium. San Antonio, TX; 2015. Abstract P4-13-25.
e70
29. Rosenberg SA, Yang JC, Restifo NP. Cancer immunotherapy: moving
beyond current vaccines. Nat Med. 2004;10:909-915.
30. Schreiber RD, Old LJ, Smyth MJ. Cancer immunoediting: integrating
immunitys roles in cancer suppression and promotion. Science. 2011;
331:1565-1570.
31. Hoos A, Parmiani G, Hege K, et al; Cancer Vaccine Clinical Trial Working
Group. A clinical development paradigm for cancer vaccines and related
biologics. J Immunother. 2007;30:1-15.
32. Milani A, Sangiolo D, Montemurro F, et al. Active immunotherapy in
HER2 overexpressing breast cancer: current status and future perspectives. Ann Oncol. 2013;24:1740-1748.
33. Clifton GT, Mittendorf EA, Peoples GE. Adjuvant HER2/neu peptide
cancer vaccines in breast cancer. Immunotherapy. 2015;7:1159-1168.
34. Peoples GE, Gurney JM, Hueman MT, et al. Clinical trial results of a
HER2/neu (E75) vaccine to prevent recurrence in high-risk breast cancer
patients. J Clin Oncol. 2005;23:7536-7545.
35. Schneble EJ, Perez SA, Murray JL, et al. Primary analysis of the prospective, randomized, phase II trial of GP2+GMCSF vaccine versus GMCSF alone administered in the adjuvant setting to high-risk breast
cancer patients. J Clin Oncol. 2014;32 (suppl 26; abstr 134).
36. Disis ML, Schiffman K, Guthrie K, et al. Effect of dose on immune response in patients vaccinated with an HER-2/neu intracellular domain
proteinbased vaccine. J Clin Oncol. 2004;22:1916-1925.
37. Morse MA, Hobeika A, Osada T, et al. Long term disease-free survival
and T cell and antibody responses in women with high-risk Her2+ breast
cancer following vaccination against Her2. J Transl Med. 2007;5:42.
38. Disis ML, Wallace DR, Gooley TA, et al. Concurrent trastuzumab and
HER2/neu-specific vaccination in patients with metastatic breast cancer. J Clin Oncol. 2009;27:4685-4692.
39. Disis ML, Dang Y, Coveler AL, et al. HER-2/neu vaccine-primed autologous T-cell infusions for the treatment of advanced stage HER-2/neu
expressing cancers. Cancer Immunol Immunother. 2014;63:101-109.

AND EPIDEMIOLOGY
Controversies in Genetic
Evaluation
CHAIR
Noah D. Kauff, MD
New York, NY
SPEAKERS
Claudine Isaacs, MD, FRCPC
Georgetown Lombardi Comprehensive Cancer Center
Washington, DC
Victoria Raymond, MS, CGC
University of Michigan Medical School
Ann Arbor, MI
LYNCE AND ISAACS
How Far Do We Go With Genetic Evaluation? Gene, Panel, and

Tumor Testing
Filipa Lynce, MD, and Claudine Isaacs, MD, FRCPC
OVERVIEW
The traditional model by which an individual was identified as harboring a hereditary susceptibility to cancer was to test for a mutation
in a single gene or a finite number of genes associated with a particular syndrome (e.g., BRCA1 and BRCA2 for hereditary breast and
ovarian cancer or mismatch repair genes for Lynch syndrome). The decision regarding which gene or genes to test for was based on
a review of the patients personal medical history and their family history. With advances in next-generation DNA sequencing
technology, offering simultaneous testing for multiple genes associated with a hereditary susceptibility to cancer is now possible.
These panels typically include high-penetrance genes, but they also often include moderate- and low-penetrance genes. A number of
the genes included in these panels have not been fully characterized either in terms of their cancer risks or their management options.
Another way some patients are unexpectedly identified as carrying a germline mutation in a cancer susceptibility gene is at the time
they undergo molecular profiling of their tumor, which typically has been carried out to guide treatment choices for their cancer. This
article first focuses on the issues that need to be considered when deciding between recommending more targeted testing of a single
or a small number of genes associated with a particular syndrome (single/limited gene testing) versus performing a multigene panel.
This article also reviews the issues regarding germline risk that occur within the setting of ordering molecular profiling of tumors.
hroughout the past several decades, we have witnessed

tremendous advances in our knowledge of evaluating
and treating patients with germline mutations in hereditary
cancer syndromes, with studies clearly demonstrating the
feasibility and clinical utility of genetic testing. Perhaps most
importantly, studies have provided convincing evidence
that implementing prevention strategies in some instances
prolongs the survival of mutation carriers. For example, for
unaffected women who carry a BRCA1 or BRCA2 mutation,
risk-reducing salpingo-oophorectomy results in a significant
reduction in all-cause mortality (3% vs. 10%; hazard ratio
[HR] 0.40; 95% CI, 0.260.6), breast cancer-specific mortality (2% vs. 6%; HR 0.44; 95% CI, 0.260.76) and ovarian
cancerspecific mortality (0.4 vs. 3%; HR 0.21; 95% CI,
0.060.8) when compared with carriers who chose not to
undergo this procedure.1 Additionally, Markov modeling suggests that a 30-year old healthy BRCA1 mutation carrier
would gain 0.2 to 1.8 years in life expectancy with riskreducing salpingo-oophorectomy and 0.6 to 2.1 years from
risk-reducing mastectomies.2,3 Given these findings, genetic
testing for hereditary cancer syndromes has now become part
of standard practice.
As set forth in the Evaluation of Genomic Applications in
Practice and Prevention (known as EGAPP) initiative4 and further
supported by the recent American Society of Clinical Oncology

(ASCO) policy statement on testing for genetic and genomic
cancer susceptibility,5 a number of criteria must be considered
when evaluating existing or emerging genetic tests. These criteria include analytical and clinical validity, clinical utility, and
the associated ethical, legal, and social issues. In the context of
genetic testing, analytic validity refers to the accuracy and reproducibility by which the assay detects the presence or absence
of a mutation. Clinical validity focuses on whether the test accurately and reproducibly predicts the clinically defined disorder.
Clinical utility can be defined as the evidence that a genetic test
results in improved health outcomes typically based on early
detection or prevention strategies, and the tests usefulness
and added value to patient management decision making. For
genetic testing, particularly for moderate-penetrance genes,
clinical utility remains the fundamental issue.5 The EGAPP
framework is key when evaluating the utility of genetic testing
for hereditary cancer syndromes. Failure to meet some of
these criteria forms the basis for many concerns regarding the
current clinical actionability of multigene panel testing.
SINGLE/LIMITED GENE TESTING

For well over a century, it has been recognized that some
families harbor a hereditary predisposition to a variety of
From the Lombardi Comprehensive Cancer Center, MedStar Georgetown University Hospital, Washington, DC; Lombardi Comprehensive Cancer Center, Georgetown University,
Washington, DC.
Corresponding author: Claudine Isaacs, MD, Lombardi Comprehensive Cancer Center, 3800 Reservoir Rd. NW, Washington, DC 20057; email: isaacsc@georgetown.edu.
e72
SINGLE GENE VERSUS MULTIGENE PANEL TESTING FOR HEREDITARY CANCER
malignancies. In 1913, Warthin described a kindred known

as Family G, in which he noted an aggregation of endometrial
carcinoma along with gastric and colorectal cancer.6 This
family, among others, formed the basis of the initial
descriptions of hereditary nonpolyposis colorectal cancer
syndrome, now more commonly known as Lynch syndrome. Similarly, astute clinicians recognized other hereditary cancer syndromes such as Li-Fraumeni and Cowden
syndromes and hereditary breast and ovarian cancer based
on the cancer phenotype of the family.7-12 By the mid-1990s,
linkage analyses and other studies resulted in the ability to pinpoint individual genes associated with some of
these hereditary cancer syndromes.13,14 These included the
identification of BRCA1 and BRCA2 associated with hereditary breast and ovarian cancer; MLH1, MSH2, MSH6, PMS2,
and more recently EPCAM associated with Lynch syndrome;
FAP with familial adenomatous polyposis; and TP53 with
Li-Fraumeni syndrome.
However, it became apparent that many families with
striking histories consistent with either a hereditary colorectal or breast/ovarian cancer syndrome are not found to
carry a mutation in one of the mismatch repair genes associated with Lynch syndrome or in BRCA1/2. For example,
studies indicate that a mutation in a mismatch repair gene is
found in approximately 40% to 80% of families that meet
the Amsterdam I criteria and only about 5% to 50% of
families meeting the Amsterdam II criteria.15 Similarly, only
about 5% to 10% of unselected patients with breast cancer16
and 20% to 25% of patients with hereditary breast cancer17
are found to carry a deleterious BRCA1 or BRCA2 mutation.
Additionally, a number of other genes associated either
KEY POINTS
Advances in technology have resulted in the ability to

test for multiple genes associated with a hereditary
predisposition to cancer.
Multigene panel testing can allow for a more
comprehensive and efficient approach to testing, but
many of the genes included in multigene panels have
not been fully characterized either in terms of their
cancer risks or management strategies. In many cases,
single/limited gene testing remains a very appropriate
testing option.
The decision to pursue single or limited gene testing is
complex and referral to clinicians with expertise in
cancer genetics is critical. Testing must be carried out
with pre- and post-test genetic counseling.
In the tumor molecular profiling setting, secondary
incidental germline mutations may be detected. ASCO
recommends that the possibility of identifying
secondary incidental germline information and the
clinical relevance, benefits, risks, and limitations of such
findings be discussed with all patients before they
undergo tumor sequencing.
with rare high-penetrance syndromes or a more moderate

penetrance were identified.
Based on these findings, the general paradigm of testing
evolved whereby the more common genes such as BRCA1
and BRCA2 were tested first, and, if negative, sequential
testing for additional gene(s) was performed if the patient
met criteria for testing for other syndromes. This process
had both advantages and disadvantages. In terms of advantages, the genes tested in this setting typically have
well-described cancer risks and often have established
management guidelines. Additionally, through the pretest
counseling process, patients undergoing this testing have
had the opportunity to fully consider the benefits, risks, and
limitations of testing in their particular situation. In terms
of disadvantages, such testing is less comprehensive than
multigene testing, and, if performed, sequential testing is
quite time-consuming and costly.
MULTIGENE PANELS
There has been a dramatic shift in the genetic testing
landscape over the past several years in large part because of
two major factors. The first is the development of nextgeneration sequencing, a high-throughput approach to DNA
sequencing that allows for massively parallel sequencing of
multiple genes more efficiently and at a lower cost than the
traditional Sanger sequencing methods. The second is the
Supreme Court decision in 2013 for Association for Molecular Pathology v. Myriad Genetics, which invalidated many
patents restricting BRCA1/2 testing. Very shortly after the
ruling, many companies and some academic institutions
announced they would offer BRCA testing in addition to the
existing genes on their multigene panels.18,19 As a result of
these two factors, offering relatively rapid turnaround times
for multigene testing in a reasonably affordable manner
became feasible.
The panels differ from company to company. They may
be comprehensive, tumor-specific, and focus only on highly
penetrant genes, or be customizable (Table 1). The price of
testing also has dropped significantly. It now can range from
$249 to $6,040, with most costing $1,500 to $6,040. The cost
varies among laboratories and differs based on the number
of genes included. In most cases, the cost of multigene panel
testing does not significantly differ from the cost of more
single/limited gene testing. Furthermore, the cost of testing
likely will continue to diminish over time. Adding to the
complexity of testing choices is that insurance companies
have different policies and may cover some but not all
choices.
A number of studies have evaluated the utility and impact
of multigene testing in a variety of settings. The key questions that must be addressed revolve around the clinical
utility or actionability of the findings from such testing,
namely (1) the numbers of patients who are found to have a
deleterious mutation in a gene for which cancer risks are
known and management strategies exist, (2) patients who
are found to have a mutation with uncertain cancer risks
e73
LYNCE AND ISAACS
TABLE 1. Examples of Genes Included on Some Next-Generation Sequencing Cancer Panels*

Company
Comprehensive
Panels
Breast/Ovarian
Panels
Genes Included
Ambry Genetics
20
CancerNext
32
APC, ATM, BARD1, BRCA1, BRCA2, BRIP1, BMPR1A, CDH1, CDK4,

CDKN2A, CHEK2, EPCAM, GREM1, MLH1, MRE11A, MSH2, MSH6,
MUTYH, NBN, NF1, PALB2, PMS2, POLD1, POLE, PTEN, RAD50,
RAD51C, RAD51D, SMAD4, SMARCA4, STK11, TP53
GeneDx
OncoGene Dx Comprehensive Cancer Panel21
32
APC, ATM, AXIN2, BARD1, BMPR1A, BRCA1, BRCA2, BRIP1, CDH1,

CDK4, CDKN2A, CHEK2, EPCAM, FANCC, MLH1, MSH2, MSH6,
MUTYH, NBN, PALB2, PMS2, POLD1, POLE, PTEN, RAD51C, RAD51D,
SCG5/GREM1, SMAD4, STK11, TP53, VHL, XRCC2
Myriad Genetics
MyRisk22
25
BRCA1, BRCA2, MLH1, MSH2, MSH6, PMS2, EPCAM, APC, MUTYH,

CDKN2A, CDK4, TP53, PTEN, STK11, CDH1, BMPR1A, SMAD4, PALB2,
CHEK2, ATM, NBM, BARD1, BRIP1, RAD51C, RAD51D
Invitae
Invitae Multi-Cancer
Panel23
79
ALK, APC, ATM, AXIN2, BAP1, BARD1, BLM, BMPR1A, BRCA1, BRCA2,
BRIP1, CASR,CDC73, CDH1, CDK4, CDKN1B, CDKN1C, CDKN2A,
CEBPA, CHEK2, DICER1, DIS3L2, EGFR, EPCAM, FH, FLCN, GATA2,
GPC3, GREM1, HOXB13, HRAS, KIT, MAX, MEN1, MET, MITF, MLH1,
MSH2, MSH6, MUTYH, NBN, NF1, NF2, PALB2,PDGFRA, PHOX2B,
PMS2, POLD1, POLE, PRKAR1A, PTCH1, PTEN, RAD50,RAD51C,
RAD51D, RB1, RECQL4, RET, RUNX1, SDHA, SDHAF2, SDHB, SDHC,
SDHD, SMAD4, SMARCA4, SMARCB1, SMARCE1, STK11, SUFU, TERC,
TERT, TMEM127, TP53, TSC1, TSC2, VHL, WRN, WT1
Ambry Genetics
BRCAplus24
BRCA1, BRCA2, CDH1, PALB2, PTEN, TP53
BreastNext25
17
ATM, BARD1, BRCA1, BRCA2, BRIP1, CDH1, CHEK2, MRE11A, MUTYH,

NBN, NF1, PALB2, PTEN, RAD50, RAD51C, RAD51D, TP53
OvaNext26
23
ATM, BARD1, BRCA1, BRCA2, BRIP1, CDH1, CHEK2, EPCAM, MLH1,

MRE11A, MSH2, MSH6, MUTYH, NBN, NF1, PALB2, PMS2, PTEN,
RAD50, RAD51C, RAD51D, SMARCA4, STK11, TP53
Breast and Gynecologic

Cancers Guidelines
Based Panel27
14
ATM, BRCA1, BRCA2, CDH1, CHEK2, EPCAM, MLH1, MSH2, MSH6,

PALB2, PMS2,PTEN, STK11, TP53
Breast Cancer Guidelines

Based Panel28
ATM, BRCA1, BRCA2, CDH1, CHEK2, PALB2, PTEN, STK11, TP53
Color
Genomics29
Color
19
ATM, BARD1, BRCA1, BRCA2, BRIP1, CDH1, CHEK2, EPCAM, MHL1,

MSH2, MSH6, NBM, PALB2, PMS2, PTEN, RAD51C, RAD51D, STK11,
TP53
GeneDx
Breast Cancer High/Moderate Risk Panel21
ATM, BRCA1, BRCA2, CDH1, CHEK2, PALB2, PTEN, STK11, TP53
Breast/Ovarian Cancer
Panel21
21
ATM, BARD1, BRCA1, BRCA2, BRIP1, CDH1, CHEK2, EPCAM, FANCC,

MLH1, MSH2, MSH6, NBN, PALB2, PMS2, PTEN, RAD51C, RAD51D,
STK11, TP53, XRCC2
Ambry Genetics
ColoNext30
17
APC, BMPR1A, CDH1, CHEK2, EPCAM, GREM1, MLH1, MSH2, MSH6,

MUTYH, PMS2, POLD1, POLE, PTEN, SMAD4, STK11, TP53
Invitae
Colorectal Cancer Guidelines Based Panel31
12
APC, BMPR1A, EPCAM, MLH1, MSH2, MSH6, MUTYH, PMS2, PTEN,

SMAD4, STK11, TP53
Myriad Genetics
COLARIS32
Invitae
Gastrointestinal
Panels
Number
of Genes
Test
COLARIS AP33
GeneDx
MLH1, MSH2, EPCAM, MSH6, PMS2, MUTYH

APC, MUTYH
Colorectal Cancer Panel21
19
APC, ATM, AXIN2, BMPR1A, CDH1, CHEK2, EPCAM, MLH1, MSH2,

MSH6, MUTYH, PMS2, POLD1, POLE, PTEN, SCG5/GREM1, SMAD4,
STK11, TP53
Lynch/Colorectal High Risk

Panel21
APC, EPCAM, MLH1, MSH2, MSH6, MUTYH, PMS2
Abbreviation: VUS, variants of uncertain significance.

*Current as of February 2, 2016.
and/or no evidence-based recommendations for management, and (3) the rate detection of variants of uncertain
significance (VUS). As summarized in Tables 2 and 3, the rate
of VUS varies between 3.3% and 42%, and many patients
were reported to have two or more VUS. The VUS rate is still
high in some reports, but it is expected to fall in the near
e74
future because of the rapid accumulation of data from

multigene panel testing.
Given the high rate of VUS and the detection of genes for
which the cancer risks are not well-defined, several registries have been created to catalog and curate these variants with the goal of advancing our knowledge about their
TABLE 2. Rate of Deleterious Mutations and VUS Reported in Multigene Testing Associated with Hereditary Breast
Cancer
Publication
No. of Participants
38
BRCA Mutations
Detected (%)
Panel Tested
Non-BRCA Mutations
Detected (%)
VUS Rate (%)
Kapoor et al
337
Ambry Genetics; different 3.6%

panel types and genes
tested
3.9%; gene mutations more often 3.3% BRCA1/2 and

detected: PALB2, CHEK2, ATM
13.4% non-BRCA1/2
Slavin et al39
348
Not provided
3.4%
16.4%
Cohort 1: Referred for

commercial BRCA1/2
gene testing and
9.3%
4.2%
41.7%
3.7%
41.6%
1%
7.6%
40
Tung et al
2,158
Cohort 2: Previously
tested negative for
BRCA1/2 mutations
Chong et al41 3,000
BRCAPlus: 6-gene panel

(BRCA1, BRCA2, TP53,
CDH1, PTEN, SKT11)
4.6%
Invitae
Kurian et al42 198 (57 BRCA1/2,
141 BRCA1/2 neg)
LaDuca et al43 2,079 (874 had
breast panel)
11.4%
Ambry Genetics
42%
2.1% VUS/average per

participant
All previously underwent 7.4% gene mutations detected: 19.8%

BRCA sequencing and
CHEK2 (19), ATM (18), PALB2
BART testing and were
(15), TP53 (4), PTEN (3), RAD50
negative
(3), RAD51C (2), BRIP1 (1),
MRE11A (1), NBN (1)
clinical utility. The Prospective Registry of Multiplex Testing

(PROMPT)34 is a multi-institutional online registry that encourages patients to self-enter information about their
genetic testing results and to complete questionnaires about
their personal medical and family histories. Others involved
in reclassifying variants include ENIGMA (Evidence-based
Network for the Interpretation of Germline Mutant Alleles)
Consortium35 and ClinVar, a peer-reviewed database funded
by the National Institutes of Health, which is a freely available
archive of reports of relationships among medically important
variants and phenotypes.36 Professional societies such as the
American Medical Association also have adopted positions in
favor of data sharing.37
SINGLE/LIMITED GENE VERSUS MULTIGENE

PANEL TESTING
Several factors guide the decision to pursue testing of a single
gene or a finite set of genes associated with a particular
syndrome versus multigene panel testing. These include

(1) the characteristics of the probands personal and family
history, (2) an individuals preferences and tolerance regarding the possibility of ambiguous results, (3) insurancerelated issues, and (4) the rapidity with which results are
needed. A publication in the New England Journal of Medicine
in 2015, authored by experts from the United States, United
Kingdom, the Netherlands, Germany, Australia, and Canada,
thoughtfully reviewed the issues that must be addressed
when considering multigene panels.45 Additionally, ASCO
released a policy statement on genetic and genomic testing
for cancer susceptibility to reflect the impact of advances in
this field.5
Single/limited gene testing remains an excellent option
when the clinical features, such as the patients personal and
family history, are strongly indicative of a particular syndrome associated with a single or finite set of genes. This
approach allows for a focused and comprehensive pretest
evaluation in which individuals have an opportunity to more
TABLE 3. Rate of Deleterious Mutations and VUS Reported in Multigene Testing Associated with Colorectal Cancer
Publication
LaDuca et al43
Yurgelun et al44
No. Patients
557
1,260
Panel Tested
LS Genes (MLH1, MSH2, MSH6,

PMS2, EPCAM) Mutations Detected (%)
Non-LS Genes Mutations

Detected (%)
VUS Rate (%)
Ambry Genetics
4.5%: MSH (7), MLH1 (7),

PMS2 (6), MSH (5)
4.7% (26): APC (6), CHEK2 (6), MUTYH

biallelic (6), SMAD4 (4), PTEN (3),
CDH1 (1), STK11 (1), TP53 (1)
15.1% (84)
Myriad MyRisk Hereditary

Cancer
9.5%
3.8% (48): BRCA1/2 (15), APC biallelic

MUTYH, PTEN, STK11, ATM,
BARD1, BRIP1, CHEK2, NBN, PALB2,
and RAD51C
44%
Abbreviations: LS, Lynch syndrome; VUS, variants of uncertain significance.
e75
LYNCE AND ISAACS
fully consider the impact of testing for a particular gene or

set of genes. Additionally, such testing minimizes the likelihood of detecting a VUS or a deleterious mutation in a gene
with limited clinical information.
Multigene panel testing is an appropriate option when
the family phenotype is not suggestive of a single specific
mutation and one or more hereditary cancer syndromes
are in the differential. Additionally, panel testing is often
considered if more focused initial testing is negative (e.g.,
BRCA1/2 testing followed by multigene breast/ovarian
panel). Multigene panel testing has a number of advantages and potential disadvantages (Table 4). The advantages include gains in efficiency both in terms of cost and
time. Such testing also would result in a more comprehensive assessment of the genes that could account for
the cancer phenotype in the family. Finally, pragmatically,
in this era of multigene panel testing, it is unclear if an
individual could obtain insurance coverage for repeat
testing if the initial, more limited testing results were
negative. In terms of disadvantages, as described previously, multigene panel testing has a higher rate of detection of VUS. Individuals undergoing testing must be
fully informed of this possibility before testing and counseled on the interpretation of such a result. Furthermore,
it is important that an individual undergoing panel testing understands it is possible that a high-penetrance mutation
in an uncommon or rare gene may be identified, even in the
absence of a classic presentation of the associated syndrome.
Consequently, aggressive interventions may be recommended,
such as consideration of prophylactic gastrectomy if a CDH1
mutation was found, even in the absence of gastric cancer in
the family. At this moment, it is unclear if the cancer risks for
patients identified through panel testing without features of
the associated syndrome are the same as quoted in the literature because of ascertainment bias. Moreover, laboratories
have varying methods by which they assure the analytic and
clinical validity and the clinical utility of the variants they report. Expertise in this area is required to ensure accurate interpretation of the clinical significance of the findings reported.
Given the panoply of testing options, this expertise is also
critical to guide the choice of which test to order and from

which laboratory. These issues further underscore the importance of ensuring that patients undergo pre- and post-test
genetic counseling by well-trained professionals, as endorsed
by ASCO and the National Comprehensive Cancer Network.46
GERMLINE FINDINGS ON MOLECULAR

PROFILING OF TUMORS
An important challenge when performing vast-scale sequencing is the potential for detecting incidental findings.
Incidental findings are defined as unexpected positive
findings. In this context, they refer to the detection of
deleterious or likely deleterious alterations in genes that
have clinical significance and are unrelated to the indication for obtaining the sequencing test. Typically, these
are germline mutations. As such, the only way to truly
determine if an identified sequence variant is somatic or
inherited is to simultaneously analyze tumor and normal
DNA. This analysis allows for a determination on which
variants are unique to the cancer (i.e., somatic) and which
are germline. Determining whether, which, and how incidental findings are returned to the patient is becoming
increasingly important and controversial. The American
College of Medical Genetics and Genomics published a
policy statement on clinical sequencing that included
exome and genome sequencing.47 The policy statement recommended that constitutional mutations from a
panel of 56 disease-associated genes be reported to the
ordering clinician, regardless of the indication for which
the clinical sequencing was ordered (Table 5). These
genes were chosen because they result in disorders for
which preventive strategies and/or treatments are
available. About half are associated with syndromes that
increase the risk of cancer; the others are primarily associated with various cardiac diseases. The American
College of Medical Genetics and Genomics also states
that the ordering clinician/team is responsible for providing the patient with comprehensive pre- and post-test
counseling.
TABLE 4. Pros and Cons of Single/Limited Gene Testing and Multigene Panels
Single/Limited Gene Testing
Multigene Panels
Advantages
Phenotype-directed testing
Cancer risks and management
options often more established
Lower likelihood of detecting VUS
More rapid turnaround time
More cost effective (less expensive per gene cost)

More time efficient
Decrease in testing fatigue for patients and providers
Efficient use of single specimen
Higher mutation detection rate, genes individually rare but
collectively significant
Disadvantages
Higher risk of loss to follow-up during sequential testing multiple

single genes (test fatigue)
Less comprehensive
Increased prevalence of VUS

Cancer risks and management options often not well-defined,
particularly for some moderate- and low-penetrance genes
Unexpected findings such as off-phenotypic-target gene
mutation
Longer turnaround time
Panels may include genes that patients dont wish to test for
e76
TABLE 5. Conditions and Genes Recommended by the American College of Medical Genetics and Genomics for
Return of Incidental Findings in Clinical Sequencing47
Phenotype
Gene
Hereditary breast and ovarian cancer
BRCA1, BRCA2
Li-Fraumeni syndrome
TP53
Peutz-Jeghers syndrome
STK11
Lynch syndrome
MLH1, MSH2, MSH6, PMS2
Familial adenomatous polyposis
APC
MYH-associated polyposis; adenomas; multiple colorectal cancers; familial amyloid

polyneuropathy type 2; colorectal adenomatous polyposis, autosomal recessive,
with pilomatricomas
MUTYH
Von Hippel-Lindau syndrome
VHL
Multiple endocrine neoplasia type 1
MEN1
Multiple endocrine neoplasia type 2
RET
Familial medullary thyroid cancer
RET
PTEN hamartoma tumor syndrome
PTEN
Retinoblastoma
RB1
Hereditary paraganglioma-pheochromocytoma syndrome
SDHD, SDHAF2, SDHC, SDHB
Tuberous sclerosis complex
TSC1, TSC2
WT1-related Wilm syndrome
WT1
Neurofibromatosis type 2
NF2
Ehlers-Danlos syndrome (vascular type)
COL3A1
Marfan syndrome, Loeys-Dietz syndrome, familial thoracic aortic aneurysms and

dissections
FBN1, TGFBR1, TGFBR2, SMAD3, ACTA2, MYLK, MYH11
Hypertrophic cardiomyopathy, dilated cardiomyopathy
MYBPC3, MYH7, TNNT2, TNNI3, TPM1, MYL3, ACTC1, PRKAG2,

GLA, MYL2, LMNA
Catecholaminergic polymorphic ventricular tachycardia
RYR2
Arrhythmogenic right-ventricular cardiomyopathy
PKP2, DSP, DSC2, TMEM43, DSG2
Romano-Ward Long QT syndrome types 1, 2, and 3; Brugada syndrome
KCNQ1, KCNH2, SCN5A
Familial hypercholesterolemia
LDLR, APOB, PCSK9
Malignant hyperthermia susceptibility
RYR1, CACNA1S
Several challenges are inherent in this process. They range

from analytic issues (e.g., determining if the variant is
germline and, if so, if it is deleterious/likely deleterious,
a VUS, or a benign polymorphism) to practical issues related
to obtaining informed consent and delivering traditional
pre- and post-testing genetic counseling. ASCO recently
published a policy statement on genetic and genomic testing
for cancer susceptibility that includes germline implications of
somatic mutation profiling.5 The policy statement recognizes
that standard pre- and post-test counseling may not be
feasible in this setting for all patients. It recommends that the
possibility of identifying secondary incidental germline information as well as the clinical relevance, benefits, risks, and
limitations of such incidental findings be discussed with all
patients before they undergo tumor sequencing. ASCO also
endorses that providers should honor patients decisions if
they elect not to receive information about such incidental
findings.
CONCLUSION
Next-generation sequencing has introduced substantial
complexity and promise in the field of cancer risk assessment.
Although multigene panel testing provides a more comprehensive and efficient approach to testing an individual
for a hereditary susceptibility to cancer, the information obtained can be challenging to interpret. Furthermore, many of the genes included in multigene panels
have not been fully characterized either in terms of
their cancer risks or management strategies. In many
cases, single/limited gene testing remains a very appropriate testing option. Presently, we live in an era in
which our technical capabilities have outstripped our
medical knowledge. A strong and continuous partnership among clinicians, individuals with genetics
expertise, and laboratory geneticists is critical to bridge
this gap.
As to the detection of incidental findings on tumor sequencing, more research is clearly necessary to better clarify
how to approach this complex area. Until such time, as
stated by ASCO, it is critical that individuals undergoing
tumor sequencing be fully apprised of the possibility,
benefits, risks, and limitations that such testing could
uncover unanticipated mutations in cancer susceptibility
genes.
e77
LYNCE AND ISAACS
References
1. Domchek SM, Friebel TM, Singer CF, et al. Association of risk-reducing
mortality. JAMA. 2010;304:967-975.
2. Schrag D, Kuntz KM, Garber JE, et al. Benefit of prophylactic mastectomy for
women with BRCA1 or BRCA2 mutations. JAMA. 2000;283:3070-3072.
3. Grann VR, Jacobson JS, Thomason D, et al. Effect of prevention strategies on
survival and quality-adjusted survival of women with BRCA1/2 mutations:
an updated decision analysis. J Clin Oncol. 2002;20:2520-2529.
4. Teutsch SM, Bradley LA, Palomaki GE, et al; EGAPP Working Group. The
Evaluation of Genomic Applications in Practice and Prevention (EGAPP)
initiative: methods of the EGAPP working group. Genet Med. 2009;11:3-14.
5. Robson ME, Bradbury AR, Arun B, et al. American Society of Clinical
Oncology policy statement update: genetic and genomic testing for
cancer susceptibility. J Clin Oncol. 2015;33:3660-3667.
6. Warthin AS. Heredity with reference to carcinoma. Arch Intern Med
(Chic). 1913;12:546-555.
7. Li FP, Fraumeni JF Jr. Soft-tissue sarcomas, breast cancer, and other
neoplasms. A familial syndrome? Ann Intern Med. 1969;71:747-752.
8. Li FP, Fraumeni JF Jr. Rhabdomyosarcoma in children: epidemiologic
study and identification of a familial cancer syndrome. J Natl Cancer
Inst. 1969;43:1365-1373.
9. Lloyd KM II, Dennis M. Cowdens disease. A possible new symptom
complex with multiple system involvement. Ann Intern Med. 1963;58:
136-142.
10. Le Dran H. Memoire avec un precis de plusieurs observations sur le
cancer. Mem Acad Chir (Paris). 1757;3:1-54.
11. Smithers DW. Family histories of 459 patients with cancer of the breast.
Br J Cancer. 1948;2:163-167.
12. Broca PP. Traite des tumeurs. Paris: P. Asselin; 1866.
13. Hall JM, Lee MK, Newman B, et al. Linkage of early-onset familial breast
cancer to chromosome 17q21. Science. 1990;250:1684-1689.
14. Nakamura Y, Lathrop M, Leppert M, et al. Localization of the genetic
defect in familial adenomatous polyposis within a small region of chromosome 5. Am J Hum Genet. 1988;43:638-644.
15. Lynch HT, de la Chapelle A. Hereditary colorectal cancer. N Engl J Med.
2003;348:919-932.
16. Antoniou A, Pharoah PDP, Narod S, et al. Average risks of breast and
ovarian cancer associated with BRCA1 or BRCA2 mutations detected in
case series unselected for family history: a combined analysis of 22
17. Nathanson KL, Wooster R, Weber BL. Breast cancer genetics: what we
know and what we need. Nat Med. 2001;7:552-556.
18. Azvolinsky A. Supreme Court ruling broadens BRCA testing options.
19. Cook-Deegan R, Niehaus A. After Myriad: genetic testing in the wake of
recent Supreme Court decisions about gene panels. Curr Genet Med
Rep. 2014;2:223-241.
20. Ambry Genetics. CancerNext. http://www.ambrygen.com/tests/cancernext.
Accessed February 2, 2016.
21. Ambry Genetics. GeneDx Oncology Genetics. http://www.genedx.com/
oncology-genetics/. Accessed February 2, 2016.
22. Myriad Pro. MyRisk FAQ. https://www.myriadpro.com/myrisk/frequentlyasked-questions/. Accessed February 2, 2016.
23. Invitae Multi-Cancer Panel. https://www.invitae.com/en/physician/tests/
01101/. Accessed February 2, 2016.
24. Ambry Genetics. BRCAPlus. http://www.ambrygen.com/tests/brcaplus.
25. Ambry Genetics. BRCANext. http://www.ambrygen.com/tests/breastnext.
e78
26. Ambry Genetics. OvaNext. http://www.ambrygen.com/tests/ovanext.

27. Invitae. Breast and Gyn Cancers Guidelines-Based Panel. https://www.
invitae.com/en/physician/tests/01204/. Accessed February 2 ,2016.
28. Invitae Breast Cancers Guidelines-Based Panel. https://www.invitae.
com/en/physician/tests/01206/. Accessed February 2, 2016.
29. Color Genomics. Get Color. https://getcolor.com/learn/the-science#gene-panel.
30. Ambry Genetics. ColoNext. http://www.ambrygen.com/tests/colonext.
31. Invitae. Colorectal Cancer Guidelines Based Panel. https://www.invitae.
com/en/physician/tests/01252/. Accessed February 2, 2016.
32. Myriad Pro. COLARIS FAQ. https://www.myriadpro.com/hereditary-cancertesting/hereditary-colon-and-gynecologic-cancers/lynch-syndrome-hnpcc/
colaris-faq/. Accessed February 2, 2016.
33. Myriad Pro. COLARIS AP FAQ. https://www.myriadpro.com/hereditarycancer-testing/hereditary-colon-and-gynecologic-cancers/polyposissyndromes-fap-afap-map/colaris-ap-faq/. Accessed February 2, 2016.
34. PROMPT. Prospective Registry of Multiplex Testing. http://www.
promptstudy.org. Accessed February 22, 2016.
35. ENIGMA. Evidence-based Network for the Interpretation of Germline
Mutant Alleles. http://www.enigmaconsortium.org. Accessed February
22, 2016.
36. Landrum MJ, Lee JM, Riley GR, et al. ClinVar: public archive of relationships among sequence variation and human phenotype. Nucleic
Acids Res. 2014;42:D980-D985.
37. American Medical Association. Support of Public Access to Genetic
Data. http://www.ama-assn.org/ama/pub/news/news/2013/2013-0618-new-ama-policies-annual-meeting.page. Accessed February 2, 2016.
38. Kapoor NS, Curcio LD, Blakemore CA, et al. Multigene panel testing detects
equal rates of pathogenic BRCA1/2 mutations and has a higher diagnostic
yield compared to limited BRCA1/2 analysis alone in patients at risk for
hereditary breast cancer. Ann Surg Oncol. 2015;22:3282-3288.
39. Slavin TP, Niell-Swiller M, Solomon I, et al. Clinical application of multigene
panels: challenges of next-generation counseling and cancer risk management. Front Oncol. 2015;5:208.
40. Tung N, Battelli C, Allen B, et al. Frequency of mutations in individuals
with breast cancer referred for BRCA1 and BRCA2 testing using nextgeneration sequencing with a 25-gene panel. Cancer. 2015;121:25-33.
41. Chong HK, Wang T, Lu HM, et al. The validation and clinical implementation of BRCAplus: a comprehensive high-risk breast cancer diagnostic assay. PLoS One. 2014;9:e97408.
42. Kurian AW, Hare EE, Mills MA, et al. Clinical evaluation of a multiplegene sequencing panel for hereditary cancer risk assessment. J Clin
Oncol. 2014;32:2001-2009.
43. LaDuca H, Stuenkel AJ, Dolinsky JS, et al. Utilization of multigene panels
in hereditary cancer predisposition testing: analysis of more than 2,000
patients. Genet Med. 2014;16:830-837.
44. Yurgelun MB, Allen B, Kaldate R, et al. Multigene panel testing in
patients suspected to have Lynch syndrome. J Clin Oncol. 2014;32:5s
(suppl; abstr 1509).
45. Easton DF, Pharoah PDP, Antoniou AC, et al. Gene-panel sequencing and
the prediction of breast-cancer risk. N Engl J Med. 2015;372:2243-2257.
46. Hall MJ, Forman AD, Pilarski R, et al. Gene panel testing for inherited
cancer risk. J Natl Compr Canc Netw. 2014;12:1339-1346.
47. Green RC, Berg JS, Grody WW, et al; American College of Medical
Genetics and Genomics. ACMG recommendations for reporting of incidental findings in clinical exome and genome sequencing. Genet Med.
2013;15:565-574.

AND EPIDEMIOLOGY
Evolving Recommendations on
Prostate Cancer Screening
CHAIR
Otis W. Brawley, MD, FASCO
Atlanta, GA
SPEAKERS
Henrik Gronberg, MD, PhD
Karolinska Institutet
Stockholm, Sweden
Ian M. Thompson Jr., MD
The University of Texas Health Science Center at San Antonio
San Antonio, TX

BRAWLEY, THOMPSON, AND GRONBERG

Otis W. Brawley, MD, FASCO, Ian M. Thompson Jr., MD, and Henrik Gronberg, MD, PhD
OVERVIEW
Results of a number of studies demonstrate that the serum prostate-specific antigen (PSA) in and of itself is an inadequate
screening test. Today, one of the most pressing questions in prostate cancer medicine is how can screening be honed to
identify those who have life-threatening disease and need aggressive treatment. A number of efforts are underway. One
such effort is the assessment of men in the landmark Prostate Cancer Prevention Trial that has led to a prostate cancer
risk calculator (PCPTRC), which is available online. PCPTRC version 2.0 predicts the probability of the diagnosis of no
cancer, low-grade cancer, or high-grade cancer when variables such as PSA, age, race, family history, and physical findings
are input. Modern biomarker development promises to provide tests with fewer false positives and improved ability to
find high-grade cancers. Stockholm III (STHLM3) is a prospective, population-based, paired, screen-positive, prostate
cancer diagnostic study assessing a combination of plasma protein biomarkers along with age, family history, previous
biopsy, and prostate examination for prediction of prostate cancer. Multiparametric MRI incorporates anatomic and
functional imaging to better characterize and predict future behavior of tumors within the prostate. After diagnosis of
cancer, several genomic tests promise to better distinguish the cancers that need treatment versus those that need
observation. Although the new technologies are promising, there is an urgent need for evaluation of these new tests in
high-quality, large population-based studies. Until these technologies are proven, most professional organizations have
evolved to a recommendation of informed or shared decision making in which there is a discussion between the doctor
and patient.
he U.S. Food and Drug Administration (FDA) first approved sale and use of a serum PSA test in 1986. It was
approved for assessing response to prostate cancer therapy
and monitoring for relapse. Large-scale prostate cancer
screening began in the United States in 1991 after publications showing serum PSA useful in finding localized prostate
cancer. Use increased dramatically after the American Cancer
Society recommended it in 1992.1 The FDA approved it as a
diagnostic test in 1994.
In the United States, use of the PSA test was rampant
throughout the 1990s, despite the lack of conclusive data to
show that screening and aggressive treatment of localized
prostate cancer saved lives. Screening spread outside of the
physician-patient relationship. There was mass screening at
shopping malls, community centers, and even on the floor of
national political conventions. The effect was that prostate
cancer incidence rates soared, and a large number of men
were treated for localized prostate cancer with radical prostatectomy, radiation therapy, or cryotherapy.2 Screening has
not been as popular in Europe, and prostate cancer incidence
rates have not risen as much.3
It is of note that the first definitive studies to show that
localized therapy reduces mortality were published in 1997.
They were studies of hormones and radiation therapy for

locally advanced disease.4,5 A prospective randomized study
demonstrating a positive effect for radical prostatectomy
was first published in 2002.6,7
Two large relatively well-designed randomized trials
assessing prostate cancer screening were published in
2009.8,9 Although relatively well designed, both have some
flaws. One suggests screening is ineffective, and one suggests a regimented screening program does reduce risk of
prostate cancer death. Both studies documented the harms
of prostate cancer screening, especially overdiagnosis and
resultant overtreatment.
The fact that there is considerable overdiagnosis associated with prostate cancer screening was confirmed in assessment of the placebo arm of the large multi-institutional
Prostate Cancer Prevention Trial (PCPT).10 This trial of men at
average risk of prostate cancer demonstrated that annual
screening could diagnose prostate cancer in upward of 14%
of men in their fifties, sixties, and seventies. Men underwent
biopsy for diagnosis when their PSA was 4 ng/mL or greater
or when their PSA changed by 1 ng/mL in 1 years time.
Planned biopsies for men who screened normal every year
for 7 years (median age 69) found 15% of these men had
From the American Cancer Society, Emory University, Atlanta, GA; The University of Texas Health Science Center at San Antonio, San Antonio, TX; Karolinska Institute, Stockholm, Sweden.
Corresponding author: Otis W. Brawley, MD, American Cancer Society, 250 Williams St., Ste. 600, Atlanta, GA 30303-1002; email: otis.brawley@cancer.org.
e80
EVOLVING RECOMMENDATIONS ON PROSTATE CANCER SCREENING
prostate cancer. This demonstrates that PSA screening missed

as much cancer as it found.
Radical prostatectomy was compared with conservative
therapy in the Prostate Intervention versus Observation Trial
(PIVOT).11 Mortality reductions were only found among
patients treated for intermediate- and high-risk disease. This
again demonstrates that a considerable portion of men with
screened diagnosed prostate cancer have disease that does
not need treatment.
By 2010, the data became very clear that PSA testing identifies a lot of cancer, and a lot of that cancer is of no risk to
the patients lives. Today some even go so far as to ask
whether low-grade disease (Gleason 6) should be considered cancer.12 Gleason score 6 and lower disease constitutes
about 50% of screen-detected prostate cancer in the United
States.
It took a number of years for the screening activity to
subside. It is estimated that more than a million American
men were diagnosed and received treatment because of PSA
screening.13 Although a significant proportion of screendiagnosed prostate cancers are of no clinical significance,
prostate cancer is a significant cause of death.
The U.S. prostate cancer mortality rate has declined by
53% since 1991.14 Although some of this decline may be due
to screening and aggressive treatment, some is certainly not
due to screening and aggressive treatment. The U.S. death
rate began declining the same year that widespread PSA
screening was introduced. One would expect screening to
reduce mortality 8 or more years after introduction. There
have been declines in prostate cancer mortality in more than
two dozen countries. PSA screening is uncommon in most of
these countries.3 There is also speculation that some of the
decline is an artifact.15 The decline in mortality also corresponded with changes in coding causes of death and the use
of PSA for assessing progression of prostate cancer. In reality, the U.S. decline may be due to all of the above,
including a screening benefit.15,16
The U.S. Preventive Services Task Force (USPSTF) is a
health advisory group for the U.S. government. In 2008,
KEY POINTS
Prostate cancer screening has clear documented harms

and some possible benefits.
Men with low-grade or Gleason 6 disease generally have
long-term survival with observation.
The benefit-risk ratio of prostate cancer screening
improves if screening finds high-grade disease.
The Prostate Cancer Prevention Trial risk calculator is
useful in predicting whether a man of a specific PSA has
no cancer, low-grade cancer, or high-grade cancer and
reduces the number of unnecessary prostate biopsies.
The Stockholm III model is a promising predictor of lowgrade versus high-grade disease that may further
decrease the number of unnecessary prostate biopsies.
USPSTF recommended against PSA-based prostate cancer

screening for men age 75 and older.17 USPSTF recommended against screening for all men in 2012.18 During this
period, other organizations moved toward informed decision making in which the decision to screen or not to screen
is made within the physician-patient relationship, with the
patient taking into account the potential harms and potential benefits (informed decision making) or the physician
and patient taking into account potential harms and potential benefits (shared decision making).19
According to the 2008 National Health Interview Survey,
40.6% of American men age 50 and older reported having
PSA screening within the previous year. By 2013, it was
20.8%. From 2010 to 2013 the prostate cancer incidence
declined by 18% in the United States. Longer follow-up is
needed to see whether these decreases are associated with
trends in mortality.
NEXT-GENERATION PROSTATE CANCER

SCREENING AND TREATMENT
The historical approach to prostate cancer screening was
based in part on the historically poor outcomes of treatment of metastatic prostate cancer and the realization that
PSA-detected tumors were often cured with radiation or
surgery. As noted in Fig. 1, all patients over a certain age
(often age 50) were recommended to undergo annual
screening. If a PSA was found to be greater than 4.0 ng/mL,
a biopsy was recommended. If biopsy found cancer,
treatment with radiation or surgery was generally recommended. This approach led to a remarkable spike
in prostate cancer detection, treatment, and the conduct of the PLCO and ERSPC studies.8,20 Recognizing the
questionable impact of such an approach when comparing the potential benefit and the known risks of
treatment, the USPSTF recommended against PSA testing in 2012.20
Twenty years of experience with this initial approach to
screening began to change with several sets of observations.
First, it was recognized that many tumors detected via PSA
screening were small and low-grade; the possible extent of
overdetection was illustrated in the end-of-study biopsies of
the PCPT in which all men with a PSA , 4.0 ng/mL were
recommended to undergo biopsy. In this group of men, 15%
were found to have cancer, a stunning risk of cancer when it
is recognized that the lifetime risk of cancer death in all men
is between 2% and 3%.21 Concurrent with this observation,
the first reports of the outcomes of active surveillance for
low-risk prostate cancer detected through PSA testing began
to appear.9 With the maturation of these series, it became
clear that low-grade (Gleason 3 + 3) and low-volume
intermediate-grade (Gleason 3 + 4) prostate cancers will
achieve high prostate cancerspecific survival at periods up
to 15 years.22 In a cohort of 993 men with such tumors, of
whom 13% had Gleason 3 + 4 tumors, combined causespecific survival at 10 years was 98.1% and at 15 years was
94.3%.23
e81

FIGURE 1. Early Approach to Prostate Cancer

Screening and Treatment
low-grade cancer, the benefit-risk ratio would more likely

argue against a biopsy. The first tool developed for this
was the PCPTRC (www.prostate-cancer-risk-calculator.
com).8
Two clinical scenarios illustrate how such a tool can be
used in next-generation screening.24
Case 1: a healthy 55-year-old white man who has a
normal examination. His current PSA is 3.0 (previous
year was 1.5). He has no family history of prostate
cancer and no prior biopsy.
Case 2: a 72-year-old black man who runs 5K races each
weekend, winning his age group times (i.e., in good
health). His father had prostate cancer, and on examination, he has a prostate nodule. He has heard
that PSA doesnt matter, but, because of his family
history, asked his physician to have one drawn. It is
3.0 ng/mL.
The risk calculator displays outcomes as green smiley faces
(no cancer), yellow uncertain faces (low-grade cancer), and
red frowny faces (high-grade cancer). As can be seen in these
two examples, whereas the patient in case 1 (Fig. 2A) has a
3% risk of a high-grade cancer (potential benefit), he also
has a 15% risk of a low-grade cancer (unlikely benefit, likely
harm) and 82% likelihood of a negative biopsy (unnecessary
with risk of pain, bleeding, and infection). In contrast, the
patient in case 2 (Fig. 2B), with precisely the same PSA
value, has an eightfold greater risk of a high-grade cancer
(24%). In clinical practice, when provided these data with the
accompanying information regarding what happens to you if
you have any of these three outcomes, most patients similar to that in case 1 opt to not undergo biopsy, whereas
most patients similar to that in case 2 opt for a biopsy. It is
also important to note that, again in clinical practice, two
patients with precisely the same risks will often reach
two different decisions, on the basis of their own personal
preferences and values that they place on the different
Abbreviation: PSA, prostate-specific antigen.
By studying the biopsy outcomes of the European Randomized Study of Prostate Cancer Screening and the PCPT,
it also has become apparent that using a single biomarker
(i.e., PSA) to determine a mans risk of prostate cancer was
naive; other measures of risk had a profound impact on
prostate cancer risk.8 Also phase III randomized clinical
trials have demonstrated very little (if any) benefit of
treatment of low-grade tumors but mortality benefit for
high-grade tumors, the clinical community recognized
that screening should seek to identify the man with highgrade cancer in whom a biopsy may have net potential
benefit from detection (and, presumably, treatment). On
the other hand, if biopsy would preferentially detect a
FIGURE 2. Prostate Cancer Prevention Trial Risk Calculator Patient Displays for (A) Case 1 and (B) Case 2
e82
outcomes of each possible finding. As can be seen at the

PCPTRC website, other biomarkers (e.g., percent free PSA) or
biomeasures (e.g., body mass index) can be incorporated to
better estimate risk.
It has also been recognized that annual PSA testing is
unnecessary for many men. In most population-based studies,
the median PSA is about 1.0 ng/mL. We have found, in a large
community-dwelling cohort, if a mans PSA is 1.0 ng/mL or
lower, his risk of a prostate cancer diagnosis within 10 years
is 3.4% and 90% of the cancers are low-grade. Thus, conceivably, half of the general population could have their PSA
performed concurrently with their colonoscopy, eliminating
an enormous burden on society.
Finally, information is now available to better help patients
decide on treatment options. As up to two-thirds of patients
will qualify for active surveillance, it is important that men
with low-risk cancers are aware of this option.18 Multiple
series now attest to the safety of this approach and the
low risk of prostate cancer mortality. Studies are ongoing
to examine how to reduce the current burden of surveillance. By comparison, phase III clinical trials have been
completed that demonstrate the importance of multimodal therapy (e.g., androgen deprivation plus radiation
or adjuvant radiotherapy after prostatectomy) for highrisk tumors to reduce cancer recurrence and to reduce
mortality.18,25 To help patients with decision making, considerable data are also available regarding the impact of
these treatments on genitourinary, sexual, and gastrointestinal functions.
It is unequivocal that using the current body of knowledge regarding whom to screen, how frequently to screen
them, what their risk is for different biopsy outcomes, incorporating active surveillance, and, as appropriate, multimodal therapies for high-risk cancers, that the harms of
screening and treatment can be dramatically reduced. It
remains to be determined whether such an intelligent
approach, summarized only briefly in Fig. 3, to screening
and treatment reduces prostate cancer mortality and
morbidity.
A NEW VIEW ON PROSTATE CANCER SCREENING

Prostate cancer mortality and morbidity is substantial in
many parts of the world, and it is the most common cancer
among males in most parts of Europe, North and South
America, Australia, and sub-Saharan Africa.26 The overarching goal of prostate cancer screening is to decrease the
morbidity and suffering from advanced prostate cancer
without causing harm to men who do not require treatment
for low-risk prostate cancer.
The first question to ask is, is there evidence that PSA
screening decreases prostate cancer mortality? The highest
scientific weight is carried by randomized trials, and two
large randomized PSA screening studies have been published, ERSPC (Europe) and PLCO (United States), unfortunately with contradictory results.21,27 The European
ERSPC trial examined the role of PSA screening in a largely
FIGURE 3. Next-Generation Approach to Prostate

Cancer Screening and Treatment
Abbreviations: HG, high grade; LG, low grade; PSA, prostate-specific antigen.
unscreened population from eight countries with different

screening and treatment strategies. After 13 years of followup, a significant 21% reduction (rate ratio 0.79; 95% CI,
0.690.91) in death from prostate cancer is found in a
predefined subgroup of men age 5569. After adjustment
for nonparticipation, the rate ratio of prostate cancer
mortality among men screened was 0.73 (95% CI, 0.610.88).
After 13 years of follow-up, there was no evidence of a
mortality benefit for organized annual PSA screening in
the PLCO trial compared with opportunistic screening
used in normal care. Men randomly assigned to the annual
screening group had a higher incidence of prostate cancer
(relative rate [RR] 1.12; 95% CI, 1.071.17) but no reduction
in prostate cancer mortality (RR 1.09; 95% CI, 0.871.36).
Differences in follow-up protocols and screening intervals
exist between the two trials. The major differences can
mostly be explained by the high screening rates (52% after
6 years) in the control arm and low compliance to biopsies
in the PLCO trial.
An interesting observation from cancer registries shows
that in a majority of countries in northern Europe and North
America, the age-adjusted mortality rates for prostate
cancer have been decreasing substantially in the last 20
years. However, this trend is not global, as several countries
in Asia and Africa record increasing mortality during the
same period (Fig. 4). Lithuania is an outlier with a very rapid
increase in mortality in the last 20 years.26 The drop in
mortality in some countries can be explained by PSA testing,
better treatment, or changes in how prostate cancer deaths
are recorded. On the basis of SEER data modeling, 50% of the
observed decrease is estimated due to earlier detection with
PSA, and 50% can be accounted to better treatment.28 More
detailed analysis from the Swedish Cancer Registries during
1990 to 2009 shows a lower prostate cancer mortality in
counties with high PSA testing as compared with less-intense
opportunistic PSA testing areas.29 Data from cancer registries also show that the incidence of advanced and metastatic prostate cancer is going down.
e83

FIGURE 4. Trends in Prostate Cancer Mortality Rates

in 10 Selected Countries (Age Adjusted to World
Standard)
So what is the way forward? Continuing to argue over the

use of the PSA test would result in polarized discussion
rather than promote development of new ideas. We suggest
an alternative route forward, getting away from the PSA test
into a more modern diagnostic pipeline using the rapidly
advancing technology in biomarkers, imaging, and genomics. The goal of such a new diagnostic pipeline is to substantially reduce the harm of current PSA testing while
maintaining or increasing the sensitivity for diagnosing highrisk cancers.
Blood-Based Biomarkers and Risk Calculators
In a recent review of early detection of prostate cancer, 16 of

26 (62%) key opinion leaders agree that PSA screening does
reduce death from prostate cancer; others thought that the
present evidence is not sufficiently conclusive.30 The consensus
was that the magnitude of the effect was uncertain and that
substantial overdiagnosis and overtreatment exists, which
needs to be reduced before recommendations for PSA
screening in the general population can be made.
With available data, guideline committees have come to
very different conclusions, ranging from not recommending
PSA testing (USPTF) to recommending population-based,
organized PSA testing (Lithuania).31 Most professional organizations are somewhere in between, recommending PSA
testing for men age 45 or age 50 to 70 years but in a context
of shared decision making.32
The currently used biomarker PSA has low specificity

(only 20%30%), which leads to many men without cancer undergoing prostate biopsies. Biopsies are unpleasant, and 2%5% of men will get septicemia after the
biopsies.
Use of PCPTRC has been previously discussed. These risk
calculators are important tools, but unfortunately not used
as much as they should be.
With the inefficient use of PSA in Stockholm,33 we initiated
the STHLM3 study in 2011 with the goal of developing a
better prostate cancer test than PSA (Fig. 5). This new prostate cancer test should at least have the same sensitivity to
identify men with high-risk prostate cancer but with much
better specificity. This improved specificity should lead
to fewer unnecessary biopsies and fewer men diagnosed
with low-risk cancer. The STHLM3 study is a prospective,
population-based, paired, screen-positive, diagnostic study
of men without prostate cancer aged 50 to 69 randomly
invited by date of birth from the Swedish Population Register. The predefined STHLM3 model included a combination
of six plasma protein biomarkers (PSA, free PSA, intact PSA,
hK2, MSMB, MIC1), genetic polymorphisms (232 SNPs) associated with prostate cancer susceptibility, and clinical
variables (age, family, history, previous prostate biopsy,
FIGURE 5. Scheme for Prostate Cancer Screening, Diagnostics, and Treatment: Today and Tomorrow
Abbreviations: PSA, prostate-specific antigen; STKHLM3, Stockholm3.
e84
FIGURE 6. Variables Used in the Stockholm III Model
sensitivity as the PSA test using a cutoff of at least 3 ng/mL

to diagnose high-risk prostate cancer, use of the STHLM3
model could reduce the overall number of biopsies by 32%
(95% CI, 24%39%) and small Gleason score 6 cancers by
17% (95% CI, 7%26%) and could avoid 44% (95% CI,
35%54%) of biopsies for men without cancer. The STHLM3
model also identifies a significant number of men with highrisk cancers in the PSA range of 1-3 ng/mL (Fig. 7).
We see the STHLM3 test as a first step in reducing the
harms of current ongoing PSA screening. The Institute of
Health Economy in Lund, Sweden, has conducted a detailed
health-economic evaluation of the STHLM3 test. Compared
with current clinical practice in Stockholm today, the new
test can save money for the health care provider and at the
same time increase the quality of life in the population.
Abbreviations: DRE, digital rectal examination; PSA, prostate-specific antigen;

STHLM3, Stockholm III
Targeted Prostate Biopsies Using MRI With Fusion

Technology
and prostate examination) with the endpoint of Gleason
score 7+ cancers.
The STHLM3 model was predefined in a test cohort of
11,130 men and then prospectively validated in an independent cohort of 47,688 men. All variables used in the
STHLM3 model were significantly associated with prostate
cancers with a Gleason score 7+ (p , .05) in a multiple
logistic regression model (Fig. 6). At the same level of
Men at increased risk of prostate cancer are traditionally

assessed using systematic prostate biopsies. The procedure
is performed under local anesthesia using antibiotic prophylaxis and includes 10 to 14 cores taken from predefined
areas of the peripheral zone of the gland as visualized by
endorectal ultrasound. While the biopsies systematically
cover the prostatic gland rather than targeting a lesion, and
as low-risk cancers are common, the risk of overdiagnosis
FIGURE 7. Comparison of Number of Intermediate- and High-Grade Prostate Cancers Diagnosed with Serum PSA
Screening Versus Stockholm III
Abbreviation: PSA, prostate-specific antigen
e85

(i.e., detection of nonsignificant tumors) is high. The risk

of nonrepresentative biopsy findings results in underestimation of Gleason grade compared with subsequent
prostatectomy in up to 40% of men undergoing surgery.
Multiparametric MRI incorporating anatomic and functional imaging has now been validated as a means of detecting and characterizing prostate tumors. In 2012, the
European Society of Urogenital Radiology established the
Prostate Imaging Reporting and Data System (PI-RADS)
guidelines aimed at standardizing the acquisition, interpretation, and reporting of prostate multiparametric MRI.
Consensus on an updated version (PI-RADS v2) has recently
been published, outlining aspects of both interpretation and
technical execution.34
Targeted biopsies of the prostate consist of imaging (MRI)
detecting a lesion and a biopsy procedure where biopsies are
targeted to that area using various devices for guidance.
Cognitive fusion is based on a skilled physician interpreting
the MRI images and directing needles solely on the basis of
the ultrasound images. The disadvantages of cognitive fusion lie in the potential for human error when attempting to
mentally fuse the MRI with ultrasound while aiming for
cancers that are often smaller than 1 cm and the inability to
track the location of each biopsy site. Hard fusion enables
proper fusion of MRI information with the ultrasound image,
increasing precision of the biopsy.
A number of studies have investigated the value of fusion
biopsies, using nonrandomized designs and nonscreening
populations. A recent meta-analysis of 16 studies compared
MRI-targeted biopsies (MRI-TBx) and ultrasound-targeted
biopsies (TRUS-Bx) for prostate cancer detection.35 MRI-TBx
and TRUS-Bx did not significantly differ in overall prostate
cancer detection (sensitivity, 0.85 and 0.81, respectively).
MRI-TBx had a higher rate of detection of significant prostate
cancer compared with TRUS-Bx (sensitivity, 0.91 vs. 0.76)
and a lower rate of detection of insignificant prostate cancer
(sensitivity, 0.44 vs. 0.83). Subgroup analysis revealed an
improvement in significant prostate cancer detection by
MRI-TBx among men with previous negative biopsy, rather
than among men with initial biopsy. The authors concluded
that, due to underlying methodological flaws of MRI-TBx, the
comparison must be regarded with caution.
The field of MRI in prostate cancer diagnostics is rapidly

evolving, and, in many centers, MRI is regarded as standard
of care, despite solid scientific evidence to support such.
There is a lack of large randomized studies comparing these
two biopsy techniques. In Stockholm, we are initiating a
randomized study (STHLM3-MRI/fusion study) during 2016,
including a population-based sample of 8,000 men age 50
to 69.
Genomic Profiling of Prostate Biopsies

The knowledge of cancer genomics has exploded with the
introduction of modern sequencing technology. New
technologies make it possible to conduct both RNA- and
DNA-based sequencing on small amounts of formalin-fixed
tissue. Several commercially available RNA-based gene
panels with some external validation are now available.
Today, the commercially available tests are Prolaris,
Oncotype DX Genomic Prostate Score, and Decipher, which
can be used to estimate disease outcome in addition to
clinical parameters. ConfirmMDx is a DNA-based epigenetic
test used to predict the results of repeat prostate biopsy
after an initial negative biopsy.36 However, there is a need
for external validation in large cohorts of men on active
surveillance to find the optimal role of these genomic tests.
CONCLUSION
Screening guidelines regarding use of PSA have, for the most
part, moved to informed or shared decision making. The PSA
and the PCPTRC 2.0 are currently available for use and can be
used in helping doctors and patients decide about biopsy
after screening results are available. More precise tests and
decision tools will be available in the future. Indeed, the use
of new biomarkers, imaging, and genomic profiles in diagnosing high-risk prostate cancer is extremely promising.
There is an urgent need for evaluation of these new tests in
high-quality, large population-based studies. In addition to
efficacy, these new modalities need to be assessed with a
health-economic perspective. If not, we might find ourselves
in the same situation as with the PSA test and most likely
using these new diagnostic tools inefficiently, only driving
costs without any gain.
References
1. Wolf AM, Wender RC, Etzioni RB, et al; American Cancer Society Prostate
Cancer Advisory Committee. American Cancer Society guideline for the early
detection of prostate cancer: update 2010. CA Cancer J Clin. 2010;60:70-98.
2015;65:5-29.
3. Center MM, Jemal A, Lortet-Tieulent J, et al. International variation in
prostate cancer incidence and mortality rates. Eur Urol. 2012;61:
1079-1092.
4. Pilepich MV, Caplan R, Byhardt RW, et al. Phase III trial of androgen
suppression using goserelin in unfavorable-prognosis carcinoma of the
prostate treated with definitive radiotherapy: report of Radiation
e86
Therapy Oncology Group Protocol 85-31. J Clin Oncol. 1997;15:

1013-1021.
5. Bolla M, Gonzalez D, Warde P, et al. Improved survival in patients with
locally advanced prostate cancer treated with radiotherapy and
goserelin. N Engl J Med. 1997;337:295-300.
6. Bill-Axelson A, Holmberg L, Ruutu M, et al; SPCG-4 Investigators. Radical
prostatectomy versus watchful waiting in early prostate cancer. N Engl J
Med. 2011;364:1708-1717.
7. Bill-Axelson A, Holmberg L, Ruutu M, et al; Scandinavian Prostate
Cancer Group Study No. 4. Radical prostatectomy versus watchful
waiting in early prostate cancer. N Engl J Med. 2005;352:1977-1984.
8. Andriole GL, Crawford ED, Grubb RL III, et al; PLCO Project Team.
Mortality results from a randomized prostate-cancer screening trial.
N Engl J Med. 2009;360:1310-1319.
9. Schroder FH, Hugosson J, Roobol MJ, et al; ERSPC Investigators.
Screening and prostate-cancer mortality in a randomized European
study. N Engl J Med. 2009;360:1320-1328.
10. Thompson IM, Goodman PJ, Tangen CM, et al. The influence of finasteride
on the development of prostate cancer. N Engl J Med. 2003;349:215-224.
11. Wilt TJ, Brawer MK, Jones KM, et al; Prostate Cancer Intervention versus
Observation Trial (PIVOT) Study Group. Radical prostatectomy versus
observation for localized prostate cancer. N Engl J Med. 2012;367:
203-213.
12. Carter HB, Partin AW, Walsh PC, et al. Gleason score 6 adenocarcinoma:
should it be labeled as cancer? J Clin Oncol. 2012;30:4294-4296.
13. Welch HG, Albertsen PC. Prostate cancer diagnosis and treatment after
the introduction of prostate-specific antigen screening: 1986-2005.
2016;66:7-30.
15. Hoffman RM, Stone SN, Hunt WC, et al. Effects of misattribution in
assigning cause of death on prostate cancer mortality rates. Ann Epidemiol. 2003;13:450-454.
16. Attard G, Parker C, Eeles RA, et al. Prostate cancer. Lancet. 2016;387:
70-82.
17. U.S. Preventive Services Task Force. Screening for prostate cancer: U.S.
Preventive Services Task Force recommendation statement. Ann Intern
Med. 2008;149:185-191.
18. Moyer VA; U.S. Preventive Services Task Force. Screening for prostate
cancer: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med. 2012;157:120-134.
19. Smith RA, Manassaram-Baptiste D, Brooks D, et al. Cancer screening in
the United States, 2015: a review of current American cancer society
guidelines and current issues in cancer screening. CA Cancer J Clin. 2015;
65:30-54.
Prostate-cancer mortality at 11 years of follow-up. N Engl J Med. 2012;
366:981-990.
Screening and prostate cancer mortality: results of the European
Randomised Study of Screening for Prostate Cancer (ERSPC) at 13 years
of follow-up. Lancet. 2014;384:2027-2035.
22. Klotz L, Vesprini D, Sethukavalan P, et al. Long-term follow-up of a large
active surveillance cohort of patients with prostate cancer. J Clin Oncol.
2015;33:272-277.
23. Do V, Choo R, De Boer G, et al. The role of serial free/total prostatespecific antigen ratios in a watchful observation protocol for men with
localized prostate cancer. BJU Int. 2002;89:703-709.
24. Thompson IM, Ankerst DP, Chi C, et al. Assessing prostate cancer risk:
results from the Prostate Cancer Prevention Trial. J Natl Cancer Inst.
2006;98:529-534.
25. Thompson IM, Pauler DK, Goodman PJ, et al. Prevalence of prostate
cancer among men with a prostate-specific antigen level , or =4.0 ng
per milliliter. N Engl J Med. 2004;350:2239-2246.
26. Torre LA, Siegel RL, Ward EM, et al. Global cancer incidence and
mortality rates and trends-an update. Cancer Epidemiol Biomarkers
Prev. 2016;25:16-27.
27. Andriole GL, Crawford ED, Grubb RL III, et al; PLCO Project Team.
Prostate cancer screening in the randomized Prostate, Lung, Colorectal,
and Ovarian Cancer Screening Trial: mortality results after 13 years of
follow-up. J Natl Cancer Inst. 2012;104:125-132.
28. Etzioni RD, Thompson IM. What do the screening trials really tell
us and where do we go from here? Urol Clin North Am. 2014;41:
223-228.
29. Stattin P, Carlsson S, Holmstrom B, et al. Prostate cancer mortality in
areas with high and low prostate cancer incidence. J Natl Cancer Inst.
2014;106:dju007.
30. Cuzick J, Thorat MA, Andriole G, et al. Prevention and early detection of
prostate cancer. Lancet Oncol. 2014;15:e484-e492.
31. Smailyte G, Aleknaviciene B. Incidence of prostate cancer in Lithuania
after introduction of the Early Prostate Cancer Detection Programme.
Public Health. 2012;126:1075-1077.
32. Heidenreich A, Bastian PJ, Bellmunt J, et al; European Association of
Urology. EAU guidelines on prostate cancer. part 1: screening, diagnosis, and local treatment with curative intent-update 2013. Eur Urol.
2014;65:124-137.
33. Gronberg H, Adolfsson J, Aly M, et al. Prostate cancer screening in men
aged 50-69 years (STHLM3): a prospective population-based diagnostic
study. Lancet Oncol. 2015;16:1667-1676.
34. Barentsz JO, Weinreb JC, Verma S, et al. Synopsis of the PI-RADS v2
guidelines for multiparametric prostate magnetic resonance imaging
and recommendations for use. Eur Urol. 2016;69:41-49.
35. Schoots IG, Roobol MJ, Nieboer D, et al. Magnetic resonance imagingtargeted biopsy may enhance the diagnostic accuracy of significant
prostate cancer detection compared to standard transrectal
ultrasound-guided biopsy: a systematic review and meta-analysis. Eur
Urol. 2015;68:438-450.
36. Bostrom PJ, Bjartell AS, Catto JW, et al. Genomic predictors of outcome
in prostate cancer. Eur Urol. 2015;68:1033-1044.
e87

AND EPIDEMIOLOGY
Refining Breast Cancer Risk

Assessment: Additional Genes,
Additional Screening
CHAIR
Allison W. Kurian, MD, MSc
Stanford, CA
SPEAKERS
Antonis C. Antoniou, PhD
University of Cambridge
Cambridge, United Kingdom
Susan M. Domchek, MD
University of Pennsylvania Perelman School of Medicine
Philadelphia, PA
KURIAN, ANTONIOU, AND DOMCHEK
Refining Breast Cancer Risk Stratification: Additional Genes,

Additional Information
Allison W. Kurian, MD, MSc, Antonis C. Antoniou, PhD, and Susan M. Domchek, MD
OVERVIEW
Recent advances in genomic technology have enabled far more rapid, less expensive sequencing of multiple genes than was
possible only a few years ago. Advances in bioinformatics also facilitate the interpretation of large amounts of genomic data.
New strategies for cancer genetic risk assessment include multiplex sequencing panels of 5 to more than 100 genes (in which
rare mutations are often associated with at least two times the average risk of developing breast cancer) and panels of
common single-nucleotide polymorphisms (SNPs), combinations of which are generally associated with more modest cancer
risks (more than twofold). Although these new multiple-gene panel tests are used in oncology practice, questions remain
about the clinical validity and the clinical utility of their results. To translate this increasingly complex genetic information for
clinical use, cancer risk prediction tools are under development that consider the joint effects of all susceptibility genes,
together with other established breast cancer risk factors. Risk-adapted screening and prevention protocols are underway,
with ongoing refinement as genetic knowledge grows. Priority areas for future research include the clinical validity and
clinical utility of emerging genetic tests; the accuracy of developing cancer risk prediction models; and the long-term
outcomes of risk-adapted screening and prevention protocols, in terms of patients experiences and survival.
ecent advances in genomic technology have enabled

far more rapid, less expensive sequencing of multiple genes than was possible only a few years ago. Advances in bioinformatics also facilitate the interpretation
of large amounts of genomic data.1 Massively parallel,
next-generation sequencing allows the simultaneous
analysis of many different genes, 2,3 and the emerging
clinical application of whole-genome sequencing.4-6 For
clinical cancer risk assessment, this next-generation
sequencing technology has been implemented primarily as germline multiplex panels that analyze from 5 to
more than 100 cancer-implicated genes. In general,
pathogenic mutations in genes on these multiplex sequencing panels confer a twofold or greater risk of developing cancer (considered in some settings to be a
threshold value for intensified screening),7,8 and are rare
in the population (, 0.5% carrier frequency).9,10 These
genes are sometimes referred to as moderate to high
penetrance cancer susceptibility genes. Focusing on
breast cancer risk, multiplex sequencing panels (MSPs)
represent an intermediate stage between the past
practice of testing BRCA1 and BRCA2 (BRCA1/2) only, 11
and a possible future of routine whole-genome sequencing.
The genes included on most MSPs are displayed in

Table 1. 12-42
Another recent development are panels of several
SNPs that correlate with breast cancer risk. Identified
largely through genome-wide association studies (GWAS)
that compared breast cancer cases and controls,43-46 SNPs
most often have minor allele frequencies of greater than
1% and are associated with a lower magnitude of cancer
risk (e.g., odds ratios of 1.0-1.5). Recent studies have explored the effect of SNP combinations,47,48 and of SNP
interactions with BRCA1/2 mutations14,49-51 and other risk
factors.52,53 As described later, SNP panels are being used
to improve the accuracy of breast cancer risk prediction
models.
At this time, no single, clinically available panel combines
all established breast cancerassociated genes and SNPs.
Practice guidelines have begun to extend management
recommendations to carriers of mutations in the higher-risk
genes on MSPs,12 but no guidelines currently exist for cancer
screening or prevention on the basis of the combined effects
of the common genetic variants (SNPs). Significant questions
remain about these emerging approaches to genetic risk
assessment.
From the Departments of Medicine and of Health Research and Policy, Stanford University School of Medicine, Stanford, CA; Centre for Cancer Genetic Epidemiology, Department of
Public Health and Primary Care, University of Cambridge, Cambridge, United Kingdom; Basser Research Center and Abramson Cancer Center, University of Pennsylvania, Philadelphia, PA.
Corresponding author: Allison W. Kurian, MD, MSc, Divisions of Oncology and Epidemiology, Stanford University School of Medicine, 150 Governors Ln., Stanford, CA 94305-5405;
email: akurian@stanford.edu.
44
EXTENDING GERMLINE TESTING IN PATIENTS WITH BREAST CANCER
ADVANTAGES OF MULTIPLEX SEQUENCING

PANELS
Multiplex sequencing panels that analyze several breast
cancerrelated genes simultaneously offer several advantages over the prior standard of testing only two genes
(BRCA1/2). In past years, the high cost of sequencing
more than two genes was prohibitive and thus limited
to patients with striking features of Cowden, Li Fraumeni, or
other rare syndromes. By contrast, MSPs of 5 to greater than
100 genes now cost approximately $1,500, less than half the
cost of sequencing two genes in past years. With a turnaround time of 3 to 4 weeks, MSPs are far more efficient than
the previous standard of care.
Multiplex sequencing panels also provide an improved
diagnostic yield. Among clinic-based studies that collectively
assessed more than 10,000 patients who tested negative for
BRCA1/2 mutations, mutation prevalence in other MSP
genes ranged from 4% to 16%.54-64 Some mutations were
clinically unexpected (e.g., a patient with triple-negative
breast cancer who was found to have a MSH6 mutation,
consistent with Lynch syndrome),56 prompting a change in
cancer screening or prevention targeted to the identified
mutation. There is growing recognition that the striking
cancer phenotypes associated with certain high-risk mutations may be too narrowly defined, and that mutation penetrance may be more variable than previously
believed.65-67
MULTIPLEX SEQUENCING PANELS

Questions of Clinical Validity
The benefits of MSPs are accompanied by potential risks and
uncertainties. These concern questions of clinical validity
(such as whether the test result correctly predicts a clinical phenotype [e.g., high cancer risk]) and clinical utility
(whether the test result enables better patient care and

outcomes).
With regard to clinical validity, a major challenge lies in
the high rate of reported variants of uncertain significance
(VUS) with MSPs. Variants of uncertain significance are
genetic alterations (frequently missense) whose disease
association is unknown, and uncertainty about their clinical
relevance may frustrate patients and physicians. With
MSPs, the prevalence of at least one VUS (among all sequenced genes) is approximately 20% to 40%.54,57,59,64,68
By contrast, VUS are rare (2%5%) when only BRCA1/2 are
sequenced, because widespread BRCA1/2 testing in diverse
populations has better defined the spectrum of normal
variation for these genes.69 Another challenge arises when
two laboratories provide conflicting interpretations of the
same genetic alteration (e.g., VUS vs. deleterious).
Questions of clinical validity also concern apparently
pathogenic (often truncating) mutations in genes whose
cancer risks are not well understood, yet are included on
MSPsfor example, whether a mutation in MRE11A confers
clinically meaningful risk of breast cancer.9,66 There is no
transparent process for gene selection on many MSPs, and
the combination of genes may seem arbitrary. Emerging
MSP delivery models that empower ordering clinicians to
select genes for inclusion on a custom panel (and to exclude other genes considered irrelevant to a patients care)
may help, although this may require a level of genetic
knowledge that is unrealistic for most clinicians. A recent
review article designated 11 well-established breast
cancer susceptibility genes (ATM, BRCA1, BRCA2, CDH1,
CHEK2, NBN, NF1, PALB2, PTEN, STK11, TP53) and advised that mutations in other genes not be used in clinical
breast cancer risk assessment.70
Questions of Clinical Utility

KEY POINTS
Recent advances in genomic technology have enabled

far more rapid, less expensive sequencing of multiple
genes than was possible only a few years ago.
Although these new multiple-gene panel tests are used
in oncology practice, questions remain about the clinical
validity and the clinical utility of their results.
Cancer risk prediction tools that consider the joint
effects of all known susceptibility genes, together with
other established breast cancer risk factors, are under
development.
Risk-adapted screening and prevention protocols are
underway, with ongoing refinement as genetic
knowledge grows.
Priority research areas include the validity of emerging
genetic tests; the accuracy of developing cancer risk
prediction models; and the long-term outcomes of riskadapted screening and prevention protocols, in terms of
patients experiences and survival.
The most pressing question about MSPs is their clinical utility.

Rather than the more typical process for a new intervention,
in which proof of clinical utility precedes widespread
adoption, MSPs diffused into oncology practice ahead of
any supporting evidence base.9,66 Causative forces include
new genomic technology, a 2013 U.S. Supreme Court decision against gene patenting that caused increased market
competition and reduced costs, and celebrity testing disclosures that raised public awareness.71-74 In recent years,
studies have evaluated the clinical experience of MSP
testing in various populations, 54-60,62-64 with some reporting subsequent changes in screening and prevention
protocols.56,64 Practice guidelines including those of the
National Comprehensive Cancer Network (NCCN) have added
management recommendations for genes with a betterestablished risk profile (e.g., ATM, CHEK2, PALB2),12,26 and
in 2015 the American Society of Clinical Oncology (ASCO)
issued a position statement on extended germline testing.75,76
However, uncertainty remains about patients preferences and
experiences of MSPs; the optimal care delivery models; and
the effectiveness of risk-reducing interventions developed for
45
TABLE 1. Cancer Risks and Guidelines for Breast CancerAssociated Genes Often Included on Multiplex Panels
Established Breast Cancer Susceptibility Genes
Gene
Breast Cancer
Relative Risk
Other Cancer Risks and Syndromes
Clinical Practice Guidelines
Reference
ATM
Two- to threefold
Ataxia telangiectasia syndrome in

homozygotes; possibly
colon, pancreas
National Comprehensive Cancer Network (NCCN): Screening

with breast MRI
12,13
BRCA1
10-fold
Ovarian
American Cancer Society (ACS) and NCCN: Screening

breast MRI, recommend risk-reducing bilateral
salpingo-oophorectomy (RRBSO),
discuss risk-reducing mastectomy (RRM)
8,12,14-16
BRCA2
10-fold
Ovarian, pancreatic, prostate;

possibly melanoma
ACS and NCCN: Screening breast MRI, recommend RRBSO,

discuss RRM
8,12,14-16
CDH1
Fivefold
Gastric
NCCN: Screening breast MRI, discuss RRM, discuss

risk-reducing gastrectomy
12,17-19
CHEK2
Two- to threefold
Possibly colon, thyroid, lung
NCCN: Screening breast MRI
12,20,21
NBN
Two- to threefold
Nijmegen breakage syndrome in

homozygotes; possibly ovarian
None
22-24
NF1
Two- to threefold
Central nervous system,

peripheral nerve sheath
No breast cancerrelated guidelines
25
PALB2
Three- to fivefold
Pancreas; possibly ovarian
NCCN: Screening breast MRI, discuss RRM
12,26
PTEN
At least fivefold
Thyroid, endometrial
12,27,28
STK11
At least fivefold
Pancreas, colon, ovarian sex

cord-stromal
12,29
TP53
At least 10-fold
Multiple: sarcoma, leukemia,

adrenocortical, brain, others
NCCN: Screening breast MRI, discuss RRM;

whole-body MRI, colonoscopy, complete
blood count, and other tests
12,30
Genes With Undefined Risk of Breast Cancer Frequently Included on Multiplex Sequencing Panels
Gene
Breast Cancer
Relative Risk
Other Cancer Risks and Syndromes
Clinical Practice Guidelines
Reference
BARD1
Undefined
Unknown
None
31,32
BLM
Undefined
Blooms syndrome in homozygotes
None
33
BRIP1
Undefined
Ovarian
NCCN: Consider RRBSO
12,34
FAM175A
Undefined
Fanconi anemia in homozygotes
None
35
FANCC
Undefined
Fanconi anemia in homozygotes
None
36
MRE11A
Undefined
Ataxia telangiectasialike disorder in

homozygotes; possibly ovarian
None
22,23,37
RAD50
Undefined
Possibly ovarian
None
22,23
RAD51C
Undefined
Ovarian
12,38
RAD51D
Undefined
Ovarian
12,39-41
XRCC2
Undefined
Unknown
None
42
BRCA1/2 mutation carriers among carriers of other gene

mutations.
UNANSWERED QUESTIONS ABOUT MULTIPLEX

SEQUENCING
Genetic Epidemiology
Determining the clinical utility of MSPs requires research in two
major disciplines: epidemiology and health service research.66
First, we must understand the prevalence, penetrance, and
phenotype of mutations in genes beyond BRCA1/2; second, we
must understand how identifying such mutations affects patients outcomes, in terms of psychosocial symptoms, cancer
detection rate and stage shift, and, ultimately, survival.
46
As discussed previously, studies have reported estimates

of the prevalence of other gene mutations among patients
tested for BRCA1/2 mutations.12,54-60,62-64 Although informative and clinically relevant, most existing studies have
focused exclusively on patients with striking personal and/or
family histories of early onset breast/ovarian cancer, and on
non-Hispanic whites. A key question is the prevalence of
mutations and VUS among patients at lower estimated risk,
including breast cancer patients unselected for age or family
history; such data are essential to learning which patients may benefit from MSPs. Of similar importance is
defining the prevalence of mutations and VUS across
racial/ethnic groups, as done for BRCA1/2.77,78 Studies in
population-based datasets are currently underway to

address these questions.
Understanding the variability in penetrance of mutations
in genes beyond BRCA1/2 is necessary to enable accurate
patient counseling. Large, collaborative studies will be required to obtain sufficient numbers of mutation carriers:
ongoing consortium efforts include the PROMPT, Breast
Cancer Risk after Diagnostic Gene Sequencing (BRIDGES),
and COMPLEXO studies (www.promptstudy.info; cordis.
europa.eu/project/rcn/193315_en.html).79 Other crucial
research questions include reclassifying clinically ambiguous
VUS to benign or deleterious, which is an aim of the ENIGMA
consortium80; identifying the pathologic and prognostic
features of tumors associated with mutations in specific
genes; and etiologic studies of cancer causation by mutations, particularly those in genes whose association with
breast cancer is less well proven.
Health Services Research

There are several crucial topics for health services research on
MSPs. One important question is how best to deliver genetic
counseling and testing services, in the challenging environment of increasingly complex test results and a shortage of
licensed genetic counselors.81 New delivery models include a
tiered and binned approach to summarizing test options by
risk categories.82 Long-distance counseling methods such as
telemedicine are also under investigation.
Another essential question is the patient experience of
MSP testing. As described before, there is concern that the
volume, complexity, and uncertainty of results may overwhelm patients. One study reported that some patients
decline MSPs because of such concerns.83 In the ongoing
Hereditary Cancer Panel prospective trial of MSPs at three
academic cancer centers, most patients reported little
distress or uncertainty after receiving their test results.84
A related question is the impact of extended germline
sequencing on patients uptake of interventions such as
preventive surgery or more intensive screening (e.g.,
magnetic resonance imaging), given the concern that VUS
might be misinterpreted as cause for unwarranted interventions. Although early results of the Hereditary Cancer
Panel trial offer no evidence that patients overuse prophylactic procedures based on test results,84 longer-term
follow-up is required. Answering these questions about the
delivery and outcomes of MSP testing will be required
to guide analyses of the incremental benefit of analyzing
more versus fewer genes, both in terms of efficacy and of
cost-effectiveness.
POLYMORPHIC BREAST CANCER RISK FACTORS

AND NEW MODELS
Despite the widespread availability of MSPs and SNP arrays,
their utility is limited by the lack of risk prediction models
that consider the multifactorial etiology of breast cancer
susceptibility. There is an urgent need for breast cancer risk
prediction models that consider the joint effects of all known
genetic susceptibility variants, together with other established risk factors. However, to construct such multifactorial
cancer risk prediction models, it is important first to understand the underlying models of genetic susceptibility for
breast cancer.
Models of Genetic Susceptibility to Breast Cancer

Epidemiologic studies have estimated that breast cancer is
approximately twice as common among first-degree relatives of breast cancer patients (defined as the familial
relative risk).85 The most important breast cancer susceptibility genes are BRCA1/2, mutations that account for
17% to 20% of the excess familial risk of breast cancer.86-88
Rare mutations in the established breast cancer susceptibility genes included on MSPs, such as PALB2, CHEK2, ATM,
TP53, PTEN, CHD1, STK11, NF1, and NBN (Table 1), account
for approximately 5%,89 and more than 100 common SNPs
account for an additional 16% of familial breast cancer.90
Together, all known breast cancer susceptibility variants
account for 38% to 41%; thus more than 50% of familial
clustering remains unexplained. Both known genetic variants and residual familial aggregation must be considered in
calculating the risk of developing breast cancer.
Combined Single-Nucleotide Polymorphisms

Associations
Individual common alleles confer modest risks and their
individual predictive utility is poor. Recently, Mavaddat
et al91 investigated all pairwise interactions between 77
breast cancerassociated SNPs and found no evidence of
deviation from a model in which SNPs combine multiplicatively. Common SNPs can thus be combined into a Polygenic Risk Score (PRS): considering 90 common susceptibility
loci, 5% of women have a greater than twofold increased risk
and 0.7% have a greater than threefold increased risk,
compared with the general populations risk of breast
cancer.90 In models that considered the joint effects of PRS
and family history, PRS stratifies risk in women without
family history and further refines genetic risk assessment in
women with family history.48,91-95 Ongoing large studies,
such as the OncoArray experiment, are expected to refine
these models.
Single-Nucleotide Polymorphisms and BRCA1/2

Mutations
The International Consortium of Investigators of Modifiers
of BRCA1/2 (CIMBA) studied the effects of common SNPs on
BRCA1/2-associated breast cancer risks,96 using different
approaches: GWAS of BRCA1/2 mutation carriers,49,97,98
meta-analysis of BRCA1/2-associated risks with other similar
phenotypes, fine-scale mapping of known risk-modifying
loci,43 and directly assessing the effects of established
breast cancer susceptibility alleles in mutation carriers.99-103
CIMBA identified 24 SNPs that modify BRCA1-associated risk
(which have stronger associations with estrogen receptor
[ER]negative breast cancer in the general population) and
20 SNPs that modify BRCA2-associated risk.104-106 A recent
47
study investigated the associations of common SNPs subdivided by the ER status of BRCA1- and BRCA2-associated
breast tumors,51,103 and found that differences in SNP associations between BRCA1 and BRCA2 carriers and noncarriers are largely explained by differences in the
prevalence of ER-positive and ER-negative tumors in mutation carriers and the heterogeneity of the associations
between common alleles and ER-positive or ER-negative
disease. These are in line with a general model of susceptibility under which BRCA1/2 mutations and common alleles
combine multiplicatively once stratified by ER status. As a
result, jointly, common SNPs can result in large differences in
the absolute lifetime risks of breast cancer among BRCA1/2
mutation carriers.97,107 Risk stratification may improve as
enhanced PRS are developed that include a large number of
SNPs. Women with BRCA1/2 mutations may be among the
first to gain a clinically meaningful benefit from the use of
common SNPs for risk assessment; however, large studies
are required to assess prospectively the enhanced PRS that
are currently under development.
Single-Nucleotide Polymorphisms and Other Rare

Gene Mutations
Little is known about the combined effects of common SNPs
and other rare mutations beyond BRCA1/2, mainly because of
the lack of large-scale cohorts tested with MSPs. However,
breast cancer risks among PALB2 and CHEK2 mutation carriers appear to be modified by other familial factors, increasing with family history and/or with bilateral breast
cancer.26,108-110 These findings suggest a multiplicative model
as with BRCA1/2, but confirmatory studies are needed.
Combined Effects of Rare Gene Mutations

Recent evidence suggests that gene-gene interactions
among CHEK2, ATM, BRCA1, and BRCA2 may not be multiplicative.111 Indeed, a multiplicative model would be implausible for BRCA1 and BRCA2, because it would predict an
extremely high breast cancer risk at very young ages. Proteins encoded by the majority of breast cancer genes play
roles in DNA repair pathways. If no elevated risks are observed in the presence of two gene mutations, the data
would be consistent with a model in which, if the pathway is
disrupted by a mutation, further disruption by a lower
penetrance mutation would not increase risk.112
Combined Effects of Genetic, Lifestyle, and Hormonal

Factors
Studies of common SNPs and other risk factors suggest that
they combine in a multiplicative fashion.113-116 There is
currently no evidence on how the risks of rare gene mutations (other than BRCA1/2) combine with lifestyle and
hormonal factors, but large ongoing studies (e.g., BRIDGES,
cordis.europa.eu/project/rcn/193315_en.html), will report
on this question soon. The evidence for pathogenic variants
in BRCA1 and BRCA2 is currently limited and conflicting, but
there are some suggestions that the relative risks of lifestyle/
hormonal factors may not necessarily be the same in both
48
BRCA1 and BRCA2 mutation carriers as in the general

population.117 For example, later age of first birth is associated with lower breast cancer risk for BRCA1 mutation
carriers, yet with higher breast cancer risk in the general
population.117
Implications for Cancer Risk Stratification

Our growing understanding of the multifactorial etiology of
breast cancer suggests that much more powerful risk prediction can be achieved by combining data on all genetic,
lifestyle, and hormonal risk factors,92,118 reaching stratification levels that can guide individual decision making and
population risk-reduction programs.93,95,119 However, the
assumptions underlying these projections require testing in
empirical studies.
Breast Cancer Risk Assessment Tools

Several breast cancer risk prediction models are available
for unaffected women; they differ in the risk factors they
consider and in their underlying statistical modeling approaches (Table 2). The two broad categories of models
are empirical and genetic. Empirical models do not
assume a specific genetic model, but instead consider
summary measures of family history: an example is the
widely used Gail model120 in the Breast Cancer Risk Assessment Tool (www.cancer.gov/bcrisktool/). Genetic
models make explicit assumptions about known and
unknown genetic effects, considering the mode of inheritance, mutation frequency, and penetrance. A major
advantage of genetic models is that they can compute
risks for families of any disease structure; however, their
adequacy depends on modeling all genetic effects, which
can be approximate only because all breast cancer susceptibility genes have not yet been discovered.121
Examples of genetic models include Claus,122,123
BRCAPRO,124 IBIS,125 and BOADICEA (Table 2).112,126,127
BRCAPRO, IBIS, and BOADICEA all consider the explicit
effects of BRCA1/2 mutations but differ in their modeling
of residual familial aggregation: IBIS assumes a hypothetical third dominant gene, whereas BOADICEA
assumes a polygenic component.
Single-Nucleotide Polymorphisms in Risk Assessment

Tools
Several studies have demonstrated improvement in the
performance of some models through consideration of
common SNPs.47,119,128,129 However, most such studies assumed that the risks predicted by existing models (e.g., IBIS,
BOADICEA) are multiplicative with those conferred by SNPs,
without adjusting for the fact that SNPs account for approximately 15% of the familial aggregation of breast cancer.
This approach warrants further validation before routine
clinical use. More accurate implementations will require the
underlying algorithms to be modified to ensure that the familial risks predicted by the models on the basis of the known
and unknown genetic effects are correctly specified.
TABLE 2. Breast Cancer Risk Assessment Tools Used in Clinical Practice: Components and Assumptions
Factor*
Gail
Claus
BRCAPRO
IBIS
BOADICEA
Family history
YES (descriptive)
YES
YES
YES
YES
BRCA1, BRCA2 mutations
NO
NO
YES
YES
YES
Common low-risk alleles
NO
NO
NO
NO
NO
Intermediate-high risk mutations (CHEK2, PALB2, ATM, etc.)
NO
NO
NO
NO
YES**
Residual non-BRCA1/2 familial aggregation
NO
NO
NO
YES; dominant
3rd gene
YES; polygenic
BRCA1/2 breast cancer pathology associations
NO
NO
YES
NO
YES
BRCA1/2 risk modification by family history
NO
NO
NO
NO
YES
Variants of uncertain significance
NO
NO
NO
NO
NO
Predicting estrogen receptor (ER)-specific risks
NO
NO
NO
NO
NO
Mammographic density
NO
NO
NO
NO
NO
Hormonal, lifestyle, and reproductive risk factors
YES
NO
NO
YES; assumes same

effect on BRCA1/2
NO
Other cancers (nonbreast or ovarian cancer)
NO
NO
YES
NO
YES
Predicting second cancer risks (contralateral breast, ovarian cancer)
NO
NO
NO
NO
YES
*Considered in the available tools for clinical use.

**Beta-version release.
Only pathology information for the proband.
Rare Gene Mutations in Risk Assessment Tools

Until recently, no model incorporated the effects of mutations in genes beyond BRCA1/2. A major difficulty had been
the lack of precise cancer risk estimates; however, a recent
extension to the BOADICEA risk assessment model includes
the effects of truncating mutations in PALB2, CHEK2, and
ATM for which reliable risk estimates are now available.70,112
A beta-testing version of this tool is available for clinical use
(ccge.medschl.cam.ac.uk/boadicea/). This extended BOADICEA model can be used to obtain cancer risks on the basis
of mutation screening in all five major breast cancer susceptibility genes (BRCA1, BRCA2, PALB2, CHEK2, ATM) of
explicit family history information and of other risk factors
(Table 2). Figure 1A demonstrates that risks for carriers of
rare mutations (e.g., an ATM truncating variant) depend on
family history. Although the average lifetime breast cancer
risk for a carrier of an ATM truncating variant is approximately 29%, the risk for an ATM carrier whose mother developed breast cancer at age 40 is estimated at
approximately 44%. These differences may have implications
for screening and preventive interventions. Figure 1B shows
predicted risks for a woman whose mother carried a BRCA1
or ATM mutation and had breast cancer, but who has herself
tested negative for the same mutation (known as a true
negative or noncarrier130). Compared to considering
family history alone, the reduction in breast cancer risk after
negative testing is greater when the mother carries a
BRCA1 mutation (predicted risk approximately 15%)
compared with an ATM mutation (predicted risk . 18%).
These results suggest that negative predictive testing for
ATM (or similarly moderate-risk mutations) may not imply
sufficient risk reduction to enable reclassification for
clinical management purposes, and emphasize the need for
further study of the breast cancer risks associated with such
newly identified mutations for both mutation carriers and

their relatives.
Unresolved Problems in Risk Modeling, Challenges,

and Future Directions
Along with BRCA1, BRCA2, PALB2, ATM, and CHEK2, there are
several other breast cancer susceptibility genes which could
be incorporated into risk prediction.70 However, more accurate estimates are required for the breast cancer risks of
mutations in these genes. Similarly, little is known about the
joint effects of rare gene mutations, about the risks for carriers
of mutations in more than one breast cancer susceptibility
gene, or how these genetic risks combine with other lifestyle
and hormonal risk factors. A number of ongoing research
programs will address these issues in the near future, such as
the PROMPT, BRIDGES, and CARRIERS programs. As discussed
before, another problem is VUS reported in BRCA1/2 and
other genes. Several efforts are underway to reclassify these
VUS through consortia like ENIGMA80 and using approaches
such as the posterior probability of causality.131,132 Future
modeling efforts should aim to include such evidence in order
to provide an integrated approach for counseling VUS carriers.
Breast cancer is a heterogeneous disease that differs
based on tumor characteristics. Tumor characteristics vary
based on genetic background, and risk factors have different
associations with different tumor subtypes.133,134 A metaanalysis of trials of selective estrogen receptor modulators
as breast cancer chemoprevention agents suggested a 38%
risk reduction135; however, a critical gap for clinical translation is how to identify which women will develop ERpositive breast cancer, because selective estrogen receptor
modulators prevent only ER-positive disease.133,136 Thus
there is need of risk models to predict ER-specific risks. Data
generated through the large scale B-CAST (Breast CAncer
49
FIGURE 1. Predicted Breast Cancer Risks Using the BOADICEA Model, Beta-Version 4.0
(A) Predicted breast cancer risk by ATM mutation status and family history. BOADICEA-predicted risks by ATM mutation status for a female in the United Kingdom born in 1995 (age
20). The three curves show the breast cancer risk in the population, the risk for the average carrier of an ATM mutation (without taking family history into account), and the risk for
a carrier of an ATM mutation whose mother also developed breast cancer at age 40.
(B) Negative predictive testing and predicted breast cancer risk. The predicted risk of breast cancer for a 20-year-old female born in 1995 in the United Kingdom, whose
mother developed breast cancer at age 40, according to her mothers mutation status. Risk estimates were obtained using the BOADICEA model, beta-version 4.0.
Stratification, http://www.b-cast.eu) program will yield valuable data for modeling disease subtypespecific risks.
Finally, it will be essential to validate risk-prediction
models in large, well-designed prospective studies, and to
develop user-friendly tools that meet the needs of clinicians
at all care levels.
RISK REFINEMENT AND STRATIFYING

SCREENING APPROACHES
Mammography
Recent updates to guidelines have triggered ongoing controversy regarding the age of initiation and interval timing
of screening mammography in women at average risk for
breast cancer.137-139 Although the specifics of guidelines
differ, the trend is for later initiation of routine mammograms, with the possibility of spacing them to every other
year. Throughout this discussion, it has been recognized that
risk stratification is key to optimizing the clinical benefit of
breast cancer screening; that is, these new recommendations are for women at average risk, but women at higher risk
may need a more intensive approach. However, most
women who develop breast cancer are not considered to be
at elevated risk before their diagnosis. Thus, as guidelines
50
favor fewer mammograms, it becomes increasingly important to improve risk stratification using factors such as reproductive and hormonal exposures, breast density, and
genetics; and to understand the complex interactions between these risk factors. Optimizing risk stratification is
essential to inform individualized breast cancer screening
recommendations.
Breast MRI
At the current time for women at elevated risk, breast MRI is
often added to yearly screening mammogram. Breast MRI is
more sensitive than mammogram for detecting cancers in
BRCA1/2 mutation carriers and women with a family history
of breast cancer.140 Compared with historical controls,
breast cancers detected by MRI are diagnosed at an earlier
stage; historical comparisons and modeling analyses predict
that MRI screening may improve survival.141-145 No randomized studies have been performed evaluating breast
MRI as an adjunct to screening mammography. Concerns
about yearly breast MRI include cost, inconvenience (the
test requires intravenous contrast) and specificity. Cost of
breast MRI is particularly high in the United States. Although
adding annual breast MRI from the ages of 35 to 54 to annual
mammograms in BRCA1 mutation carriers has been shown

to be cost-effective at $100,000 per quality-adjusted lifeyear, the cost-effectiveness of MRI screening decreases at
lower levels of risk. For example, Plevritis et al calculated a
cost-effectiveness ratio (CER) of $266,344/quality-adjusted
life-year for annual MRI in addition to mammogram screening of BRCA2 mutation carriers from ages 35 to 54.146 CERs
in women with mutations conferring a lower risk of breast
cancer will be even less favorable.
Regardless of concerns about cost, multiple guidelines
recommend annual breast MRI in women at elevated risk
of breast cancer.7,12,147-149 Guidelines differ in the
stipulated level of risk that should trigger the addition of
breast MRI, but in the United States, the American
Cancer Society, and NCCN recommend adding annual
breast MRI to mammogram in women with a lifetime risk
of greater than 20% based on a family history model8,12;
this recommendation exists despite the absence of
empirical evidence that breast MRI screening improves
outcomes. Existing guidelines do not specify what is
meant by lifetime riskwhether cumulative lifetime risk
or remaining risk at a womans current age. Therefore,
depending on the models used, different recommendations may be made for the same woman.150
Average lifetime breast cancer risks are estimated to
approach or exceed 30% with mutations in PALB2, ATM,
NBN, and CHEK2 (for CHEK2, excluding the p.I157T and p.
S428F mutations). Therefore, carriers of mutations in
these genes could be considered for breast MRI screening
on the basis of existing U.S. guidelines. Importantly, two
missense mutations in CHEK2, p.I157T and p.S428F, are
associated with lower levels of breast cancer risk (relative
risk [RR] , 2), and the corresponding lifetime risks are
unlikely to exceed 20% in the absence of a family history of
breast cancer.151,152 Likewise, there is insufficient evidence at this time to recommend breast MRI screening in
carriers of mutations in BARD1, BRIP1, MRE11A, RAD51C,
and RAD51D in the absence of a family history.70 In those
women who do meet existing guidelines for MRI surveillance, the appropriate age at which to initiate screening is currently unclear. Because of the lower risks and
later median onset of breast cancer associated with mutations in moderate penetrance genes compared with
BRCA1/2, it seems reasonable to initiate breast MRI
screening in carriers of mutations in moderate penetrance
genes later than for BRCA1/2 mutation carriers, for whom
MRI screening begins at age 25; however, screening
recommendations for any individual patient will depend
on family cancer history.
Currently, breast MRI screening often begins 5 to
10 years before the earliest age at diagnosis of breast
cancer in the family; however, a more rational approach
may be to focus on the absolute risk of breast cancer over
the next 5 years. As evidence accumulates, the ages for
initiation of surveillance and the type of surveillance
must be re-examined. Because guidelines recommend
less for women at average risk, women identified as
carriers of mutations in moderate penetrance genes

should continue annual screening mammograms. Recently, Mandelblatt et al examined eight different
mammographic screening strategies and determined
that for women with a two- to fourfold increase over the
average breast cancer risk, annual screening from age
40 years had similar harms and benefits as did screening
average-risk women biennially from age 50 to 74 years.153
Finally, emerging technologies such as breast tomosynthesis
and rapid breast MRI may alter screening approaches in
the future.
Prevention Options
Women at elevated risk for breast cancer also have
options for prevention. All women, regardless of estimated risk level, should be counseled on the harmful
impact that obesity, lack of exercise, and alcohol use
have on breast cancer risk. In addition, pharmacologic
breast cancer risk reduction with tamoxifen, raloxifene,
or an aromatase inhibitor is available. In randomized
clinical trials, tamoxifen taken for 5 years confers a 30%
to 50% reduction in the risk of developing breast cancer, 136,154 with continued benefit at 16 years of followup (albeit no documented survival benefit to date). 136
Raloxifene also confers a reduction in the risk of breast
cancer. When directly compared with tamoxifen in a
randomized clinical trial, in long-term follow-up raloxifene had 76% of the effectiveness of tamoxifen in
preventing invasive breast cancer, yet far less toxicity in
terms of the risk of endometrial cancer.155 Exemestane
(an aromatase inhibitor) has also been shown to reduce
breast cancer risk in a randomized controlled trial, with a
hazard ratio of 0.35 (95% CI, 0.180.70; p = .002). 156
Although eligibility criteria differed somewhat for these
studies, most included women age 60 or older, those
with a Gail modelbased estimate of 5-year risk of developing breast cancer of greater than 1.66%, and/or
those with lobular carcinoma in situ. Data are not
available regarding the effectiveness of chemoprevention
among carriers of mutations in moderate penetrance
breast cancer susceptibility genes. CHEK2 mutations do
appear to be associated with ER-positive breast cancer,
but data regarding chemoprevention in this setting are
unavailable.
There is no threshold risk that mandates risk-reducing
mastectomy in any unaffected woman, regardless of her
mutation type. It is unlikely that mastectomy will provide a
survival advantage to women with moderate penetrance
mutations, given their level of risk and the effectiveness of
breast cancer screening and treatment. Elicitation of individual preferences and communication about levels of risk
are critical in these conversations.
Ovarian Cancer
Unfortunately, ovarian cancer screening has been shown not
to reduce mortality.157 Therefore, risk-reducing salpingooophorectomy (RRSO) is the standard recommendation for
51
women with BRCA1/2 mutations. Large case-control studies

have demonstrated that women with mutations in BRIP1,
RAD51C, and RAD51D are at increased risk of ovarian cancer.
The absolute risks associated with mutations in these genes
are intermediate between those of a woman with a BRCA1/2negative first-degree relative with ovarian cancer (RR 2.2,
absolute lifetime risk 3%5%) and of a woman carrying
BRCA2 mutation (absolute lifetime risk approximately
17%).15,158 Studies have not proven increased ovarian
cancer risks in carriers of mutations in other moderate
penetrance genes including PALB2, ATM, CHEK2, BARD1,
MRE11A, NBN, and RAD51B, although mutations in these
genes have been observed in families with both breast and
ovarian cancers.
Risk-reducing salpingo-oophorectomy is not routinely
recommended for women whose only ovarian cancer risk
factor is an affected first-degree relative. A womans cumulative risk should therefore at least be similar to the risk
of a woman with an affected BRCA1/2-negative first-degree
relative with ovarian cancer before RRSO is recommended. It
is unlikely that carriers of mutations in BRIP1, RAD51C, or
RAD51D would cross this risk threshold before age 50.
However, family history of ovarian cancer may influence this
risk and thus may alter decision making about RRSO timing.
Pancreas Cancer
Mutations in PALB2 and ATM have been associated with
familial pancreas cancer159,160; however, the absolute risks
of pancreas cancer associated with such mutations are
unknown. There are no proven effective screening or prevention measures for pancreas cancer, although there is
much interest in endoscopic ultrasonography and magnetic
resonance cholangiopancreatography as potential screening
approaches. Carriers of ATM or PALB2 mutations who have a
family history of pancreas cancer may be candidates for
clinical trials of screening strategies.
Communication
Communicating information about breast cancer risk and
the risks and benefits of management options can be
challenging in the clinical setting. As knowledge increases
about the genetic epidemiology of breast cancer risk,
communication becomes even more important and challenging. Partly owing to the op-ed pieces by celebrity
Angelina Jolie in The New York Times, more women are
aware of the implications of BRCA1/2 mutations. Physicians
must be careful to explain how mutations in moderate
penetrance genes differ from BRCA1/2 mutations in terms of
the magnitude of associated risks.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
52
Collins FS, Varmus H. A new initiative on precision medicine. N Engl J

Med. 2015;372:793-795.
Shendure J, Ji H. Next-generation DNA sequencing. Nat Biotechnol.
2008;26:1135-1145.
Tucker T, Marra M, Friedman JM. Massively parallel sequencing: the
next big thing in genetic medicine. Am J Hum Genet. 2009;85:142-154.
Kidd JM, Cooper GM, Donahue WF, et al. Mapping and sequencing of
structural variation from eight human genomes. Nature. 2008;453:56-64.
Yang Y, Muzny DM, Xia F, et al. Molecular findings among patients
referred for clinical whole-exome sequencing. JAMA. 2014;312:
1870-1879.
Dewey FE, Grove ME, Pan C, et al. Clinical interpretation and implications of whole-genome sequencing. JAMA. 2014;311:1035-1045.
Mainiero MB, Lourenco A, Mahoney MC, et al. ACR Appropriateness
Criteria Breast Cancer Screening. J Am Coll Radiol. 2013;10:11-14.
Saslow D, Boetes C, Burke W, et al; American Cancer Society Breast
Cancer Advisory Group. American Cancer Society guidelines for breast
screening with MRI as an adjunct to mammography. CA Cancer J Clin.
2007;57:75-89.
Domchek SM, Bradbury A, Garber JE, et al. Multiplex genetic testing
for cancer susceptibility: out on the high wire without a net? J Clin
Oncol. 2013;31:1267-1270.
Kurian AW, Kingham KE, Ford JM. Next-generation sequencing for
hereditary breast and gynecologic cancer risk assessment. Curr Opin
Obstet Gynecol. 2015;27:23-33.
Couch FJ, Nathanson KL, Offit K. Two decades after BRCA: setting
paradigms in personalized cancer care and prevention. Science. 2014;
343:1466-1470.
Daly MB, Pilarski R, Axilbund JE, et al; National comprehensive cancer
network. Genetic/familial high-risk assessment: breast and ovarian,
version 1.2014. J Natl Compr Canc Netw. 2014;12:1326-1338.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
Renwick A, Thompson D, Seal S, et al; Breast Cancer Susceptibility

Collaboration (UK). ATM mutations that cause ataxia-telangiectasia
are breast cancer susceptibility alleles. Nat Genet. 2006;38:873-875.
Mavaddat N, Peock S, Frost D, et al; EMBRACE. Cancer risks for BRCA1
and BRCA2 mutation carriers: results from prospective analysis of
EMBRACE. J Natl Cancer Inst. 2013;105:812-822.
Antoniou A, Pharoah PD, Narod S, et al. Average risks of breast and
ovarian cancer associated with BRCA1 or BRCA2 mutations detected
in case Series unselected for family history: a combined analysis of 22
Chen S, Parmigiani G. Meta-analysis of BRCA1 and BRCA2 penetrance.
J Clin Oncol. 2007;25:1329-1333.
Pharoah PD, Guilford P, Caldas C; International Gastric Cancer Linkage
Consortium. Incidence of gastric cancer and breast cancer in CDH1
(E-cadherin) mutation carriers from hereditary diffuse gastric cancer
families. Gastroenterology. 2001;121:1348-1353.
Kaurah P, MacMillan A, Boyd N, et al. Founder and recurrent CDH1
mutations in families with hereditary diffuse gastric cancer. JAMA.
2007;297:2360-2372.
van der Post RS, Vogelaar IP, Carneiro F, et al. Hereditary diffuse
gastric cancer: updated clinical guidelines with an emphasis on
germline CDH1 mutation carriers. J Med Genet. 2015;52:361-374.
Weischer M, Bojesen SE, Ellervik C, et al. CHEK2*1100delC genotyping
for clinical assessment of breast cancer risk: meta-analyses of 26,000
patient cases and 27,000 controls. J Clin Oncol. 2008;26:542-548.
Meijers-Heijboer H, van den Ouweland A, Klijn J, et al; CHEK2-Breast
Cancer Consortium. Low-penetrance susceptibility to breast cancer
due to CHEK2(*)1100delC in noncarriers of BRCA1 or BRCA2 mutations. Nat Genet. 2002;31:55-59.
Bartkova J, Tommiska J, Oplustilova L, et al. Aberrations of the MRE11RAD50-NBS1 DNA damage sensor complex in human breast cancer:
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
MRE11 as a candidate familial cancer-predisposing gene. Mol Oncol.

2008;2:296-316.
Hsu HM, Wang HC, Chen ST, et al. Breast cancer risk is associated with
the genes encoding the DNA double-strand break repair Mre11/
Rad50/Nbs1 complex. Cancer Epidemiol Biomarkers Prev. 2007;16:
2024-2032.
Zhang G, Zeng Y, Liu Z, et al. Significant association between Nijmegen
breakage syndrome 1 657del5 polymorphism and breast cancer risk.
Tumour Biol. 2013;34:2753-2757.
Seminog OO, Goldacre MJ. Age-specific risk of breast cancer in women
with neurofibromatosis type 1. Br J Cancer. 2015;112:1546-1548.
Antoniou AC, Casadei S, Heikkinen T, et al. Breast-cancer risk in
families with mutations in PALB2. N Engl J Med. 2014;371:497-506.
Tan MH, Mester JL, Ngeow J, et al. Lifetime cancer risks in individuals
with germline PTEN mutations. Clin Cancer Res. 2012;18:400-407.
Pilarski R, Eng C. Will the real Cowden syndrome please stand up
(again)? Expanding mutational and clinical spectra of the PTEN
hamartoma tumour syndrome. J Med Genet. 2004;41:323-326.
Hearle N, Schumacher V, Menko FH, et al. Frequency and spectrum of
cancers in the Peutz-Jeghers syndrome. Clin Cancer Res. 2006;12:
3209-3215.
Gonzalez KD, Noltner KA, Buzin CH, et al. Beyond Li Fraumeni Syndrome: clinical characteristics of families with p53 germline mutations. J Clin Oncol. 2009;27:1250-1256.
Karppinen SM, Barkardottir RB, Backenhorn K, et al. Nordic collaborative study of the BARD1 Cys557Ser allele in 3956 patients
with cancer: enrichment in familial BRCA1/BRCA2 mutation-negative
breast cancer but not in other malignancies. J Med Genet. 2006;43:
856-862.
Stacey SN, Sulem P, Johannsson OT, et al. The BARD1 Cys557Ser
variant and breast cancer risk in Iceland. PLoS Med. 2006;3:e217.
Broberg K, Huynh E, Schlawicke Engstrom K, et al. Association between polymorphisms in RMI1, TOP3A, and BLM and risk of cancer, a
case-control study. BMC Cancer. 2009;9:140.
Seal S, Thompson D, Renwick A, et al; Breast Cancer Susceptibility
Collaboration (UK). Truncating mutations in the Fanconi anemia J gene
BRIP1 are low-penetrance breast cancer susceptibility alleles. Nat
Genet. 2006;38:1239-1241.
Solyom S, Aressy B, Pylkas K, et al. Breast cancer-associated Abraxas
mutation disrupts nuclear localization and DNA damage response
functions. Sci Transl Med. 2012;4:122ra23.
Barroso E, Pita G, Arias JI, et al. The Fanconi anemia family of genes and
its correlation with breast cancer susceptibility and breast cancer
features. Breast Cancer Res Treat. 2009;118:655-660.
Walsh T, Lee MK, Casadei S, et al. Detection of inherited mutations for
breast and ovarian cancer using genomic capture and massively
parallel sequencing. Proc Natl Acad Sci USA. 2010;107:12629-12633.
Meindl A, Hellebrand H, Wiek C, et al. Germline mutations in breast
and ovarian cancer pedigrees establish RAD51C as a human cancer
susceptibility gene. Nat Genet. 2010;42:410-414.
Loveday C, Turnbull C, Ramsay E, et al; Breast Cancer Susceptibility
Collaboration (UK). Germline mutations in RAD51D confer susceptibility to ovarian cancer. Nat Genet. 2011;43:879-882.
Thompson ER, Rowley SM, Sawyer S, et al. Analysis of RAD51D in
ovarian cancer patients and families with a history of ovarian or breast
cancer. PLoS One. 2013;8:e54772.
Baker JL, Schwab RB, Wallace AM, et al. Breast cancer in a RAD51D
mutation carrier: case report and review of the literature. Clin Breast
Cancer. 2015;15:e71-e75.
Kuschel B, Auranen A, McBride S, et al. Variants in DNA double-strand
break repair genes and breast cancer susceptibility. Hum Mol Genet.
2002;11:1399-1407.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
Bojesen SE, Pooley KA, Johnatty SE, et al. Multiple independent

variants at the TERT locus are associated with telomere length and
risks of breast and ovarian cancer. Nat Genet. 2013;45:371-384,
384e371-372.
Garcia-Closas M, Couch FJ, Lindstrom S, et al. Genome-wide association studies identify four ER negative-specific breast cancer risk loci.
Nat Genet. 2013;45:392-398, 398e391-392.
Ghoussaini M, Fletcher O, Michailidou K, et al; Netherlands Collaborative Group on Hereditary Breast and Ovarian Cancer (HEBON);
Familial Breast Cancer Study (FBCS); Gene Environment Interaction of
Breast Cancer in Germany (GENICA) Network; kConFab Investigators;
Australian Ovarian Cancer Study Group. Genome-wide association
analysis identifies three new breast cancer susceptibility loci. Nat
Genet. 2012;44:312-318.
Michailidou K, Hall P, Gonzalez-Neira A, et al. Large-scale genotyping
identifies 41 new loci associated with breast cancer risk. Nat Genet.
2013;45:353-361, 361e351-352.
Dite GS, MacInnis RJ, Bickerstaffe A, et al. Breast cancer risk prediction
using clinical models and 77 independent risk-associated SNPs for
women aged under 50 years: Australian Breast Cancer Family Registry.
Cancer Epidemiol Biomarkers Prev. 2015.
Pharoah PD, Antoniou AC, Easton DF, et al. Polygenes, risk prediction,
and targeted prevention of breast cancer. N Engl J Med. 2008;358:
2796-2803.
Antoniou AC, Wang X, Fredericksen ZS, et al; EMBRACE; GEMO Study
Collaborators; HEBON; kConFab; SWE-BRCA; MOD SQUAD; GENICA. A
locus on 19p13 modifies risk of breast cancer in BRCA1 mutation
carriers and is associated with hormone receptor-negative breast
cancer in the general population. Nat Genet. 2010;42:885-892.
Gaudet MM, Kuchenbaecker KB, Vijai J, et al; KConFab Investigators;
Ontario Cancer Genetics Network; HEBON; EMBRACE; GEMO Study
Collaborators; GENICA Network. Identification of a BRCA2-specific
modifier locus at 6p24 related to breast cancer risk. PLoS Genet. 2013;
9:e1003173.
Mulligan AM, Couch FJ, Barrowdale D, et al; Breast Cancer Family
Registry; EMBRACE; GEMO Study Collaborators; HEBON; kConFab
Investigators; Ontario Cancer Genetics Network; SWE-BRCA; CIMBA.
Common breast cancer susceptibility alleles are associated with tumour subtypes in BRCA1 and BRCA2 mutation carriers: results from
the Consortium of Investigators of Modifiers of BRCA1/2. Breast
Cancer Res. 2011;13:R110.
Wacholder S, Hartge P, Prentice R, et al. Performance of common
genetic variants in breast-cancer risk models. N Engl J Med. 2010;362:
986-993.
Mealiffe ME, Stokowski RP, Rhees BK, et al. Assessment of clinical
validity of a breast cancer risk model combining genetic and clinical
information. J Natl Cancer Inst. 2010;102:1618-1627.
Tung N, Battelli C, Allen B, et al. Frequency of mutations in individuals
with breast cancer referred for BRCA1 and BRCA2 testing using nextgeneration sequencing with a 25-gene panel. Cancer. 2015;121:25-33.
Castera L, Krieger S, Rousselin A, et al. Next-generation sequencing for
the diagnosis of hereditary breast and ovarian cancer using genomic
capture targeting multiple candidate genes. Eur J Hum Genet. 2014;
22:1305-1313.
Kurian AW, Hare EE, Mills MA, et al. Clinical evaluation of a multiplegene sequencing panel for hereditary cancer risk assessment. J Clin
Oncol. 2014;32:2001-2009.
LaDuca H, Stuenkel AJ, Dolinsky JS, et al. Utilization of multigene
panels in hereditary cancer predisposition testing: analysis of more
than 2,000 patients. Genet Med. 2014;16:830-837.
Maxwell KN, Wubbenhorst B, DAndrea K, et al. Prevalence of
mutations in a panel of breast cancer susceptibility genes in
53
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
54
BRCA1/2-negative patients with early-onset breast cancer. Genet

Med. 2015;17:630-638.
Selkirk CG, Vogel KJ, Newlin AC, et al. Cancer genetic testing panels for
inherited cancer susceptibility: the clinical experience of a large adult
genetics practice. Fam Cancer. 2014;13:527-536.
Couch FJ, Hart SN, Sharma P, et al. Inherited mutations in 17 breast
cancer susceptibility genes among a large triple-negative breast
cancer cohort unselected for family history of breast cancer. J Clin
Oncol. 2015;33:304-311.
Churpek JE, Walsh T, Zheng Y, et al. Inherited predisposition to breast
cancer among African American women. Breast Cancer Res Treat.
2015;149:31-39.
Cybulski C, Lubinski
J, Wokoorczyk D, et al. Mutations predisposing to
breast cancer in 12 candidate genes in breast cancer patients from
Poland. Clin Genet. 2015;88:366-370.
Thompson ER, Rowley SM, Li N, et al. Panel testing for familial breast
cancer: calibrating the tension between research and clinical care.
J Clin Oncol. Epub 2016 Jan 19.
Desmond A, Kurian AW, Gabree M, et al. Clinical actionability of
multigene panel testing for hereditary breast and ovarian cancer risk
assessment. JAMA Oncol. 2015;1:943-951.
Idos GE, Kurian AW, McDonnell K, et al. Interim analysis of multiplex
gene panel testing for inherited susceptibility to breast cancer. Abstract presented at the San Antonio Breast Cancer Symposium; 2015;
San Antonio, Texas.
Kurian AW, Ford JM. Multigene panel testing in oncology practice:
how should we respond? JAMA Oncol. 2015;1:277-278.
Kurian AW, Ford JM. Multiple-gene panels and the future of genetic
testing. Curr Breast Cancer Rep. 2015;7:98-104.
Lincoln SE, Kobayashi Y, Anderson MJ, et al. A systematic comparison
of traditional and multigene panel testing for hereditary breast and
ovarian genes in more than 1000 patients. J Mol Diagn. 2015;17:
533-544.
Hall MJ, Reid JE, Burbidge LA, et al. BRCA1 and BRCA2 mutations in
women of different ethnicities undergoing testing for hereditary
breast-ovarian cancer. Cancer. 2009;115:2222-2233.
Easton DF, Pharoah PD, Antoniou AC, et al. Gene-panel sequencing
and the prediction of breast-cancer risk. N Engl J Med. 2015;372:
2243-2257.
Offit K, Bradbury A, Storm C, et al. Gene patents and personalized
cancer care: impact of the Myriad case on clinical oncology. J Clin
Oncol. 2013;31:2743-2748.
Evans DG, Barwell J, Eccles DM, et al; FH02 Study Group; RGC teams.
The Angelina Jolie effect: how high celebrity profile can have a major
impact on provision of cancer related services. Breast Cancer Res.
2014;16:442.
Juthe RH, Zaharchuk A, Wang C. Celebrity disclosures and information
seeking: the case of Angelina Jolie. Genet Med. 2015;17:545-553.
Lebo PB, Quehenberger F, Kamolz LP, et al. The Angelina effect
revisited: Exploring a media-related impact on public awareness.
Cancer. 2015;121:3959-3964.
Robson ME, Bradbury AR, Arun B, et al. American Society of Clinical
Oncology Policy Statement Update: Genetic and Genomic Testing for
Cancer Susceptibility. J Clin Oncol. 2015;33:3660-3667.
Yu PP, Vose JM, Hayes DF. Genetic cancer susceptibility testing: increased
technology, increased complexity. J Clin Oncol. 2015;33:3533-3534.
John EM, Miron A, Gong G, et al. Prevalence of pathogenic BRCA1
mutation carriers in 5 US racial/ethnic groups. JAMA. 2007;298:
2869-2876.
Kurian AW. BRCA1 and BRCA2 mutations across race and ethnicity:
distribution and clinical implications. Curr Opin Obstet Gynecol. 2010;
22:72-78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
Southey MC, Park DJ, Nguyen-Dumont T, et al; COMPLEXO.

COMPLEXO: identifying the missing heritability of breast cancer via
next generation collaboration. Breast Cancer Res. 2013;15:402.
Spurdle AB, Healey S, Devereau A, et al; ENIGMA. ENIGMAevidencebased network for the interpretation of germline mutant alleles: an
international initiative to evaluate risk and clinical significance associated with sequence variation in BRCA1 and BRCA2 genes. Hum
Mutat. 2012;33:2-7.
Stadler ZK, Schrader KA, Vijai J, et al. Cancer genomics and inherited
risk. J Clin Oncol. 2014;32:687-698.
Bradbury AR, Patrick-Miller L, Long J, et al. Development of a tiered
and binned genetic counseling model for informed consent in the era
of multiplex testing for cancer susceptibility. Genet Med. 2015;17:
485-492.
Bradbury AR, Patrick-Miller LJ, Egleston BL, et al. Patient feedback and
early outcome data with a novel tiered-binned model for multiplex
breast cancer susceptibility testing. Genet Med. 2016;18:25-33.
Kurian AW, Idos GE, McDonnell K, et al. The patient experience in a
prospective trial of multiplex gene panel testing for cancer risk.
Abstract presented at the San Antonio Breast Cancer Symposium.
2015; San Antonio, Texas.
Collaborative Group on Hormonal Factors in Breast Cancer. Familial
breast cancer: collaborative reanalysis of individual data from 52
epidemiological studies including 58,209 women with breast cancer
and 101,986 women without the disease. Lancet. 2001;358:
1389-1399.
Anglian Breast Cancer Study Group. Prevalence and penetrance of
BRCA1 and BRCA2 mutations in a population-based series of breast
cancer cases. Br J Cancer. 2000;83:1301-1308.
Peto J, Collins N, Barfoot R, et al. Prevalence of BRCA1 and BRCA2 gene
mutations in patients with early-onset breast cancer. J Natl Cancer
Inst. 1999;91:943-949.
Mavaddat N, Pharoah PD, Blows F, et al; SEARCH Team. Familial
relative risks for breast cancer by pathological subtype: a populationbased cohort study. Breast Cancer Res. 2010;12:R10.
Bahcall OG. iCOGS collection provides a collaborative model. Foreword. Nat Genet. 2013;45:343.
Michailidou K, Beesley J, Lindstrom S, et al; BOCS; kConFab Investigators; AOCS Group; NBCS; GENICA Network. Genome-wide
association analysis of more than 120,000 individuals identifies 15
new susceptibility loci for breast cancer. Nat Genet. 2015;47:373-380.
Mavaddat N, Pharoah PD, Michailidou K, et al. Prediction of breast
cancer risk based on profiling with common genetic variants. J Natl
Cancer Inst. 2015;107:djv036.
Garcia-Closas M, Gunsoy NB, Chatterjee N. Combined associations of
genetic and environmental risk factors: implications for prevention of
breast cancer. J Natl Cancer Inst. 2014;106:dju305.
Pashayan N, Guo Q, Pharoah PD. Personalized screening for cancers:
should we consider polygenic profiling? Per Med. 2013;10(6).
Hall AE, Chowdhury S, Hallowell N, et al. Implementing risk-stratified
screening for common cancers: a review of potential ethical, legal and
social issues. J Public Health (Oxf). 2014;36:285-291.
Burton H, Chowdhury S, Dent T, et al. Public health implications from
COGS and potential for risk stratification and screening. Nat Genet.
2013;45:349-351.
Chenevix-Trench G, Milne RL, Antoniou AC, et al; CIMBA. An international initiative to identify genetic modifiers of cancer risk in
BRCA1 and BRCA2 mutation carriers: the Consortium of Investigators
of Modifiers of BRCA1 and BRCA2 (CIMBA). Breast Cancer Res. 2007;9:
104.
Couch FJ, Wang X, McGuffog L, et al; kConFab Investigators; SWEBRCA; Ontario Cancer Genetics Network; HEBON; EMBRACE; GEMO
98.
99.
100.
101.
102.
103.
104.
105.
106.
107.
108.
109.
Study Collaborators; BCFR; CIMBA. Genome-wide association study in

BRCA1 mutation carriers identifies novel loci associated with breast
and ovarian cancer risk. PLoS Genet. 2013;9:e1003212.
Gaudet MM, Kirchhoff T, Green T, et al; GEMO Study Collaborators;
HEBON Study Collaborators; OCGN; kConFab; EMBRACE. Common
genetic variants and modification of penetrance of BRCA2-associated
breast cancer. PLoS Genet. 2010;6:e1001183.
Antoniou AC, Hardy R, Walker L, et al. Predicting the likelihood of
carrying a BRCA1 or BRCA2 mutation: validation of BOADICEA,
BRCAPRO, IBIS, Myriad and the Manchester scoring system using data
from UK genetics clinics. J Med Genet. 2008;45:425-431.
Antoniou AC, Sinilnikova OM, McGuffog L, et al; Kathleen Cuningham
Foundation Consortium for Research into Familial Breast Cancer;
OCGN; HEBON; EMBRACE; GEMO; Breast Cancer Family Registry;
CIMBA. Common variants in LSP1, 2q35 and 8q24 and breast cancer
risk for BRCA1 and BRCA2 mutation carriers. Hum Mol Genet. 2009;18:
4442-4456.
Antoniou AC, Kartsonaki C, Sinilnikova OM, et al; SWE-BRCA; HEBON;
EMBRACE; CEMO Study Collaborators; Breast Cancer Family Registry;
kConFab investigators; CIMBA. Common alleles at 6q25.1 and 1p11.2
are associated with breast cancer risk for BRCA1 and BRCA2 mutation
carriers. Hum Mol Genet. 2011;20:3304-3321.
Antoniou AC, Kuchenbaecker KB, Soucy P, et al; CIMBA, SWE-BRCA;
HEBON; EMBRACE; GEMO Collaborators Study; kConFab Investigators. Common variants at 12p11, 12q24, 9p21, 9q31.2 and in
ZNF365 are associated with breast cancer risk for BRCA1 and/or
BRCA2 mutation carriers. Breast Cancer Res. 2012;14:R33.
Kuchenbaecker KB, Neuhausen SL, Robson M, et al; Breast Cancer
Family Registry; EMBRACE Study; GEMO Study Collaborators; HEBON;
KConFab Investigators; CIMBA. Associations of common breast cancer
susceptibility alleles with risk of breast cancer subtypes in BRCA1 and
BRCA2 mutation carriers. Breast Cancer Res. 2014;16:3416.
Mavaddat N, Barrowdale D, Andrulis IL, et al; HEBON; EMBRACE;
GEMO Study Collaborators; kConFab Investigators; SWE-BRCA Collaborators; Consortium of Investigators of Modifiers of BRCA1/2.
Pathology of breast and ovarian cancers among BRCA1 and BRCA2
mutation carriers: results from the Consortium of Investigators of
Modifiers of BRCA1/2 (CIMBA). Cancer Epidemiol Biomarkers Prev.
2012;21:134-147.
Lakhani SR, Van De Vijver MJ, Jacquemier J, et al. The pathology of
familial breast cancer: predictive value of immunohistochemical
markers estrogen receptor, progesterone receptor, HER-2, and p53 in
patients with mutations in BRCA1 and BRCA2. J Clin Oncol. 2002;20:
2310-2318.
Lakhani SR, Reis-Filho JS, Fulford L, et al; Breast Cancer Linkage
Consortium. Prediction of BRCA1 status in patients with breast cancer
using estrogen receptor and basal phenotype. Clin Cancer Res. 2005;
11:5175-5180.
Antoniou AC, Beesley J, McGuffog L, et al; Ontario Cancer Genetics
Network; SWE-BRCA; HEBON; EMBRACE; GEMO; GEMO; Breast
Cancer Family Registry; kConFab; CIMBA. Common breast cancer
susceptibility alleles and the risk of breast cancer for BRCA1 and
BRCA2 mutation carriers: implications for risk prediction. Cancer Res.
2010;70:9742-9754.
Meijers-Heijboer H, van den Ouweland A, Klijn J, et al; CHEK2-Breast
Cancer Consortium. Low-penetrance susceptibility to breast cancer
due to CHEK2(*)1100delC in noncarriers of BRCA1 or BRCA2 mutations. Nat Genet. 2002;31:55-59.
CHEK2 Breast Cancer Case-Control Consortium. CHEK2*1100delC and
susceptibility to breast cancer: a collaborative analysis involving
10,860 breast cancer cases and 9,065 controls from 10 studies. Am J
Hum Genet. 2004;74:1175-1182.
110. Fletcher O, Johnson N, Dos Santos Silva I, et al. Family history, genetic
testing, and clinical risk prediction: pooled analysis of CHEK2 1100delC
in 1,828 bilateral breast cancers and 7,030 controls. Cancer Epidemiol
111. Turnbull C, Seal S, Renwick A, et al; Breast Cancer Susceptibility
Collaboration (UK), EMBRACE. Gene-gene interactions in breast
cancer susceptibility. Hum Mol Genet. 2012;21:958-962.
112. Lee A, Cunningham AP, Tischkowitz M, et al. Incorporating Truncating
Variants in PALB2, CHEK2 and ATM into the BOADICEA Breast Cancer
Risk Model. Genet Med. 2016. In press.
113. Nickels S, Truong T, Hein R, et al; Genica Network; kConFab; AOCS
Management Group. Evidence of gene-environment interactions
between common breast cancer susceptibility loci and established
environmental risk factors. PLoS Genet. 2013;9:e1003284.
114. Campa D, Kaaks R, Le Marchand L, et al. Interactions between genetic
variants and breast cancer risk factors in the breast and prostate
cancer cohort consortium. J Natl Cancer Inst. 2011;103:1252-1263.
115. Rudolph A, Fasching PA, Behrens S, et al. A comprehensive evaluation
of interaction between genetic variants and use of menopausal
hormone therapy on mammographic density. Breast Cancer Res.
2015;17:110.
116. Rudolph A, Milne RL, Truong T, et al; kConFab Investigators; AOCS
Group; GENICA-Network. Investigation of gene-environment interactions between 47 newly identified breast cancer susceptibility
loci and environmental risk factors. Int J Cancer. 2015;136:E685-E696.
117. Friebel TM, Domchek SM, Rebbeck TR. Modifiers of cancer risk in
BRCA1 and BRCA2 mutation carriers: systematic review and metaanalysis. J Natl Cancer Inst. 2014;106:dju091.
118. Sieh W, Rothstein JH, McGuire V, et al. The role of genome sequencing
in personalized breast cancer prevention. Cancer Epidemiol Biomarkers Prev. 2014;23:2322-2327.
119. Brentnall AR, Evans DG, Cuzick J. Distribution of breast cancer risk
from SNPs and classical risk factors in women of routine screening age
in the UK. Br J Cancer. 2014;110:827-828.
120. Gail MH, Brinton LA, Byar DP, et al. Projecting individualized probabilities of developing breast cancer for white females who are being
examined annually. J Natl Cancer Inst. 1989;81:1879-1886.
121. Antoniou AC, Easton DF. Risk prediction models for familial breast
cancer. Future Oncol. 2006;2:257-274.
122. Claus EB, Risch N, Thompson WD. The calculation of breast cancer risk
for women with a first degree family history of ovarian cancer. Breast
123. Claus EB, Risch N, Thompson WD. Autosomal dominant inheritance of
early-onset breast cancer. Implications for risk prediction. Cancer.
1994;73:643-651.
124. Parmigiani G, Berry D, Aguilar O. Determining carrier probabilities for
breast cancer-susceptibility genes BRCA1 and BRCA2. Am J Hum
Genet. 1998;62:145-158.
125. Tyrer J, Duffy SW, Cuzick J. A breast cancer prediction model incorporating familial and personal risk factors. Stat Med. 2004;23:
1111-1130.
126. Antoniou AC, Cunningham AP, Peto J, et al. The BOADICEA model of
genetic susceptibility to breast and ovarian cancers: updates and
extensions. Br J Cancer. 2008;98:1457-1466.
127. Lee AJ, Cunningham AP, Kuchenbaecker KB, et al; Consortium of
Investigators of Modifiers of BRCA1/2; Breast Cancer Association
Consortium. BOADICEA breast cancer risk prediction model: updates
to cancer incidences, tumour pathology and web interface. Br J
Cancer. 2014;110:535-545.
128. Gail MH. Discriminatory accuracy from single-nucleotide polymorphisms in models to predict breast cancer risk. J Natl Cancer Inst.
2008;100:1037-1041.
55
129. Darabi H, Czene K, Zhao W, et al. Breast cancer risk prediction and
individualised screening based on common genetic variation and
breast density measurement. Breast Cancer Res. 2012;14:R25.
130. Kurian AW, Gong GD, John EM, et al. Breast cancer risk for noncarriers
of family-specific BRCA1 and BRCA2 mutations: findings from the
Breast Cancer Family Registry. J Clin Oncol. 2011;29:4505-4509.
131. Easton DF, Deffenbaugh AM, Pruss D, et al. A systematic genetic
assessment of 1,433 sequence variants of unknown clinical significance in the BRCA1 and BRCA2 breast cancer-predisposition genes.
Am J Hum Genet. 2007;81:873-883.
132. Lindor NM, Guidugli L, Wang X, et al. A review of a multifactorial
probability-based model for classification of BRCA1 and BRCA2 variants of uncertain significance (VUS). Hum Mutat. 2012;33:8-21.
133. Rosner B, Glynn RJ, Tamimi RM, et al. Breast cancer risk prediction
with heterogeneous risk profiles according to breast cancer tumor
markers. Am J Epidemiol. 2013;178:296-308.
134. Yang XR, Chang-Claude J, Goode EL, et al. Associations of breast cancer risk
factors with tumor subtypes: a pooled analysis from the Breast Cancer
Association Consortium studies. J Natl Cancer Inst. 2011;103:250-263.
135. Cuzick J, Sestak I, Bonanni B, et al; SERM Chemoprevention of Breast
Cancer Overview Group. Selective oestrogen receptor modulators in
prevention of breast cancer: an updated meta-analysis of individual
participant data. Lancet. 2013;381:1827-1834.
136. Cuzick J, Sestak I, Cawthorn S, et al; IBIS-I Investigators. Tamoxifen for
prevention of breast cancer: extended long-term follow-up of the IBISI breast cancer prevention trial. Lancet Oncol. 2015;16:67-75.
137. Oeffinger KC, Fontham ET, Etzioni R, et al; American Cancer Society.
Breast cancer screening for women at average risk: 2015 guideline
update from the american cancer society. JAMA. 2015;314:1599-1614.
138. Nelson HD, Tyne K, Naik A, et al. Screening for breast cancer: an
update for the U.S. Preventive Services Task Force. Ann Intern Med.
2009;151:727-737, W237-742.
139. Myers ER, Moorman P, Gierisch JM, et al. Benefits and harms of breast
cancer screening: A systematic review. JAMA. 2015;314:1615-1634.
140. Warner E, Messersmith H, Causer P, et al. Systematic review: using
magnetic resonance imaging to screen women at high risk for breast
cancer. Ann Intern Med. 2008;148:671-679.
141. Warner E, Hill K, Causer P, et al. Prospective study of breast cancer
incidence in women with a BRCA1 or BRCA2 mutation under surveillance with and without magnetic resonance imaging. J Clin Oncol.
2011;29:1664-1669.
142. Evans DG, Kesavan N, Lim Y, et al; MARIBS Group. MRI breast
screening in high-risk women: cancer detection and survival analysis.
143. Heijnsdijk EA, Warner E, Gilbert FJ, et al. Differences in natural history
between breast cancers in BRCA1 and BRCA2 mutation carriers and
effects of MRI screening-MRISC, MARIBS, and Canadian studies
combined. Cancer Epidemiol Biomarkers Prev. 2012;21:1458-1468.
144. Saadatmand S, Obdeijn I-M, Rutgers EJ, et al. Survival benefit in
women with BRCA1 mutation or familial risk in the MRI screening
study (MRISC). Int J Cancer. 2015;137:1729-1738.
145. Kurian AW, Sigal BM, Plevritis SK. Survival analysis of cancer risk reduction
strategies for BRCA1/2 mutation carriers. J Clin Oncol. 2010;28:222-231.
56
146. Plevritis SK, Kurian AW, Sigal BM, et al. Cost-effectiveness of screening
BRCA1/2 mutation carriers with breast magnetic resonance imaging.
JAMA. 2006;295:2374-2384.
147. National Collaborating Centre for Cancer. Familial breast cancer:
Classification and care of people at risk of familial breast cancer and
management of breast cancer and related risks in people with a family
history of breast cancer. Cardiff (U.K.): National Collaborating Centre
for Cancer, United Kingdom; 2013.
148. Netherlands Breast Cancer Screening Guidelines, version 2.0. 2012;
Integraal Kankercentrum Nederland.
149. Thomssen C, Harbeck N, on behalf of the AGO Breast Committee.
Update 2010 of the German AGO Recommendations for the Diagnosis
and Treatment of Early and Metastatic Breast Cancer - Chapter B:
Prevention, Early Detection, Lifestyle, Premalignant Lesions, DCIS,
Recurrent and Metastatic Breast Cancer. Breast Care (Basel). 2010;5:
345-351.
150. Quante AS, Whittemore AS, Shriver T, et al. Practical problems with
clinical guidelines for breast cancer prevention based on remaining
lifetime risk. J Natl Cancer Inst. 2015;107:djv124.
151. Han FF, Guo CL, Liu LH. The effect of CHEK2 variant I157T on cancer
susceptibility: evidence from a meta-analysis. DNA Cell Biol. 2013;32:
329-335.
152. Shaag A, Walsh T, Renbaum P, et al. Functional and genomic approaches reveal an ancient CHEK2 allele associated with breast cancer
in the Ashkenazi Jewish population. Hum Mol Genet. 2005;14:
555-563.
153. Mandelblatt JS, Stout NK, Schechter CB, et al. Collaborative modeling
of the benefits and harms associated with different U.S. breast cancer
screening strategies. Ann Oncol. 2016;164:215-225.
154. Fisher B, Costantino JP, Wickerham DL, et al. Tamoxifen for prevention
of breast cancer: report of the National Surgical Adjuvant Breast and
Bowel Project P-1 Study. J Natl Cancer Inst. 1998;90:1371-1388.
155. Vogel VG, Costantino JP, Wickerham DL, et al; National Surgical
Adjuvant Breast and Bowel Project. Update of the National Surgical
Adjuvant Breast and Bowel Project Study of Tamoxifen and Raloxifene
(STAR) P-2 Trial: Preventing breast cancer. Cancer Prev Res (Phila).
2010;3:696-706.
156. Goss PE, Ingle JN, Ale s-Martnez JE, et al; NCIC CTG MAP.3 Study
Investigators. Exemestane for breast-cancer prevention in postmenopausal women. N Engl J Med. 2011;364:2381-2391.
157. Jacobs IJ, Menon U, Ryan A, et al. Ovarian cancer screening and
mortality in the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS): a randomised controlled trial. Lancet. 2016;387:
945-956.
158. Jervis S, Song H, Lee A, et al. Ovarian cancer familial relative risks by
tumour subtypes and by known ovarian cancer genetic susceptibility
variants. J Med Genet. 2014;51:108-113.
159. Grant RC, Selander I, Connor AA, et al. Prevalence of germline mutations in cancer predisposition genes in patients with pancreatic
cancer. Gastroenterology. 2015;148:556-564.
160. Zhen DB, Rabe KG, Gallinger S, et al. BRCA1, BRCA2, PALB2, and
CDKN2A mutations in familial pancreatic cancer: a PACGENE study.
Genet Med. 2015;17:569-577.
Access to Cancer Therapeutics

in Low- and Middle-Income
Countries
CHAIR
Paul Ruff, MBBCh, MMed, FCP(SA)
University of Witwatersrand Faculty of Health Sciences
Johannesburg, South Africa
SPEAKERS
Lawrence N. Shulman, MD
Abramson Cancer Center of the University of Pennsylvania
Philadelphia, PA
Sana Al-Sukhun, MD, MSc
The University of Jordan
Amman, Jordan
RUFF ET AL
Access to Cancer Therapeutics in Low- and Middle-Income

Countries
Paul Ruff, MBBCh, MMed, FCP(SA), Sana Al-Sukhun, MD, MSc, Charmaine Blanchard, MPhil, MBBCh, BSc,
and Lawrence N. Shulman, MD
OVERVIEW
Cancer is rapidly becoming a major health care problem, especially in developing countries, where 60% of the worlds total
new cases are diagnosed. The success of new antineoplastic medicines and modern radiation devices to cure a good
proportion of patients with cancer and to alleviate the suffering of many more has been achieved at a dramatic cost.
Therefore, it has become mandatory for health care authorities and pharmaceutical companies to cooperate to use and
develop resources in an efficient manner to improve health care delivery to patients with cancer worldwide. Regulatory
harmonization is an important key to overcome delays in the approval process, whether for antineoplastic and pain control
medicines or for essential medical devices. More emphasis on the significant role of opiates in pain control among patients
with cancer is needed to overcome the ingrained belief in their potential for addiction. The World Health Organization (WHO)
serves an important role in guiding priorities for health care and efficiently allocating resources by providing essential
medicine lists (EMLs) and device lists. However, the financial challenge for access to health care is multi-tiered and requires
collaboration between key stakeholders including pharmaceutical industry, local national health authorities, WHO, and
other nonprofit, patient-oriented organizations.
ince the turn of the century, we have seen a paradigm

shift in the way we treat cancer, with the advent of
targeted therapies, especially monoclonal antibodies and
small-molecule kinase inhibitors. Indeed the development of
imatinib in chronic myeloid leukemia (CML) and rituximab in
B-cell lymphomas have been considered among the greatest
breakthroughs in cancer care in the past 50 years. Unfortunately, these and many other therapies that have significantly improved outcomes in patients with cancer are not
available to everyone who needs them, especially in lowand middle-income countries (LMICs). The cost of new
anticancer medicines is increasing, with the average monthly
cost of some newly released molecules being over $12,000
per month for oral kinase inhibitors and $150,000 for
a course of monoclonal antibodies. These costs have long
been out of the range of most LMICs and are now becoming
excessive even for European and North American patients.
The treatment of cancer has always been costly and difficult to access, be it surgery, radiation therapy, systemic
medicines, or palliative care. Radiation machines are costly
and limited in access in LMICs, whereas chemotherapy drugs
have always been difficult to procure since they first became
available in the 1950s. Surgical care is limited by lack of skills

and facilities, whereas access to palliative medicines, especially for pain control, is limited by regulatory and legal
restrictions, costs, and storage, as well as cultural attitudes
to end-of-life issues. Radiation machines should be procured
and maintained in LMICs at reasonable prices without
sacrificing safety and efficacy, whereas surgical skills should
be enhanced by improving local facilities and training cancer
surgeons on site or at international centers.
There are no easy solutions to drug costs because pharmaceutical companies should make profits to be able to
develop new medicines, knowing that not all medicines
undergoing study come to market and that antineoplastic
agents, especially biologic medicines, are difficult and costly
to produce. The current model of oncology drug development, distribution, and marketing must change before it is
too late. There must be collaboration between academia and
individual pharmaceutical companies, as well as with regulatory authorities and governments, to avoid the situation
being experienced in renal cancer, where we now have more
than 10 agents on the market from multiple pharmaceutical
companies. Collaboration with regulatory authorities can
From the University of Jordan, Amman, Jordan; University of Witwatersrand Faculty of Health Sciences, Johannesburg, South Africa; Abramson Cancer Center, University of
Pennsylvania, Philadelphia, PA.
Corresponding author: Paul Ruff, MBBCh, MMed, FCP(SA), University of Witwatersrand Faculty of Health Sciences, 7 York Rd., Johannesburg, 2193, South Africa; email: pruff@
iafrica.com.
58
CULTURAL AND REGULATORY BARRIERS TO TREATING CANCER PAIN
reduce research and development costs by streamlining the

regulatory burden on clinical research and expediting registration of medicines with significant benefits over those
with limited benefits. Development of medicines with
limited benefits should be halted as early as possible.
Evergreening (enhanced intellectual property protection)
must also be prevented; extended patents are detrimental
to underfunded patients in both high-income countries and
LMICs. International collaboration in the development of
good quality generics and biosimilars will help make these
molecules available more rapidly to patients around the
world. Development of EMLs, both in individual countries
and internationally by WHO, will make the use of cancer
medicines more cost effective and less wasteful.
Unless we all work together to find a solution, cancer care
will only be available to the very wealthy. Indeed, we are
seeing the development of more and more exciting treatments for fewer and fewer people.1
DRUG ACCESS AND APPROVAL IN LOW- AND

MIDDLE-INCOME COUNTRIES
Cancer is rapidly becoming a major health care problem,
especially in LMICs. The success of antineoplastic medicines
to cure patients and alleviate their suffering has been achieved at
a dramatic cost. Therefore, it has become mandatory for health
care authorities to cooperate to use resources efficiently to
improve health care delivery to patients with cancer worldwide.
The initial step in the process of access to antineoplastic
medicines is a pharmaceutical companys application for
approval by the responsible local health regulatory authorities. The expected revenues of a medicine depend on
the volume of drug sold and the average income of targeted
KEY POINTS
The majority of patients with cancer reside in LMICs with

limited access to essential medicines and services,
resulting in a disproportionate increase in cancer
mortality.
Obstacles to cancer care are numerous and include
costs, regulatory and cultural barriers, and limited
availability of health care practitioners.
The financial challenge for access to care is multi-tiered
and requires collaboration between all stakeholders
including the pharmaceutical industry, national and
local health authorities, nongovernmental
organizations, and the World Health Organization.
Regulatory, financial, and cultural barriers limit access to
chemotherapy agents, supportive care medicines,
radiation therapy, and complementary and alternative
medicines required for good cancer pain management in
LMICs.
Collaborative partnerships and regulatory
harmonization can overcome delays in treatment access
and approval and facilitate the development of quality
cancer care delivery.
populations.2 For cancer, expected revenues in LMICs may

not justify investment because the target populations
average income is low. This lack of incentive compelling
regulation to register medicines results in excessive delays
in access to newer medicines.
Once an application for a new medicine is filed, the
complex regulatory approval process starts to ensure
availability of high-quality, safe, and effective medicines.3
Although some countries can assess a new drug themselves,
based on the scientific dossier provided by the manufacturer
(usually high-income countries), middle-income countries
with varying levels of development and drug regulatory
capabilities, or low-income countries with very limited or no
drug regulatory capability cannot undertake full assessments of new pharmaceutical products. LMICs depend to
differing extents on assessment made by a foreign drug
regulatory authority, for example the U.S. Food and Drug
Administration (FDA) or European Medicines Agency (EMA).
Drug approval by one of these major authorities provides
a valid basis for marketing a product in these LMICs. In an
attempt to facilitate drug registration in LMICs, WHO has
developed a certification scheme to provide quality assurance for imported medicines by means of standard forms
confirming the registration of the product in the country of
manufacture (Certificate of Pharmaceutical Product [CPP])
and approval of good manufacturing practice conditions
based on inspections and quality analysis of the product.4,5
However, even after 2 decades of existence, the WHO
certification scheme is not widely used. Regulatory procedures still vary significantly among nations, with certificate
format, conditions, and wording varying from one country to
another.6 Each country requires different documentation,
but almost all require at least one CPP from the manufacturing
site, the packaging and release site, or both. The process for
registering medicines remains complex and burdened with
inefficiencies, duplications, delay, and, in some instances,
corruption.7,8 For instance, the manufacturing process may
take place in several countries, complicating the process of
obtaining a satisfactory CPP for the regulatory authorities in
importing countries. This will potentially delay regulatory approval if the regulatory authority requests different CPPs than
what was already provided. Some products are manufactured and approved in a certain country for export only,
leading to concern that the competent authorities in the
manufacturing country are less stringent with regard to
products not to be used in their own territory. It is logical for
some importing countrys regulatory authority to request
more reassuring documentation (e.g., formal declaration
signed by the manufacturers responsible good manufacturing
practice person in addition to the Export CPP), to ensure
follow-up on complaints from the importing countrys authority regarding product defects or inaccuracy of the declarations contents. Still, the regulatory authority could refuse
to register that product.9,10 To further illustrate the complexity
of implementing the WHO certification scheme, assessment of quality of active pharmaceutical ingredients was
also recommended by the International Conference of
59
RUFF ET AL
Drug Regulatory Authorities.7,11 Such recommendations

added to the confusion and complexity of drug approval
processes in LMICs, especially because many do not yet
have the regulatory capacity required to fully implement
the WHO certification scheme, let alone comply with
those recommendations.12
After a medicines approval by EMA or FDA, it takes a year
or more to register that medicine in LMICs. Some countries
adopt fast-track for the registration of priority medicines
fulfilling an unmet medical need.13 Even in North America, it
was estimated that more than 250,000 life-years are lost per
year of delay in access to drugs that were shown to prolong
overall survival in phase III clinical trials.14 Even more lifeyears are lost if we consider the possible estimate in LMICs
where the delay is more of a challenge and the disease
burden is escalating. Regulatory harmonization (i.e., either
allowing for one centralized approval for drug registration in
multiple countries, or mutual recognition once a drug has
been registered in certain countries) is the key to using
resources more efficiently and especially to speeding up the
process to facilitate access to patients.7
Once a medicine is registered, access varies from one
country to another. Rarely, registration entails availability
for patients treated in public health care systems. In some
countries, even if the medicine were to be available, it takes
an additional 1 to 2 years to obtain.13 In most countries,
medicines will only be available for those who can pay out of
pocket, creating a financial access obstacle, probably the
most challenging in the access process. In many LMICs, as
much as 90% of the population purchase medicines on an
out-of-pocket basis,15,16 with spending often disproportionate to personal or family income. The recently updated
WHO EML for cancer could potentially serve as a valuable
asset for advocacy to influence the local health authorities in
LMICs to provide those medicines free of charge or make
them available for patients at affordable cost.17
The financial access challenge creates another serious
concern of drug quality. Many reports have described patients receiving substandard or counterfeit drugs.18,19 This is
mainly the result of importing cheaper, poorly manufactured
rogue generics. Although some governments have begun
to promote the development of generic drug manufacturing
capacity within their own bordersas in India, Brazil, South
Africa, and, recently, Jordanmany countries rely solely on
importing drugs. In its effort to combat this challenge faced
in every field of medicine, WHO developed a prequalification
program for pharmaceuticals and active pharmaceutical
ingredients,20 initially developed to help international
procurement agencies, but lately used by LMICs to guide
bulk purchase of medicines. Drugs passing quality control
appear on the WHO prequalification website; however,
antineoplastic medicines are not there yet. Conversely,
delayed market entry of generics because of evergreening in
countries that joined the World Trade Organization further
increases cost.21 Such countries should consider increasing
spending on public health to offset the adverse impact on
patients of strengthening its intellectual property protection
60
relevant to medicines, in addition to long-term plans to control

the price increase as a result of that commitment. Compulsory
licensing allowing generic versions of medications to be
produced despite the existence of a patent to provide medicines of need in a society at affordable prices, is a plausible
policy already adopted in India, South Africa, and Thailand.22
Direct price control measures could help reduce the price
by an average of 20%.23 Control measures could be price
setting included in registration and procurement through
a central government agency. Because most of antineoplastic medicines are not procured centrally, they are only
available in the private sector for out-of-pocket purchase at
much higher prices.7 If price is adjusted for affordability,
LMICs pay a much higher price compared with high-income
countries.24 The prices set for the recently developed biologics are prohibitive not only for out-of-pocket purchase
but also for central procurement, creating the case for
differential international pricing. Concerns related to this
approach (e.g., parallel trading and use in reference pricing)
do seem legitimate, but the experience with HIV treatment
does not support those concerns.22 However, this differential pricing is beneficial if the different pricing levels indeed reflect the ability of the target population to pay. This is
amplified by The World Banks country income classification; although it was designed for World Bank lending
decisions, its improper adoption to inform health-related
decisions increases the cost of medicines. It does not reflect
health system capacity of governments to invest more in
health, and it does not take into account income inequality
within a nation. Many countries have graduated to become
upper-middle income resulting from statistical recalculations, although 15% to 55% live at or below national poverty
lines.25 This graduation will further complicate drug pricing
policies and availability of expanded drug access programs
for patients. Regulatory harmonization is the key to overcome delays in the approval process and efficiently use
resources. The financial challenge for access is multitiered
and requires collaboration between key stakeholders
including pharmaceutical industry, local national health
authorities, WHO, and other nonprofit, patient-oriented
organizations.
CULTURAL AND REGULATORY BARRIERS TO

PAIN MANAGEMENT INCLUDING ANTICANCER
DRUGS, RADIOTHERAPY, AND SUPPORTIVE
MEDICINES IN LOW- AND MIDDLE-INCOME
COUNTRIES
More than half of all patients with cancer and more than
three-quarters of patients with advanced disease will suffer
from pain at some time in the course of the illness. The pain is
usually moderate to severe, affecting the patients quality of
life, and it can become chronic in long-term cancer survivors.26
The management of cancer pain is a multidisciplinary
process that has many barriers confronting patients and
their families, as well as health care professionals. There are
many difficult aspects to the management of cancer pain in
both high-income countries and LMICs developing. Despite

the high prevalence of cancer pain, it is often poorly treated
for several reasons. To manage cancer pain effectively, it is
important to assess the type of pain and its effect on quality
of life and to decide on a management plan using appropriate
interventions, analgesics, and supportive care as required.
Interventions can include disease-modifying therapies such as
chemotherapy, radiotherapy, or surgery. Analgesics include
nonopioid and opioid drugs, as well as co-analgesic (adjuvant)
drugs for pain that is either not responsive or partially responsive to opioids. Interventions and analgesics have side
effects, which should be treated with further medications.
Many societies and cultures use traditional or complementary and alternative medications and nondrug measures to control pain.
SIDEBAR. Medicines Used for Management of

Cancer Pain
Disease-Modifying Agents
Cancer chemotherapy plays a critical role in reducing pain in

patients with cancer by shrinking and, in certain malignancies, curing the cancer. As mentioned before, access to
cancer chemotherapy has always been limited by costs and
by slow regulatory approval in many LMICs. In South Africa,
approval of a new chemical entity or generic medicine may
take as many as 3 years unless a fast-track status is obtained,
owing to an unmet medical need or public health requirement, which will reduce the time to about 18 months.
Dossiers for registration have to go through three to four
committees, including the Naming and Scheduling Committee, the Pharmaceutical and Analytic Committee, the Central
Clinical Committee, and, in some cases, the Biologic Committee, all of which are understaffed. Many doctors in LMICs
are reluctant to refer patients for chemotherapy because of
myths concerning the side effects, and patients themselves
have similar views often perpetuated by the media.
Opioid analgesics are the mainstay of cancer pain control but are frequently inadequately used because of
doctors and patients fears concerning addiction and abuse.
Longer-acting opioids including slow-release morphine and
fentanyl patches are limited by slow regulatory approval
of generics and high costs of the originator. Opiophobia
is a significant barrier to pain control with opioid analgesics in LMICs. 27 Patient-related factors include fear of
psychological dependence and stigma related to opioid
use, especially where opioids are associated with criminal activity and gang violence.28,29 Health professional
related opiophobia, including beliefs that opioids cause
addiction, tolerance, or difficult-to-control side effects, is
also a significant barrier to adequate pain treatment with
opioids.30
Tricyclics, anticonvulsants, and particularly a2d ligands
gabapentin and pregabalin play an important role in the
treatment of neuropathic pain but are often not available in
LMICs because of slow regulation of generics and high costs
of originators.
Osteoclast inhibitors including bisphosphonates and RANKligand inhibitors play an important role in the management
Paracetamol
Nonsteroidal anti-inflammatory drugs
Weak opioids: codeine, tramadol.
Strong opioids: morphine, fentanyl
Local anesthesia
Co-analgesics (Adjuvants)
ORTHODOX MEDICINES
Chemotherapeutic agents
Analgesics
Tricyclic antidepressants
Anticonvulsants
a2d ligands: gabapentin and pregabalin
Baclofen
Benzodiazepines
Other Essential Medications
Laxatives
Antiemetics
Proton-pump inhibitors
Corticosteroids
Bisphosphonates and RANK-ligand inhibitors
of bone pain. Unfortunately, their availability is also limited

by slow regulatory approvals and high costs (Sidebar).
RADIATION THERAPY FACILITIES: LINEAR

ACCELERATOR VERSUS COBALT 60
Linear accelerator access is limited by high up-front and
maintenance costs, with less effective but low maintenance
and cheaper cobalt 60 being supported in some LMICs by the
International Atomic Energy Agency. In addition, cultural
beliefs and use of traditional medicines likely affect the acceptability of radiation therapy for patients in LMICs. A study
in San Francisco found that Chinese and other Asian women
received less radiation or other adjuvant treatment of breast
cancer than Japanese or white women, most likely because of
cultural beliefs about disease and death, body image, medical decision making, and use of alternative medicines.31
TRADITIONAL OR COMPLEMENTARY AND

ALTERNATIVE MEDICINES
The use of traditional or complementary and alternative medicines (TCAMs) varies across the world from 7% to 64%.32 In LMICs,
TCAMs are a natural element of traditional health practice.
TCAMs are categorized by the National Centre for Complementary and Alternative Medicine in the United States (Table 1).33
There is limited evidence for the effectiveness of TCAMS in
treating cancer pain. Differing responses to multidisciplinary
pain programs have been noted where the mood symptoms
improved in all patient groups but improvement in pain
was culturally determined.34 The effectiveness of TCAMs in
managing cancer pain is dependent on cultural beliefs and
expectations. However, reliance on TCAMs and resistance to
61
RUFF ET AL
TABLE 1. Categories of Traditional or Complementary

and Alternative Medicines
Category
Example
Alternative medical systems
Traditional Chinese, Ayurvedic,

and African traditional
medicine
Mind-body (spirit)
interventions
Meditation, prayer, faith healing,

support groups, music therapy,
and hypnosis
Biologically based therapies
Herbal medicines, dietary

supplements, vitamins
Manipulation and
body-based therapies
TENS, acupuncture massage,

chiropractic, osteopathy,
and reflexology
Energy therapies
Qi Gong, Reiki, and magnetic field

therapy
conventional interventions and opioids may exacerbate pain

in LMICs.
CANNABINOIDS
Historically, the use of cannabis as medicine dates back to
before the Christian era in Asia and spread to the Middle East
in the 10th century, to Africa in the 15th century, to South
America in the 16th century, and to Europe and the United
States in the 19th century.35 A meta-analysis of several
controlled clinical trials supports the use of cannabinoids
(dronabinol and nabilone) for chemotherapy-induced nausea and vomiting, but not for pain. There is no good evidence
supporting the use of inhaled or oral extracts of cannabis for
any cancer-related side effects.36 Country- and state-specific
barriers to legalizing cannabis for symptom control exist
largely because of lack of adequate controls and concerns
about abuse.
CULTURAL BARRIERS TO CANCER PAIN

MANAGEMENT IN LOW- AND MIDDLE-INCOME
COUNTRIES
Many studies on cultural barriers, reported in English, are
largely associated with pain management in Western
countries with cultural minorities, usually of ethnic groups
originating from LMICs. These minority groups are often
marginalized and from lower socioeconomic levels, which
may confound the results because these groups often do not
have equitable access to health care. In these countries,
language barriers may also play a role in inaccurate pain
assessment. It has been shown that nurses who share the
same language as Arabic patients when assessing pain
usually assigned similar ratings to the patients compared
with non-Arabicspeaking nurses.37 However, with this in
mind, these studies may provide a proxy for studying cultural
aspects of cancer pain management in LMICs.
Two reviews of ethnic and cultural differences in pain and
pain management by Kwok and Bhuvanakrishna38 and by
Campbell and Edwards34 reveal the following potential
62
cultural barriers to pain management: (1) differences in

perception of pain, health beliefs, and the meaning of pain
with normalizing of the cancer experience; (2) reluctance to
report pain because of negative stigma associated with
cancer progression and fatalism, and a belief in accepting
pain with stoicism; and (3) different coping practices with
belief in traditional remedies, faith healing, prayer, or
positive thinking.
There appear to be differences in the experience of
physical pain, in the reporting of pain, and in the emotional
response to pain, which may reflect differences in the
meaning of pain in different cultures. There are cultural
taboos and fears that limit access to pain-relieving medications, but there are also traditional practices that are used
to manage pain. Although it is important to ensure that pain
medications and modalities to treat pain are accessible to all,
it is important to consider different cultural attitudes toward
illness and pain and its management.
WORLD HEALTH ORGANIZATION LIST OF

ESSENTIAL MEDICINES AND DEVICES:
MAXIMIZING CANCER CARE VALUE AND
SUPPORTING THE DEVELOPMENT OF CANCER
DELIVERY PROGRAMS
Cancer is a group of very disparate malignant diseases, with
the diagnostics and treatment varying widely among them.
In addition, many patients require sophisticated surgery,
radiation, and systemic therapies. High-quality pathology is
required to make accurate diagnoses in almost all patients,
and imaging is frequently necessary for cancer staging and
assessment of treatment response and follow-up. There are
many essential cancer medications with different cancers
requiring different noninterchangeable agents. Many
countries and ministries of health face daunting obstacles in
establishing functional and quality services in all of these
areas. Toward this end, the WHO has several processes
aimed at aiding countries in the establishment and enhancement of cancer services.
The Essential Medicines List

In 1977 WHO published its first EML addressing medicines
felt to be critical to treating many diseases, including cancer.
In the intervening years, additional medicines were evaluated and added to the EML. In the case of cancer medicines,
agents were added one at a time, without clear instruction
about where the benefit for the medication existed. In 2012,
the Union International for Cancer Control and the DanaFarber Cancer Institute filed applications with WHO to add
trastuzumab and imatinib to the EML for cancer. At that
time, there were no targeted therapies or on-patent drugs
on the EML for cancer. The applications argued that these
two medicines dramatically alter the outcomes and survival
rates for patients with HER2-positive breast cancer and CML,
respectively, and that there were no less-costly alternatives
having similar benefits. WHO deferred judgment and asked
Union International for Cancer Control and Dana-Farber
Cancer Institute to lead a comprehensive review of the

cancer EML.
In 2014, a process was undertaken to identify cancers
where systemic therapies had significant impact. Cancers
were chosen if they were of high burden, if systemic therapies had at least some benefit (such as nonsmall cell lung
cancer) and if they had lower disease burden (such as CML),
and if systemic therapies (e.g., imatinib) had a major impact
on survival. In all, 27 diseases were evaluated, with separate
applications for early-stage and metastatic disease for breast
cancer and colorectal cancer. Nearly 100 oncologists from all
continents were recruited to help in the process. For each
disease, a document was created that included: (1) an executive summary; (2) public health relevance; (3) requirements
for diagnosis, treatment, and monitoring; (4) overview of
regimens; (5) review of benefits and harms (including references and systematic reviews); and (6) recommendations for
additions to the EML.
Each disease-based document was initially written by
a cancer expert or team of experts and reviewed and critiqued by at least two other experts or teams. A central
committee then collated the work of all three groups to form
a consensus-based document. The central committee added
the section on public health relevance and references focusing on important phase III studies and systematic reviews, and in some cases costing information. Importantly,
each document contained a section on diagnostics and
specific needs for treatment and monitoring of patients.
The documents were designed to provide critical information to governments and ministries to understand
better what was needed to treat the particular disease
and make administration of the listed medications safe
and effective.
The approach in each document was regimen-based,
rather than based on individual medications. For some
diseases, such as CML, the regimen was a single drug; in this
case imatinib, but for many diseases the regimen consisted
of multiple drugs such as doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) for Hodgkin lymphoma. In
addition, the incremental benefit of the medicines was
quantitated. As an example for diffuse large B-cell lymphoma, where surgery is required for biopsy and diagnosis
but does not add at all to remission or cure rates, the
medicines alone accounted for all benefits. In this case, the
administration of cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) would result in a 55% long-term
remission rate, whereas the addition of rituximab to this
regimen (R-CHOP) would provide an incremental benefit of
15%, bringing the long-term remission rate to 70%. In the
case of early-stage breast cancer, where surgery is required
and will result in cure for many patients, the medicines
would have incremental benefit above surgery. In a hypothetical patient with node-negative, estrogen receptor
positive, HER2-negative breast cancer, surgery alone might
result in a long-term remission rate of 70%. The addition of
hormone therapy such as tamoxifen might give 15% incremental benefit to the long-term remission rate, bringing
it to 85%, and the addition of adjuvant chemotherapy might

add 5% more to the long-term remission rate, bringing it
to 90%.
The disease-based documents contained specific information on dosing of each medication as well as schedule and
duration of therapy. With this information, ministries of
health could estimate annual drug needs by knowing the
number of patients with each disease and the dose of each
drug needed to treat a patient.
Drafts of the 29 disease-based documents were reviewed by
an expert committee at an in-person meeting at WHO in
November 2014, and finalized documents were submitted to
WHO in December 2014. They were posted on the WHO
website for public comment in January 2015. In April 2015,
a WHO Expert Committee reviewed all documents, and in May
2015 made public their decisions. The documents had provided support for the 30 existing medications on the EML and
recommended the addition of 22 medicines. WHO approved
the 30 existing medications and added 16 new medicines to
the list, whereas six were rejected. Approved medications
included, for the first time, patented, costly agents (i.e.,
trastuzumab and imatinib). The rejected medicines included
two second-generation tyrosine kinase inhibitors for CML
(insufficient data to support them as more essential than
imatinib), two EGFR antagonists for the treatment of non
small cell lung cancer (molecular testing frequently unavailable with modest incremental benefit), arsenic trioxide for the
treatment of acute promyelocytic leukemia (insufficient data
to support it as an essential medication), and diethylstilbestrol
for the treatment of metastatic prostate cancer (mostly unavailable and toxic).39
The final documents and the revised EML for cancer are
now available on the WHO website. Each medicine on the
EML can be referenced back to a disease-based document
supporting and delineating its contribution to treatment of
the disease, something not previously available on the EML.
Planning is underway to develop a mechanism for periodic
reviews of the EML for cancer as new scientific data become
available.
ESSENTIAL MEDICAL DEVICES FOR CANCER

In April 2015, WHO convened a group of experts with the
goal of detailing needs for services critical to the diagnosis,
evaluation, and treatment of patients with cancer. Based on
the advice of this group, a steering committee of experts was
formed and met at WHO in September 2015 to better outline
a plan to accomplish the work. Committees were formed to
delineate needs for pathology and laboratory services, radiology services, surgical services, administration of systemic
therapies, and radiation therapy. Committee members
were recommended by the steering committee and met
via teleconference throughout 2015.
This work is also disease-based, using a subset of cancers
to complete lists of essential medical devices for each of the
categories listed before. The work is still underway, with
plans to complete the lists in the first half of 2016.
63
RUFF ET AL
THE ROLE OF WHO IN DEVELOPMENT OF

CANCER TREATMENT PROGRAMS
WHO exists to serve its member states in the arena of health
care. In many LMICs, ministries of health struggle with health
care priorities as advances in control of infectious diseases
and other important areas such as maternal-child health,
noncommunicable diseases become increasing health burdens. In acknowledgment of this, the United Nations convened a high-level meeting in September 2011 to address
the needs of noncommunicable diseases. The work of WHO
flows from the recommendations of this meeting and followup meetings.
THE ROLE OF COUNTRIES AND MINISTRIES

WHO is providing guidance to countries and ministries on
essential medicines and devices for the diagnosis and
treatment of cancer. It remains, though, for individual
countries to develop cancer plans specific to their needs and
environment. In some LMICs, this is well underway, and in
others, much remains to be done. Many countries have little
or no in-country cancer care expertisefew, if any, pathologists, medical oncologists, radiation oncologists, and
skilled nurses. Many countries have very limited resources
and must prioritize their needs as best they can. Partnerships
between cancer specialists from high-income countries and
LMICs can help. External funding would help immensely to
move the global cancer agenda forward, funding such as was
raised for the Global Fund to Fight AIDS, Tuberculosis, and

Malaria.
All of this work aims at bringing affordable and high-quality
cancer care to those who currently have little or no access.
Patients who have treatable cancers with a high likelihood
for cure in high-income countries are, in many locations,
dying without treatment or with inadequate and poorquality treatment. We should work together to erase this
social and medical inequity.
CONCLUSION
Improving patient outcomes is not only achieved by the
success of medicines or procedures in large phase III
clinical trials. It is achieved when patients worldwide have
ready access to those successful interventions and the
infrastructure and human capacity to use those interventions safely and effectively. The challenges to such
access are many, with regulatory, financial, and cultural
barriers among the most important barriers to overcome.
Success can only be accomplished by constant collaboration between key stakeholders, including the pharmaceutical industry, local and national health authorities,
the WHO, and other nonprofit, patient-oriented organizations. The oncology community in high-income
countries should have a humanitarian obligation to accompany those in LMICs to achieve these goals to the
betterment of all.
References
1. Ruff P. Achieving treatment for all. ASCO Connection: The Best of ASCO
Connection Commentary 2015; Special Edition 2014-2015: 35.
2. Villa S, Compagni A, Reich MR. Orphan drug legislation: lessons for
neglected tropical diseases. Int J Health Plann Manage. 2009;24:27-42.
3. World Health Organization. Policy Perspectives on Medicine. http://
whqlibdoc.who.int/hq/2003/WHO_EDM_2003.2.pdf. Accessed January 5, 2016.
4. World Health Organization. Model Certificate of a Pharmaceutical
Product. http://www.who.int/medicines/areas/quality_safety/
regulation_legislation/certification/modelcertificate/en/. Accessed
January 5, 2016.
5. World Health Organization. Certification scheme on the quality of
pharmaceutical products moving in international commerce. http://
www.who.int/medicines/areas/quality_safety/regulation_legislation/
certification/en/. Accessed January 5, 2016.
6. World Health Organization. Guidelines on the implementation of the
WHO certification scheme on the quality of pharmaceutical products
moving in international commerce. http://www.who.int/medicines/areas/
quality_safety/regulation_legislation/certification/guidelines/en/index1.
html. Accessed January 5, 2016.
7. World Health Organization. The World Medicines Situation 2011.
http://apps.who.int/medicinedocs/en/m/abstract/Js20054en/.
8. World Health Organization. Measuring transparency in medicines registration, selection and procurement: four country assessment studies. http://www.who.int/medicines/areas/policy/goodgovernance/
Transparency4CountryStudy.pdf. Accessed January 6, 2016.
64
9. Ministry of Public Health. Guidelines on Good Storage and Distribution

Practices of Pharmaceutical Products in Lebanon. http://www.moph.
gov.lb/Drugs/Documents/Guidelines-GSDPLebanon%28English%29.pdf.
10. World Health Organization. Proceedings of the Ninth International
Conference of Drug Regulatory Authorities (ICDRA) - Berlin, Germany
25-April 29, 1999. http://apps.who.int/medicinedocs/en/d/Js4925e/.
11. World Health Organization. 15th International Conference of Drug
Regulatory Authorities (ICDRA) - Tallinn, Estonia, 23-October 26, 2012,
http://apps.who.int/medicinedocs/documents/s20142en/s20142en.
12. Doua JY, Van Geertruyden JP. Registering medicines for low-income
countries: How suitable are the stringent review procedures of the World
Health Organisation, the US Food and Drug Administration and the
European Medicines Agency? Tropic Med Int Health. 2014;19:23-36.
13. Registration Requirements of the Pharmaceutical Products with Jordan
FDA. http://www.nobles.com.jo/PharmaceuticalProducts/FormNo21Product/reg_req_of_the_pharmacutical_products_2-1.pdf. Accessed
January 8, 2016.
14. Stewart DJ, Stewart AA, Wheatley-Price P, et al. Impact of time to drug
approval on potential years of life lost: The compelling need for improved trial and regulatory efficiency. 16th World Conference on Lung
Cancer. Presented September 7, 2015.
15. Cameron A, Ewen M, Ross-Degnan D, et al. Medicine prices, availability,
and affordability in 36 developing and middle-income countries:
a secondary analysis. Lancet. 2009;373:240-249.
16. ODonnell O, van Doorslaer E, Rannan-Eliya RP, et al. Who pays for
health care in Asia? J Health Econ. 2008;27:460-475.
17. World Health Organization. WHO Model List of Essential Medicines.
2015; http://www.who.int/entity/medicines/publications/essentialmedicines/en/index.html. Accessed January 15, 2016.
18. Caudron JM, Ford N, Henkens M, et al. Substandard medicines in
resource-poor settings: a problem that can no longer be ignored. Trop
Med Int Health. 2008;13:1062-1072.
19. Gautam CS, Utreja A, Singal GL. Spurious and counterfeit drugs:
a growing industry in the developing world. Postgrad Med J. 2009;85:
251-256.
20. World Health Organization. Prequalification Programme. A United
Nations Programme Managed by WHO. http://apps.who.int/prequal/.
21. Abbott RB, Bader R, Bajjali L, et al. The price of medicines in Jordan:
the cost of trade-based intellectual property. J Generic Med. 2012;9:
75-85.
22. Hoen Et, Law L-M. Access to cancer treatment. http://apps.who.int/
medicinedocs/documents/s21758en/s21758en.pdf. Accessed March 2,
2016.
23. Schut FT, Van Bergeijk PA. International price discrimination: the
pharmaceutical industry. World Dev. 1986;14:1141-1150.
24. Machado M, OBrodovich R, Krahn M, et al. International drug price
comparisons: quality assessment. Rev Panam Salud Publica. 2011;29:46-51.
25. The World Bank. Data. http://data.worldbank.org/country. Accessed
January 10, 2016.
26. Blanchard C, Chetty S, Ganca L, et al. Guide to the treatment of cancer
pain in South Africa. http://dx.doi.org/10.6084/m9.figshare.1612170.
27. Krakauer EL, Wenk R, Buitrago R, et al. Opioid inaccessibility and its
human consequences: reports from the field. J Pain Palliat Care
Pharmacother. 2010;24:239-243.
28. Bagivan G, Tosun N, Komurcu S, et al. Analysis of patient-related

barriers in cancer pain management in Turkish patients. J Pain Symptom
Manage. 2009;38:727-737.
29. Kim YC, Ahn JS, Calimag MM, et al; ACHEON Working Group. Current
practices in cancer pain management in Asia: a survey of patients and
physicians across 10 countries. Cancer Med. 2015;4:1196-1204.
30. Peker L, Celebi N, Canbay O, et al. Doctors opinions, knowledge and
attitudes towards cancer pain management in a university hospital.
Agri. 2008;20:20-30.
31. Prehn AW, Topol B, Stewart S, et al. Differences in treatment patterns
for localized breast carcinoma among Asian/Pacific islander women.
Cancer. 2002;95:2268-2275.
32. Cassileth BR. Complementary therapies: overview and state of the art.
Cancer Nurs. 1999;22:85-90.
33. Molassiotis A, Fernadez-Ortega P, Pud D, et al. Use of complementary
and alternative medicine in cancer patients: a European survey. Ann
Oncol. 2005;16:655-663.
34. Campbell CM, Edwards RR. Ethnic differences in pain and pain management. Pain Manag. 2012;2:219-230.
35. Zuardi AW. History of cannabis as a medicine: a review. Rev Bras
Psiquiatr. 2006;28:153-157.
36. National Cancer Institute. Cannabis and CannabinoidsHealth Professional Version http://www.cancer.gov/about-cancer/treatment/
cam/hp/cannabis-pdq. Accessed January 23, 2016.
37. Al-Atiyyat N. Cultural diversity and cancer pain. J Hosp Palliat Nurs.
2009;11:154-164.
38. Kwok W, Bhuvanakrishna T. The relationship between ethnicity and the
pain experience of cancer patients: a systematic review. Indian J Palliat
Care. 2014;20:194-200.
39. Shulman LN, Wagner CM, Barr R, et al. Proposing essential medicines to
treat cancer: Methodologies, processes, and outcomes. J Clin Oncol.
2016;34:69-75.
65
Integrating Patient-Reported
Outcomes Into Real-Life Medical
Decisions
CHAIR
Paul G. Kluetz, MD
U.S. Food and Drug Administration
Kensington, MD
SPEAKERS
Diana T. Chingos, MS, MFA
Patient Advocate
Studio City, CA
Ethan M. Basch, MD, MSc
Chapel Hill, NC
Sandra A. Mitchell, PhD, CRNP, FAAN
National Cancer Institute
Rockville, MD
MEASURING SYMPTOMATIC ADVERSE EVENTS WITH PRO-CTCAE
Patient-Reported Outcomes in Cancer Clinical Trials:

Measuring Symptomatic Adverse Events With the National
Cancer Institutes Patient-Reported Outcomes Version of
the Common Terminology Criteria for Adverse Events
(PRO-CTCAE)
Paul G. Kluetz, MD, Diana T. Chingos, MS, MFA, Ethan M. Basch, MD, MSc, and
Sandra A. Mitchell, PhD, CRNP, FAAN
OVERVIEW
Systematic capture of the patient perspective can inform the development of new cancer therapies. Patient-reported outcomes
(PROs) are commonly included in cancer clinical trials; however, there is heterogeneity in the constructs, measures, and analytic
approaches that have been used making these endpoints challenging to interpret. There is renewed effort to identify rigorous
methods to obtain high-quality and informative PRO data from cancer clinical trials. In this setting, PROs are used to address specific
research objectives, and an important objective that spans the product development life cycle is the assessment of safety and
tolerability. The U.S. Food and Drug Administrations (FDA) Office of Hematology and Oncology Products (OHOP) has identified
symptomatic adverse events (AEs) as a central PRO concept, and a systematic assessment of patient-reported symptomatic AEs can
provide data to complement clinician reporting. The National Cancer Institutes Patient-Reported Outcomes version of the Common
Terminology Criteria for Adverse Events (PRO-CTCAE) is being evaluated by multiple stakeholders, including the FDA, and is
considered a promising tool to provide a standard yet flexible method to assess symptomatic AEs from the patient perspective. In this
article, we briefly review the FDA OHOPs perspective on PROs in cancer trials submitted to the FDA and focus on the assessment of
symptomatic AEs using PRO-CTCAE. We conclude by discussing further work that must be done to broaden the use of PRO-CTCAE as a
method to provide patient-centered data that can complement existing safety and tolerability assessments across cancer clinical trials.
he intent of this educational manuscript is to discuss the importance of PRO assessments in cancer trials, identify strengths
and limitations of currently used PRO strategies, and focus on
the potential utility of a rigorous and systematic assessment of
symptomatic AEs as a component of a broader PRO strategy.
As part of this effort, we have been fortunate to have a patient
advocate included to introduce the manuscript by providing her
personal perspective on the inclusion of PROs in cancer trials.
Diana Chingos is a 20-year survivor of early onset breast cancer
whose advocacy work extends to membership on the National
Cancer Institutes (NCI) Investigational Drugs Steering Committee and the National Clinical Trials Networks Core Correlative Sciences Committee, as well as participation on a data and
safety monitoring board for the California Cancer Consortium
(a phase I/II clinical trials group) and an institutional review
board. Her experience as a patient and caregiver coupled with
extensive work as an advisor to cancer studies brings a unique

combination of patient focus and understanding of the complexities of clinical trial design and conduct.
THE PATIENT PERSPECTIVE ON PATIENTREPORTED OUTCOMES IN CANCER TRIALS

Patients, and human beings more generally, are accustomed
to providing our views in many aspects of our lives. Until
recently, the act of seeking out feedback from those who
use the health care system was rare. One can argue that the
health care user experience should reign supreme over all
other contexts, when the quality and quantity of our lives is
at stake, sometimes at great financial and logistical expense.
Gratefully, at least from this patients view, PRO questionnaires have been developed to bring our perspective
From the Office of Hematology and Oncology Products, Center for Drug Evaluation and Research, U.S. Food and Drug Administration, Silver Spring, MD; Independent Cancer Patient
Advocate, Los Angeles, CA; Lineberger Comprehensive Cancer Center, University of North Carolina, Chapel Hill, NC; Outcomes Research Branch, National Cancer Institute, Rockville, MD.
Corresponding author: Sandra Mitchell, PhD, CRNP, FAAN, Division of Cancer Control and Population Sciences, Outcomes Research Branch, National Cancer Institute, 9609 Medical
Center Blvd., Rockville, MD 20850; email: mitchlls@mail.nih.gov.
67
KLUETZ ET AL
into the research setting more systematically. In my opinion,

soliciting patient reports espouses equity and autonomy,
enabling patients to speak for themselves, without the filter
of health care providers. Capturing a patients self-report
using well-validated measures offers a direct indicator of
change in symptoms, function, or well-being during treatment, providing additional information to supplement the
clinicians evaluation of tumor response and toxicity.
Many patients with cancer are willing to play their part in
research through answering questionnaires. Unless patients
are very ill and/or have cognitive deficits, many patients with
cancer like the opportunity to contribute to research, especially meaningful research that can improve care for
patients in the future. It can be difficult for some patients to
assess the value of specific research studies, but everyone
understands treatment toxicity and side effects. You live it
and usually have something to say about it.
From the informed patients perspective, the patients
voice has been a key missing element in the current system
of drug safety assessment. Assessing patient-reported
symptomatic AEs can fill this need. Published studies have
demonstrated discordance between physician and patient
reports, with underreporting of patients symptoms and
their severity being common.1,2 Something is getting lost
in the translation. PRO measures provide an opportunity
for the patient to directly report side effects and their intensity from the perspective of the person experiencing it.
The PRO-CTCAE has generated considerable interest in the
broader stakeholder community as a PRO tool that could be
used across the therapeutic development process to address
important questions related to the tolerability profile of a
specific therapy.
KEY POINTS
68
There is a need to strengthen the rigor of PROs in cancer

trials.
The FDAs Office of Hematology and Oncology Products
has identified the systematic assessment of
symptomatic adverse events reported by patients as
an opportunity to better describe the safety and
tolerability of an investigational product across the
drug development life cycle.
The NCIs PRO-CTCAE is a PRO measurement system
that includes a library of questions that measure
symptomatic adverse events from the patient
perspective.
PRO-CTCAE holds promise as a rigorous and flexible
approach to the longitudinal assessment of
symptomatic adverse events in cancer clinical trials.
Challenges and knowledge gaps exist with respect to
trial designs, implementation, and interpretation of
PRO-CTCAE. Effort is ongoing to identify the best
approaches to make PRO-CTCAE scores available,
alongside CTCAE grading, in published reports and
FDA drug labels.
I welcome the ability to report the side effects I have

experienced, and I encourage others to do so. However, it is
important to mention that although most patients want to
describe their experiences, it can become tedious, depending
on the time required. Survey fatigue is real and is magnified by
medication-related fatigue, cancer-related fatigue, and the
unfortunate synergy of physical, emotional, psychologic, and
environmental strains involved with being a patient with cancer
undergoing therapy. Given this, it is important to consider the
length of the survey, the relevance of the questions, and the
time points for assessment, while trying to reduce the redundancy of questions.
Some patients need encouragement and validation to
participate in this process, that their voice is valued and
integral to the clinical trial. Patient engagement helps the
systemit does not hinder it. Empowering patients to articulate their experiences can provide value to those who
develop cancer therapies by more accurately characterizing a drugs effect on the patient and also may enhance
research participation, a necessity if we are to speed up the
rate of knowledge generation.
Past Efforts at Collecting Patient-Reported Outcomes

in Cancer Trials
To date, the most common PRO strategy for oncology has
been to assess the broad multidomain concept of healthrelated quality of life (HRQOL), utilizing instruments largely
developed in a different therapeutic era.35 These existing
HRQOL measures have strengths, including translations
across multiple languages and a familiarity with their use
among the cancer therapeutic development community.
Many instruments have also been expanded to include
disease-specific modules in an effort to better capture
disease- and treatment-related symptoms.6,7 Substantial
data have been accumulated using many of these measures,
and some offer the advantage of well-established cut scores,
minimally important difference thresholds, and normative
values that can aid in interpretation.
However, although the PRO measures commonly used in
oncology trials to date address a broad range of important and common symptoms and functional domains, they
typically include the same questions irrespective of disease
stage or the therapy under study. This can lead to questions
that may be less relevant to the trial context and/or miss the
assessment of important symptoms (e.g., toxicities not
currently included in existing static instruments). This limitation is becoming more evident in the current drug development era of molecularly targeted agents with wide-ranging
side effect profiles. Investigators could benefit from a more
flexible toolbox of PRO measures that can adapt to differing
disease and treatment contexts.
From the FDA OHOP perspective, all PRO data will be taken
into consideration as part of the overall data package to
inform the benefits and risks of a therapy under review.
However, not all data reviewed in an application can be
included in the FDA drug label. The FDA is tasked with
providing information in the product label that is useful to

prescribers in treating their patients and must ensure that
the information is easily interpreted, unbiased, and not
misleading. Broad concepts such as HRQOL are more
challenging to define, with some domains such as social wellbeing farther removed from a therapys direct effect on the
patient. Submitted HRQOL data can be further complicated
by trial design limitations, missing data, and lack of prespecified analyses. For this reason, PRO data have rarely
been included in FDA labeling of cancer therapies. When
PRO data have been included in the FDA product label for
cancer products, it has predominantly relied on well-defined
measures of specific symptoms or functional measures that
relate directly to the disease under study.8
There continues to be wide variability in the type and
quality of PRO data acquired from cancer trials. This lack
of standardization includes heterogeneity in the PRO instruments used, as well as the assessment frequencies, and
the approach to data analysis and reporting. High levels of
missing data have also been a common challenge that can
adversely affect the interpretability of PRO findings. Many
stakeholders, including the FDA OHOP, have actively engaged the oncology drug development community to
evaluate existing instruments and identify emerging opportunities to improve the PRO strategy in cancer trials to
provide more rigorous patient-centered data to all those
who weigh the risks and benefits of cancer therapies.9
FOCUSING ANALYSES ON CORE

PATIENT-REPORTED OUTCOME CONCEPTS
INCLUDING SYMPTOMATIC ADVERSE EVENTS
In most late-phase randomized clinical trials, concepts
such as HRQOL and social, emotional, and cognitive domains are expected by many stakeholders to provide a
reasonably comprehensive picture of the patients experience of their disease and treatment. These data will
be reviewed by FDAs OHOP as important supportive data.
However, in an effort to increase the amount of patientcentered data in FDA labeling, OHOP has recently proposed that our key PRO analyses focus on three core
symptom and functional concepts (Fig. 1).10 Symptomatic
AEs, physical function, and disease-related symptoms are
considered by OHOP to be more well-defined and closer to
the effect of a therapy on the patient and their disease.
The careful collection and analysis of these core concepts
can provide PRO data that may be more consistent with
FDA requirements for labeling.
It should be acknowledged that cancer trials are
designed for differing purposes along the therapeutic
development continuum from first in-human exploration
of dose and safety, to exploratory therapeutic trials, to
trials designed to demonstrate substantial evidence of
safety, tolerability, and efficacy to support a regulatory
submission. Whereas a comprehensive PRO strategy may
be expected to address the needs of multiple stakeholders
in the later phases of therapeutic development, a focused
PRO strategy may be more appropriate and efficient in

early clinical development. This is an important distinction,
as the classic phases of drug development are blurring, and
precision medicine trials may have multiple study arms
designed to simultaneously gauge antitumor activity11
with some single-arm cohorts potentially demonstrating
notable antitumor activity suitable for accelerated approval.12 Safety is an important trial objective in all phases
of therapeutic product development. Inclusion of a PRO
measure of symptomatic AEs can improve our understanding
of safety, tolerability, and dose selection and thus is applicable
to a broad range of clinical trial contexts.
THE USE OF PATIENT-REPORTED OUTCOMES TO

INFORM TREATMENT TOLERABILITY
Safety assessment in cancer clinical trials has been standardized through the use of the CTCAE.13 Currently, in
version 4, the CTCAE provides a widely accepted lexicon
and associated criteria for grading AEs of cancer treatment.
CTCAE is routinely updated and used in conjunction with
the Medical Dictionary for Regulatory Activities by regulatory agencies. Using CTCAE, AEs are graded on a scale
from grade 1 to 5, in which, by convention, grade 5 is death,
and grade 4 reflects toxicities that are life-threatening and
warrant urgent intervention. Grades 13 represent progressive worsening in severity or frequency of the toxicity,
interference with self-care and the performance of daily
activities, and the need for clinical intervention. Although
this method of safety reporting provides for standardization and efficiency in data collection and analysis, the
assessment of AEs using CTCAE is elicited by health care
providers, including symptomatic AEs such as nausea or
sensory neuropathy. It has been stated by the FDA and
others that the patient is best positioned to report his or
her own symptoms.14 Thus, the systematic assessment of
symptomatic AEs using a PRO provides additional information that is complementary to existing safety assessments reported by clinicians using the CTCAE.
The assessment of symptomatic AEs may be of increasing
importance as we enter into a new therapeutic era in malignant hematology and oncology. An expanding number of
mechanistic drug classifications have produced a more diverse range of potential toxicities.15 Many of the molecularly
targeted agents are administered orally, often require
prolonged treatment duration, and may produce less severe
but more chronically bothersome side effects.16 A systematic longitudinal assessment of relevant symptomatic AEs
using a PRO measure may provide informative patientcentered data on symptomatic side effects that may otherwise have been considered low grade by standard clinician
report.17
Existing HRQOL measures and their disease modules
evaluate a more limited range of side effects, many of
which were selected based on therapies that were used at
the time they were developed (e.g., cytotoxic chemotherapies). Given the different therapeutic landscape
69
KLUETZ ET AL
FIGURE 1. U.S. Food and Drug Administration Core Concepts for Patient-Reported Outcomes Analysis in Cancer Trials
In an effort to increase the amount of informative patient-centered data in product labeling, FDA OHOP focuses its PRO analyses on the core concepts of symptomatic
AEs, physical function, and disease-related symptoms. Although these well-defined concepts are more in line with the regulatory framework of the FDA for labeling
considerations, all submitted PRO data will be taken into consideration as important supportive data. The three core concepts are not the only PRO measures to assess
in a trial to support drug approval, as broader domains and HRQOL remain important exploratory measures. Reprinted from Kluetz et al.10
Abbreviations: OHOP, Office of Hematology and Oncology Products; PRO, patient-reported outlets; AEs, adverse events; HRQOL, health-related quality of life.
today, this can lead to measurement of irrelevant symptoms not considered part of the toxicity profile of the
newer drug and/or potentially miss the assessment of
important unique side effects of contemporary therapies.
Furthermore, the limited assessment frequency used to
date in many PRO corollary studies may not be optimal to
adequately gauge tolerability.
Contemporary drug development requires a more flexible
PRO approach to ensure an unbiased assessment of the most
important symptomatic treatment side effects based on the
anticipated toxicity profile of the therapies under study.
Selection of symptomatic AEs from a large library of options
would therefore be desirable. Recently, the NCI has developed and tested a measurement system to capture
symptomatic toxicities directly from patients.18 Comprised of both a library of 124 questions reflecting 78
symptomatic toxicities drawn from the CTCAE and an
electronic system for survey administration, reminders,
central monitoring, and alerts, NCI PRO-CTCAE is designed
70
to provide a standard yet flexible tool to assess symptomatic AEs.
Development of the NCI PRO-CTCAE Measurement

System
The PRO-CTCAE measurement system has been developed
by NCI as a companion to the CTCAE. It was designed to
improve the validity, reliability, and precision with which
symptomatic adverse effects of treatment are evaluated
in patients on cancer clinical trials. PRO-CTCAE was developed
by a multidisciplinary team of trialists, methodologists, clinicians, informatics experts, patients, and regulators18 and
has been tested and refined in a consortium of academic and community-based cancer-treatment sites and
in the NCI-sponsored clinical trials network (NCI contracts
HHSN261200800043C and HHSN261200800063C) and by
more than 120 early adopters in 10 countries. The PRO-CTCAE
item library consists of 124 discrete items representing 78
symptomatic AEs that are common in oncology clinical trials

and included in the CTCAE. PRO-CTCAE items were created
with substantial input from patients, clinicians, and PRO
methodologists and underwent refinement through cognitive
interviews with patients to establish content validity.19 Subsequently, the quantitative measurement properties, including validity, reliability, and responsiveness, were evaluated
in a large and diverse sample of patients receiving cancer
treatment in six sites around the United States.20
Each of the 78 symptom terms included in the PRO-CTCAE
item library is assessed relative to one or more distinct
attributes, including presence/absence, frequency, severity,
and/or interference with usual or daily activities.18 Responses are provided on a five-point Likert scale. The generic
PRO-CTCAE item structures for the frequency, severity, and
interference attributes are listed below:
Frequency item: How OFTEN did you have __________?
(Never / Rarely / Occasionally / Frequently / Almost
constantly)
Severity item: What was the SEVERITY of your
__________ at its WORST?
(None / Mild / Moderate / Severe / Very severe)
Interference item: How much did __________ INTERFERE
with your usual or daily activities?
(Not at all / A little bit / Somewhat / Quite a bit / Very
much)
The standard PRO-CTCAE recall period is the past 7 days.
A recent study suggests that longer recall periods (2-, 3-,
or 4-week recall) are associated with small but successively increasing measurement error, which must be
considered if recall periods longer than 1 week are used
in a trial for logistical reasons (TR Mendoza, AV Bennett,
SA Mitchell et al, unpublished data, March 2016). Administration of PRO-CTCAE via different modes including
web, interactive voice response, and paper offers flexibility for patients and study operations personnel, and
there is psychometric evidence to justify comparison of
results and pooled analyses across studies that use different PRO-CTCAE modes of administration.21 A pediatric
version of PRO-CTCAE is also currently in development.22
For more information about PRO-CTCAE, visit http://
healthcaredelivery.cancer.gov/pro-ctcae.
Early Adoption of PRO-CTCAE and Lessons Learned

The PRO-CTCAE has been implemented in multiple cancer
clinical trials. Patients are generally willing and able to selfreport this information weekly via the web or automated
telephone systems though electronic reminders; central
monitoring and personnel for backup data collection were
also needed to optimize response rates. Clinicians note
finding this information to be meaningful and valuable for
clinical decision-making and AE reporting.23
Selecting which symptomatic AEs to assess and determining the time points for measurement are critical
trial-design decisions. In trials developed to date, items
from the PRO-CTCAE and time points of measurement
have generally been specified using an approach similar to

that used to define the AE surveillance plan for the trial
more broadly. That is, the study team reviews published
data, as well as data from earlier phase trials or animal
models, if available, and incorporates information about
the known or anticipated on- and off-target effects of
agents in a similar mechanistic class to identify those
symptomatic AEs likely to be associated with the regimens
in the trial.24 PRO-CTCAE items corresponding to these
symptomatic AEs are loaded into a software platform.
Patients are trained to use the software and are asked to
self-report either from home on a regular basis or at clinic
visits. After establishing a pretreatment baseline, more
frequent PRO-CTCAE administration is generally warranted during the first few cycles of therapy (e.g., weekly
reporting during the initial several months of therapy). Thereafter, the assessment intervals may be extended
(e.g., monthly or quarterly, depending on the regimen
under study), particularly in trial contexts in which the
duration of investigational treatment is prolonged. However, the time points of measurement should reflect the
anticipated pattern of toxicity and scientific objectives of
the trial.
Moving forward, several challenges and knowledge gaps
must be addressed. Although multiple translations are in
progress (for example, Italian, Korean, Chinese, and Swedish), and linguistically validated language versions are
available in Spanish, German, Japanese, and Danish,25-27
translation and linguistic validation of the PRO-CTCAE item
library in other languages is needed. Second, as is the case
for the collection of PRO data in any trial, there are personnel and infrastructure requirements. Investments will
be required to develop and refine strategies that achieve
efficient PRO data collection and ensure data completeness
(e.g., central monitoring and backup data collection). Potential concerns about workload for clinical research staff
will also need to be addressed, and analyses are in progress that will yield specific estimates of the resource requirements associated with collecting and analyzing PRO-CTCAE
data in NCI-sponsored clinical trials. Third, there is limited experience with respect to how to optimally analyze and interpret
patient-reported symptomatic AE data. Efforts are underway to
determine how PRO-CTCAE scores should be interpreted to
assign a corresponding CTCAE grade. Additional work will
be required to identify the most informative ways to display
symptomatic toxicity scores descriptively alongside CTCAE
grades in both published reports as well as potential product
labels.
Notably, FDA OHOP is committed to working with the
NCI as well as commercial sponsors early in programs to
identify opportunities to include PRO-CTCAE in clinical
trials. Data generated using PRO-CTCAE offer complementary descriptive patient-reported information about
symptomatic side effects that may further inform patients
and providers. There is also interest in using PRO-CTCAE
as a component to support comparative tolerability
trial designs. More work must be done to explore this
71
KLUETZ ET AL
opportunity, including identifying an appropriate approach

to capture the overall side effect burden of treatment. In
addition, FDA OHOP, in collaboration with Clinical Outcome
Assessment staff and the Office of Biostatistics, has initiated
several internal working groups to explore different analysis
and data presentation methods.
There is a reasonable concern that stacking new instruments on top of existing lengthy HRQOL and disease
modules as well as utility measures could lead to duplication
and pose additional respondent burden. FDA OHOP is fostering international collaboration to review existing HRQOL
instruments and their disease modules to identify a collection of new and existing or modified existing instruments
to meet the needs of all those who will use these data to
make treatment, regulatory, and health policy decisions. The
goal remains to identify approaches that increase the relevance and interpretability of PRO data while minimizing the
burden to patients to ensure that the patient can be queried
at an assessment frequency that provides the best picture of
the patient experience while on therapy.
There has been a call to improve the quality of PRO data
captured from cancer clinical trials and integrate more of
these data in the FDA product label.28 To realize this goal,
more attention must be paid to PRO measures in both trial
design and conduct to improve overall data quality, regardless of the symptoms or domains being measured. The
assessment of symptomatic AEs using a rigorously developed item library such as the PRO-CTCAE can increase the
likelihood of inclusion of descriptive patient-centered data
in product labels, complementing the standard safety assessment of a therapy.
CONCLUSION
Patients and their clinicians would benefit from improved
data about the effects of anticancer therapy on how an individual feels and functions. Traditional PRO strategies are
being revisited as patient-focused drug development has
generated multistakeholder interest to optimize the collection and interpretation of PRO data to satisfy the needs of the
many end users of this information. As a key component of a
broader PRO strategy, the systematic longitudinal assessment
of patient-reported symptomatic AEs can provide additional
complementary tolerability data to inform dose selection and
the overall benefit: risk assessment of a cancer therapy. The
PRO-CTCAE has been developed as a standardized measurement system that can provide a flexible fit-for-purpose
approach to assess relevant symptomatic AEs across a broad
range of cancer therapies. It is anticipated that the NCI PROCTCAE item library will continue to be iteratively refined as
novel symptomatic toxicities are identified and a deeper
understanding of its measurement properties emerges. There
is vigorous and ongoing international collaboration among
trialists, methodologists, regulators, and patients to address
these and other challenges in study design, implementation,
and interpretation to evolve a standard method to obtain
well-defined, descriptive patient-centered data on the safety
and tolerability of cancer therapies.
References
1. Basch E, Iasonos A, McDonough T, et al. Patient versus clinician
symptom reporting using the National Cancer Institute Common Terminology Criteria for Adverse Events: results of a questionnaire-based
2. Xiao C, Polomano R, Bruner DW. Comparison between patient-reported
and clinician-observed symptoms in oncology. Cancer Nurs. 2013;36:
E1-E16.
3. Cella DF, Tulsky DS, Gray G, et al. The Functional Assessment of Cancer
Therapy scale: development and validation of the general measure.
J Clin Oncol. 1993;11:570-579.
4. Niezgoda HE, Pater JL. A validation study of the domains of the core
EORTC quality of life questionnaire. Qual Life Res. 1993;2:319-325.
5. Srensen JB, Klee M, Palshof T, et al. Performance status assessment in
cancer patients. An inter-observer variability study. Br J Cancer. 1993;
67:773-775.
6. European Organization for Research and Treatment of Cancer (EORTC)
Disease Specific Modules. http://groups.eortc.be/qol/why-do-weneed-modules. Accessed February 25, 2016.
7. Functional Assessment of Cancer Therapy (FACT) Cancer Specific
Measures. http://www.facit.org/FACITOrg/Questionnaires. Accessed
February 25, 2016.
8. Rock EP, Kennedy DL, Furness MH, et al. Patient-reported outcomes
supporting anticancer product approvals. J Clin Oncol. 2007;25:
5094-5099.
72
9. Basch E, Geoghegan C, Coons SJ, et al. Patient-reported outcomes in

cancer drug development and US regulatory review: perspectives from
industry, the Food and Drug Administration, and the patient. JAMA
Oncol. 2015;1:375-379.
10. Kluetz PG, Slagle A, Papadopoulos E, et al. Focusing on core patientreported outcomes in cancer clinical trials: symptomatic adverse
events, physical function, and disease-related symptoms. Clin Cancer
Res. 2016;22:1553-1558.
11. Biankin AV, Piantadosi S, Hollingsworth SJ. Patient-centric trials for
therapeutic development in precision oncology. Nature. 2015;526:
361-370.
12. Theoret MR, Pai-Scherf LH, Chuk MK, et al. Expansion cohorts in first-inhuman solid tumor oncology trials. Clin Cancer Res. 2015;21:4545-4551.
13. U.S. Department of Health and Human Services NIH, National Cancer
Institute. Common Terminology Criteria for Adverse Events (CTCAE)
Version 4.0. http://evs.nci.nih.gov/ftp1/CTCAE/CTCAE_4.03_2010-0614_QuickReference_5x7.pdf. Accessed February 27, 2016.
14. U.S. Food and Drug Administration. Guidance for Industry PatientReported Outcome Measures: Use in Medical Product Development to
Support Labeling Claims. http://www.fda.gov/downloads/Drugs/
Guidances/UCM193282.pdf. Accessed February 27, 2016.
15. Michot JM, Bigenwald C, Champiat S, et al. Immune-related adverse
events with immune checkpoint blockade: a comprehensive review. Eur
J Cancer. 2016;54:139-148.
16. Klastersky JA. Adverse events of targeted therapies. Curr Opin Oncol.
2014;26:395-402.
17. Thanarajasingam G, Hubbard JM, Sloan JA, et al. The imperative for a
new approach to toxicity analysis in oncology clinical trials. J Natl Cancer
Inst. 2015;107:djv216.
18. Basch E, Reeve BB, Mitchell SA, et al. Development of the National
Cancer Institutes patient-reported outcomes version of the common
terminology criteria for adverse events (PRO-CTCAE). J Natl Cancer Inst.
2014;106:dju244.
19. Hay JL, Atkinson TM, Reeve BB, et al; NCI PRO-CTCAE Study Group.
Cognitive interviewing of the US National Cancer Institutes PatientReported Outcomes version of the Common Terminology Criteria for
Adverse Events (PRO-CTCAE). Qual Life Res. 2014;23:257-269.
20. Dueck AC, Mendoza TR, Mitchell SA, et al; National Cancer Institute
PRO-CTCAE Study Group. Validity and reliability of the US National
Cancer Institutes Patient-Reported Outcomes version of the Common
Terminology Criteria for Adverse Events (PRO-CTCAE). JAMA Oncol.
2015;1:1051-1059.
21. Bennett AV, Dueck AC, Mitchell SA, et al; National Cancer Institute PROCTCAE Study Group. Mode equivalence and acceptability of tablet
computer-, interactive voice response system-, and paper-based administration of the U.S. National Cancer Institutes Patient-Reported
Outcomes version of the Common Terminology Criteria for Adverse
Events (PRO-CTCAE). Health Qual Life Outcomes. 2016;14:24.
22. Reeve BB, Withycombe JS, Baker JN, et al. The first step to integrating
the childs voice in adverse event reporting in oncology trials: a content
23.
24.
25.
26.
27.
28.
validation study among pediatric oncology clinicians. Pediatr Blood

Cancer. 2013;60:1231-1236.
Bruner DW, Hanisch LJ, Reeve BB, et al. Stakeholder perspectives on
implementing the National Cancer Institutes patient-reported outcomes version of the Common Terminology Criteria for Adverse Events
(PRO-CTCAE). Transl Behav Med. 2011;1:110-122.
Cella D, Wagner L. Re-personalizing precision medicine: is there a role
for patient-reported outcomes? J Community Support Oncol. 2015;13:
275-277.
Arnold B, Lent L, Mendoza T, et al; PRO-CTCAE Spanish Translation
and Linguistic Validation Study Group. Linguistic validation of
the Spanish version of the National Cancer Institutes PatientReported Outcomes version of the Common Terminology Criteria
for Adverse Events (PRO-CTCAE). Support Care Cancer. Epub 2016
Feb 2.
Kirsch M, Mitchell SA, Dobbels F, et al. Linguistic and content validation
of a German-language PRO-CTCAE-based patient-reported outcomes
instrument to evaluate the late effect symptom experience after allogeneic hematopoietic stem cell transplantation. Eur J Oncol Nurs.
2015;19:66-74.
Baeksted C, Pappot H, Bidstrup PE, et al. Danish translation and linguistic validation of the Patient-Reported Outcomes version of the
Common Terminology Criteria for Adverse Events (PRO-CTCAE). J Pain
Symptom Manage. In press.
Basch E. Toward patient-centered drug development in oncology.
N Engl J Med. 2013;369:397-400.
73
Optimizing Team-Based Oncology

Care: Building the Village
CHAIR
Lee Schwartzberg, MD, FACP
The West Cancer Center
Memphis, TN
SPEAKERS
Lillie D. Shockney, RN, BS, MAS
Johns Hopkins Breast Center
Baltimore, MD
Wendy H. Vogel, MSN, FNP, AOCNP
Wellmont Cancer Institute
Kingsport, TN
INFORMING THE TEAM WITH PATIENT-REPORTED OUTCOMES
Electronic Patient-Reported Outcomes: The Time Is Ripe for

Integration Into Patient Care and Clinical Research
Lee Schwartzberg, MD, FACP
OVERVIEW
In the emerging team-based approach to delivering cancer care, collecting patient-reported outcomes (PROs) provides
longitudinal monitoring of treatment adverse effects, disease complications, functional statuses, and psychological states
throughout the cancer continuum for all providers to use. Electronic systems offer added capabilities, including easy
quantitation of individual symptom items and aggregated scales, standardization, and longitudinal tracking of patient
surveys for trend analysis over time. An ideal electronic PRO (ePRO) platform is clinically relevant, validated, and reliable and
would offer patient usability. Additionally, it should allow for automated responses to and from patients, have scheduling
functionality, and send real-time alerts to site personnel and patients. Clinical interfaces should be easy to read and integrated into the electronic medical record. Multiple ePRO systems, often using electronic tablets, have been created and are
beginning to be widely deployed. The Patient Care Monitor is one example of a system that has evolved into a comprehensive patient engagement platform, with a complete review of systems survey and capabilities for mobile health
usage. Recent clinical trials have established ePRO systems as an effective method of providing information, which aids
improved patient outcomes, including reduced health resource utilization and longer time on therapy. ePROs are also
increasingly incorporated into clinical trials, where they can provide more thorough reporting of adverse events than can be
captured by alternative methods. Mobile devices have the potential to become the method by which all members of the
provider team communicate with patients both at the point-of-care and between clinic visits to optimize care delivery.
chieving optimal patient-centered cancer care depends

in large part on the crucial clinician-patient interaction
at the point of care. In an era of increasing demands on the
physicians time, declining reimbursement, a surge in documentation requirements, and ever-increasing patient
volumes, the very nature of the patient-provider relationship is threatened. One fundamental goal of the patient
interview is to obtain relevant information about the present
health status to guide treatment and symptom control. A
thorough review of systems and evaluation of psychosocial
status and physical and social functioning is ideally accomplished at the clinic visit.
The complexities of modern oncology care require a team
approach to deliver quality management of these multiple
dimensions. Having a record of PROs, which are quantitative,
validated, easily captured, standardized, and recorded in the
electronic health record, has the potential to add huge value.
All of these characteristics can be achieved by using a variety
of electronic systems that have been developed to enhance
the interaction between the clinical staff and the patient.
A patient-reported outcome is defined as a measurement
of the patients condition, reported directly by the patient
without interpretation by a clinician or any other individual.

Collecting PROs allows clinicians to longitudinally monitor a
patients tolerance of therapy, response to therapy, and
symptoms that result from the underlying disease or
treatment. Psychological and functional statuses can be
assessed in the same fashion, which allows examination of
the interplay between treatment and global quality of life
and identification of issues that require referral to other
members of the team. The importance of subjective monitoring is firmly established; pain as the fifth vital sign is an
integral assessment measure for the Quality Oncology
Practice Initiative program, and the Commission on Cancer
now requires distress screening as the sixth vital sign to be
measured and addressed at every visit. A systematic approach to assess distress can be enhanced through the use of
an ePRO system, which allows self-identification of the
problem without prejudice.
In a distributed care model, a permanent, searchable
record of the longitudinal health status allows all members
of the care team to access, interpret, and act on the findings.
It is well documented that patients with cancer are often
reluctant to discuss important toxicities with providers for a
From The West Cancer Center, Memphis, TN.

Corresponding author: Lee Schwartzberg, MD, FACP, 7945 Wolf River Blvd., Germantown, TN 38138; email: lschwartzberg@westclinic.com.
e89
LEE SCHWARTZBERG
variety of reasons.1 Completion of an electronic survey

appears to reduce that reluctance.2
Empowering patients to self-report provides the best information on subjective symptoms. Multiple investigations
have clearly documented under-reporting when information
is provided by clinicians, although nurses appear better at
eliciting symptoms than physicians.3,4 Reasons for underreporting of toxicities may include a judgment call by the
clinician when the symptom is detected but not reported.
This circumstance might occur when symptoms are preexisting, attributable to the disease, or mild and not requiring intervention.5 Symptoms that are not reported
because of inadequate communication are potentially more
damaging. With the advent of many new drugs, some with
novel toxicities, the possibility of missing unusual but important adverse effects that are not familiar to the provider
looms large in the presence of poor symptom communication. Standardized routine assessment is recognized
among medical policymakers as a key component of ensuring high-quality patient-centered care.6
staff, typically by email or text message. Many of the newer

systems are designed for long-term monitoring across the
cancer care continuum. Flexibility of system reporting allows
ePROs to be used effectively by various stakeholders of the
care team, including clinicians, navigators, psychologists,
care support counselors, and financial aid associates. One
limitation of most ePRO systems to date is a lack of the ability
of patients to add self-identified concerns, which would
require the use of free text and patient input into selection of
the items of most importance to them at a given moment in
the cancer trajectory.
In order for an ePRO system to be widely disseminated and
routinely used, value to both the provider and the patient
must be demonstrated. Evidence exists to support integration of ePROs into clinical workflow in a manner that is
relatively unobtrusive and that provides the potential for
improved symptom control and patient-clinician interaction.8 Table 1 displays a proposed feature set that
serves as a good framework for elements of optimal
platforms.
USE OF ePRO SYSTEMS
DEVELOPMENT OF AN ePRO SYSTEM
A review of ePRO systems used in cancer clinical care was

recently published.7 Thirty-three systems were identified;
the majority are used in the United States, and one-third are
used in a single academic institution, with the majority used
at a single clinical entity. Most systems were developed to be
used in the clinic at the point of care, and most are webbased. The focus of these systems is use predominantly
during treatment and follow-up care, to monitor chemotherapy and other cancer therapy. Only half of the systems
had greater than 500 users. The majority of systems sent
real-time alerts of the patients response to providers and
KEY POINTS
e90
PROs provide the most thorough and reliable

information on subjective symptoms, overcoming the
well-documented under-reporting by clinicians.
ePRO systems, which are web-based and are typically
delivered on tablets, computers, or other mobile
devices, are now readily available.
One example of an ePRO system, the Patient Care
Monitor, is a full patient engagement system, which
includes a complete review of system survey and a
flexible architecture that allows deployment of custom
surveys and patient alerts to relevant clinical staff.
The use of an ePRO survey for patients with advanced
cancer has been shown in a randomized controlled trial
to improve health-related quality of life, reduce
emergency department visits, and improve 1-year
survival compared with usual clinical care.
A subset of the Common Terminology Criteria for
Adverse Events library is now available as an ePRO
platform and has been validated as an alternative
method of eliciting symptom reports in clinical trials.
Our group began developing an electronic tablet-based

system and an instrument to routinely monitor patient
symptoms and health statuses more than 15 years ago.
Assessment of items pertaining to physical symptoms,
emotional reactions, and functional statuses were identified
by a team of medical oncologists, clinical psychologists
working in oncology, oncology nurses, and patients. Items
were assembled by reviewing previously developed qualityof-life instruments from the research setting and by determining the importance via face validity in routine patient
care. Each item was presented as a 0-to-10 Likert scale of
severity, in which 0 represented no severity and 10 represented the most severe, with a timeframe of the past 7 days.
The instrument, termed the Patient Care Monitor (PCM;
initially called Cancer Care Monitor) consisted of 38 items,
grouped into seven constructs of physical symptoms,
treatment adverse effects, distress, despair, impaired ambulation, impaired performance, and global quality of life.
Validation studies established psychometric validity, scale
consistency, and reliability.9 Alternative form reliability
that was based on comparisons with paper forms was high,
and patient preference was for the tablet-administered
version. In subsequent studies, item validity by PCM was
compared with a blinded structured interview conducted by
an experienced oncology nurse, and good concordance was
established between the structured interview and patient
self-report.10
An expanded version of the PCM, termed PCM 2.0, was
subsequently created to address clinicians desires to have a
complete review of systems provided by patient self-report.
The gender-specific survey consisted of 86 items (79 for
men) that encompassed 13 organ sitespecific systems (e.g.,
gastrointestinal, cardiovascular). A validation study was
performed for PCM 2.0 and showed good correlation between patient- and nurse-reported severities.11 The median
TABLE 1. Desirable Features of ePRO Systems in

Clinical Care
Elements
Attributes
Symptom Questions/
Items
Demonstrated validity and reliability

Known clinical significance of score changes
Relevance in the selected patient population
and clinical context
Appropriate recall period (no longer than
1 week)
Patient Interface
Demonstrated usability in patients

Web-based and available in multiple modes
for data entry via internet, handheld device, tablet app
Allowance for in-clinic and/or between-visit
reporting
Inclusion of a scheduling/calendar component to specify when patients are expected
to report
Automated reminders to patients to report
(e.g., email, text message, automated
telephone call)
Allowance for use of simple usernames/
passwords that are not overly complex or
cumbersome to patients
Availability of staff to train and assist patients
Allows patients to request additional help
Clinician Interface
Includes displays of longitudinal patientreported symptoms numerically and

graphically
Integrated with electronic health record
Allows clinicians to report symptoms/
comment on patient reports
Alerts/Notifications
Sends real-time automated alerts (e.g., email,

text message) to clinicians when patients
report symptom severities above
predetermined thresholds or experience
symptom worsening that exceeds a
predetermined amount
Triggers automated notifications to site
personnel when patients do not report at
a scheduled time point
Provides automated messages to patients
and triggers automated patient education
when symptoms exceed predetermined
absolute thresholds or score-change
thresholds
Back-up Data
Collection
Includes predetermined methods to contact

patients who do not self-report on
schedule
Abbreviations: ePRO, electronic patient-reported outcome.

Adapted from Basch and Abernethy.8
time for patients to complete PCM 2.0 was 11 minutes. This

version of the platform, which included a wireless tablet
system linked to a central server, was eventually deployed in
more than 100 community practice sites for use in practice
performance improvement projects and as part of routine
clinical care.
PCM 2.0 results were available both in a paper format
and integrated into the electronic medical record (EMR)
of commercial oncology EMR products through an HL-7
interface. In the EMR, the items formed part of the permanent record located on the flowsheet for a given outpatient visit. The paper version displayed demographics,
columns of the last three surveys of each item, and visual
prompts to allow the clinician to quickly scan the report for
severe endorsements and potentially clinically significant
differences ($ 3-point change) from the last survey, signified
by up and down arrows. The surveys are transmitted to the
clinician immediately after completion and serve the
function of focusing the patient interview on the most
severe/changed item endorsements.
PCM 2.0 was installed in a major academic cancer center in
2006. A pilot trial established patient satisfaction with PCM
usage and acknowledgment that the system facilitated
symptom reporting to the clinicians.12 Validation was accomplished in a larger trial of 275 patients with breast,
gastrointestinal, or lung cancer.13 Good internal consistency
was demonstrated; Cronbachs a coefficients were similar to
those determined in the community oncology validation
study. Pearson correlatives for PCM subscales were good,
and there was good factor concordance with the Functional
Assessment of Cancer TherapyGeneral (FACT-G), Functional Assessment of Chronic Illness Therapy-Fatigue (FACIT-F),
and MD Anderson Symptom Inventory (MDASI) instruments
widely used in the research setting. PCM 2.0 thus served a dual
functionality as a clinical and a research tool, which allowed
individual clinicians to use the data to inform the patient visit
and became part of a longitudinal repository of standardized
PRO items well suited to research. A larger data set of 5,624
patients who completed the PCM survey was recently analyzed. The instrument exhibited high internal consistency, and
investigators were able to accurate distribute the items into
three constructs of emotional function, physical function, and
physical symptoms.14 This large-scale experience firmly
established the practicality of systematic collection of ePROs
as part of the standard of care.
The widespread adoption of smartphones, tablets, and
other portable computing devices led to the development
of a new architecture for the PCM system in 2014, when it
was redeployed as a cloud-based, web-enabled application
with an open architecture to meet the growing need for
additional surveys at the practice level. The core instrument
was condensed somewhat to 56 items on the basis of the
evaluation of rarely reported symptoms and those that did
not contribute meaningfully to the aggregate scale scores.
A complete 13-system review to meet Centers for Medicare
& Medicaid Services guidelines for the highest-level visit
was retained.
This revised PCM core version is backward compatible with
earlier versions and has concordance with several other
commonly used instruments for core symptoms. The new
system is delivered via standard commercial tablets provided by the oncology practice rather than as the previously
used self-contained dedicated system. Acceptance by patients and providers has been excellent. All told, greater than
100,000 patients have completed PCM surveys since its
inception. Additional modules that have been added to the
e91
LEE SCHWARTZBERG
core survey include a radiation therapyspecific survey, a

falls risk assessment and learning needs assessment survey,
and a tobacco use and readiness to quit module.
FIGURE 1. Cumulative Incidence of Emergency

Department Visits
BENEFIT OF ePROS
Despite the value attributed to PROs in reporting and
addressing routine symptoms, the impact of collecting this
data has not been settled. Systematic reviews have largely
failed to demonstrate a clear benefit of collecting PROs,
largely because of methodologic issues with the ways
studies are structured, such as including of several sources of
bias or widely varying endpoints.15 A prospective randomized controlled trial conducted at two academic medical
centers met its primary endpoint of demonstrating that selfreporting of ePROs available to the clinician at the clinic visit
resulted in more discussion of the problem with the patient
than usual care when the symptom intensity was higher
than a predefined threshold.16 Patient-provider visits were
not significantly longer in duration despite having the PROs
to address, and most clinicians surveyed as part of the study
felt that the PROs served a clinical utility function.
A recent multicenter randomized controlled trial reported
by Basch et al17 assigned patients who received chemotherapy to report a small number of symptoms via tabletcomputer with weekly email prompts or to receive usual
care. Symptom assessment was delivered either to providers
at the outpatient visit or to the nursing staff via email when
patients reported severe or worsening symptoms. The primary endpoint of the study was change in health-related
quality of life (QOL) at 6 months compared with baseline.
Secondary endpoints included emergency department visits,
hospitalizations, and overall survival. Patients were additionally separated into subgroups on the basis of their prior
experiences with computer systems.
The trial revealed that more patients in the ePRO group at
6 months had improvements in health-related QOL (34% vs.
18%), fewer had a decrease (38% vs. 53%, p , .001), and the
mean health-related QOL score declined less in the ePRO group
than in the routine care group (1.4 vs. 7.1, p , .001). Importantly, significantly fewer patients who routinely reported
PROs had an emergency department visit during the period of
study (Fig. 1), and those who used the electronic system received chemotherapy longer. Overall survival at 1 year was
better for the ePRO group than the usual care group (75% vs.
69%, p = .05). Interestingly, computer-inexperienced patients
appeared to derive more benefit from the system with regard
to reduction in emergency department visits, perhaps because
this subgroup was older, less educated, and may have been
under-reporting symptoms previously for cultural reasons.
This study highlights the power of ePRO reporting to influence tangible outcome measures for patients with advanced cancer, including reduced resource utilization, better
quality of life, and evenpossiblyimproved survival. Why
would this be? Routine reporting of cancer symptoms and a
system of planned interactions to address them appear to
improve patient well-being and allow patients to tolerate
treatment better, which might influence survival. In the trial
e92
The incidence of patients visiting the emergency department is shown, with

death as a competing event: all patients (A); computer-experienced patients (B);
and computer-inexperienced patients (C).
Abbreviation: STAR, Symptom Tracking and Reporting (a web-based self-reporting
system as the study intervention).
Adapted and used with permission.17
conducted by Basch et al,17 time on therapy was approximately 2 months longer in the ePRO group, which supports
this assertion. The results also mirror those of patients with
lung cancer who were referred early to palliative care,18 in part
because patients symptoms in both instances were taken
seriously and addressed. The patient-centered medical home
model, endorsed by the Centers for Medicare & Medicaid
Services (via its Oncology Care Model), calls for improved
interactions between the clinic staff and patients, including
mechanisms for patients to report problems in a real-time way
without having to resort to hospitalization or emergency
department visits. An ePRO system is an excellent tool to assist
implementation of the Oncology Care Model program.
PROS IN CLINICAL RESEARCH

Inclusion of patients self-reported symptoms and adverse
events is fundamental to evaluating new drugs in oncology.
The use of PROs is endorsed by all major entities involved in
conducting clinical trials.19 A multidisciplinary group was
charged by the Center for Medical Technology Policy to
create an evidence guidance document with recommendations for PRO use in comparative effectiveness research.20
The group devised 15 recommendations related to three
broad domains of PROs: measure selection, implementation
measures, and data analysis/reporting. Chief among these
were the importance of selecting a validated tool that is fit
for the purpose of the study. Twelve symptoms that share
commonality among PRO instruments were highlighted as
having particular value in assessing wellbeing and healthrelated QOL (Table 2). Implementation recommendations
focused on obtaining information rapidly, preferably electronically, when this method had been adequately compared with other standard methods such as paper forms,
and on collecting information as frequently as necessary

for the proposed research without overburdening patients.
In using PROs for comparative effectiveness research, the
evidence guidance document recommended a priori power
calculations of PRO endpoints, a plan for handling missing
data, and a plan for reporting individual measure changes
proportionally and as mean group measure changes. The
document also endorsed simultaneous publication of the
PRO data analysis with the other endpoints of the research trial.
Standard adverse event reporting in clinical trials is assessed
by the Common Terminology Criteria for Adverse Events
(CTCAE), created by the National Cancer Institute. Adverse
events may reflect laboratory abnormalities, physical examination changes, functional status variations, and patientelicited toxicities. Traditionally, symptoms are prompted
from the patient by a clinical observer involved in the trial.
However, evidence suggests that adverse event symptoms are
often under-reported by clinical investigators.21,22 The National Cancer Institute has developed a library of items, termed
the PRO-CTCAE, which consists of 124 PROs that reflect 78
different adverse events. Recently, a validity and reliability
assessment of the PRO-CTCAE tool has been published.23 The
study included 975 adult patients receiving chemotherapy or
radiation therapy who completed the instrument on tablets in
clinic waiting rooms at nine U.S. cancer centers and community oncology practices. PRO-CTCAE were compared with
the clinician-reported Eastern Cooperative Oncology Group
(ECOG) performance status and a reference instrument, the
European Organization for Research and Treatment of Cancer
Core Quality of Life Questionnaire (QLQ-C30).
A broad sample of patients was enrolled, including many
who had ECOG performance statuses of 2 to 4. Symptom
TABLE 2. Availability in Existing PRO Instruments of Common Cancer Symptoms

PRO Instrument*
Symptom
ESAS
FACT
LASA
MDASI
MSAS
PCM
PRO-CTCAE
PROMIS
QLQ-C30
RSCL
SDS
Anorexia
Anxiety
Constipation
Depression
Diarrhea
Dyspnea
Fatigue
Insomnia
Nausea
Pain
Neuropathy
Vomiting
Abbreviations: CER, comparative effectiveness research; ESAS, Edmonton Symptom Assessment Scale; FACIT, Functional Assessment of Chronic Illness Therapy; FACT-G, Functional
Assessment of Cancer Therapy-General; LASA, Linear Analog Self-Assessment; MDASI, MD Anderson Symptom Inventory; MSAS, Memorial Symptom Assessment Scale; PCM, Patient
Care Monitor; PRO, patient-reported outcome; PRO-CTCAE, PRO version of the Common Terminology Criteria for Adverse Events; PROMIS, PRO Measurement Information System;
QLQ-C30, European Organization for Research and Treatment of Cancer Quality of Life Questionnaire; RSCL, Rotterdam Symptom checklist; SDS, Symptom Distress Scale.
Adapted and used with permission.20
*Instruments are listed alphabetically. Dash signifies symptom not measured by that instrument; X, measured by that instrument. Most measurement systems include additional
symptom items beyond these 12 symptoms.
e93
LEE SCHWARTZBERG
endorsement was compared with several clinical anchors,

including performance status, type of therapy, supportive
care measures initiated, expert consensus, and patient input. A median (range) of 23 (091) PROs were endorsed.
Ninety-six percent of patients reported five or more
symptoms at the first visit, and 768 (81.7%) of 940 patients
reported at least one symptom as frequent, severe, and/or
interfering quite a bit with daily activities. Validity of the
instrument was achieved; all 124 items were endorsed in the
same direction as the QLQ-C30 reference instrument, and
114 items demonstrated meaningful correlations, with a
Pearson coefficient r of 0.1 or greater. Of these 114 items,
111 were significant (p , .05).
Responsiveness of items to the patient-reported global
impression of change (GIC) from visit one to two was
evaluated in a core set of 20 symptomatic adverse events
with high face validity for relation to change in QOL, physical
condition, and/or emotional state. Thirteen items achieved
excellent correlation with GIC scores, which validated this
symptom set as critical in approximating patient-reported
health statuses. This trial represents the culmination of more
than a decade of study comparing patient assessment of
toxicity to clinician assessment in clinical trials. The challenge
now will be to incorporate these assessments into routine
adverse event reporting when new therapeutic entities are
evaluated.
Longitudinal collection of ePROs offers opportunities to
integrate symptom attribution into retrospective studies of
treatment patterns in the real-world setting. Such analyses
have been performed with the PCM to address the effect of
disease progression or various chemotherapy regimens on

PROs in metastatic breast cancer, prostate cancer, and lung
cancer.24-27
EXPANDING ePROS TO FULL PATIENT

ENGAGEMENT PLATFORM
A 2013 Institute of Medicine report, Delivering High-Quality
Cancer Care: Charting a New Course for a System in Crisis,
stated that cancer care is often not as patient-centered,
accessible, coordinated, or evidence-based as it should be.28
Part of the solution is to actively engage patients to interact
with an adequately staffed and trained workforce. To meet
the needs of an increasing number of patients, cancer clinics
must embrace and expand nonphysician members of the
care team to deliver high-quality multidimensional care.
Enhancing the patient-clinician interaction is another aspect
of best practice cancer care organizations.
The electronic platforms that serve as the basis for ePRO
acquisitions are being expanded to address the growing
needs of the cancer patient. Flexible system design can begin
to deliver functionality beyond information gathering.
Configured with advanced scheduling functions, branching
question capabilities, easy loading of additional surveys, and
two-way communication loops between patient and practice, patient engagement platforms can facilitate patient
decision support and shared decision making. These platforms can triage patients automatically to appropriate
members of the care delivery team with minimal interruption of normal clinic processes.
FIGURE 2. Workflow for Smoking Cessation Program
Tobacco Use Cessaon

PCM Screening
PCM
PCM will present NEW
paents with quesons
about tobacco use at the
rst visit following NCCN
guidelines for smoking
cessaon.
PCM will present all other
paents the tobacco use
quesons on a rolling basis
over the first year to create
a sustainable work ow.
May we contact you to

support your eorts to quit
using tobacco?
Yes
No
If YES, paent will be contacted via

phone, or e-mail or mail and will be provided
with resources via phone, email ,or postal
mail
Invitaon to aend weekly informaon
sessions to receive informaon about
smoking, resources, and support
Screen for Case Finding
Currently use Tobacco

Frequency and quanty
Of tobacco use
Ready to Quit
Individual assessments
The opportunity to have an individual
session where personalized quit plans
will be provided
Informaon on pharmacotherapy
Resources on Quit lines
A dedicated email and phone will for the
smoking cessaon team will be provided as
a way for those interested to contact for
addional informaon
Request for Informaon
The workflow is based on patient engagement platforms to survey patients on tobacco use.
Abbreviation: PCM, Patient Care Monitor.
e94
A Psycho-oncology Behavioral
Health Iniave
If NO, quesons will be
presented again in 6
months and at 1 year.
Our institution has recently launched a smoking cessation

project that identifies patients through a series of questions
on the PCM about their current and past tobacco use statuses and their readiness to quit (Fig. 2). If the patient is
smoking but does not self-identify as ready to quit, they will
be scheduled to be automatically recontacted at 6-month
intervals. If ready to quit and requesting more information,
an email message is sent to members of the psychology
team, who then contact the patient and enroll them in an
evidence-based smoking cessation program. Tobacco use
habits can also be appropriately coded to trigger email alerts
to clinicians for consideration of low-dose CT screening for
early detection of lung cancer.
With the advent of mobile technology, studies have addressed postoperative symptom severity after cancer surgery at home, and an automated system has shown
superiority in controlling symptom threshold events in a
randomized trial.29 Multiple pilot studies are underway to
address adherence to oral medications, such as aromatase
inhibitors in early-stage breast cancer and tyrosine kinase
use in advanced colorectal cancer, through frequent
mobile surveying of emerging symptoms with triggers of
emails/texts to nursing or navigator staff, who then contact

the patient directly. The goal is to alleviate severe toxicities
that emerge between visits by using electronic communication with the platform.
CONCLUSION
ePROs have emerged as an effective method of capturing
patient symptoms that can lead to a more thorough and
consistent evaluation of patient physical and emotional
health status and QOL. Numerous studies have validated
multiple assessment instruments and platforms that can
now be used in routine clinical practice. New data support an
ePRO solution to collecting patient symptoms consistent
with regulatory criteria in the clinical trial assessment of new
therapies. The widespread use of mobile devices provides an
excellent opportunity to expand patient engagement beyond the point of care to routinely monitor patients for
emerging problems with minimal investment of time and
resources by the care delivery system, thus enhancing the
overarching goal of high-quality patient-centered cancer
care.
References
1. Forcina JM. Re: National Institutes of Health State-of-the-Science
Conference Statement: Symptom Management in Cancer: Pain, Depression, and Fatigue, July 15-17, 2002. J Natl Cancer Inst. 2004;96:
1109-1110, author reply 1110.
2. Gwaltney CJ, Shields AL, Shiffman S. Equivalence of electronic and
paper-and-pencil administration of patient-reported outcome measures: a meta-analytic review. Value Health. 2008;11:322-333.
3. Basch E, Iasonos A, McDonough T, et al. Patient versus clinician
symptom reporting using the National Cancer Institute Common Terminology Criteria for Adverse Events: results of a questionnaire-based
4. Cirillo M, Venturini M, Ciccarelli L, et al. Clinician versus nurse symptom
reporting using the National Cancer Institute Common Terminology
Criteria for Adverse Events during chemotherapy: results of a comparison based on patients self-reported questionnaire. Ann Oncol.
2009;20:1929-1935.
5. Di Maio M, Basch E, Bryce J, et al. Patient-reported outcomes in the
evaluation of toxicity of anticancer treatments. Nat Rev Clin Oncol.
Epub 2016 Jan 20.
6. Bergeson SC, Dean JD. A systems approach to patient-centered care.
JAMA. 2006;296:2848-2851.
7. Jensen RE, Snyder CF, Abernethy AP, et al. Review of electronic patientreported outcomes systems used in cancer clinical care. J Oncol Pract.
2014;10:e215-e222.
8. Basch E, Abernethy AP. Supporting clinical practice decisions with realtime patient-reported outcomes. J Clin Oncol. 2011;29:954-956.
9. Fortner B, Okon T, Schwartzberg L, et al. The Cancer Care Monitor:
psychometric content evaluation and pilot testing of a computer administered system for symptom screening and quality of life in adult
cancer patients. J Pain Symptom Manage. 2003;26:1077-1092.
10. Fortner B, Baldwin S, Schwartzberg L, et al. Validation of the Cancer Care
Monitor items for physical symptoms and treatment side effects using expert
oncology nurse evaluation. J Pain Symptom Manage. 2006;31:207-214.
11. Schwartzberg L, Fortner B, Houts A. Use of technology to enhance

doctor-patient interactions. Am Soc Clin Oncol Educ Book. 2007;27:686691.
12. Abernethy AP, Herndon JE II, Wheeler JL, et al. Feasibility and
acceptability to patients of a longitudinal system for evaluating cancerrelated symptoms and quality of life: pilot study of an e/tablet datacollection system in academic oncology. J Pain Symptom Manage. 2009;
37:1027-1038.
13. Abernethy AP, Zafar SY, Uronis H, et al. Validation of the Patient Care
Monitor (Version 2.0): a review of system assessment instrument for
cancer patients. J Pain Symptom Manage. 2010;40:545-558.
14. Samsa GP, Wolf S, LeBlanc TW, et al. An exploratory factor analysis of
the scale structure of the Patient Care Monitor Version 2.0. J Pain
Symptom Manage. 2016;51:776-783.
15. Kotronoulas G, Kearney N, Maguire R, et al. What is the value of the
routine use of patient-reported outcome measures toward improvement of patient outcomes, processes of care, and health service
outcomes in cancer care? A systematic review of controlled trials. J Clin
Oncol. 2014;32:1480-1501.
16. Berry DL, Blumenstein BA, Halpenny B, et al. Enhancing patient-provider
communication with the electronic self-report assessment for cancer:
a randomized trial. J Clin Oncol. 2011;29:1029-1035.
17. Basch E, Deal AM, Kris MG, et al. Symptom monitoring with patientreported outcomes during routine cancer treatment: a randomized
controlled trial. J Clin Oncol. 2015;34:557-565.
18. Temel JS, Greer JA, Muzikansky A, et al. Early palliative care for patients with
metastatic nonsmall-cell lung cancer. N Engl J Med. 2010;363:733-742.
19. Lipscomb J, Gotay CC, Snyder CF. Patient-reported outcomes in cancer:
a review of recent research and policy initiatives. CA Cancer J Clin. 2007;
57:278-300.
20. Basch E, Abernethy AP, Mullins CD, et al. Recommendations for incorporating patient-reported outcomes into clinical comparative effectiveness research in adult oncology. J Clin Oncol. 2012;30:4249-4255.
e95
LEE SCHWARTZBERG
21. Di Maio M, Gallo C, Leighl NB, et al. Symptomatic toxicities experienced

during anticancer treatment: agreement between patient and physician
reporting in three randomized trials. J Clin Oncol. 2015;33:910-915.
22. Fromme EK, Eilers KM, Mori M, et al. How accurate is clinician reporting
of chemotherapy adverse effects? A comparison with patient-reported
symptoms from the Quality-of-Life Questionnaire C30. J Clin Oncol.
2004;22:3485-3490.
23. Dueck AC, Mendoza TR, Mitchell SA, et al; National Cancer Institute
PRO-CTCAE Study Group. Validity and Reliability of the US National
Cancer Institutes patient-reported outcomes version of the Common
Terminology Criteria for Adverse Events (PRO-CTCAE). JAMA Oncol.
2015;1:1051-1059.
24. Schwartzberg LS, Cobb P, Walker MS, et al. Patient and practice impact
of capecitabine compared to taxanes in first-/second-line chemotherapy for metastatic breast cancer. Support Care Cancer. 2009;17:
1081-1088.
e96
25. Walker MS, Masaquel AS, Kerr J, et al. Early treatment discontinuation
and switching in first-line metastatic breast cancer: the role of patientreported symptom burden. Breast Cancer Res Treat. 2014;144:673-681.
26. Stepanski EJ, Houts AC, Schwartzberg LS, et al. Second- and third-line
treatment of patients with non-small-cell lung cancer with erlotinib in
the community setting: retrospective study of patient healthcare utilization and symptom burden. Clin Lung Cancer. 2009;10:426-432.
27. Walker MS, Reyes C, Kerr J, et al. Treatment patterns and outcomes
among patients with metastatic melanoma treated in community
practice. Int J Dermatol. 2014;53:e499-e506.
28. Levit L, Balogh E, Nass S, et al. Delivering High-Quality Cancer Care:
Charting a New Course for a System in Crisis. Washington, DC, Institute
of Medicine, National Academies Press; 2013.
29. Cleeland CS, Wang XS, Shi Q, et al. Automated symptom alerts reduce
postoperative symptom severity after cancer surgery: a randomized
controlled clinical trial. J Clin Oncol. 2011;29:994-1000.
ADVANCED PRACTITIONERS IN ONCOLOGY CARE
Oncology Advanced Practitioners Bring Advanced Community

Oncology Care
Wendy H. Vogel, MSN, FNP, AOCNP
OVERVIEW
Oncology care is becoming increasingly complex. The interprofessional team concept of care is necessary to meet projected
oncology professional shortages, as well as to provide superior oncology care. The oncology advanced practitioner (AP) is a
licensed health care professional who has completed advanced training in nursing or pharmacy or has completed training
as a physician assistant. Oncology APs increase practice productivity and efficiency. Proven to be cost effective, APs may
perform varied roles in an oncology practice. Integrating an AP into an oncology practice requires forethought given to the
type of collaborative model desired, role expectations, scheduling, training, and mentoring.
uperior oncology care requires a superior team. As oncology care becomes increasingly complex, the interprofessional team concept and collaborative care is no
longer an option but a necessity. A team is a group of two or
more people who function interdependently to achieve a
shared goal. Oncology APs are an integral part of the oncology team.
DEFINITION OF ADVANCED PRACTITIONER

An AP is defined as a licensed health care professional who
has completed advanced training in nursing (such as a nurse
practitioner or clinical nurse specialist) or pharmacy or has
completed training as a physician assistant.1 Table 1 lists the
various APs and educational requirements.
2,400 pharmacists who specialized in oncology.1 The National Commission on Certification of Physician Assistants4
Statistical Profile of Recently Certified Physician Assistants
noted over 102,000 certified physician assistants in the
United States, across all heath fields. Data extrapolated from
the American Academy of Physician Assistants Physician
Assistants Annual Survey Report estimate that about 2.4% of
physician assistants work in oncology (approximately 2,000).
Based on these data, there could be over 11,000 oncology
APs, but it is impossible to give an exact number because
some oncology APs may belong to multiple societies, not all
APs have certification in oncology, and not all APs are
members of one of the previously mentioned societies.
SCOPE OF PRACTICE
STATISTICS
The American Society of Clinical Oncologys (ASCO) The State
of Cancer Care in America: 2015 notes that the numbers of
APs in oncology are increasing. This may assist in meeting the
growth in demand for oncologists despite the fact that the
number of oncologists remains static.2 However, it is difficult
to estimate the exact number of APs in oncology. The 28th
Annual Advanced Practice Registered Nurse Legislative
Update3 notes there are more than 306,000 advanced
practice nurses in the United States, across all medical fields.
ASCO2 estimates that there are about 3,000 APs working in
oncology and that this number is growing. The Oncology
Nursing Society counts 2,601 nurse practitioners (NPs) and
1,173 clinical nurse specialists in the membership.1 The
Hematology/Oncology Pharmacy Organization notes about
Advanced practitioners legal scope of practice varies from

state to state,3 as noted in Table 1. The scope of practice may
also differ because of the APs type, level of experience,
and facility.1,5 The majority of APs have prescriptive
privileges, with differing requirements/rules for controlled
substances.3
COLLABORATIVE CARE MODELS IN ONCOLOGY

The majority of oncology care (80%) is provided in the
community setting.6 With the expected shortfall of oncologists by 2020, collaborative practice will be required to
meet the demands of oncology care.7,8 Collaborative
practice requires the involvement of all interdisciplinary
team members to maximize positive patient outcomes.1
Advanced practitioners are an integral part of the oncology
From the Wellmont Cancer Institute, Kingsport, TN.

Corresponding author: Wendy H. Vogel, MSN, FNP, AOCNP, WMA Hematology & Oncology at Kingsport, Wellmont Cancer Institute, 4485 West Stone Dr., Suite 200, Kingsport, TN
37660; email: wendyvogel55@gmail.com.
e97
WENDY H. VOGEL
TABLE 1. Education, Licensure, Professional Organizations, and Certification of Advanced Practitioners in Oncology
Advanced
Practice Role
Entry-Level
Degree
Board Certification
and Certifying Body
Professional
Organizations
Prescriptive
Authority Eligibility*
Nurse
Practitioner
Doctorate (DNP or PhD)

preferred, previously
masters level or
postmasters
certificate
State Board of Nursing

NP: adult, pediatric,
acute care, primary
care, family practice
General certification: American

Nurses Association and State
Board of Nursing Oncology
Certification: AOCNP-ONCC
AOCN-ONCC
AANP, ANA,
APSHO, ONS
Scope defined by
individual states
Physician
Assistant
Masters degree
State Board requires

PA National Certifying
Examination,
administered by NCCPA
PA-C, NCCPA
AAPA, APAO,
APSHO
Yes, requires
supervising
physician
agreement
Clinical
Pharmacist
4-year professional
degree (Doctor
of Pharmacy)
State Board
of Pharmacy
The Board of Pharmacy Specialties

provides testing and maintains
the Board Certified Oncology
Pharmacist status
ACCP, APSHO,
ASHP, HOPA
Scope defined by
individual states
Clinical Nurse
Specialist
Masters degree
State Board of Nursing

Advanced Practice Nurse
State Board of
Nursing: AOCNS-ONS
APSHO, ONS
Scope defined by
individual states**
Licensure
Abbreviations: DNP, Doctor of Nursing Practice; AOCNP, advanced oncology certified nurse practitioner; ONS, Oncology Nursing Society; AOCN, advanced oncology certified nurse;
AANP, American Academy of Nurse Practitioners; ANA, American Nurses Association; APSHO, Advanced Practitioner Society for Hematology and Oncology; NCCPA, National
Commission on Certification of Physician Assistants; APAO, Association of Physician Assistants in Oncology; AAPA, American Academy of Physician Assistants; ACCP, American College
of Clinical Pharmacy; ASHP, American Society of Health-System Pharmacists; HOPA, Hematology/Oncology Pharmacy Organization; AOCNS, advanced oncology clinical nurse
specialist.
*Variability by state.
**From the National Council of State Boards of Nursings APRN campaign for Consensus: Moving Toward Uniformity in State Laws (https://www.ncsbn.org/5410.htm).
Adapted from Vogel5 and from Kurtin SE et al.1
team. The ASCO Workforce Advisory Group suggested that

increasing the use of and maximizing the potential of APs
would improve practice efficiency.7 Since this 2007 publication, other studies and publications have examined various collaborative oncology practice models.8
Three collaborative practice models have been described in
oncology practices: the incident-to visit model, the shared
visit model, and the independent visit model (Table 2).8,9 The
ASCO study of collaborative practice arrangements8 noted
that more than 75% of practices use the incident-to visit
model. This is likely to maximize reimbursement. Most of the
practices in this study (73%) were physician-owned private
KEY POINTS
e98
An advanced practitioner is defined as a licensed health

care professional who has completed advanced training
in nursing or pharmacy or has completed training as a
physician assistant.
Advanced practitioner legal scope of practice varies
from state to state, but most have prescriptive
privileges, with differing requirements/rules for
controlled substances.
Three collaborative care models are identified in
oncology practices: incident-to visit model, the shared
visit model, and the independent visit model.
Barriers to efficient collaborative practice include
disallowing the AP the full scope of practice and
inadequate training and mentoring.
Advanced practitioners are cost effective and an integral
part of the future of oncology care.
practices. The collaborative model and billing practices in

academic oncology practices are likely to be different.10 In
the ASCO study, both the incident-to model and the independent model consist of the physician and AP working
independently; however, the primary difference is billing, the
amount of consultation required, and the autonomy of the
AP. Other disciplines, such as primary care, describe practice
models of the AP (an NP) working completely independent
of a physician. Currently in 15 states, NPs are regulated by a
Board of Nursing and have full autonomous practice and
prescriptive authority without requirement for physician
supervision, delegation, consultation, or collaboration.3
In some oncology practices, the AP works exclusively with
one oncologist; in other practices, the AP works with all
practice oncologists.8 The ASCO Study of Collaborative
Practice agreements noted that about 60% of APs work with
all practice oncologists and not exclusively with one oncologist. Interestingly, the practices with APs working with
all practice oncologists had higher productivity.8
When choosing a collaborative practice model, the
practice needs and patient volume must be considered, as
well as both the physician and APs training, strengths, skills,
and expert knowledge. The ability to meet regulatory
constraints for AP billing must be carefully considered to
avoid fraud. If the AP is new to oncology, a thorough training
and mentoring program must be planned.1,5,11
Barriers to efficient practice must also be considered. A
recent survey of APs in oncology noted the most common
barrier to the AP role was time spent on tasks that could
be delegated to non-AP staff.1,12 Other common barriers
included insufficient support, mentoring, and insufficient
autonomy and training to meet role expectations. Most APs
are trained as generalists with limited time devoted to
ADVANCED PRACTITIONERS IN ONCOLOGY CARE
TABLE 2. Collaborative Practice Models8-10,18

Model
Description
Incident-To (Mixed) Visit Model
AP sees patient independently, following physician plan of care; physician consult available if needed;
physician generally in office suite; billing is incident-to at full physician fee schedule; however,
must meet CMS regulations on alternate visit schedules
Shared Visit Model
Both the AP and the physician see patient. The physician completes the visit note and bills for services.
Independent Visit Model
AP sees patient independently, consulting at critical treatment decision points (such as initial treatment
plans, at progression, end-of-life decisions); billed under NPP own provider number, at 85% of
physician fee schedule
Abbreviations: AP, advanced practitioner; CMS, Center for Medicaid & Medicare Services; NPP, nonphysician provider.
oncology.1,13 In a survey of physician assistants in oncology

practices,13 the majority reported that it took 1 to 2 years of
practice to become fully competent in the practice of
oncology. PAs in this survey also described the most important mode of obtaining their oncology knowledge base
was direct physician mentoring. The second most important
mode was self-study. A similar study of oncology NPs reported that the majority of NPs felt inadequately prepared to
provide cancer care.14 These barriers are substantial, noting
that most oncology practices use informal, on-the-job
training for new APs.8 Sidebar 1 lists organizations that
provide AP oncology education.
Adequate and thorough training and direct mentoring by
the oncologist would enable the oncology AP to become
much more efficient and productive. Ensuring that the AP is
functioning at their highest scope of practice will maximize
the skills of the AP, allowing maximum productivity. It is
important to understand that adequate AP productivity
measurements are lacking. To only measure AP productivity
by relative value units (RVUs) will underestimate the value
the AP brings to a practice.15
practice and prescriptive authority in some states.5 Physicians and APs must work together to change restrictive
legislation.
MYTHS DEBUNKED
There was a fear among physicians that patients might not
be aware that their care was provided by an AP. However,
the ASCO Study of Collaborative Practice Arrangements8
found that 98% of patients with cancer were aware that an
AP was providing their care. There was also a belief that
patients would be unhappy with the care of the AP. In fact,
SIDEBAR 2. Oncology Advanced Practitioner

Functions and Roles*1,5,8,13
ROLES OF ADVANCED PRACTITIONERS IN

ONCOLOGY
Advanced practitioners in oncology perform a variety of
functions (Sidebar 2). Roles vary according to legal scope of
practice, practice policy, and experience/expertise of the
AP. To maximize role productivity and efficiency, the AP
must be allowed to work at their full scope of practice as
independently as possible. The biggest obstacle to the most
effective and efficient utilization of the AP is the lack of full
SIDEBAR 1. Organizations Providing Oncology

Education for the Advanced Practitioner
Association of Physician Assistants in Oncology

(www.apao.org)
American Society of Clinical Oncology (www.asco.org)
Advanced Practitioner Society for Hematology and Oncology (www.apsho.org)
National Cancer Institute (www.cancer.gov)
Oncology Nursing Society (www.ons.org)
History taking
Physical examination
Diagnostic testing: ordering, interpretation, discussion
Treatment selection
Prescribing (including chemotherapy)
Cancer risk assessment, screening, and management
Symptom management
Medication management
In-patient hospital rounds
On-call services
Patient education
Staff/peer education
Survivorship care
Palliative care
Hospice care
Pain management
Research (including roles as principal or subinvestigator)
Emergent care
Case management
Cancer patient navigator
Genetic services
Lung nodule clinic
Community education
Faculty positions outside institution of employment
Procedures including (but not limited to) bone marrow
aspiration/biopsy, paracentesis, thoracentesis, central line
placement, lumbar puncture, intrathecal chemotherapy
administration, punch biopsy, suturing, surgical assist
*This list is not all inclusive.
e99
WENDY H. VOGEL
this same study found that 92.5% of patients were extremely

satisfied with AP services. Other disciplines have documented similar results.16,17
One of the most common reasons that oncology practices
have cited as reasons for not adding an AP to the practice is
physician lack of interest in working with an AP.8 However,
data show that not only is practice productivity increased
with the addition of an AP, professional satisfaction of the
collaborative oncologist increases as well.8,15
There is also misunderstanding about the cost effectiveness
of an AP, especially in terms of AP billing. APs can see and bill
for new patients and can bill all levels of services. Data
consistently show the cost effectiveness of the AP.16,17
INTEGRATING THE ADVANCED PRACTITIONER

INTO THE ONCOLOGY TEAM
Oncology team members have a shared goal of improving
patient outcomes. Other goals of collaboration might include increased productivity of the practice, improved patient access to timely care, among others.18 Successful
collaboration takes time and deliberate planning to fully
engage each members unique strengths and skills. Measurement of productivity must include not only individual
RVUs, but also team-based RVUs (combined physician and
AP RVUs). Productivity measures should also include valueadded functions (that might be nonrevenue generating),
patient satisfaction, and professional satisfaction.
Proactive and thoughtful scheduling of patient visits will
enable the oncologist to offer new patients rapid access.18
Planning for urgent care time slots decreases the use of the
emergency department, thus increasing patient satisfaction.
Efficient scheduling takes planning in advance and thorough
training of scheduling staff and patient education, as well as
the flexibility of the AP and oncologist.
Adequate training and mentoring of the AP will enable the
highest level of productivity in the timeliest manner. This will
also lower burnout rates and increase professional satisfaction. Thoughtful preplanning for this education and
mentoring is required. It is also important to plan for the
lengthy time period that could be required for license and
credentialing. Ongoing education is critical for the AP to stay
current in this rapidly changing field.
CONCLUSION
Collaborative care is required to meet the increasing demands of oncology care. The oncology AP is a cost-effective,
integral part of the professional oncology team.
References
1. Kurtin SE, Peterson M, Goforth P, et al. White Paper: The Advanced
Practitioner and Collaborative Practice in Oncology. J Adv Pract Oncol.
2015;6:515-527.
2. American Society of Clinical Oncology. The State of Cancer Care
in America: 2015. http://www.asco.org/sites/www.asco.org/files/
2015ascostateofcancercare.pdf. Accessed January 16, 2016.
3. Phillips SJ. 28th annual APRN legislative update: advancements continue for APRN practice. Nurse Pract. 2016;41:21-48.
4. National Commission on Certification of Physician Assistants. 2014
Statistical Profile of Recently Certified Physician Assistants. https://
www.nccpa.net/Uploads/docs/RecentlyCertifiedReport2014.pdf.
5. Vogel WH. Advanced practitioners in oncology: meeting the challenges.
J Adv Pract Oncol. 2010;1:13-18.
6. Levit L, Smith AP, Benz EJ, et al. Ensuring quality cancer care through the
oncology workforce. J Oncol Pract. 2010;6:7-11.
oncologists : challenges to assuring access to oncology services. J Oncol
Pract. 2007;3:79-86.
8. Towle EL, Barr TR, Hanley A, et al. Results of the ASCO Study of Collaborative Practice Arrangements. J Oncol Pract. 2011;7:278-282.
9. Buswell LA, Ponte PR, Shulman LN. Provider practice models in ambulatory oncology practice: analysis of productivity, revenue, and provider
and patient satisfaction. J Oncol Pract. 2009;5:188-192.
10. Hinkel JM, Vandergrift JL, Perkel SJ, et al. Practice and productivity of
physician assistants and nurse practitioners in outpatient oncology
e100
11.
12.
13.
14.
15.
16.
17.
18.
clinics at national comprehensive cancer network institutions. J Oncol

Pract. 2010;6:182-187.
Kurtin S, Viale P, Hylton H, et al. The APSHO practice survey. Paper
presented at: 3rd JADPRO LIVE at Advanced Practitioner Society for
Hematology and Oncology Annual Meeting; November 2015; Phoenix,
AZ Poster JL316.
McCorkle R, Engelking C, Knobf MT, et al. Transition to a new cancer
care delivery system: opportunity for empowerment of the role of the
advanced practice provider. J Adv Pract Oncol. 2012;3:34-42.
Ross AC, Polansky MN, Parker PA, et al. Understanding the role of
physician assistants in oncology. J Oncol Pract. 2010;6:26-30.
Nevidjon B, Rieger P, Miller Murphy C, et al. Filling the gap: development of the oncology nurse practitioner workforce. J Oncol Pract.
2010;6:2-6.
Kosty M, Acheson AK, Tetzlaff ED. Clinical oncology practice 2015:
preparing for the future. Am Soc Clin Oncol Educ Book. 2015;35:e622e627.
Martin-Misener R, Harbman P, Donald F, et al. Cost-effectiveness of
nurse practitioners in primary and specialised ambulatory care: systematic review. BMJ Open. 2015;5:e007167.
Donald F, Kilpatrick K, Reid K, et al. A systematic review of the costeffectiveness of nurse practitioners and clinical nurse specialists: what is
the quality of the evidence? Nurs Res Pract. 2014;2014:896587.
Shulman LN. Efficient and effective models for integrating advanced
practice professionals into oncology practice. Am Soc Clin Oncol Educ
Book. 2013;33:e377-e379.
Quality and Value: Measuring and

Utilizing Both in Your Practice
CHAIR
Lowell E. Schnipper, MD
Beth Israel Deaconess Medical Center
Boston, MA
SPEAKERS
Anne C. Chiang, MD, PhD
Yale Cancer Center
New Haven, CT
Steven L. DAmato, BCOP, RPh
New England Cancer Specialists
Scarborough, ME
ANNE C. CHIANG
Why the Quality Oncology Practice Initiative Matters: Its Not

Just About Cost
Anne C. Chiang, MD, PhD
OVERVIEW
The nature and cost of cancer care is evolving, affecting more patients and often involving expensive treatment options. The
upward cost trends also coincide with a national landscape of increasing regulatory mandates that may demand improved
outcomes and value, but that often require significant up-front investment in infrastructure to achieve safety and quality.
Oncology practices participating in the American Society of Clinical Oncology (ASCO) Institute for Qualitys Quality Oncology
Practice Initiative (QOPI) and the QOPI Certification Program (QCP) continue to grow in number and reflect changing
demographics of the provision of cancer care. QOPI and QCP benchmarking can be used to achieve quality improvement and
to build collaborative quality communities. These programs may be useful tools for oncology practices to comply with new
legislation such as the Medicare Access and CHIP Reauthorization Act (MACRA).
he cost of cancer in the United States has risen significantly over the past decade. According to the Centers
for Disease Control and Prevention, 8.5% of the adult U.S.
population has been diagnosed with cancer, totaling over
20 million patients in 2012 and more than 29 million
ambulatory visits to physician offices, hospital outpatient
offices, and emergency departments.1 The estimated direct
medical costs for cancer care in the United States in 2011
were $88.7 billion, with 50% of that cost for physician or
hospital outpatient office visits, 35% for inpatient stays, and
11% for prescription drugs.2 Projections of growth from 2010
to 2020 based on 13.8 and 18.1 million cancer survivors, respectively, estimated the costs associated with cancer care to
be $124.57 and $157.77 billion, respectively, reflecting a 27%
increase in medical costs.3
The increase in costs is likely multifactorial, involving
more patients, more expensive drugs, and changes in
cancer care delivery. The upward cost trends also coincide
with a national landscape of increasing government and
regulatory mandates that often require significant upfront investment in equipment and infrastructure to
achieve. Such government mandates include the American Recovery and Reinvestment Act of 2009 enacted on
February 17, 2009, that established incentive payments to
promote the adoption and meaningful use of qualified
electronic health records (EHRs) and interoperable health
information technology.4 The U.S. Pharmacopeia Chapter
797 describes enforceable sterile compounding standards
that apply to chemotherapy preparation, with even more
comprehensive detail in the recently released Chapter 800

regarding not only storage and preparation but also the
administration of hazardous drugs.5 The American College
of Surgeons has recently mandated implementation of
distress screening and survivorship care plans for hospitals
to achieve and maintain accreditation of their cancer
programs.6
More recently, MACRA was passed to reform Medicare
payments. This act is intended to focus on payment for value
and better care by incentivizing providers to participate in
quality programs through either a Merit-Based Incentive Payment System (MIPS) or Alternative Payment Model (APM). The
MIPS program involves three components: the physician
quality reporting system (PQRS), the value modifier (valuebased payment modifier), and the Medicare EHR incentive
program. Alternative Payment Models include accountable
care organizations, patient-centered medical homes, and
bundled payment models.7
In such an evolving landscape, it is critical to foster a
culture of self-examination and quality improvement to
maintain a focus on achieving quality and patient-centered
care. To this end, the principles and evolution of ASCOs
quality programs, QOPI and QCP, make them attractive tools
for practices to comply with legislation, such as MACRA, that
require not only benchmarking of standards but also built-in
quality improvement mechanisms. QOPI is deemed by the
Centers for Medicare & Medicaid Services (CMS) as a
Qualified Clinical Data Registry (QCDR), with clinical measures that are highly relevant to contemporary ambulatory
From the Yale University School of Medicine, New Haven, CT.

Corresponding author: Anne C. Chiang, MD, PhD, Yale University School of Medicine, 200 York St., New Haven, CT 06520; email: anne.chiang@yale.edu.
e102
WHY QOPI MATTERS: ITS NOT JUST ABOUT COST
practice.8 Participants abstract and submit data for a set of

core measures as well as additional modules (e.g., diseasespecific) that they select. They receive detailed reports of
their performance on quality metrics, which allows them to
compare with national benchmark data, to identify areas for
improvement and develop focused improvement plans.
While QOPI focuses on quality metrics, QCP validates a
practices processes, demonstrating to patients, payers, and
colleagues a commitment to quality. Through practice and
policy review, including an on-site survey, QCP verifies
practice standards for safe administration of chemotherapy.
Some health plans recognize QOPI certification. QCP uses
national benchmarking standards developed by ASCO, the
Oncology Nursing Society (ONS), and other organizations
that safeguard the practice and patients.
CURRENT STATE OF QOPI

QOPI has been ASCOs signature quality program for almost
10 years, and since 2006, over half of all U.S.-based medical
oncologists have participated, numbering over 1,000 practices representing 7,000 oncologists. In addition, international
practices within the European Union, Argentina, Brazil, India,
and Saudi Arabia were recently allowed to participate in QOPI,
with 16 practices participating since fall 2015. A total of 994
practices submitted over 32,000 charts for the three most
recent abstraction rounds (fall 2013, spring 2014, and spring
2015). These results were compared with data from 5 years
ago (fall 2009, spring 2009, and fall 2010) and demonstrate
notable changes in the demographics, the practice model of
employment, and the range of services provided (Table 1).
Of note, the QOPI fall 2014 round did not take place because
of a change in vendor.
Several interesting trends were noted when comparing
data from 2009 to 2010 with data from 2013 to 2015. First,
KEY POINTS
ASCOs quality programsQOPI and QOPI QCPare

attractive tools for practices to comply with new
legislation, such as MACRA, that require not only
benchmarking of standards but also built-in quality
improvement mechanisms.
Oncology practices participating in QOPI and QCP
continue to grow in number and reflect changing
demographics of the provision of cancer care.
QOPI and QCP benchmarking can be used to identify
areas for improvement, provide comparative metrics for
quality initiative projects, and build collaborative quality
communities.
QCP provides hands-on expert gap analysis assessment
of a practice and a road map toward quality through
ASCO certification status.
ASCO has multiple offerings as vehicles for diverse
practices to join a quality community working toward
improving cancer care.
the mean number of medical oncologists per practice increased from 6.7 to 7.8. The participating practices had
similar geographic distribution across the United States, but
overall, there was an increase in the number of practices that
are high volume (e.g., with more than 1,500 new patients
seen per year). Interestingly, the number of QOPI practices
with employed physicians increased from 15% (82) to 28.7%
(273), and the number of private independent physician
practices decreased from 63.8% (344) to 47.9% (455). The
percentage of practices identifying as academic full time
increased slightly from 7.2% (39) to 10.5% (100), while the
percentage of practices with academic affiliation was stable
at 7%. Roughly 6% of practices have a fellowship program.
QOPI participation correlates with improvement in scores
over time.9 The expectations of the nature of cancer care is
constantly changing, as evidenced by new standards mandated by the American College of Surgeons regarding distress screening and use of survivorship care plans as well as
the CMS requirement that practices deliver care using an
electronic medical record (EMR). These mandates parallel
the shift from oncologists being self-employed to employed;
hospital-affiliated practices certainly may have increased access to other services. In fact, patient access to these interdisciplinary services such as social work increased from 45%
(188) to 55.5% (466), nutrition from 38.2% (159) to 52.3%
(324), and genetic counseling from 31.9% (132) to 38.7% (324).
Furthermore, 90.7% (564) of practices now have an EMR.
FUTURE STATE OF QOPI

QOPI continues to offer new opportunities and enhancements in its services to aid practices in assessing quality and
to comply with new federal requirements. CMS has approved
QOPI as such a registry for oncologists to satisfy PQRS measures. Oncology practices registered with QOPI can fulfill CMS
PQRS requirements by manually or electronically submitting
data through QOPI for the oncology measures group or by
submitting data through QOPI via a QCDR reporting pathway,
which offers 20 measures chosen by ASCO.10 Since QOPI has
been deemed as a QCDR, it is anticipated that QOPI will
become the primary mechanism for ASCO members to report
their quality as required by MIPS and in APMs.
In fall 2015, ASCO began piloting eQOPI to decrease the
burden of manual chart abstraction by implementing
technology to automate data abstraction. The technology
enables the required data elements in the EHR to be mapped
and extracted (or in some cases, data can be pushed); this
allows the majority of the elements required for the clinical
quality measures to be captured automatically, leaving
only a small percentage for manual abstraction. The differences in the scope and reporting pathways for classic
QOPI and eQOPI are highlighted in Table 2. Many of the
performance measures in QOPI are already based on ASCO
clinical practice guidelines, and as guideline scope and
volume increase, so will the available measures yet to be
incorporated into QOPI, with an increased emphasis on outcomes measures and measures that reflect patient experience,
e103
ANNE C. CHIANG
TABLE 1. ASCO 2015 QOPI Practice Demographic Data

Fall 2009Fall 2010 (609)
Fall 2013Spring 2015 (994)
Mean (SD) or N (%)

Number of Medical Oncologists
Mean (SD) or N (%)
6.7
(10.0)
7.8
(13.6)
< 500
96
(31.8)
251
(31.5)
501-1,500
136
(45.0)
326
(40.8)
> 1,500
70
(23.2)
221
(27.7)
Midwest
198
(32.5)
302
(30.4)
Northeast
111
(18.2)
192
(19.3)
South
195
(32.0)
308
(31.0)
West
105
(17.3)
192
(19.3)
Academic Full Time
39
(7.2)
100
(10.5)
Private With Academic Affiliation
40
(7.4)
67
(7.1)
Employee
82
(15.2)
273
(28.7)
Private Independent
344
(63.8)
455
(47.9)
Number of New Patients
Census Region
Site Affiliation
34
(6.4)
55
(5.8)
Social Work
Fellowship Program
188
(45.1)
466
(55.5)
Dietician/Nutritionist
159
(38.2)
442
(52.3)
Genetic Counselor
132
(31.9)
324
(38.7)
Quality Improvement Training
320
(52.5)
Electronic Medical Records
564
(90.7)
as well as patient-reported outcomes. For instance, QOPI began

pilot testing value measures in 2015, based on lists of tests
and procedures ASCO developed as part of its participation
in the Choosing Wisely campaign that are not evidence-based.
As the technology that powers QOPI becomes more sophisticated to decrease the workload on practices required for
quality reporting, QOPI-based quality assessment will become
a key and value-added component of ASCOs rapid-learning
system, CancerLinQ, where abstraction of data elements from
the EHR will be fully automated.11,12
QOPI CERTIFICATION PROGRAM

Although QOPI focuses on data regarding individual patients
and providers only, certification of practices by ASCO
requires a minimum QOPI Quality Score, submission of
documentation for 20 standards, and an on-site survey.13
The certification process provides a comprehensive evaluation of patient care including examination of documentation, policies and procedures, and administration of
chemotherapy. A surveyor travels to multiple sites in the
practice and has the opportunity to observe and evaluate
processes directly. The surveyor and ASCO QCP staff have
the opportunity to share feedback and best practices between sites to help address standards that are initially not in
accordance. The benefits of participating in the QCP program include identifying areas for quality improvement and
achieving QOPI certification status by ASCO, as well as payor
benefits in some cases.
e104
A total of 298 practices have achieved QOPI certification,

out of 525 applicants representing 336 unique practices.
Subsequently, 189 practices have applied for recertification, with 96% receiving certification. Most certified practices
are small to midsized with one to three full-time equivalents
(FTEs; 30%) or four to seven FTEs (35%). The rest of the
practices are larger with seven to 15 FTEs (23%), 25 to 50
FTEs (4%), and even 50 to 100 FTEs (1%). Only 25% of the
practices identify themselves as private independent and
17% as academic full-time practices. The largest group of
practices (35%) identify themselves as employed, and an
additional 14% are hospital-employed. The practices are
distributed across the United States, with 23.4% in the
South, 28.2% in the East, 33.1% in the Midwest, and
15.2% in the West (Fig. 1; unpublished data, T. Gilmore,
ASCO).
In general, most practices performed well, with an average
of 4.1 missed standards out of 20. Eight standards demonstrated 100% concordance in the last round. In rounds 9,
10, and 11, the top six missed standards were qualifications,
chart documentation, double check in administration,
documentation for each visit, oral chemotherapy monitoring, and patient education. The top six most commonly met
standards include both process and policy measures: consent policy, double check of order, day of treatment assessments, referrals, toxicity assessment available prior to
writing chemotherapy order, and cumulative dosing (unpublished data, T. Conti-Kalchik, ASCO).
TABLE 2. Differences in Scope and Reporting

Pathways for Classic QOPI and eQOPI
Classic QOPI
191 Measures
7 Modules (Existing)
1 Module (New)
3 Palliative Care
3 Reporting Pathways (New)
3 QOPI Certification Program
3 PQRS Oncology Measures Group
3 Patient-Centered Oncology Practice
eQOPI
38 Measures
6 Modules (Existing)
2 Reporting Pathways (New)
3 QOPI Certification Program
3 Qualified Clinical Data Registry
ASCO continues to expand and improve the QCP program

by exploring medical liability insurance discounts and new
measures to respond to MACRA MIPS/APM requirements
and a pilot program to allow international practices to apply
for QOPI certification for the first time in 2016.
QOPI IN ACTION
QOPI and QCP have succeeded in spurring local quality
improvement efforts that have led to score improvements,
achievement of standards, increased discussion of quality,
and a positive effect on practices. To understand the effect

on practices, a survey in 2013 was sent to 1,450 participants
at 850 practices to assess which measure modules/standards
were selected by QOPI/QCP participants as the basis for local
quality initiative efforts and to understand the nature of the
improvement initiatives.14
Out of 89 respondents, 96% (85/89) reported that QOPI/
QCP led to quality improvement efforts. Respondents were
asked to select module(s) that spurred subsequent quality
improvement activities: core measures (57%; 45), symptom/
toxicity management (48%; 38), end-of-life care (38%; 29),
breast cancer (13%; 10), colorectal cancer (10%; 8), nonHodgkin lymphoma (6%; 5), and nonsmall cell lung cancer
(4%; 3). Related to the QCP structural safety standards,
participants reported quality improvement projects as follows: chemotherapy planning/chart documentation (39%;
31), general chemotherapy standards (30%; 24), monitoring
and assessment (29%; 23), chemotherapy administration
(27%; 21), chemotherapy orders (23%; 18), staffing (16%;
13), and drug preparation (9%; 7).
Practices reported that QOPI measures improved in
subsequent rounds as a result of specific projects (22/25;
88%); 100% felt that these quality improvement projects
affected their practices for the better. Quality improvement
project results were presented primarily in practice meetings (74%; 26), hospital or community forums (17%; 6), ASCO
Quality Care Symposium, or other meetings (9%; 3). No
projects reached publication. Of note, 17 of 31 respondents
who reported practice status indicated achieving QOPI
certification. Specific quality improvement projects included
building oral chemotherapy order sets and modified antiemetic regimens in chemotherapy protocols, improving
FIGURE 1. QOPI Certification Program Growth, December 2015
e105
ANNE C. CHIANG
documentation of distress, pain, performance status, smoking

status and referrals, chemotherapy consent, and intent of
treatment. Practices also improved safety by initiating
semiannual review and gap analysis of adherence to ASCO/
ONS Chemotherapy Safety Standards, improving the medication reconciliation process, and training for oral chemotherapy.
ASCO continues to build quality improvement resources
and practical training, such as the Quality Training Program,
launched in 2013 as part of its Institute for Quality.15 The
Quality Training Program uses in-person training sessions
and coaching to teach quality improvement curriculum and
skills throughout an improvement project and to develop
quality leaders and champion quality improvement. Since
2013, 60 individuals have participated in this program
(unpublished data, E. Holton, ASCO).
A QUALITY IMPROVEMENT COMMUNITY

QOPI has served as a platform to create a quality community
in two published reports by the Michigan Oncology Quality
Consortium and the National Cancer Institute Community
Cancer Centers Program (NCCCP). In each case, a network of
diverse practices agreed to benchmark their aggregate QOPI
and collaborate on quality improvement initiatives to improve quality of care and to identify best practices.16,17 The
NCCCP practices demonstrated a measurable increase in
each sites QOPI scores from fall 2010 to fall 2013, with the
exception of one parameter.
Another example illustrates how QOPI can be used to
achieve hospital integration of a network of community
practices. The Smilow Cancer Hospital provides cancer
services by Yale Cancer Center clinicians in New Haven,
Connecticut, or in one of nine Care Center community locations. Started in 2012 by the acquisition of community
practices, all Smilow Cancer Hospital Care Centers are fully
integrated practice sites using a provider-based model and
unified EMR baseline data across all centers. They created a
quality council to focus on clinical quality and patient safety
and to coordinate quality improvement activities. In spring
2012, 49% (217/445) of the patients at Smilow Cancer
Hospital Care Centers had documented emotional assessments, with compliance in two sites below 12%. Distress
screening was implemented in these two clinics through a
Quality Training Program pilot by using quality improvement
tools and subsequently expanded in 2015 to 2016 to all
ambulatory clinics throughout Smilow to comply with the

American College of Surgeons mandate for psychosocial
distress assessment.18 Other areas of noncompliance have
been identified and addressed. An oral chemotherapy adherence program including specialty pharmacy and EMR
workflows was recently established and implemented
through all clinics (unpublished data, K. Adelson). QOPI
certification was achieved for the Smilow Cancer Hospital
network and continues to spur quality improvement projects (e.g., pain assessment and immunotherapy toxicity
education and management).
CONCLUSION
QOPI and QCP continue to be ASCOs signature quality
programs, with continuing growth in the number and extent
of participating practices. With the increasing emphasis on
value because of the rising costs of cancer care, it is essential
to have constant development and validation of quality
measures that reflect the changing nature of oncology (e.g.,
oral chemotherapy adherence, Choosing Wisely measures,
and response to the demands of MACRA). QOPI benchmarking can be used to identify areas for improvement,
provide comparative metrics for quality improvement
projects, and build collaborative quality communities.
Widespread EMR adoption will allow eQOPI to help reduce
the time burden of abstraction, and delivery of the QOPI
quality assessments through CancerLinQ will take advantage
of the fully automated data abstraction capabilities of that
platform. QCP provides hands-on expert gap analysis assessment of a practice and a road map toward quality
through ASCO certification status. The ASCO Institute for
Quality also offers the Quality Training Program to help train
and coach interdisciplinary teams in quality improvement
initiatives and develop leadership. In summary, ASCO has
multiple offerings as vehicles for diverse practices to join a
quality community working toward improving cancer care.
ACKNOWLEDGMENT
I would like to acknowledge ASCO staff members Terry
Gilmore, Tara Conti-Kalchik, Elaine Holton, Stephanie Crist,
and Robert S. Miller, MD, for their help in providing data and
in reviewing the manuscript. I would to acknowledge Joe
Jacobson, MD, for his help in reviewing the manuscript.
References
1. Centers for Disease Control and Prevention. Cancer. http://www.cdc.
gov/nchs/fastats/cancer.htm. Accessed February 10, 2016.
2. American Cancer Society. Economic Impact of Cancer. http://www.
cancer.org/cancer/cancerbasics/economic-impact-of-cancer. Accessed
February 10, 2016.
3. Mariotto AB, Yabroff KR, Shao Y, et al. Projections of the cost of cancer care
in the United States: 2010-2020. J Natl Cancer Inst. 2011;103:117-128.
e106
4. Centers for Medicare & Medicaid Services. Electronic Health Records

(EHR) Incentive Programs. https://www.cms.gov/Regulations-and-Guidance/Legislation/EHRIncentivePrograms/index.html?redirect=/ehrincentiveprograms. Accessed February 10, 2016.
5. U.S. Pharmacopeial Convention. USP Compounding Standards & Resources.
http://www.usp.org/usp-healthcare-professionals/compounding. Accessed
February 10, 2016.
6. American College of Surgeons. Cancer Program Standards (2016

Edition). https://www.facs.org/quality%20programs/cancer/coc/
standards. Accessed February 10, 2016.
7. Hahn J, Blom KB. H.R. 2: The Medicare Access and CHIP Reauthorization Act
of 2015. Congressional Research Service. http://democrats.waysandmeans.
house.gov/sites/democrats.waysandmeans.house.gov/files/documents/
CRS%20report%20on%20HR%202.pdf. Accessed February 10, 2016.
8. Centers for Medicare & Medicaid Services. Qualified Clinical Data
Registry Reporting. https://www.cms.gov/medicare/quality-initiativespatient-assessment-instruments/pqrs/qualified-clinical-data-registryreporting.html. Accessed February 10, 2016.
9. Neuss MN, Malin JL, Chan S, et al. Measuring the improving quality of
outpatient care in medical oncology practices in the United States. J Clin
Oncol. 2013;31:1471-1477.
10. ASCO Institute for Quality. QOPI Certification Standards. http://www.
instituteforquality.org/qopi/qopi-certification-standards. Accessed February
10, 2016.
11. Jacobson JO, Neuss MN, Hauser R. Measuring and improving value of care in
oncology practices: ASCO programs from Quality Oncology Practice Initiative
to the rapid learning system. Am Soc Clin Oncol Educ Book. 2012;e70-e76.
12. Shah A, Stewart AK, Kolacevski A, et al. Building a rapid learning health
care system for oncology: why CancerLinQ collects identifiable health
information to achieve its vision. J Clin Oncol. 2016;34:756-763.
13. Gilmore TR, Schulmeister L, Jacobson JO. Quality Oncology Practice

Initiative Certification Program: measuring implementation of chemotherapy administration safety standards in the outpatient oncology
setting. J Oncol Pract. 2013;9(2 Suppl)14s-18s.
14. Chiang AC, McNiff K, Kadlubek P, et al. Assessment of quality improvement efforts by ASCOs Quality Oncology Practice Initiative
(QOPI) participants. J Clin Oncol. 2013;31:15s (suppl; abstr 55).
15. Kamal, AH, Quinn, D, Gilligan TD, et al. Feasibility and effectiveness of a
pilot program to facilitate quality improvement learning in oncology:
experience of the American Society of Clinical Oncology Quality Training
Program. J Oncol Pract. Epub 2015 Aug 18.
16. Blayney DW, Stella PJ, Ruane T, et al. Partnering with payers for success:
quality oncology practice initiative, blue cross blue shield of michigan,
and the michigan oncology quality consortium. J Oncol Pract. 2009;5:
281-284.
17. Siegel RD, Castro KM, Eisenstein J, et al. Quality improvement in the
national cancer institute community cancer centers program: the
quality oncology practice initiative experience. J Oncol Pract. 2015;11:
e247-e254.
18. Chiang AC, Buia Amport S, Corjulo D, et al. Incorporating patientreported outcomes to improve emotional distress screening and assessment in an ambulatory oncology clinic. J Oncol Pract. 2015;11:
219-222.
e107
The Oncology Care Model: The

Man Behind the Curtain
CHAIR
Christian A. Thomas, MD
New England Cancer Specialists
Scarborough, ME
SPEAKERS
Jeffery C. Ward, MD
Swedish Cancer Institute Edmonds
Edmonds, WA
Ronald Michael Kline, MD
Center for Medicare & Medicaid Innovation
Washington, DC
THE ONCOLOGY CARE MODEL: A CRITIQUE
The Oncology Care Model: A Critique

Christian A. Thomas, MD, and Jeffrey C. Ward, MD
OVERVIEW
Rapidly increasing national health care expenditures are a major area of concern as threats to the integrity of the health care
system. Significant increases in the cost of care for patients with cancer are driven by numerous factors, most importantly the
cost of hospital care and escalating pharmaceutical costs. The current fee-for-service system (FFS) has been identified as a
potential driver of the increasing cost of care, and multiple stakeholders are interested in replacing FFS with a system that
improves the quality of care while at the same time reducing cost. Several models have been piloted, including a Center for
Medicare & Medicaid Innovation (CMMI)sponsored medical home model (COME HOME) for patients with solid tumors
that was able to generate savings by integrating a phone triage system, pathways, and seamless patient care 7 days a week to
reduce overall cost of care, mostly by decreasing patient admissions to hospitals and referrals to emergency departments.
CMMI is now launching a new pilot model, the Oncology Care Model (OCM), which differs from COME HOME in several
important ways. It does not abolish FFS but provides an additional payment in 6-month increments for each patient on active
cancer treatment. It also allows practices to participate in savings if they can decrease the overall cost of care, to include all
chemotherapy and supportive care drugs, and fulfill certain quality metrics. A critical discussion of the proposed model,
which is scheduled to start in 2016, will be provided at the 2016 American Society of Clinical Oncology (ASCO) Annual Meeting
with practicing oncologists and a Centers for Medicare & Medicaid Services (CMS) representative.
ational health care expenditures (HCEs) are a substantial

part of the national spending and currently account for
17.1% of the U.S. gross domestic product, significantly more
than other developed countries, which typically spend between 8% and 12% of their gross domestic product on health
care. Despite high spending, several important health outcomes measuresfor example, infant mortalitydo not reflect the current level of health care spending, supporting the
notion that there is a disconnect between money spent and
the quality of care produced by the system. When national
health care systems are compared according to country, the
United States currently ranks 54th when factors such as efficiency, life expectancy, and health care cost are considered.1
A major concern is the significant growth rate of HCEs that
outpace other national financial benchmarks and continue
to rise at annual levels of greater than 5%. The inflation of
health carerelated expenditures is mainly driven by two
major factors: hospital-related expenditures and drug prices.
U.S. expenditures on pharmaceuticals are more than twice
that of any other member country of the Organisation for
Economic Cooperation and Development.2
Almost no other area of medicine is affected more by the
steep increase in the cost of pharmaceuticals than cancer
care where drug prices can easily top $10,000 per month per
patient.3
There is broad consensus that this trend is not sustainable

at the current rate. Private and public stakeholders are
discussing mechanisms to curb spending while at the same
time maintaining high-quality care. Because physician behavior can be influenced by financial incentives,4 it has been
postulated that tying payment directly to the quantity of
performed services, the backbone of the FFS system, incentivizes physicians to perform more services to improve
financial gain. The flipside of the argument is the Health
Maintenance Organization experience of the early 1990s
where FFS was replaced by a fixed monthly payment for each
patient but not tied to patient encounters or specific services. This led to inadequate access to care with poor patient
care and satisfaction.
So how is a rational system of providing the right health
care to the right patient at the right time to be designed?
If FFS is abandoned for an improved way to deliver and pay
for health care, the following elements have to be met:
1. A payment system that adequately covers the true
cost of carethe current Medicare fee schedule is
underfunding the cost of care by 20% to 30%, forcing
providers to use other resources to cover the cost of
care for Medicare beneficiaries such as increasing their
fees for private payers;
From the New England Cancer Specialists, Scarborough, ME; Swedish Cancer Institute, Edmonds, WA.
Corresponding author: Christian A. Thomas, MD, New England Cancer Specialists, 100 Campus Dr., Suite 108, Scarborough, ME 04074; email: thomac@newecs.org.
e109
THOMAS AND WARD
2. A system that allows providers and their staff to care

for patients in a variety of ways including phone calls,
emails, and other forms of contact in addition to
traditional face-to-face encounters;
3. A system that allows easy and unrestricted access for
patients;
4. A system that ties successful adherence to nationally
accepted quality standards and patient satisfaction
directly to physician reimbursement;
5. A system that provides a mechanism for controlling
the cost for drug and hospital spending.
THE ONCOLOGY CARE MODEL

Federal legislation now mandates CMS to fund alternate
payment models to deliver health care, and CMS has established the Center for Medicare & Medicaid Innovation
(CMMI) to explore options. From 2013 to 2015, a $20 million
grant was provided to 12 medical oncology practices for
the COME HOME project, which used three novel mechanisms to control the cost of care for patients with common
cancers:
1. The systematic use of a phone triage system to determine the acuity of a patients complaint and offer
them one of three options: advise over phone for
minor complaints, a same-day appointment in the
medical oncology office for more urgent issues, or a
direct referral to the nearest emergency department
for the sickest patients;
2. Extended office hours on evenings and weekends so
that patients requiring an assessment or office services
such as antiemetics, hydration, antibiotics, or other
injections would not have to visit a hospital or emergency department;
3. The use of treatment pathways that were created by
the participating practices to create predictable, costeffective, but high-quality care.
Although savings varied geographically, as a whole
participating COME HOME practices were able to show
substantial savings with lower costs largely attributable
to avoiding costly hospital admissions and emergency
KEY POINTS
e110
The rapidly increasing cost of health care threatens to

destabilize the health care system.
New systems are needed, and CMS has proposed a new
pilot model, the Oncology Care Model, that provides
payments to practices in addition to the traditional FFS if
practices can decrease overall cost of care and fulfill
quality criteria.
This article and its corresponding Education Session at
the 2016 ASCO Annual Meeting are aimed at providing
a critical overview of the model by practicing
physicians and a CMS representative.
department referrals for patients who could be adequately cared for in the office.5 This example of increasing
the value of care by reducing cost while at the same time
improving the quality of care and leading to higher patient
satisfaction is an encouraging example of positive health
care reform.
Unfortunately, CMMI decided to discontinue further funding for COME HOME and has presented a new model aimed
at reducing cost of care for medical oncology patients while
at the same time attempting to improve the quality of care,
the Oncology Care Model (OCM). This model uses two types
of additional payments in addition to the usual FFS payments and other payments, for example, for chemotherapy
administration: a performance-based incentive as well as
bundled monthly payments for patients undergoing chemotherapy. In exchange, oncology practices are obligated to
establish systems aimed at providing a higher level of care
with additional reporting requirements to CMMI. As detailed
further below, practices will receive an additional payment
of $160 per beneficiary per month (the bundle) for any 6month period during which a patient is undergoing active
chemotherapy treatment in a medical oncology practice.
In return, practices agree to strict reporting and quality
standards:
Provide patient navigation;
Document a care plan that contains the 13 components
as outlined by the Institute of Medicine;
Provide 24-hour-a-day, 7-day-a-week patient access to
an appropriate clinician who has real-time access to the
practices medical records;
Treat patients with therapies consistent with nationally
recognized clinical guidelines;
Use data to drive continuous quality improvement;
Use an ONC-certified electronic health record and attest
to stage 2 of meaningful use by the end of the third
model performance year.
In addition, private payers were invited to apply, extending
the potential reach of the project to patients not covered
by CMS.
The OCM is often characterized as an episode-of-care or
bundled reimbursement model but is in reality a FFS and
shared savings model. The shared savings are predicated
on a calculated bundle, but the reimbursement application
would more honestly be called an Oncology Accountable
Care Organization.
There are theoretical advantages to a true bundle for both
payers and providers. The payers costs are predictable for
an individual patient for a set period time and the economic
incentive toward high-cost, high-quantity care, inherent in
FFS medicine, is replaced by physician accountability for
costs and quality they can control. The provider has freedom
from payer micromanagement and, ideally, adequate and
predictable payments. Then, released from the constraints
of practicing to billing codes, practices have the flexibility
necessary to seek high-quality care with fewer resources
through innovative care redesign. But the key to these
advantages is that the bundle has replaced FFS reimbursement.6 Unfortunately, OCM fails to do that.
The Accountable Care Organization model has yet to
deliver on its promise of high-quality, low-cost care,7 perhaps because simply adding incentives to decrease costs on
top of the existing payment system that incents increased
costs does not solve the problem. At the best, it requires a
very complicated balancing act of incentives and quality
accountability. At its worst, it is the hope that two wrongs
make a right. Skeptics might suggest that participants,
constrained by continued dependence on FFS reimbursement from capitalizing on a redesign of the way they deliver
services, will be best served financially by collecting their FFS
billings and management fee, go through the motions of
CMS dictates, but avoid real innovative efforts and hope that
health care inflation and a bit of luck will result in some
additional booty.
LOGISTICAL HURDLES OF THE ONCOLOGY CARE

MODEL BUNDLE
However, it is the bundle construct that makes OCM both
novel and controversial. Additionally, because one can easily
envision that the CMMI pilot model is the precursor of a real
bundled reimbursement model, it requires careful scrutiny.
The OCM takes a simple approachall spending on all care
provided to the patient with a particular cancer is in a bundle
that starts with the initiation of chemotherapy and lasts for
6 months. This construct has raised a number of concerns.
The OCM bundle reimbursement is constructed based on
an individual practices historical costs when available and
on regional or national data adjusted for geographic variation when historical costs are not available. The focus on
competing against your own data guarantees that over time
opportunities for new cost sharing diminishes. Although
CMMI may not choose to ratchet down the reimbursement
for a bundle during the pilot model, the history of Medicares
oldest bundle, dialysis reimbursement, strongly suggests
that is exactly what will happen.
The choice of 6 months is arbitrary. In other specialties,
CMS and commercial payers have defined an episode of
care to be a full course of treatment of a specified treatment. In this scenario, adjuvant concurrent fluorouracil
(5-FU) and radiation for rectal cancer is a shorter episode and
FOLFOX (folinic acid, 5-FU, and oxaliplatin) for colon cancer
is a longer episode. In the OCM model, the first patient is
paid a 6-month bundle for 5 weeks of treatment and the
second patient is paid the same bundle for 24 weeks of
therapy. Unless, of course, the second patient has some
delays in treatment, then the practice may be paid two
6-month bundles for 27 weeks of therapy. A study commissioned by CMS and performed by the RAND Corporation
demonstrated that gaps of this sort in courses of the same
therapy are common. One can envision that the longer the
bundle the greater the variation of practice costs within a
bundle and greater the opportunity for selection bias, either
intentionally or inadvertently. Critiques have suggested a
sequence of shorter repeating bundles. ASCO has put

forward a monthly bundle in its own payment reform
proposal.8
The all-inclusive nature of the bundle has also been
controversial. Conceptually, the aim of a bundle is to transfer
risks associated with high levels of spending from payers to
the providers doing the spending. Presumably, the provider
then decreases the spending. However, within an allinclusive bundle, there is spending, or risks, that the provider can control and risks that he cannot. Economists refer
to these two categories of financial risk as technical risk and
probability risk.
Technical risk is a function of the practices clinical skill and
efficiency in managing the diagnosis and treatment or other
aspects of care that are under the physicians control and
that he is able to impact either by controlling utilization or
cost, preferably both. Probability risks are random or unpredictable events that we classically buy insurance to
mitigate. In theory, a bundle should only include technical
risks and the probability risks should remain with the payer.
Practically, this can be very difficult.
When ASCOs payment reform work group attempted to
split patient care into oncology related versus other medical
care, it quickly became apparent that adjudicating which CPT
and ICD-10 codes should be attributed to the oncologist
was simply impractical. Furthermore, several pilot projects
have now demonstrated that medical oncology homes
can dramatically lower emergency department and hospital
costs,9-11 broadly indicating that there is substantial technical risk within patient care. CMS has indicated that the
probability risk that does exista patient with higher than
average comorbidities or the patient who is in an automobile accident on the way to their chemotherapy appointment, for examplecan be mitigated through risk
adjustments based on patient and service characteristics
in the first example and through reinsurance or stop-loss
provisions in the latter case. Of course, the most important
mitigation of risk in the pilot model is the provision of onesided risk, clearly a carrot that will ultimately be replaced
with a stick.
THE ECONOMICS OF DRUGS IN A BUNDLE

The RAND Corporation estimated that 25% to 40% of the
OCM bundle is chemotherapy and supportive care drugs.12
Among key opinion leaders and policymakers, the inclusion
of drugs in the oncology bundle seems a foregone conclusion. It is assumed that (1) physicians, not patients, control
demand; (2) practices generate substantial revenue from
drugs via the buy-and-bill system; (3) if oncologists are at
financial risk for the drugs, they will choose the least costly
regimens when efficacies are similar; and (4) because oncologists will become more price sensitive, they will seek out
better prices through negotiations with manufacturers
and group purchasing organizations. Although theoretically
compelling, there is reason to question the practical validity
of these claims.
e111
THOMAS AND WARD
If drugs are to be appropriately included in a bundle,

analysis should conclude that drug spending is a technical
risk. Empirical evidence of this is limited. One often-cited
example of drugs being a technical risk is the Zaltrap story.13
Zaltrap was launched in 2012 at twice the price of Avastin, a
drug with similar outcomes. Physicians at Memorial Sloan
Kettering Cancer Center announced in a New York Times
Op-Ed piece that they would not prescribe Zaltrap, and its
price was reduced by 50%. Although this may have emboldened oncologists who have increasingly decried highpriced drugs, the average monthly launch prices of cancer
drugs since then suggest that the pharmaceutical industry
has been unfazed.
Another expectation if drugs are a technical risk is that
practices should have leverage over the acquisition prices of
pharmaceuticals. Although there are examples of volume
discounts for large purchasers and group purchasing organizations, this applies only to generic medications and the
infrequent circumstance in which there are multiple infusable competitors with similar efficacy within a therapeutic
class. Oncology practices have no control over the price of
novel agents, and oral drug prices are negotiated by payers.
For these reasons, the prices of new cancer drugs, those that
drive costs the most, should be considered a probability risk.
Proponents of drugs in the bundle will argue that the
inability to control price is outweighed by the oncologists
ability to impact utilization. They argue that the 20% to 30%
of annual spending on off-label drugs represents tremendous waste in the system, driven by FFS medicine and buyand-bill drug reimbursement. The evidence for this is sparse.
Brooks et al14 reported in an analysis of FFS Medicare data
that only 10% of regional variation in spending was a result of
drug choice, and that drug costs were only 16% of the total
spend. This was dwarfed by the 67% of regional spending
variation related to acute hospital care, accounting for 48%
of the total spend.
A real-life example of the principles of probability and
technical risks as they pertain to oncology may be found in
the recently reported United Health Care (UHC) demonstration project. Margins on drugs were replaced by a
management fee to remove profit via drug choice, and
practices were incentivized to decrease hospital and
emergency department services through shared savings
and a bundling of inpatient evaluation and management
services. Drug costs increased, but the demonstration was
determined a success because of impressive decreases in
acute-care costs.9 One interpretation of these data is that
the probability risks (drug costs) could not be controlled, but
the technical risks associated with hospital and emergency
department services were.
MANAGING A BUNDLE ON DRUGS

The risks of a bundle are dependent on its variability. There
are three ways to manage the variable risk. One way is
to define bundles with enough granularity to minimize
the variability between patients and their associated costs.
e112
Today, that means developing bundles based on the genetic

profiles of cancers. In a fairly short time, nonsmall cell lung
cancer has evolved from one monolithic cancer and four
equally efficacious treatment regimens to at least four
different cancers with very different therapies and associated personalized costs. Unfortunately, it would be naive to
assume that any payment system will be able to keep up with
the personalized medicine revolution. It is little wonder that
UHC and The University of Texas MD Anderson Cancer
Center have chosen head and neck cancer as their first
effort at a bundled payment contract. It is a disease for
which drug therapy and treatment options are limited.
Even then, they have eight distinct treatment/payment
bundles.15 Due to the impracticality of managing granular
bundles, it is anticipated that the OCM bundles will be
fairly generic buckets.
In that circumstance, the other way to manage the risk
of the bundle will be for practices to get big. Large volumes
of patients can then mitigate the risks associated with any
one individual. This would be expected to further the consolidation of oncology into larger groups through mergers,
acquisitions, and affiliations both in independent and hospitalbased practices. This may prove acceptable to health policy
makers but could have important unintended consequences;
there is very limited evidence that consolidation improves
either quality or access to care.
The third way to mitigate the risk of drugs in the bundle is
to move beyond efficiency in drug choice and utilization to
the realm of stenting on care. OCM promises to monitor
quality of care and to hold practices accountable for it, but
the task is not a simple one. It is hypocritical to believe that
buy-and-bill drives overutilization and to ignore the incentive to underuse inherent in drugs in a bundle.
THE ALTERNATIVE: PATHWAYS

An alternative to drugs in the bundle, one that may still
assure payers that drugs are being used appropriately, are
value-based treatment pathways.16 Evidence-based with
the purpose of improving value, quality, and safety, pathways have been shown to reduce variance in drug regimens,
doses, schedules, and treatment duration. Several studies
have shown savings with the utilization of what must be
considered rudimentary pathways and less than stellar pathway compliance.17,18 One project reduced the 12-month
average costs for chemotherapy and supportive care drugs
for patients with lung cancer by 37% or $9,747.18
Another advantage to pathways over bundles is that,
as opposed to bundles developed by payers, pathways can
be developed by oncologists through transparent, peerreviewed, committee-selected regimens that incorporate
efficacy, toxicity, and cost considerations. ASCO recently
published a policy statement providing direction and recommendations regarding the development and utilization of
pathways in clinical practice.19 As pathway development matures, they will incorporate more sophisticated value models
that will account not only for drug costs but also the total
costs of a treatment regimen to include toxicity, ancillary

support services, acute care, and patient productivity loss.
ONCOLOGY CARE MODEL II: A PATHWAY/BUNDLE

HYBRID
Criticisms of OCMs reliance on competition against oneself,
an arbitrary 6-month episode, and fuzzy details on establishing limits on financial risk are details that can most certainly be overcome if CMMI is introspective in its evaluation
and willing to make incremental improvements.
However, the model has two Achilles heels that it must
consider if OCM is to become a win-win-win for patients,
providers, and payers. First, it must divorce itself from the
reliance on FFS as its backbone of reimbursement. Only then
can it align incentives with innovative solutions that can
produce both high-quality and efficient cancer care.
Then it must consider that drugs in a bundle may mitigate
cost and variability for payers, but this is only accomplished
by putting patients and practices at unacceptable risks.
Bundled drugs are structured to require the doctor to run a
gauntlet between rational care and rationed care through
the use of hidden incentives. Pathways, alternatively, use
transparency, consensus, and evidence-based practice
benchmarks to assist the doctor.
OCM II should not wait 7 years for completion of the
current OCM pilot model. It should be a hybrid model that
brings together a true bundle without FFS to reimburse the

care that an attentive and skilled physician can control and a
robust pathways program that will negate a perceived need
to change the way we pay for drugs.
ASCO is to be commended to convene an Education
Session at its 2016 Annual Meeting during which CMMI will
be able to present its model and have an open discussion
with oncology practice representatives from various backgrounds about the models feasibility. It is clear that any
practices participating in the CMS model will have to institute significant structural changes to fulfill these criteria
and only the future will tell if the investment of time and
resources will benefit patients while at the same time
bending the cost curve for oncology. In reality, the model is
not designed to achieve at least the second goal, reducing
the cost of care. This is because the main drivers of expensive care are not addressed in the model (i.e., the cost
of hospital care and the ever-increasing cost of cancer
drugs). In addition, other expensive cost centers for patients
with cancer such as biopsies, surgical procedures, or radiation therapy are also not part of the model, making it
impossible for medical oncology practices to affect the
overall cost of care.
Unless CMS addresses all price centers contributing to the
escalating cost of cancer care, its reforms will fail to make
good on their promise.
References
1. Chen S, Wong S. Singapore Beats Hong Kong in Health Efficiency:
Southeast Asia. http://www.bloomberg.com/visual-data/best-andworst/most-efficient-health-care-2014-countries. Accessed March
29, 2016.
2. Organisation for Economic Cooperation and Development. OECD
Health Statistics 2014. How Does the United States Compare? http://
www.oecd.org/unitedstates/Briefing-Note-UNITED-STATES-2014.pdf.
3. Light DW, Kantarjian H. Market spiral pricing of cancer drugs. Cancer.
2013;119:3900-3902.
4. Lungren MP, Amrhein TJ, Paxton BE, et al. Physician self-referral:
frequency of negative findings at MR imaging of the knee as a marker
of appropriate utilization. Radiology. 2013;269:810-815.
5. Burns J. COME HOME Program Set to Save $33.5M Over 3 Years. http://
www.onclive.com/publications/oncology-business-news/2014/August2014/COME-HOME-Program-Set-to-Save-335M-Over-3-Years. Accessed
March 29, 2016.
6. American Medical Association. Center for Healthcare Quality and
Payment Reform: A Guide to Physician Focused Alternative Payment
Models. http://www.chqpr.org/downloads/Physician-FocusedAlternative
PaymentModels.pdf?utm_source=Distribute+Physician-Focused+APM+
Report&utm_campaign=Bundling+Badly&utm_medium=email. Accessed
March 29, 2016.
7. Burns LR, Pauly MV. Accountable care organizations may have difficulty
avoiding the failures of integrated delivery networks of the 1990s.
Health Aff (Millwood). 2012;31:2407-2416. Accessed March 29, 2016.
8. American Society of Clinical Oncology. Patient-Centered Oncology
Payment. Payment Reform to Support Higher Quality, More Affordable
9.
10.
11.
12.
13.
14.
15.
Cancer Care. http://www.chqpr.org/downloads/ASCO_Patient-centered_

Oncology_Payment.pdf. Accessed March 29, 2016.
Newcomer LN, Gould B, Page RD, et al. Changing physician incentives for
affordable, quality cancer care: results of an episode payment model.
J Oncol Pract. 2014;10:322-326.
Association for Value-Based Cancer Care. Value-Based Cancer Care:
Oncology Medical Home Shows Significant Cost-Savings, Improved
Care Delivery. http://www.valuebasedcancer.com/vbcc-issues/2015/
december-2015-vol-6-no-11/26472-oncology-medical-home-showssignificant-cost-savings-improved-care-delivery. Accessed March 29,
2016.
Sprandio JD. Oncology patient-centered medical home. J Oncol Pract.
2012;8:47s-49s (suppl 3).
CMS Alliance to Modernized Healthcare. Specialty Payment Model
Opportunities and Assessment: Oncology Design Report. Santa Monica,
CA: RAND Corporation; 2014.
Conti RM, Berndt ER. Winners and losers of the Zaltrap price discount.
Health Affairs weblog. February 22, 2013. http://healthaffairs.org/blog/
2013/02/20/winners-and-losers-from-the-zaltrap-price-discountunintended-consequences/. Accessed March 29, 2016.
Brooks GA, Li L, Uno H, et al. Acute hospital care is the chief driver of
regional spending variation in Medicare patients with advanced cancer.
Health Aff (Millwood). 2014;33:1793-1800.
The University of Texas MD Anderson Cancer Center. MD Anderson,
UnitedHealthcare Launch New Cancer Care Payment Model. MD
Anderson News Release December 16, 2014. http://www.mdanderson.
org/newsroom/news-releases/2014/md-anderson-unitedhealthcarenew-cancer-payment.html. Accessed March 29, 2016.
e113
THOMAS AND WARD
16. Conti R, Ward JC, Page R, et al. A pathway through the bundle jungle.
J Oncol Pract. In press.
17. Kreys ED, Koeller JM. Documenting the benefits and cost savings of a
large multistate cancer pathway program from a payers perspective.
J Oncol Pract. 2013;9:e241-e247.
e114
18. Neubauer MA, Hoverman JR, Kolodziej M, et al. Cost effectiveness of

evidence-based treatment guidelines for the treatment of non-smallcell lung cancer in the community setting. J Oncol Pract. 2010;6:12-18.
19. Zon RT, Frame JN, Neuss MN, et al. American Society of Clinical Oncology Policy
Statement on Clinical Pathways in Oncology. J Oncol Pract. 2016;12:261-266.
Multidisciplinary Management of
Brain Metastases
CHAIR
Cleveland Clinic
Cleveland, OH
SPEAKERS
Anna S. Berghoff, MD, PhD
University of Vienna
Vienna, Austria
Jeffrey Scott Wefel, PhD
The University of Texas MD Anderson Cancer Center
Houston, TX
Morris D. Groves, MD
Texas Oncology/The US Oncology Network
Austin, TX
Paul D. Brown, MD
Houston, TX
Riccardo Soffietti, MD
University and San Giovanni Battista Hospital
Torino, Italy
BERGHOFF ET AL
Immune Checkpoint Inhibitors in Brain Metastases: From

Biology to Treatment
Anna S. Berghoff, MD, PhD, Vyshak A. Venur, MBBS, Matthias Preusser, MD, and
OVERVIEW
Cancer immunotherapy has been a subject of intense research over the last several years, leading to new approaches for
modulation of the immune system to treat malignancies. Immune checkpoint inhibitors (antiCLTA-4 antibodies and
antiPD-1/PD-L1 antibodies) potentiate the hosts own antitumor immune response. These immune checkpoint inhibitors
have shown impressive clinical efficacy in advanced melanoma, metastatic kidney cancer, and metastatic nonsmall cell lung
cancer (NSCLC)all malignancies that frequently cause brain metastases. The immune response in the brain is highly
regulated, challenging the treatment of brain metastases with immune-modulatory therapies. The immune microenvironment in brain metastases is active with a high density of tumor-infiltrating lymphocytes in certain patients and, therefore,
may serve as a potential treatment target. However, clinical data of the efficacy of immune checkpoint inhibitors in brain
metastases compared with extracranial metastases are limited, as most clinical trials with these new agents excluded
patients with active brain metastases. In this article, we review the current scientific evidence of brain metastases biology
with specific emphasis on inflammatory tumor microenvironment and the evolving state of clinical application of immune
checkpoint inhibitors for patients with brain metastases.
rain metastases are a long-known devastating complication of advanced malignancies that lead to substantial
morbidity. As patients with metastatic cancer are expected
to live longer with newer systemic therapeutic agents and
with better and more sensitive imaging studies, it is expected
that more patients will be diagnosed with brain metastases. In
the past, traditional treatment approaches included surgery
or radiation therapy, or a combination of the two. Chemotherapeutic agents were generally used for refractory disease.
However, in the last few years, several treatment advances in
targeted therapy and immunotherapy have changed the landscape of the management of brain metastases. Recently, the
American Society of Clinical Oncology (ASCO) announced cancer
immunotherapy as the advance of the year,1 and its role in
the management of brain metastases is undergoing active
investigation. There has been a paucity of research focused on
brain metastases; as a result, there are several unanswered
questions. In this article, we present the available evidence
regarding the interactions between the immune system and
brain parenchyma in brain metastases, the clinical application
of immune checkpoint inhibitors in the management of brain
metastases, and the potential of combining radiation with
immune checkpoint inhibitors.
THE INFLAMMATORY MICROENVIRONMENT OF

BRAIN METASTASES
Immune responses in the brain parenchyma are tightly regulated to prevent overwhelming and potentially destroying
immune reactions in this organ, which has little recovery
capacity.2 Importantly, the brain parenchyma is not an immune privileged organ that actively suppresses any immune
response, but, rather, it initiates and regulates immune
responses. Therefore, targeting the inflammatory tumor
microenvironment of brain metastases as a therapeutic
target must consider several unique factors compared with
the inflammatory tumor microenvironment of extracranial
malignancies.3
Immune escape is an emerging hallmark of cancer.4 A brain
metastasisinitiating cell has to facilitate the process of
immune escape several times, first during initiation of the
primary tumor, then during travel through the bloodstream,
and, finally, during the extravasation and metastatic outgrowth process within the brain parenchyma. The process of
metastatic spread may actually be supported by the immune
system, as tumor-associated macrophages were shown to
facilitate intravasation as well as extravasation from the
vascular system.5,6 Further, cytokines promote site-specific
From the German Cancer Research Center, University of Heidelberg, Heidelberg, Germany; Department of Medicine I, Medical University of Vienna, Vienna, Austria; Comprehensive
Cancer Center Vienna CNS Unit, Vienna, Austria; Division of Hematology and Oncology, Department of Internal Medicine, University of Iowa Hospitals and Clinics, Iowa City, IA; Burkhardt
Brain Tumor and Neuro-Oncology Center, Neurologic Institute, Cleveland Clinic, Cleveland, OH; Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH.
Corresponding author: Manmeet S. Ahluwalia, MD, Cleveland Clinic, 9500 Euclid Ave., S73, Cleveland, OH 44195; email: ahluwam@ccf.org.
e116
IMMUNE CHECKPOINT INHIBITIORS IN BRAIN METASTASES
metastatic behavior as the chemokine pair CXCR4/CXCL12

was shown to promote adhesion of the brain metastasis
initiating cells to brain vessels and further facilitated the
migration through endothelial cells.7,8
In established brain macrometastases, the inflammatory
tumor microenvironment is composed of the innate
immune system, namely microglia and blood-derived myeloid
cells/macrophages, and the adaptive immune system, which
is mainly represented by T cells.9
The density of microglia in and around human and in vivo
brain metastases were shown to be highly heterogeneous.10,11
Importantly, the differences in microglia accumulation were
observed early during the process of extravasation in preclinical models, indicating that microglia may be involved in
the process of extravasation and survival in the perivascular
niche.11 The functions of microglia within the brain metastasis
tumor microenvironment include antigen presentation, cytotoxicity via expression of nitric oxide (NO) and superoxide, and
phagocytic function.12 Expression of HMGB1 by microglia
cells, a factor involved in antigen presentation and activation
of the adaptive immune system, was widely observed, indicating that in general the tumor microenvironment is able to
alter an adaptive immune response. However, microglia and
macrophages were also shown to express immunosuppressive factors like PD-L1.13
Infiltration with T cells or tumor-infiltrating T-lymphocytes
(TILs) was shown to be highly heterogeneous, varying from
total absence to very dense infiltration.10,14 Therefore, the
inflammatory microenvironment of brain metastases differs
between patients, showing active tumor-directed function
in some, but not all, cases. Here, expression of the immune
suppressive factor PD-L1 could be one mechanism whereby
cancer cells facilitate an immune escape. PD-L1 tends to
have higher expression in NSCLC-derived brain metastases
KEY POINTS
The inflammatory tumor microenvironment of brain

metastases differs from extracranial metastases, as the
brain is a highly regulated organ in terms of immune
response, and is active in the majority of cases with
dense infiltration of tumor-infiltrating lymphocytes.
The blood-brain barrier presents different challenges
and unique opportunities for the treatment of patients
with brain metastases.
Immune checkpoint inhibitors have shown intracranial
activity in early clinical trials for patients with brain
metastases from melanoma and nonsmall cell lung
cancer.
Several retrospective case series suggest immune
checkpoint inhibitors can be safely combined with
radiation therapies.
As the prevalence of brain metastases increases,
concerted research in this field of limited knowledge is
of essence. Meanwhile, a multidisciplinary approach for
the management of brain metastases is recommended.
compared with the matched primary tumor.15 Further, the

composition of TIL subtypes differs. Infiltration with immune
suppressive FOXP3+ TILs, as well as so-called exhausted
PD-1+ TILs, has been observed in a large portion of specimens.14 Characteristics of the primary tumor probably influence the T cell response in brain metastases, as a denser
TIL infiltration was observed in melanoma-derived brain
metastases compared with those that are derived from
breast cancer.14 Importantly, patients with brain metastases who present with dense infiltration with effector
CD3+, cytotoxic CD8+, or memory CD45RO+ TILs had an
improved survival prognosis from diagnosis compared
with patients with little or absent infiltration of TILs.14 This
finding is well in line with the effect of the inflammatory
tumor microenvironment on survival prognosis in extracranial malignancies, as high density of TILs is associated
with improved survival prognosis in various malignancies
that frequently cause brain metastases, such as triple-negative
breast cancer, NSCLC, or melanoma.16-18
Importantly, patients with extracranial metastatic melanoma and an active inflammatory tumor microenvironment,
as defined by dense infiltration with TILs, have shown increased response rates to CTLA-4 immune checkpoint inhibitors.19 Further, expression of PD-L1 on tumor cells may
be associated with a higher chance of response to a PD1axis modulating immune checkpoint inhibitor.20 Therefore, the potential tumor microenvironment preconditions
for response to an immune checkpoint inhibitor are present
in brain metastases, underscoring the importance of further
clinical investigation of immune checkpointbased therapy
strategies in patients with brain metastases.
The inflammatory microenvironment is further influenced
by the vascular properties in the brain parenchyma. The
blood-brain barrier is composed of endothelial cells with
tight junctions and astrocytes. Here, astrocytes are known
to facilitate pro-inflammatory as well as anti-inflammatory
aspects.21 However, data of the interaction of astrocytes
and TILs in the inflammatory microenvironment of brain
metastases is yet not available. Further, the vascular endothelium regulates the infiltration of T cells. Presence of socalled high endothelial venules, which are specialized blood
vessels for lymphocyte extravasation, in melanoma specimens were shown to be associated with dense infiltration
of TILs.22 Although neovascularization is a characteristic of
brain metastases, with differing extent between primary
tumor types, no functional studies have yet investigated the
interaction of the blood-tumor barrier or the blood-brain
barrier in the infiltration zone of a brain metastasis concerning their restriction of T-cell infiltration.
CLINICAL EFFICACY OF IMMUNE THERAPY IN

CANCER
The clinical utility of immune-based therapies in the fight
against cancer has been a fascinating story with ebb and
flow. Historically, melanoma and kidney cancer were the two
major malignancies that were thought to be susceptible to
e117
BERGHOFF ET AL
immune modulation. Indeed, melanoma has high mutational burden, with production of several neoantigens,
making it an ideal candidate for immunotherapy. Interleukin
(IL)-2 was one of the first immunotherapeutic agents to be
used in melanoma. IL-2 is a cytokine that is produced by
human T-lymphocytes and plays a pivotal role in expansion
and activation of T cells.23 In the 1990s, high-dose IL-2 was
studied in advanced melanoma and renal cell carcinomas
with modest success.24 An extracranial complete response
rate of 5% was seen, and some patients experienced substantial adverse effects, including death.24 Two single institution retrospective experiences have shown intracranial
activity of IL-2; however, because of concerns of vascular
leak syndrome and potential life-threatening thrombocytopenia, prospective studies excluded patients with brain
metastases.25,26
AntiCTLA-4 Antibody: Ipilimumab

With the advent of immune checkpoint inhibitors, immunotherapy has made substantial inroads in the management
of advanced malignancies. Ipilimumab is a fully human
monoclonal antibody against CTLA-4, which plays a pivotal
role in downregulating the production of cytotoxic T cells.27
The intracranial activity of ipilimumab was first noted in the
post hoc analysis of a phase III trial of ipilimumab with or
without gp100 peptide vaccine compared with gp100 vaccine alone.28 This double-blinded randomized trial included
82 patients with asymptomatic central nervous system (CNS)
metastases, of whom, 73 had received prior treatment for
CNS metastases. Ipilimumab was administered to 61 of the
82 patients; 46 patients also received gp100 and 15 received
ipilimumab alone. The hazard ratio (HR) for death in patients
with brain metastases was 0.70 (95% CI, 0.411.20) in the
ipilimumab plus vaccine group compared with 0.76 (95% CI,
0.381.54) in the ipilimumab alone group. This led to a
single-agent, open-label phase II study of ipilimumab for
patients with melanoma-derived brain metastases.29 Ipilimumab was administered at a dose of 10 mg/kg given every
3 weeks as a 12-week induction phase, followed by maintenance therapy every 12 weeks. A total of 72 patients were
enrolled across 10 centers in the United States.30 The trials
comprised of two cohorts: cohort A included 51 asymptomatic patients with active brain metastases who were not
on corticosteroids, and cohort B included 21 patients with
symptomatic melanoma-derived brain metastases taking a
corticosteroid. A response rate of 18% (nine out of 51 patients) in cohort A and 5% (one out of 21 patients) in cohort B
were reported using modified World Health Organization
(WHO) criteria, but, when immune-related response criteria
were used, the response rate improved to 25% (13 patients)
and 10% (two patients) in cohorts A and B, respectively. The
median overall survival (OS) was 7.0 months in cohort A and
3.7 months in cohort B. Ipilimumab was well tolerated and
no unexpected adverse events were noted. Fatigue, nausea,
vomiting, diarrhea, and rash were the common side effects
noted in the study. As a part of the expanded access program,
e118
CA 184-045, 165 patients with stable brain metastases were

treated with 10 mg/kg of ipilimumab. The preliminary results
from this study reported a 1-year OS rate of 20%, similar to that
reported in the open-label, phase II study.31
Another open-label, single-arm, phase II trial evaluated the
combination of ipilimumab and fotemustine in advanced
metastatic melanoma with and without brain metastases
(NIBIT-M1).32 This Italian study enrolled a total of 83 patients; 20 of whom had asymptomatic brain metastases.
Eleven of the 20 patients completed the induction phase
of ipilimumab (four doses of 10 mg/kg every 3 weeks).
Immune-related disease control was observed in 10 patients; five had undetectable brain lesions and five had
stable disease. The median progression-free survival (PFS) in
patients with brain metastases was 3.0 months (range, 0 to
10.9 months). Adverse reactions were reported in 18 out
of 20 patients with brain metastases. Central nervous
system events like intracranial hemorrhage, seizures, and
headaches were reported in five patients, all of whom had
intracranial disease progression. A 3-year follow-up of this
trial was presented at the European Society for Medical
Oncology (ESMO) Congress in 2015, and the group with brain
metastases had a median OS of 12.7 months (95% CI,
2.722.7 months) and a 3-year survival rate of 27.8%.33 With
the promising results from the phase II trial, a phase III trial
of patients with melanoma-derived brain metastases
(NIBIT-M2) is planned.34 The trial was initially designed with
two arms; arm A with single-agent fotemustine and arm B
with fotemustine and ipilimumab. However, with recent
reports of high clinical activity with a combination of ipilimumab and the antiPD-1 monoclonal antibody nivolumab, the protocol was amended to include a third arm,
arm C, with a combination of nivolumab and ipilimumab.35
Table 1 reviews the published studies of immune checkpoint inhibitors in brain metastases.
AntiPD-1/PD-L1 Antibodies
Another promising immune checkpoint target is the
PD-1/PD-L1 pathway. AntiPD-1/PD-L1 antibodies such as
nivolumab and pembrolizumab have been approved for
advanced melanoma, kidney cancer, and NSCLC. All of the
early trials of antiPD-1/PD-L1 agents excluded patients with
brain metastases. In 2015, interim results of a single-center,
phase II study of pembrolizumab of patients with melanoma
and NSCLC with brain metastases was presented at the 2015
ASCO Annual Meeting.36,37 Patients with newly diagnosed
asymptomatic or progressing brain metastases who did not
require immediate treatment with steroids are being enrolled in this study. By December 2014, 17 patients with
metastatic melanoma and 10 patients with NSCLC were
enrolled. These patients were treated with 10 mg/kg of
pembrolizumab administered every 2 weeks. Among the
12 evaluable patients in the melanoma arm, three were
observed to have partial response and two had stable disease
with a 6-month OS of 47%. In the NSCLC arm, nine patients
were evaluated for response and four had partial response.
TABLE 1. Studies With Immunotherapy in Melanoma Patients With Brain Metastases
Study
Design
Margolin et al29
Phase II
Special Trial
Characteristics
Therapy
Tested
Concurrent
Local
Radiation
Treatment
Cohort A: no
steroids
Ipilimumab
None
Cohort B:
steroids
Intracranial
Response
No. of Patients Rate
PFS
OS
(Months)
Cohort A: 51
24%
NR
7.0
Cohort B: 21
10%
3.7
Di Giacoma
et al32
Phase II
Asymptomatic
Ipilimumab/
None
fotemustine
20
50%
NR
12.7
Patel R et al49
Retrospective
comparative
analysis
SRS vs. SRS/

ipilimumab
Ipilimumab
SRS (given
within
4 months of
initiation of
ipilimumab)
SRS only: 34
NR
1-year
LC:
92.3%
12-month:
38%
1-year
LC:
71.4%
12-month:
37.1%
SRS/
ipilimumab:
20
Weber et al50
Retrospective
Ipilimumab/
analysis of
budesonide
phase II trial
(asymptomatic
BM)
Ipilimumab
None
12
41.6%
NR
14.0
Knisely et al41
Retrospective
comparative
analysis
SRS vs. SRS/

ipilimumab
Ipilimumab
SRS
SRS: 17
NR
NR
4.9
Retrospective
comparative
analysis
SRS vs. SRS/

ipilimumab
Ipilimumab
Silk et al40
SRS/
ipilimumab:
11
SRS/WBRT
No
ipilimumab:
37
21.3
NR
Ipilimumab: 33
42
Mathew et al
Ahmed et al43
Retrospective
comparative
analysis
SRS vs. SRS/

ipilimumab
Ipilimumab
Retrospective
analysis
SRS/nivolumab
Nivolumab
SRS
SRS: 33
NR
3.3
5.3
2.7
18.3
NR
6-month: 56%
SRS/
ipilimumab:
25
SRS
26
6-month:45%
NR
NR
11.8
Abbreviations: BM, brain metastases; PFS, progression-free survival; OS, overall survival; NR, not reported; SRS, stereotactic radiosurgery; LC, local control; WBRT, whole-brain
radiotherapy.
Only one patient in both arms experienced grade 3 adverse

events related to pembrolizumab (liver function abnormalities). The investigators presented updated results at the
16th World Conference on Lung Cancer in September 2015;
18 patients had sufficient follow-up time for response evaluation. The median OS was 7.7 months (95% CI, 3.5 months
to not reached).38 These early results appear promising for
the utility of pembrolizumab in patients with asymptomatic
brain metastases. Another multicenter phase II clinical trial
of ipilimumab plus nivolumab for patients with advanced
melanoma with brain metastases (NCT02320058) is currently
accruing patients.39 This study is planned to treat patients
with asymptomatic brain metastases from melanoma with an
induction phase of four cycles of 1 mg/kg of nivolumab and
3 mg/kg of ipilimumab every 3 weeks, followed by maintenance therapy of 3 mg/kg of nivolumab every 2 weeks. Table 2
provides a list of other important clinical trials in this setting.
Immunotherapy and Radiation

Radiation therapy has been the cornerstone of treatment for brain metastases. Stereotactic radiosurgery and
whole-brain radiation therapy (WBRT) with or without surgery plays a crucial role in the management of symptomatic
brain metastases.
Several retrospective single-center case series have shown
that stereotactic radiation and ipilimumab can be combined
safely in the management of brain metastases derived from
melanoma.40-42 The safety of combining nivolumab with
stereotactic radiation was reported in a recent single-center
study.43 This retrospective report included patients who
were treated in the context of a clinical trial of nivolumab
for advanced melanoma. Twenty-six patients with brain metastases were treated with stereotactic radiation, with, prior to,
or after nivolumab. The combination of nivolumab and stereotactic radiation was well tolerated. Distant intracranial
recurrence and OS were improved compared with historical
controls treated with radiation alone.
There are three important aspects to combining radiation
with immune checkpoint inhibitors:
1. Can immunotherapy potentiate radiation therapy
(radio-sensitizing effect) leading to better intracranial
disease control?
e119
BERGHOFF ET AL
TABLE 2. Select Clinical Trials Currently Accruing Patients

Study
Design
Therapy tested
Primary Cancer
Estimated
Enrollment
NCT0246006839
Multicenter, randomized, phase III
Fotemustine
vs. fotemustine/ipilimumab
vs. ipilimumab/nivolumab
Melanoma
168
NCT0262151551
Multicenter, phase II
Nivolumab in symptomatic brain metastases
Melanoma
70
NCT0232005852
Nivolumab/ipilimumab followed by nivolumab
Melanoma
110
NCT0237424253
Nivolumab/ipilimumab for melanoma brain metastases
Melanoma
75
NCT0208507054
Single-center, phase II
Pembrolizumab
Melanoma,
nonsmall cell
lung cancer
64
2. Will the combination of radiation therapy and immunotherapy lead to better extracranial disease control
(abscopal effect) and, thus, improved OS?
3. If the radiation potentiates the immunotherapy, what
is the best fractionation schedule to achieve the most
benefit?
The first question is relatively straightforward, and several
clinical trials are currently underway to answer the potential
intracranial responses of combining immunotherapy and
radiation. The major confounder in this endeavor is the use
of steroids. Relatively high doses of steroids have been used
with radiation, especially WBRT, to reduce the intracranial
edema. The detrimental effect of the concomitant use of
steroids and immunotherapy was observed in the phase II
trial of ipilimumab for patients with asymptomatic,
melanoma-derived brain metastases.29 Since this study, the
required steroid dose used with stereotactic radiation
therapy has decreased, and they are used often for a shorter
duration. In addition, many centers are now moving away
from using steroids routinely for all patients with brain
metastases.44
This second aspect of combining radiation and immunotherapy, the abscopal effect, is interesting and more
challenging to evaluate in clinical trials. It has been well
established that radiation causes DNA damage that leads to
cell death. However, the interaction of radiation and the
immune system is not well understood. Several preclinical
data have reported that cancer cells release numerous
chemokines and cytokines in response to radiation. The
damaged cancer cells have also been shown to upregulate
MHC-1 expression, hence increasing the interaction with
CD8 T cells.45 In addition, radiation leads to the upregulation
of PD-L1 on cancer cells, limiting their interaction with CD8
T cells.46,47 Nevertheless, the activated CD8 T cells can elicit a
systemic antitumor effect, thereby leading to reduction in
the tumor burden. Methods to reliably produce and potentiate this abscopal effect are a subject of intense research. However, increasing the interaction of CD8 T cells
with radiated tumor cells by the use of immune checkpoint
inhibitors may enhance the abscopal effect. A retrospective
case series tried to evaluate this effect by following the
e120
single largest extracranial lesion after stereotactic radiation

to the brain metastases and ipilimumab.44
The third aspect of combining radiation and immunotherapy and choosing the preferred fractionation schedule
is being evaluated in preclinical models. Single session and
fractioned radiation schedules with and without immunotherapy have been tested in a preclinical mouse model.48
TSA breast carcinoma and MCA38 colon cancer mouse
models were used to test the various radiation fractionation
schemes with and without antiCTLA-4 antibody, 9H10. The
mice were injected with tumor cells at two sites; a primary
and a secondary site. They were then randomly assigned
to receive radiation to the primary site at three different
fractionation schedules (one cycle of 20 Gy, three cycles
of 8 Gy, or five cycles of 5 Gy on consecutive days), immunotherapy alone, or a combination of radiation and
immunotherapy. Only fractionated radiation with immunotherapy to the primary site produced abscopal effect at
the secondary site. However, strong clinical data to support this preclinical model are lacking. Innovative clinical
trials are needed to improve our understanding of the interaction of radiation and immunotherapy and the abscopal
effect.
CONCLUSION
In summary, some brain metastases harbor an immune
active microenvironment that in theory can be targeted by
immune-modulating therapies like immune checkpoint inhibitors. However, a deeper insight of the specific mechanisms in the highly regulated microenvironment of the brain
is needed to understand and overcome potential resistance
mechanisms. Early results of immune checkpoint inhibitors
in clinical trials have shown intracranial activity in brain
metastases from melanoma and NSCLC. There are several
ongoing clinical trials investigating the role of immune
checkpoint blockade in brain metastases. Numerous retrospective case series suggest immune checkpoint inhibitors
can be safely combined with radiation. Prospective studies
are needed to further confirm the safety of such approaches
and define the timing and the dose of the optimal radiation
modality.
References
1. Dizon DS, Krilov L, Cohen E, et al. Clinical cancer advances 2016: annual
report on progress against cancer from the American Society of Clinical
Oncology. J Clin Oncol. 2016;34:987-1011.
2. Galea I, Bechmann I, Perry VH. What is immune privilege (not)? Trends
Immunol. 2007;28:12-18.
3. Berghoff AS, Preusser M. The inflammatory microenvironment in brain
metastases: potential treatment target? Chin Clin Oncol. 2015;4:21.
4. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation.
Cell. 2011;144:646-674.
5. Noy R, Pollard JW. Tumor-associated macrophages: from mechanisms
to therapy. Immunity. 2014;41:49-61.
6. Wyckoff JB, Wang Y, Lin EY, et al. Direct visualization of macrophageassisted tumor cell intravasation in mammary tumors. Cancer Res.
2007;67:2649-2656.
7. Hinton CV, Avraham S, Avraham HK. Role of the CXCR4/CXCL12 signaling
axis in breast cancer metastasis to the brain. Clin Exp Metastasis. 2010;
27:97-105.
8. Lee BC, Lee TH, Avraham S, et al. Involvement of the chemokine receptor CXCR4 and its ligand stromal cell-derived factor 1 alpha in breast
cancer cell migration through human brain microvascular endothelial
cells. Mol Cancer Res. 2004;2:327-338.
9. Fridman WH, Pages
` F, Sautes-Fridman
`
C, et al. The immune contexture
in human tumours: impact on clinical outcome. Nat Rev Cancer. 2012;
12:298-306.
10. Berghoff AS, Lassmann H, Preusser M, et al. Characterization of the
inflammatory response to solid cancer metastases in the human brain.
Clin Exp Metastasis. 2013;30:69-81.
11. Lorger M, Felding-Habermann B. Capturing changes in the brain microenvironment during initial steps of breast cancer brain metastasis.
Am J Pathol. 2010;176:2958-2971.
12. Graeber MB, Streit WJ. Microglia: biology and pathology. Acta Neuropathol. 2010;119:89-105.
13. Berghoff AS, Kiesel B, Widhalm G, et al. Programmed death ligand 1
expression and tumor-infiltrating lymphocytes in glioblastoma. Neuro
Oncol. 2015;17:1064-1075.
14. Berghoff AS, Fuchs E, Ricken G, et al. Density of tumor-infiltrating
lymphocytes correlates with extent of brain edema and overall
survival time in patients with brain metastases. OncoImmunology.
2015;5:e1057388.
15. Berghoff AS, Inan C, Ricken G, et al. Tumor-infiltrating lymphocytes
(TILs) and PD-L1 expression in non-small cell lung cancer brain metastases (BM) and matched primary tumors (PT). Ann Oncol. 2014;25:
iv426-iv470 (suppl 4; abstr 1324P).
16. Adams S, Gray RJ, Demaria S, et al. Prognostic value of tumor-infiltrating
lymphocytes in triple-negative breast cancers from two phase III randomized adjuvant breast cancer trials: ECOG 2197 and ECOG 1199. J Clin
Oncol. 2014;32:2959-2966.
17. Tao H, Mimura Y, Aoe K, et al. Prognostic potential of FOXP3 expression
in non-small cell lung cancer cells combined with tumor-infiltrating
regulatory T cells. Lung Cancer. 2012;75:95-101.
18. Clemente CG, Mihm MC Jr, Bufalino R, et al. Prognostic value of tumor
infiltrating lymphocytes in the vertical growth phase of primary cutaneous melanoma. Cancer. 1996;77:1303-1310.
19. Ji RR, Chasalow SD, Wang L, et al. An immune-active tumor microenvironment favors clinical response to ipilimumab. Cancer Immunol
Immunother. 2012;61:1019-1031.
20. Taube JM, Klein A, Brahmer JR, et al. Association of PD-1, PD-1 ligands,
and other features of the tumor immune microenvironment with response to anti-PD-1 therapy. Clin Cancer Res. 2014;20:5064-5074.
21. Gimsa U, Mitchison NA, Brunner-Weinzierl MC. Immune privilege as

an intrinsic CNS property: astrocytes protect the CNS against T-cellmediated neuroinflammation. Mediators Inflamm. 2013;2013:320519.
22. Martinet L, Le Guellec S, Filleron T, et al. High endothelial venules
(HEVs) in human melanoma lesions: Major gateways for tumorinfiltrating lymphocytes. OncoImmunology. 2012;1:829-839.
23. Rosenberg SA. IL-2: the first effective immunotherapy for human
cancer. J Immunol. 2014;192:5451-5458.
24. Atkins MB, Lotze MT, Dutcher JP, et al. High-dose recombinant interleukin 2 therapy for patients with metastatic melanoma: analysis of
270 patients treated between 1985 and 1993. J Clin Oncol. 1999;17:
2105-2116.
25. Guirguis LM, Yang JC, White DE, et al. Safety and efficacy of high-dose
interleukin-2 therapy in patients with brain metastases. J Immunother.
2002;25:82-87.
26. Powell S, Dudek AZ. Single-institution outcome of high-dose interleukin2 (HD IL-2) therapy for metastatic melanoma and analysis of favorable
response in brain metastases. Anticancer Res. 2009;29:4189-4193.
27. Weber J. Review: anti-CTLA-4 antibody ipilimumab: case studies of
clinical response and immune-related adverse events. Oncologist.
2007;12:864-872.
28. Hodi FS, ODay SJ, McDermott DF, et al. Improved survival with ipilimumab
in patients with metastatic melanoma. N Engl J Med. 2010;363:711-723.
29. Margolin K, Ernstoff MS, Hamid O, et al. Ipilimumab in patients with
melanoma and brain metastases: an open-label, phase 2 trial. Lancet
Oncol. 2012;13:459-465.
30. Margolin K. Ipilimumab in a phase II trial of melanoma patients with
brain metastases. Oncoimmunology. 2012;1:1197-1199.
31. Heller KN, Pavlick AC, Hodi FS, et al. Safety and survival analysis of
ipilimumab therapy in patients with stable asymptomatic brain metastases. J Clin Oncol. 2011;29 (suppl; abstr 8581).
32. Di Giacomo AM, Ascierto PA, Pilla L, et al. Ipilimumab and fotemustine in
patients with advanced melanoma (NIBIT-M1): an open-label, singlearm phase 2 trial. Lancet Oncol. 2012;13:879-886.
33. Di Giacomo AM, Ascierto PA, Queirolo P, et al. Three-year follow-up
of advanced melanoma patients who received ipilimumab plus fotemustine in the Italian Network for Tumor Biotherapy (NIBIT)-M1 phase
II study. Ann Oncol. 2015;26:798-803.
34. Lyle M, Long GV. The role of systemic therapies in the management of
melanoma brain metastases. Curr Opin Oncol. 2014;26:222-229.
35. Di Giacomo AM, Margolin K. Immune checkpoint blockade in patients
with melanoma metastatic to the brain. Semin Oncol. 2015;42:
459-465.
36. Kluger HM, Goldberg SB, Sznol M, et al. Safety and activity of pembrolizumab in melanoma patients with untreated brain metastases.
J Clin Oncol. 2015;33 (suppl; abstr 9009).
37. Goldberg SB, Gettinger SN, Mahajan A, et al. Activity and safety of
pembrolizumab in patients with metastatic non-small cell lung cancer
with untreated brain metastases. J Clin Oncol. 2015;33 (suppl; abstr
8035).
38. Seto T, Shukuya T, Yamanaka T, et al. Chemotherapy developments for
lung cancer. J Thorac Oncol. 2015;10:S173-S260.
39. Margolin KA, Tawbi HA-H, Ernstoff MS, et al. A multi-center phase II
open-label study (CheckMate 204) to evaluate safety and efficacy of
nivolumab (NIVO) in combination with ipilimumab (IPI) followed by
NIVO monotherapy in patients (pts) with melanoma (MEL) metastatic to
the brain. J Clin Oncol. 2015;33 (suppl; abstr TPS9080).
40. Silk AW, Bassetti MF, West BT, et al. Ipilimumab and radiation therapy
for melanoma brain metastases. Cancer Med. 2013;2:899-906.
e121
BERGHOFF ET AL
41. Knisely JP, Yu JB, Flanigan J, et al. Radiosurgery for melanoma brain
metastases in the ipilimumab era and the possibility of longer survival.
J Neurosurg. 2012;117:227-233.
42. Mathew M, Tam M, Ott PA, et al. Ipilimumab in melanoma with limited
brain metastases treated with stereotactic radiosurgery. Melanoma
Res. 2013;23:191-195.
43. Ahmed KA, Stallworth DG, Kim Y, et al. Clinical outcomes of melanoma
brain metastases treated with stereotactic radiation and anti-PD-1
therapy. Ann Oncol. 2016;27:434-441.
44. Schoenfeld JD, Mahadevan A, Floyd SR, et al. Ipilimumab and cranial
radiation in metastatic melanoma patients: a case series and review.
J Immunother Cancer. 2015;3:50.
45. Reits EA, Hodge JW, Herberts CA, et al. Radiation modulates the peptide
repertoire, enhances MHC class I expression, and induces successful
antitumor immunotherapy. J Exp Med. 2006;203:1259-1271.
46. Dovedi SJ, Adlard AL, Lipowska-Bhalla G, et al. Acquired resistance to
fractionated radiotherapy can be overcome by concurrent PD-L1
blockade. Cancer Res. 2014;74:5458-5468.
47. Deng L, Liang H, Burnette B, et al. Irradiation and anti-PD-L1 treatment
synergistically promote antitumor immunity in mice. J Clin Invest. 2014;
124:687-695.
48. Dewan MZ, Galloway AE, Kawashima N, et al. Fractionated but not
single-dose radiotherapy induces an immune-mediated abscopal effect
e122
49.
50.
51.
52.
53.
54.
when combined with anti-CTLA-4 antibody. Clin Cancer Res. 2009;15:

5379-5388.
Patel KR, Shoukat S, Oliver DE, et al. Ipilimumab and stereotactic
radiosurgery versus stereotactic radiosurgery alone for newly diagnosed melanoma brain metastases. Am J Clin Oncol. Epub 2015
May 16.
Weber JS, Amin A, Minor D, et al. Safety and clinical activity of ipilimumab
in melanoma patients with brain metastases: retrospective analysis
of data from a phase 2 trial. Melanoma Res. 2011;21:530-534.
NCT02621515. Nivolumab in Symptomatic Brain Metastases (CA209322). https://clinicaltrials.gov/ct2/show/record/NCT02621515. Accessed
February 20, 2016.
NCT02320058. A Multi-Center Phase 2 Open-Label Study to Evaluate
Safety and Efficacy in Subjects With Melanoma Metastatic to the Brain
Treated With Nivolumab in Combination With Ipilimumab Followed by
Nivolumab Monotherapy (CheckMate 204). https://clinicaltrials.gov/
ct2/show/NCT02320058. Accessed February 20, 2016.
NCT02374242. Anti-PD 1 Brain Collaboration for Patients With Melanoma Brain Metastases (ABC). https://clinicaltrials.gov/ct2/show/
NCT02374242. Accessed February 20, 2016.
NCT02085070. MK-3475 in Melanoma and NSCLC Patients With
Brain Metastases. https://clinicaltrials.gov/ct2/show/NCT02085070.
TARGETED THERAPY IN BRAIN METASTASES
Targeted Therapy in Brain Metastases: Ready for Primetime?

Vyshak A. Venur, MD, and Manmeet S. Ahluwalia, MD
OVERVIEW
Brain metastasis is a serious complication of cancer that causes significant morbidity for patients. Over the last decade,
numerous new driver somatic mutations have been recognized and targeted therapies are changing the landscape of
treatment in lung cancer, breast cancer, and melanoma, which are also the three most common cancers that result in
brain metastases. The common actionable mutations include the EGFR mutation and anaplastic lymphoma kinase (ALK)
translocations in nonsmall cell lung cancer, the HER2 mutation in breast cancer, and the BRAF mutation in melanoma.
However, most of the early trials with targeted agents excluded patients with brain metastases. With a better understanding
of the biology, several recent trials of targeted therapy that focus on brain metastases have been reported and others are
ongoing. Novel agents with better penetration across the bloodbrain barrier are currently being investigated for patients
with brain metastases. In this review, we discuss the current state of use and future directions of targeted therapies in brain
metastases.
he discovery of driver mutations and agents targeting these

mutations and pathways has revolutionized the treatment
of patients with advanced malignancies in the past decade. The
incidence of brain metastases has increased, although newer
targeted agents have improved the systemic control of malignancy and hence survival. It is estimated that approximately 40%
of patients with metastatic cancer will develop a symptomatic
brain metastasis.1 Lung cancer, breast cancer, and melanoma are
the three common malignancies that lead to brain metastases. In
the past, the cornerstone for treatment of brain metastases included whole-brain radiation therapy (WBRT), stereotactic radiation (SRS), and surgery. The evolution of systemic therapy in the
management of brain metastases has been restricted by the
presence of the bloodbrain barrier and efflux pumps. Most initial
studies of drugs for the management of advanced cancer excluded
patients with brain metastases, further limiting progress. Hence,
chemotherapy has had a limited role in the management of brain
metastases. Several new targeted agents with bloodbrain barrier
penetration are being developed, with emphasis on treatment
patients with brain metastases.2 In this article, we review the
current evidence on the role of agents targeting the various driver
mutations and pathways in management of brain metastases from
lung cancer, breast cancer, and melanoma (Table 1).
TARGETED THERAPY FOR LUNG CANCER WITH

BRAIN METASTASES
Lung cancer is the most common cancer that results in brain
metastases. A high proportion of patients with small cell lung
cancer develop brain metastases fairly early in the course of

the disease. They are usually treated with WBRT. Patients
with nonsmall cell lung cancer (NSCLC) are a more heterogeneous group, and 25%40% of patients with NSCLC will
develop a brain metastasis during the course of the disease.3
The common actionable mutations in nonsquamous NSCLC
include EGFR mutation and ALK gene rearrangements.
The frequency of somatic EGFR mutations varies from 30% to
50% among East Asians to approximately 10% among Caucasians.4 Exon 19 (in-frame) deletions and point mutations (L858R)
at exon 21 comprise 90% of all known EGFR mutations in
NSCLC.5 Studies suggest that patients with an EGFR mutation
develop brain metastases more frequently, have several small
lesions with minimal peritumoral edema, and may have better
overall survival (OS) compared with patients with lung cancer
brain metastases without any mutation.6 Erlotinib and gefitinib are two first-generation tyrosine kinase inhibitors (TKIs)
that target EGFR mutations and are approved by the U.S. Food
and Drug Administration (FDA) for use in lung cancer treatment.7
Both erlotinib and gefitinib have distinct pharmacologic characteristics in the central nervous system (CNS). [11C]erlotinib
was shown to have effective CNS distribution in the presence
of brain metastases but not in the normal brain.8 However,
OSI-420, an active metabolite of erlotinib, is a substrate of
P-glycoprotein (P-gp), a drug efflux protein of the ATP-binding
cassette transporter family at the brain endothelial cells of the
bloodbrain barrier.9 Gefitinib is known to inhibit P-gp activity,
and none of its known metabolites are substrates of P-gp.10
From the Division of Hematology and Oncology, Department of Internal Medicine, University of Iowa Hospitals and Clinics, Iowa City, IA; Burkhardt Brain Tumor and Neuro-Oncology
Center, Department of Medicine, Neurologic Institute, Cleveland Clinic, Cleveland, OH; Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH.
Corresponding author: Manmeet. S. Ahluwalia, MD, Cleveland Clinic, 9500 Euclid Ave., S73, Cleveland, OH 44195; email: ahluwam@ccf.org.
e123
VENUR AND AHLUWALIA
However, both of these drugs have limited bloodbrain barrier

penetration, and limited cerebral spinal fluid concentrations of
erlotinib (approximately 5%) and gefitinib (approximately 2.5%)
have been reported.11,12
A Japanese phase II study evaluated the efficacy of gefitinib among 41 patients with brain metastases.13 The overall
response rate was 87.8%, with a progression-free survival
(PFS) of 14.5 months (95% CI, 10.218.3 months) and a
median OS of 21.9 months (95% CI, 18.530.1 months).13
Patients with an EGFR exon 19 deletion had significantly
better PFS and OS compared with those with an exon 21
L858R mutation. In one study, 31 Korean patients who had
never smoked and had EGFR-mutant NSCLC and synchronous asymptomatic brain metastases were treated with
gefitinib or erlotinib.14 The objective response rate was
96.9%, PFS was 7.1 months, and OS was 18.8 months.
Several preclinical studies with human cell lines have indicated that EGFR TKIs enhance radiation therapy.15 Numerous clinical trials and retrospective reports have shown
increased efficacy of combining TKIs with radiation therapy
in the context of brain metastases from EGFR-mutant NSCLC.
In a prospective phase II study, 40 patients with brain
metastases received erlotinib for 1 week prior to WBRT,
concurrently with WBRT followed by maintenance erlotinib.16 Nine of the 40 patients carried an EGFR mutation, and
an overall response rate of 89.9% with a median OS of
19.1 months was noted in that cohort. In a randomized
phase II study, 80 patients with brain metastases from NSCLC
received either erlotinib or placebo with WBRT.17 In the
erlotinib arm, 35 of 40 patients had wild-type EGFR; no
KEY POINTS
e124
Brain metastases are a major cause of morbidity for

patients with advanced malignancies, and the incidence
of brain metastases is increasing.
For patients with nonsmall cell lung cancer with a EGFR
mutation and an anaplastic lymphoma kinase mutation
with brain metastases, several agents have shown
intracranial activity. Newer EGFR-targeting agents with
better central nervous system penetration are in
development.
Lapatinib in combination with capecitabine and
neratinib has been tested in clinical trials for patients
with brain metastases from breast cancer harboring a
HER2 mutation.
Approximately 50% of patients with metastatic
melanoma carry a BRAF mutation. Small
moleculetargeted therapies such as vemurafenib and
dabrafenib have shown efficacy in BRAFV600-positive
melanoma brain metastases.
Innovative drug development and clinical trials for brain
metastases are needed to improve treatment options
and consequently outcomes of patients with brain
metastases.
significant difference in PFS and OS was noted between the

two groups.
The second-generation EGFR TKI afatinib is an irreversible
inhibitor of EGFR.18 It also inhibits HER2 and ERBB4 receptors and was initially designed to overcome T790M
mutation.18 One-hundred patients with brain metastases
from EGFR-mutant NSCLC received afatinib in a compassionate use program.19 All of the patients had been treated
previously with either erlotinib or gefitinib. For the 35
evaluable patients, time to tumor progression of 3.6 months
with an overall cerebral response rate of 35% was reported.
However, the CNS efficacy of afatinib has not been compared with either erlotinib or gefitinib.
A number of third-generation EGFR TKIs, including osimertinib (AZD-9291), rociletinib (CO-1686), ASP-8273, and
HM-61713, are in various phases of clinical investigation.20-23
These agents spare wild-type EGFRs and target mutant EGFRs,
including T790M.24 The activity of these newer agents against
brain metastases is currently being investigated. Initial studies
with the third-generation EGFR TKIs included several patients
with brain metastases. For example, 162 of 411 patients in the
phase I/II trial of AZD-9291 had CNS metastases,25 and 186 of
450 patients (41%) enrolled in the phase I/II trial of rociletinib
had a history of CNS involvement.26 In a report of patients who
were treated with osimertinib after progression on rociletinib,
all three patients with brain metastases showed a response.27
AZD-3759 is a novel EGFR TKI designed to cross the blood
brain barrier; however, it lacks activity against T790M mutation.2 Preclinical studies show that AZD-3759 is not a substrate of P-gp efflux pumps and has significantly higher
penetration across the bloodbrain barrier. Preliminary
results of the ongoing phase I trial of AZD-3759 demonstrated that the drug was well tolerated, with early evidence
of intracranial activity.28
Approximately 3%7% of patients with nonsquamous NSCLC
have ALK gene rearrangement, resulting in a novel fusion gene
containing portions of echinoderm microtubule-associated
protein like-4 and the ALK gene.29 This rearrangement leads
to constitutive activity of the kinase domain of the ALK, leading
to activation of the phosphoinositide 3-kinase and RAS pathways.30 Crizotinib is a first-generation ALK inhibitor that also
has activity against ROS1 and MET tyrosine kinases. Although
crizotinib does not have good bloodbrain barrier penetration,
CNS activity has been seen in several case reports and in
retrospective subanalyses of clinical trials. In a retrospective
pooled analysis of 388 patients enrolled in the PROFILE 10051007 clinical trial of crizotinib, 275 patients had asymptomatic
brain metastases.31 Of the 275 patients with brain metastases,
166 had received prior treatment. The intracranial disease
control rate was similar among those with prior treatment
compared with treatment-naive patients. However, the median intracranial time to tumor progression was 7 months for
patients with previously untreated brain metastases compared
with 13.2 months for patients with previously treated brain
metastases. In addition, the patients who continued crizotinib
beyond progression had better OS compared with those
who discontinued therapy at progression. A multi-institutional
TABLE 1. Selected Studies With Targeted Therapy in Brain Metastases
Study
Welsh et al
16
Lee et al17
No. of
Patients
Intracranial
Response
Rate (%)
Intracranial
ProgressionFree Survival
(Months)
Overall
Survival
(Months)
86
11.8
Specific Trial
Characteristics
Therapy
Tested
Concurrent
Local
Treatment
Phase II
Erlotinib started
a week prior to
and continued
through and
after WBRT
Erlotinib
WBRT
40
Phase II
A randomized trial
predominantly
among patients
with the EGFR
wild type; arm A:
WBRT alone;
arm B: WBRT plus
erlotinib
Erlotinib
WBRT
Arm A: 40
1.6
2.9
Arm B: 40
1.6
3.4
Primary
Malignancy
Design
NSCLC
NSCLC
Ceresoli et al81
NSCLC
Phase II
Single-arm study,
included both
newly diagnosed
and old BMs
Gefitinib
None
41
27
Sperduto et al82
(RTOG-0320)
NSCLC
Phase
III
3-arm study; arm A:

WBRT plus SRS;
arm B: WBRT plus
SRS plus temozolomide; arm
C: WBRT plus SRS
plus erlotinib
Erlotinib
WBRT plus
SRS
Arm A: 44
8.1
13.4
Arm B: 40
4.6
6.3
Arm C: 41
4.8
6.1
Bachelot et al59
(LANDSCAPE)
HER2positive
breast
cancer
Phase II
Newly diagnosed
BMs
Lapatinib
plus
capecitabine
None
45
65.9
5.5
17
Corte s et al61
HER2positive
breast
cancer
Phase II
3-arm study; arm A:

afatinib; arm B:
afatinib plus
vinorelbine;
arm C: investigators choice
Afatinib
None
Arm A: 40
30 (12 of 40
patients)*
Arm B: 38
34.2 (13 of 38
patients)*
Arm C: 43
41.9 (18 of 43
patients)*
Freedman et al63
HER2positive
breast
cancer
Phase II
Single-arm study
in previously
treated patients
Neratinib
None
40
1.9
8.7
Long et al71
BRAFpositive
melanoma
Phase II
Cohort A: no prior
treatment of
BMs; cohort B:
previously
treated BMs
Dabrafenib
None
Cohort A,
V600E: 74
39.2
4 (16.1 weeks)
8.2 (33.1
weeks)
Cohort A,
V600K: 15
6.7
2 (8.1 weeks)
7.7 (31.4
weeks)
Cohort B,
V600E: 65
30.8
4.1 (16.6
weeks)
4 (16.3
weeks)
Cohort B,
V600K: 18
22.2
4 (15.9 weeks)
5.2 (21.9
weeks)
Cohort 1: 90
18
3.7
Cohort 2: 56
20
3.94
9.53
Kefford et al73
BRAFpositive
melanoma
Phase II
Cohort 1: previously
untreated;
cohort 2: previously treated
Vemurafenib
None
Abbreviations: NSCLC, nonsmall cell lung cancer; WBRT, whole-brain radiation therapy; BM, brain metastasis; SRS, stereotactic radiation.
*Patient benefit (defined as intracranial or extracranial progression-free survival, new neurologic signs or symptoms related to tumor, or increased corticosteroid use) at
12 weeks.
retrospective study included 90 patients with brain metastases

from ALK-positive lung cancer.32 The majority of patients (84 of
90) received radiation therapy. Crizotinib was administered to
84 patients and a second-generation ALK inhibitor was given to
41 patients (21 received ceritinib, 16 were treated with brigatinib, two received alectinib, and two were administered
X-396). A median OS of 49.5 months from the diagnosis of a
brain metastasis was reported. The investigators concluded

that absence of extracranial disease, a Karnofsky performance
score (KPS) of 90 or greater, and no prior ALK inhibitor therapy
were independent prognostic factors for patients with brain
metastases from ALK-positive lung cancer. Several recent
studies have suggested CNS treatment failure and CNSpredominant progression among patients treated with
e125
VENUR AND AHLUWALIA
crizotinib.33-35 Aggressive tumor biology, poor CNS penetration

of crizotinib, and mutation in the binding domain of crizotinib
have been suggested as the likely reasons.36,37
Ceritinib and alectinib are second-generation ALK inhibitors that can be used to overcome drug resistance to
crizotinib.38,39 The phase I trial of ceritinib (ASCEND-1) accrued 246 patients, 124 of which had brain metastases at
baseline.40 Ninety-eight patients had received ALK inhibitors
in the past, whereas 26 patients were ALK-inhibitor naive. Of
the 14 patients with measurable brain metastases, seven
patients showed an intracranial response and three patients
had stable disease. Unlike crizotinib and ceritinib, alectinib is
not a substrate for P-gp efflux pumps, which may result in
a higher brain to plasma concentration of alectinib.41 The
phase I/II study of alectinib enrolled 47 patients who progressed or were intolerant to crizotinib.42 Eleven of the 21
patients with asymptomatic brain metastases had an objective response (six complete responses, five partial responses). Brigatinib is another ALK inhibitor that has shown
intracranial activity in an early phase I/II trial.43 A phase I trial
of brigatinib included 46 patients with brain metastases.
Thirteen patients had evaluable brain metastases; nine
patients (69%) had regression of the intracranial lesions,
including four patients with a complete response and two
with a partial response.44 The median intracranial PFS was
97 weeks. The phase II trial of this agent is currently accruing
patients and includes a cohort of patients with brain
metastases.44
TARGETED THERAPY FOR BREAST CANCER WITH

BRAIN METASTASES
Breast cancer is the second leading cause of brain metastases. The propensity of brain metastases varies with
subtypes of breast cancer. Patients with metastatic triplenegative breast cancer typically have brain metastases early
in the course of their disease, and they frequently present
with simultaneous progression of intracranial and extracranial disease.45,46 By contrast, patients with HER2 overexpression are often treated with targeted therapies that
include monoclonal antibodies, and they develop CNS metastases relatively later in the disease process, often with
stable extracranial disease. Approximately 30%55% of
patients with HER2-positive breast cancer develop brain
metastases.47,48 The increased incidence of brain metastases in this subtype of breast cancer is attributed to two main
reasons. First, anti-HER2 agents such as trastuzumab are
effective in controlling the systemic disease, prolonging
survival, and, in turn, unmasking otherwise asymptomatic
brain metastases. Second, the monoclonal antibodies directed against HER2, like trastuzumab, have poor CNS
penetration, especially for patients with an intact blood
brain barrier, thereby leading to brain metastases without
systemic progression of the disease.49 In the absence of
brain metastases, the CNS concentration of trastuzumab is
low compared with systemic concentrations.50 However,
among patients with brain metastases, studies using
e126
radiolabeled trastuzumab have shown increased CNS bioavailability.51,52 Although no prospective clinical trial has yet
reported on the intracranial activity of trastuzumab, several
retrospective studies have shown superior outcomes with
trastuzumab after radiation.53 In one such retrospective
study, the median OS of patients with HER2-overexpressing
breast cancer brain metastases was 21 months with WBRT
followed by trastuzumab, compared with 9 months with
WBRT followed by chemotherapy.53 Two-thirds of patients
with brain metastases still die of the systemic disease, and
better control of the primary cancer and extracranial metastases may be the reason for this observed survival benefit
with the use of trastuzumab. Hence, for a patient with
isolated CNS progression, it is reasonable to continue antiHER2 therapy as long as the extracranial disease is well
controlled.
HER2 (ERBB2) is shown to enhance EGFR signaling and vice
versa.54,55 Therefore, it is postulated that dual inhibition
of HER2 and EGFR by agents such as lapatinib may confer
improved outcomes among this patient population.56
Lapatinib in combination with capecitabine is approved for
treatment of patients with HER2-overexpressing metastatic
breast cancer who have progressed while taking trastuzumab. Lapatinib is a small molecule inhibitor; however, its
bioavailability in the CNS is limited by breast cancer resistance protein and P-gp efflux pumps.57 Lapatinib was
evaluated in an initial phase II study of patients with progressive brain metastases from HER2-positive breast cancer.58 The trial accrued 242 patients and all had received
prior radiation therapy and trastuzumab. The median PFS
and OS were 2.4 months and 6.37 months, respectively. This
study was amended to allow 50 additional patients who
were treated with lapatinib and capecitabine after progression with lapatinib. In this cohort, the median PFS was
3.65 months and a response rate of 20% was seen. The
pivotal study for the utility of lapatinib for patients with
breast cancer with brain metastases was the LANDSCAPE
trial, which investigated the combination of lapatinib and
capecitabine as a first-line combination therapy prior to
radiation.59 An intracranial response of 66% was seen among
the 45 patients enrolled in this phase II trial. The WBRT was
delayed by 8.6 months for patients with oligometastatic
asymptomatic brain metastases treated with lapatinib and
capecitabine. A median OS of 17.0 months was reported
among 44 patients who had measurable intracranial disease.
There is an ongoing randomized cooperative group study
evaluating WBRT in combination with lapatinib for patients
with brain metastases from HER2-positive breast cancer
(NCT01622868).
The drug-antibody conjugate T-DM1 (trastuzumab plus
emtansine) is approved by the FDA for treatment of patients with HER2-positive breast cancer. In a recent retrospective report, an intracranial response rate of 20% was
seen for 10 patients with asymptomatic or progressive
brain metastases.60 Prospective trials are needed to better
understand the clinical activity of T-DM1 among patients
with brain metastases.
Afatinib, a second-generation inhibitor of the ERBB family,

is approved by the FDA for treatment of patients with NSCLC
harboring an EGFR mutation.18 In a phase II study of 121
patients with HER2-overexpressing breast cancer with brain
metastases, 40 patients were treated with afatinib, 38 received afatinib and vinorelbine, and another 43 were treated
with the regimen of the investigators choice.61 Trastuzumab
and vinorelbine (11 of 43 patients) and lapatinib plus capecitabine (eight patients) were the most common treatments
chosen by the treating investigator. The primary study endpoint (absence of CNS or extracranial disease progression,
new neurologic symptoms, or new corticosteroid use [termed
as patient benefit]) at 12 weeks was seen in 12 of 40 patients
(30%) in the afatinib arm, 13 of 38 patients (34%) in the
afatinib plus vinorelbine arm, and 18 of 43 patients (42%) in
the investigator choice arm.
Neratinib is a potent irreversible inhibitor of EGFR, HER2,
and ERBB4 transmembrane tyrosine kinase receptors.62 A
phase II single-arm study evaluated neratinib among patients with HER2-overexpressing breast cancer with brain
metastases.63 The primary trial endpoint was intracranial
response rate, which also included neurocognitive and qualityof-life (QoL) assessments. The intracranial response rate among
patients enrolled in the study was 8% (3 of 40 patients failed to
meet the primary endpoint of success). The rapid accrual of
patients and the neurocognitive and QoL endpoints of the trial
were noteworthy.
Pertuzumab, an antibody against the extracellular dimerization domain of the HER2 receptor, has been shown to
increase efficacy when used with trastuzumab and docetaxel
among patients with untreated HER2-overexpressing metastatic breast cancer.64 A case report has shown intracranial
activity from the combination of pertuzumab, trastuzumab,
and docetaxel.65
Patients with luminal breast cancer have a lower frequency of brain metastases.66 It has been suggested that
hormone receptors are lost or altered during the process of
acquiring brain metastases.67 Newer agents such as mTOR
inhibitors and CDK4/6 have shown extracranial activity in
advanced luminal breast cancer; however, their role in the
treatment of patients with brain metastases is yet to be
defined. There have been limited novel agents available for
the treatment of patients with triple-negative breast cancer
with brain metastases who have a dismal prognosis.
TARGETED THERAPY FOR BRAIN METASTASES

FROM MELANOMA
Targeted agents against driver oncogenic mutations among cell
signaling pathways have shown benefit in metastatic melanoma. BRAF, NRAS, and KIT are the mutually exclusive driver
mutations commonly seen in metastatic melanoma.68 The
presence of these mutations increases the risk of CNS metastases at the diagnosis of stage IV melanoma. In a 2012 study, a
higher prevalence of melanoma brain metastases was observed
in BRAF- and NRAS-mutant melanoma (24% and 23%, respectively) compared with wild type (12%).69 BRAF mutation is
seen for approximately 50% of patients with metastatic melanoma and plays an important role in the MAPK (RAS-RAF-MEKERK) pathway. Dabrafenib and vemurafenib are agents active
against BRAF V600E or V600K mutation that are approved by
the FDA for treatment of advanced metastatic melanoma. In the
phase I trial with dabrafenib, 8 of 10 patients with untreated
asymptomatic brain metastases showed an intracranial
response (four had a complete response and four had a partial
response).70 This initial observation led to a phase II multicenter open-label study of dabrafenib among 172 patients with
asymptomatic brain metastases with a BRAFV600E or BRAFV600K
mutation and at least one measurable brain metastasis (BREAKMB).71 Cohort A consisted of 89 patients who had never received any local treatment and cohort B comprised 83 patients
with a history of prior radiation therapy for brain metastases.
A modified version of the RECIST criteria with up to five intracranial and extracranial index lesions was used to measure
the response to treatment. In cohorts A and B, patients with a
BRAFV600E mutation had intracranial response rates of 39%
(29 of 74) and 31% (20 of 65), respectively. Progression-free
survival of 16.1 weeks and 16.6 weeks and OS of 33.1 and
31.4 weeks were noted in cohorts A and B, respectively. Patients with a BRAFV600K mutation had lower intracranial response rates of 7% (1 of 15) and 22% (4 of 18) in cohorts A and B,
respectively. This trial demonstrated that dabrafenib is active in
BRAF-mutant melanoma brain metastases, specifically among
patients with a BRAFV600E mutation with acceptable toxicity.
Vemurafenib was noted to have intracranial activity in an openlabel pilot trial of 24 untreated patients with melanoma brain
metastases harboring a BRAFV600 mutation.72 Of 19 patients with
measurable intracranial disease, seven had tumor regression of
more than 30% and three achieved a partial response. Median
PFS and OS were 3.9 and 5.3 months, respectively. Initial results
of a phase II study of 146 patients with BRAFV600 melanoma brain
metastases treated with vemurafenib were reported at the 2013
Society of Melanoma Research Congress.73 Cohort 1 consisted of
90 patients with newly diagnosed melanoma brain metastases
and cohort 2 included 56 patients with melanoma brain metastases who progressed after initial local therapy. Both cohorts
had similar intracranial response rates at 18% and 20%, respectively. Median PFS and OS were 3.7 and 6.5 months in cohort
1 and 4.0 and 6.4 months in cohort 2, respectively.
The feasibility, safety profile, and effectiveness of combining BRAF inhibitors and radiation have not yet been
tested in a prospective setting. There is concern regarding an
increased incidence of dermatitis with concurrent use of
radiation and BRAF inhibitors, specifically in the extracranial
setting.74 A retrospective analysis evaluated the effectiveness
of combining vemurafenib and radiation in BRAFV600 melanoma brain metastases.75 Six patients were treated with SRS:
two received WBRT, one received SRS and WBRT, and the
remaining three received surgery and radiation. All 12 patients
received vemurafenib. Of the 48 index lesions, 36 showed a
response, with 23 (48%) complete responses and 13 (27%)
partial responses. Several other small retrospective case series
have been reported on outcomes of the combination of targeted agents with SRS/WBRT.76-78
e127
VENUR AND AHLUWALIA
CONCLUSION
Targeted therapies have improved survival in a subset of
patients with NSCLC, breast cancer, and melanoma with
actionable driver mutations. For patients with brain metastases, these agents provide not only intracranial control of the
disease, but they also help manage systemic cancer. The
clinical and pathobiology of brain metastases for patients with
mutations such as EGFR may differ from other patients with
NSCLC without such mutations.79 Some of the newer targeted
agents have shown promising results in the treatment of brain
metastases, despite limitations of the bloodbrain barrier and

efflux pumps. Over the last few years, our understanding of
molecular drivers of brain metastases, the bloodbrain barrier, and CNS penetration of drugs has improved tremendously.80 Improved understanding of biology, multimodality
treatment, and innovative drug development can help lead to
improvement in the management of brain metastases. In the
future, increasingly targeted therapies will have a significant
role in the treatment of patients with brain metastases with
actionable mutations.
References
1. Tabouret E, Chinot O, Metellus P, et al. Recent trends in epidemiology of
brain metastases: an overview. Anticancer Res. 2012;32:4655-4662.
2. Kim DW, Yang JCH, Chen K, et al. AZD3759, an EGFR inhibitor with blood
brain barrier (BBB) penetration for the treatment of non-small cell lung
cancer (NSCLC) with brain metastasis (BM): preclinical evidence and
clinical cases. J Clin Oncol. 2015;33 (suppl; abstr 8016).
3. DAntonio C, Passaro A, Gori B, et al. Bone and brain metastasis in lung
cancer: recent advances in therapeutic strategies. Ther Adv Med Oncol.
2014;6:101-114.
4. Shigematsu H, Lin L, Takahashi T, et al. Clinical and biological features
associated with epidermal growth factor receptor gene mutations in
lung cancers. J Natl Cancer Inst. 2005;97:339-346.
5. Lynch TJ, Bell DW, Sordella R, et al. Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-smallcell lung cancer to gefitinib. N Engl J Med. 2004;350:2129-2139.
6. Eichler AF, Kahle KT, Wang DL, et al. EGFR mutation status and survival
after diagnosis of brain metastasis in nonsmall cell lung cancer. Neuro
Oncol. 2010;12:1193-1199.
7. Tan CS, Cho BC, Soo RA. Next-generation epidermal growth factor
receptor tyrosine kinase inhibitors in epidermal growth factor receptor
-mutant non-small cell lung cancer. Lung Cancer. 2016;93:59-68.
8. Weber B, Winterdahl M, Memon A, et al. Erlotinib accumulation in brain
metastases from non-small cell lung cancer: visualization by positron
emission tomography in a patient harboring a mutation in the epidermal
growth factor receptor. J Thorac Oncol. 2011;6:1287-1289.
9. Elmeliegy MA, Carcaboso AM, Tagen M, et al. Role of ATP-binding
cassette and solute carrier transporters in erlotinib CNS penetration and
intracellular accumulation. Clin Cancer Res. 2011;17:89-99.
10. Kitazaki T, Oka M, Nakamura Y, et al. Gefitinib, an EGFR tyrosine kinase
inhibitor, directly inhibits the function of P-glycoprotein in multidrug
resistant cancer cells. Lung Cancer. 2005;49:337-343.
11. Togashi Y, Masago K, Fukudo M, et al. Cerebrospinal fluid concentration
of erlotinib and its active metabolite OSI-420 in patients with central
nervous system metastases of non-small cell lung cancer. J Thorac
Oncol. 2010;5:950-955.
12. Wakeling AE, Guy SP, Woodburn JR, et al. ZD1839 (Iressa): an orally
active inhibitor of epidermal growth factor signaling with potential for
cancer therapy. Cancer Res. 2002;62:5749-5754.
13. Iuchi T, Shingyoji M, Sakaida T, et al. Phase II trial of gefitinib alone
without radiation therapy for Japanese patients with brain metastases from EGFR-mutant lung adenocarcinoma. Lung Cancer. 2013;82:
282-287.
14. Kim JE, Lee DH, Choi Y, et al. Epidermal growth factor receptor tyrosine
kinase inhibitors as a first-line therapy for never-smokers with adenocarcinoma of the lung having asymptomatic synchronous brain
metastasis. Lung Cancer. 2009;65:351-354.
e128
15. Dempke WC, Edvardsen K, Lu S, et al. Brain metastases in NSCLC - are

TKIs changing the treatment strategy? Anticancer Res. 2015;35:
5797-5806.
16. Welsh JW, Komaki R, Amini A, et al. Phase II trial of erlotinib plus
concurrent whole-brain radiation therapy for patients with brain
metastases from non-small-cell lung cancer. J Clin Oncol. 2013;31:895902.
17. Lee SM, Lewanski CR, Counsell N, et al. Randomized trial of erlotinib plus
whole-brain radiotherapy for NSCLC patients with multiple brain metastases. J Natl Cancer Inst. 2014;106:dju151.
18. Sequist LV, Yang JC, Yamamoto N, et al. Phase III study of afatinib or
cisplatin plus pemetrexed in patients with metastatic lung adenocarcinoma with EGFR mutations. J Clin Oncol. 2013;31:3327-3334.
19. Hoffknecht P, Tufman A, Wehler T, et al. Efficacy of the irreversible ErbB
family blocker afatinib in epidermal growth factor receptor (EGFR)
tyrosine kinase inhibitor (TKI)-pretreated non-small-cell lung cancer
patients with brain metastases or leptomeningeal disease. J Thorac
Oncol. 2015;10:156-163.
20. Cross DA, Ashton SE, Ghiorghiu S, et al. AZD9291, an irreversible EGFR
TKI, overcomes T790M-mediated resistance to EGFR inhibitors in lung
cancer. Cancer Discov. 2014;4:1046-1061.
21. Sequist LV, Soria JC, Goldman JW, et al. Rociletinib in EGFR-mutated
non-small-cell lung cancer. N Engl J Med. 2015;372:1700-1709.
22. Sakagami H, Konagai S, Yamamoto H, et al. ASP8273, a novel mutantselective irreversible EGFR inhibitor, inhibits growth of non-small cell
lung cancer (NSCLC) cells with EGFR activating and T790M resistance
mutations. Cancer Res. 2014;74 (suppl; abstr 1728).
23. Lee KO, Cha MY, Kim M, et al. Discovery of HM61713 as an orally
available and mutant EGFR selective inhibitor. Cancer Res. 2014;74
(suppl; abstr LB-100).
24. Janne PA, Yang JC, Kim DW, et al. AZD9291 in EGFR inhibitor-resistant
non-small-cell lung cancer. N Engl J Med. 2015;372:1689-1699.
25. Ahn MJ, Tsai CM, Yang JCH, et al. 3083 AZD9291 activity in patients with
EGFR-mutant advanced non-small cell lung cancer (NSCLC) and brain
metastases: data from phase II studies. Eur J Cancer. 2015;51:
S625-S626.
26. Bedikian AY, Papadopoulos NE, Kim KB, et al. A phase IB trial of intravenous INO-1001 plus oral temozolomide in subjects with unresectable stage-III or IV melanoma. Cancer Invest. 2009;27:756-763.
27. Sequist LV, Piotrowska Z, Niederst MJ, et al. Osimertinib responses after
disease progression in patients who had been receiving rociletinib.
JAMA Oncol. 2016;2:541-543.
28. Zeng Q, Wang J, Cheng Z, et al. Discovery and evaluation of clinical
candidate AZD3759, a potent, oral active, central nervous system-penetrant,
epidermal growth factor receptor tyrosine kinase inhibitor. J Med Chem.
2015;58:8200-8215.
29. Bohn JP, Pall G, Stockhammer G, et al. Targeted therapies for the
treatment of brain metastases in solid tumors. Target Oncol. Epub 2016
Jan 29.
30. Zhang I, Zaorsky NG, Palmer JD, et al. Targeting brain metastases in
ALK-rearranged non-small-cell lung cancer. Lancet Oncol. 2015;16:
e510-e521.
31. Costa DB, Shaw AT, Ou SH, et al. Clinical experience with crizotinib in
patients with advanced ALKS-rearranged non-small-cell lung cancer and
brain metastases. J Clin Oncol. 2015;33:1881-1888.
32. Johung KL, Yeh N, Desai NB, et al. Extended survival and prognostic
factors for patients with ALK-rearranged non-small-cell lung cancer and
brain metastasis. J Clin Oncol. 2016;34:123-129.
33. Maillet D, Martel-Lafay I, Arpin D, et al. Ineffectiveness of crizotinib on
brain metastases in two cases of lung adenocarcinoma with EML4-ALK
rearrangement. J Thorac Oncol. 2013;8:e30-e31.
34. Riess JW, Nagpal S, Neal JW, et al. A patient with anaplastic lymphoma
kinase-positive non-small cell lung cancer with development of leptomeningeal carcinomatosis while on targeted treatment with crizotinib. J Natl Compr Canc Netw. 2013;11:389-394.
35. Chun SG, Choe KS, Iyengar P, et al. Isolated central nervous system
progression on crizotinib: an Achilles heel of non-small cell lung cancer
with EML4-ALK translocation? Cancer Biol Ther. 2012;13:1376-1383.
36. Costa DB, Kobayashi S, Pandya SS, et al. CSF concentration of the
anaplastic lymphoma kinase inhibitor crizotinib. J Clin Oncol. 2011;29:
e443-e445.
37. Choi YL, Soda M, Yamashita Y, et al; ALK Lung Cancer Study Group.
EML4-ALK mutations in lung cancer that confer resistance to ALK inhibitors. N Engl J Med. 2010;363:1734-1739.
38. Shaw AT, Kim DW, Mehra R, et al. Ceritinib in ALK-rearranged non-smallcell lung cancer. New Engl J Med. 2014;370:1189-1197.
39. Seto T, Kiura K, Nishio M, et al. CH5424802 (RO5424802) for patients
with ALK-rearranged advanced non-small-cell lung cancer (AF-001JP
study): a single-arm, open-label, phase 1-2 study. Lancet Oncol. 2013;
14:590-598.
40. Shaw A, Mehra R, Tan DSW, et al. BM-32 Ceritinib (LDK378) for
treatment of patients with ALK-rearranged (ALK+) non-small cell lung
cancer (NSCLC) and brain metastases (BM) in the ASCEND-1 trial. Neuro
Oncol. 2014;16:v39.
41. Kodama T, Hasegawa M, Takanashi K, et al. Antitumor activity of the
selective ALK inhibitor alectinib in models of intracranial metastases.
Cancer Chemother Pharmacol. 2014;74:1023-1028.
42. Gadgeel SM, Gandhi L, Riely GJ, et al. Safety and activity of alectinib
against systemic disease and brain metastases in patients with crizotinibresistant ALK-rearranged non-small-cell lung cancer (AF-002JG): results
from the dose-finding portion of a phase 1/2 study. Lancet Oncol. 2014;
15:1119-1128.
43. Camidge DR, Bazhenova L, Salgia R, et al. Safety and efficacy of brigatinib (AP26113) in advanced malignancies, including ALK+ non-small
cell lung cancer (NSCLC). J Clin Oncol. 2015;33 (suppl; abstr 8062).
44. Kerstein D, Gettinger S, Gold K, et al. LBA4: evaluation of anaplastic
lymphoma kinase (ALK) inhibitor brigatinib [AP26113] in patients with
ALK+ non-small cell lung cancer (NSCLC) and brain metastases. Ann
Oncol. 2015;26:i60-i61.
45. Dawood S, Lei X, Litton JK, et al. Incidence of brain metastases as a first
site of recurrence among women with triple receptor-negative breast
cancer. Cancer. 2012;118:4652-4659.
46. Lin NU, Claus E, Sohl J, et al. Sites of distant recurrence and clinical
outcomes in patients with metastatic triple-negative breast cancer: high
incidence of central nervous system metastases. Cancer. 2008;113:
2638-2645.
47. Brufsky AM, Mayer M, Rugo HS, et al. Central nervous system metastases in patients with HER2-positive metastatic breast cancer:
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
incidence, treatment, and survival in patients from registHER. Clin

Cancer Res. 2011;17:4834-4843.
Olson EM, Abdel-Rasoul M, Maly J, et al. Incidence and risk of central
nervous system metastases as site of first recurrence in patients with
HER2-positive breast cancer treated with adjuvant trastuzumab. Ann
Oncol. 2013;24:1526-1533.
Pestalozzi BC, Holmes E, de Azambuja E, et al. CNS relapses in patients
with HER2-positive early breast cancer who have and have not received
adjuvant trastuzumab: a retrospective substudy of the HERA trial (BIG
1-01). Lancet Oncol. 2013;14:244-248.
Stemmler HJ, Kahlert S, Siekiera W, et al. Characteristics of patients with
brain metastases receiving trastuzumab for HER2 overexpressing
metastatic breast cancer. Breast. 2006;15:219-225.
Dijkers EC, Oude Munnink TH, Kosterink JG, et al. Biodistribution
of 89Zr-trastuzumab and PET imaging of HER2-positive lesions in
patients with metastatic breast cancer. Clin Pharmacol Ther. 2010;
87:586-592.
Tamura K, Kurihara H, Yonemori K, et al. 64Cu-DOTA-trastuzumab PET
imaging in patients with HER2-positive breast cancer. J Nucl Med. 2013;
54:1869-1875.
Bartsch R, Rottenfusser A, Wenzel C, et al. Trastuzumab prolongs overall
survival in patients with brain metastases from Her2 positive breast
cancer. J Neurooncol. 2007;85:311-317.
Worthylake R, Opresko LK, Wiley HS. ErbB-2 amplification inhibits
down-regulation and induces constitutive activation of both ErbB-2 and
epidermal growth factor receptors. J Biol Chem. 1999;274:8865-8874.
Graus-Porta D, Beerli RR, Daly JM, et al. ErbB-2, the preferred heterodimerization partner of all ErbB receptors, is a mediator of lateral signaling. EMBO J. 1997;16:1647-1655.
Konecny GE, Pegram MD, Venkatesan N, et al. Activity of the dual
kinase inhibitor lapatinib (GW572016) against HER-2-overexpressing
and trastuzumab-treated breast cancer cells. Cancer Res. 2006;66:
1630-1639.
Polli JW, Olson KL, Chism JP, et al. An unexpected synergist role
of P-glycoprotein and breast cancer resistance protein on the
central nervous system penetration of the tyrosine kinase inhibitor
lapatinib (N-3-chloro-4-[(3-fluorobenzyl)oxy]phenyl-6-[5-([2-(methylsulfonyl)
ethyl]aminomethyl)-2-furyl]-4-quinazolinamine; GW572016). Drug Metab
Dispos. 2009;37:439-442.
Lin NU, Dieras V, Paul D, et al. Multicenter phase II study of lapatinib in
patients with brain metastases from HER2-positive breast cancer. Clin
Cancer Res. 2009;15:1452-1459.
Bachelot T, Romieu G, Campone M, et al. Lapatinib plus capecitabine in
patients with previously untreated brain metastases from HER2-positive
metastatic breast cancer (LANDSCAPE): a single-group phase 2 study.
Lancet Oncol. 2013;14:64-71.
Bartsch R, Berghoff AS, Vogl U, et al. Activity of T-DM1 in Her2-positive
breast cancer brain metastases. Clin Exp Metastasis. 2015;32:729-737.
Cortes J, Dieras V, Ro J, et al. Afatinib alone or afatinib plus vinorelbine
versus investigators choice of treatment for HER2-positive breast
cancer with progressive brain metastases after trastuzumab, lapatinib,
or both (LUX-Breast 3): a randomised, open-label, multicentre, phase 2
Zhang X, Munster PN. New protein kinase inhibitors in breast cancer:
afatinib and neratinib. Expert Opin Pharmacother. 2014;15:1277-1288.
Freedman RA, Gelman RS, Wefel JS, et al. Translational Breast Cancer
Research Consortium (TBCRC) 022: a phase II trial of neratinib for
patients with human epidermal growth factor receptor 2-positive breast
cancer and brain metastases. J Clin Oncol. 2016;34:945-952.
Swain SM, Baselga J, Kim SB, et al; CLEOPATRA Study Group. Pertuzumab, trastuzumab, and docetaxel in HER2-positive metastatic breast
e129
VENUR AND AHLUWALIA
65. Senda N, Yamaguchi A, Nishimura H, et al. Pertuzumab, trastuzumab

and docetaxel reduced the recurrence of brain metastasis from breast
cancer: a case report. Breast Cancer. 2016;23:323-328.
66. Arvold ND, Oh KS, Niemierko A, et al. Brain metastases after breastconserving therapy and systemic therapy: incidence and characteristics
by biologic subtype. Breast Cancer Res Treat. 2012;136:153-160.
67. Duchnowska R, Dziadziuszko R, Trojanowski T, et al. Conversion of
epidermal growth factor receptor 2 and hormone receptor expression
in breast cancer metastases to the brain. Breast Cancer Res. 2012;14:
R119.
68. Long GV, Margolin KA. Multidisciplinary approach to brain metastasis
from melanoma: the emerging role of systemic therapies. Am Soc Clin
Oncol Educ Book. 2013;33:393-398.
69. Jakob JA, Bassett RL Jr, Ng CS, et al. NRAS mutation status is an independent prognostic factor in metastatic melanoma. Cancer. 2012;
118:4014-4023.
70. Falchook GS, Long GV, Kurzrock R, et al. Dabrafenib in patients with
melanoma, untreated brain metastases, and other solid tumours: a phase
1 dose-escalation trial. Lancet. 2012;379:1893-1901.
71. Long GV, Trefzer U, Davies MA, et al. Dabrafenib in patients with
Val600Glu or Val600Lys BRAF-mutant melanoma metastatic to the
brain (BREAK-MB): a multicentre, open-label, phase 2 trial. Lancet Oncol.
2012;13:1087-1095.
72. Dummer R, Goldinger SM, Turtschi CP, et al. Vemurafenib in patients
with BRAF(V600) mutation-positive melanoma with symptomatic brain
metastases: final results of an open-label pilot study. Eur J Cancer. 2014;
50:611-621.
73. Kefford R, Malo M, Arance A, et al. Vemurafenib in metastatic melanoma patients with brain metastases: an open label, single arm, phase
e130
74.
75.
76.
77.
78.
79.
80.
81.
82.
2, multicenter study. Paper presented at: 2013 Society for Melanoma

Research Congress; November 2013; Philadelphia, PA.
Satzger I, Degen A, Asper H, et al. Serious skin toxicity with the combination of BRAF inhibitors and radiotherapy. J Clin Oncol. 2013;31:
e220-e222.
Narayana A, Mathew M, Tam M, et al. Vemurafenib and radiation
therapy in melanoma brain metastases. J Neurooncol. 2013;113:
411-416.
Ahmed KA, Freilich JM, Sloot S, et al. LINAC-based stereotactic radiosurgery to the brain with concurrent vemurafenib for melanoma metastases. J Neurooncol. 2015;122:121-126.
Wolf A, Zia S, Verma R, et al. Impact on overall survival of the combination of BRAF inhibitors and stereotactic radiosurgery in patients
with melanoma brain metastases. J Neurooncol. Epub 2016 Feb 6.
Patel BG, Ahmed KA, Johnstone PA, et al. Initial experience with combined
BRAF and MEK inhibition with stereotactic radiosurgery for BRAF mutant
melanoma brain metastases. Melanoma Res. Epub 2016 Feb 29.
Eichler AF, Chung E, Kodack DP, et al. The biology of brain metastasestranslation to new therapies. Nat Rev Clin Oncol. 2011;8:344-356.
Berghoff AS, Preusser M. The future of targeted therapies for brain
metastases. Future Oncol. 2015;11:2315-2327.
Ceresoli GL, Cappuzzo F, Gregorc V, et al. Gefitinib in patients with brain
metastases from non-small-cell lung cancer: a prospective trial. Ann
Oncol. 2004;15:1042-1047.
Sperduto PW, Wang M, Robins HI, et al. A phase 3 trial of whole brain
radiation therapy and stereotactic radiosurgery alone versus WBRT and
SRS with temozolomide or erlotinib for non-small cell lung cancer and 1
to 3 brain metastases: Radiation Therapy Oncology Group 0320. Int J
Radiat Oncol Biol Phys. 2013;85:1312-1318.
The Future of Immunotherapy in

Glioblastoma
CHAIR
E. Antonio Chiocca, MD, PhD
Brigham and Womens Hospital
Boston, MA
SPEAKERS
Michael Lim, MD
The Johns Hopkins Hospital
Baltimore, MD
Michael Weller, MD
University Hospital Zurich
Zurich, Switzerland
LIM, WELLER, AND CHIOCCA
Current State of Immune-Based Therapies for Glioblastoma

Michael Lim, MD, Michael Weller, MD, and E. Antonio Chiocca, MD, PhD
OVERVIEW
Glioblastoma is one of the most aggressive solid tumors, and, despite treatment options such as surgery, radiation, and chemotherapy, its prognosis remains grim. Novel approaches are needed to improve survival. Immunotherapy has proven efficacy for
melanoma, lung cancer, and kidney cancer and is now a focus for glioblastoma. In this article, glioblastoma-mediated immunosuppression will be discussed and two exciting immune approaches, checkpoint inhibitors and viral-based therapies, will be
reviewed.
lioblastoma has been studied as a paradigmatic tumor

for cancer-associated immunosuppression for more
than 3 decades. Initial observations included the characterization of decreased immune responsiveness of peripheral blood cells harvested from patients with glioblastoma.1
This remote effect of a locally growing neoplasm could only
be explained by soluble factors released by the tumor in
sufficient quantities to induce systemic immunosuppression
(Fig. 1). The search for soluble factors produced by cultured
glioblastoma cells resulted in the identification of, among
others, transforming growth factor (TGF)-b as a key immunosuppressive cytokine that has remained a therapeutic
target until today. Yet, local application of antisense oligonucleotides failed presumably because of poor target
coverage, whereas systemic application of TGF-b receptor
antagonists has its limits in nonhematologic toxicity. TGF-b
is a member of a large family of cytokines that interacts with
heterodimeric receptors. There are three TGF-b isoforms in
humans that have nonoverlapping functions, at least in
development, as demonstrated in knockout mouse models.
In contrast, the three isoforms are coregulated in glioblastomas, and no isotype-specific roles have been identified so
far in the biology of glioblastoma.2 Other soluble immunosuppressive factor candidates attributed a role in shaping
the immunosuppressive microenvironment in glioblastoma
include interleukin 10 and prostaglandin E2.
Admittedly, it has not been clarified definitively that the
peripheral immunosuppression encountered among patients
with glioblastoma is caused exclusively or even mainly by elevated levels of soluble mediators systemically. Alternatively,
it is conceivable that immune modulatory cells that encounter the glioblastoma microenvironment are forced to develop into an immunosuppressive immune cell population that
indirectly mediates systemic immunosuppression. In fact, it

has recently been recognized that glioblastoma cells are
capable of reshaping the phenotype of tumor-infiltrating
host cells, which comprise a large proportion of cells within
the microenvironment of glioblastoma, to support their
growth and maintain any immunosuppressive milieu.3
In addition to the soluble immunosuppressive molecule
candidates mentioned above, immune-relevant molecules
expressed at the cell surface of glioblastoma have attracted
interest because there is major infiltration of glioblastomas
by host cells, although the cells are mainly of macrophagemonocyte lineage. These surface molecules include the CD95
ligand, which may induce apoptosis in susceptible cells that
express the receptor, CD95, previously referred to as Fas or
APO-1. In the context of glioblastoma, cellular targets for
CD95-mediated apoptosis are probably mainly T cells. Additional candidate cell surface molecules with immunosuppressive properties, but with uncertain significance in
glioblastoma, include regeneration and tolerance factor (RTF),
lectin-like transcript 1, and HLA-E and HLA-G.4
More recently, major emphasis has been placed on the
aberrant expression of PD-L1 by glioblastoma cells. PD-L1 is
the ligand for PD-1, a cell surface molecule expressed mainly
on T cells now referred to as an immune checkpoint, together with CTLA-4. These two molecules, PD-1 and CTLA-4,
mediate inactivation of T cells, and antibody-mediated
therapeutic neutralization of these molecules with agents
such as nivolumab or pembrolizumab that target PD-1, or
ipilimumab that targets CTLA-4, can be considered the
greatest innovation in medical oncology in the past decade.5
They are currently being explored as novel agents across the
full spectrum of human cancers and have already been integrated into the standard of care for malignant melanoma.
From The Johns Hopkins University, Baltimore, MD; University Hospital Zurich, Zurich, Switzerland; Institute for the Neurosciences at the Brigham and Womens/Faulkner Hospital,
Center for Neuro-Oncology, Dana-Farber Cancer Institute, Boston, MA; The University of Chicago, Chicago, IL.
Corresponding author: Michael Lim, MD, Johns Hopkins Hospital, Neurosurgery-Phipps 123, 600 North Wolfe St., Baltimore, MD 21287; email: mlim3@jhmi.edu.
e132
CURRENT STATE OF IMMUNE-BASED THERAPIES FOR GLIOBLASTOMA
Furthermore, antibodies targeting the ligand PD-L1 rather

than the receptor, such as atezolizumab, also are being explored in various tumor entities. The extent that glioblastomas
express PD-L1 in vivo has remained controversial,6 but PD-1
and CTLA-4 inhibition are remarkably active in mouse glioma
models, both as single agents and in combination. Importantly, inhibition of immune checkpoints probably confers a
nonspecific state of immune activation that will be of benefit
to patients if their particular cancers carry a high mutational
load, rendering them potentially recognizable as altered.
Whether the observed benefit can be translated to glioblastoma remains to be clarified. Several clinical trial initiatives
to explore immune checkpoint inhibitors in glioblastoma are
now underway.
Because of disappointing results with traditional cancer
therapy (radiotherapy, chemotherapy, and angiogenesis
inhibition by promising agents, such as VEGF pathway inhibitors bevacizumab and cediranib) in this disease, various
approaches to immunotherapy have gained a lot of interest.
As of early 2016, several phase II and phase III clinical trials of
immunotherapy are underway or have already been completed, including but not limited to (1) the rindopepimut
vaccine that targets epidermal growth factor variant III,
with a completed phase III trial in the newly diagnosed
setting (ACT IV) and encouraging activity in phase II in the
recurrent setting (Re-ACT); (2) the six-candidate peptide
cocktail ICT-107, which is used to generate a vaccine by
ex vivo stimulation of patient-derived dendritic cells, with a
completed randomized phase II trial that indicated activity in
patients who were HLA-A2 positive and a phase III program
scheduled to start in early 2016; and (3) the DCVax phase III
program that is near completion, with high logistical
complexityit uses autologous tumor to stimulate autologous dendritic cells.
These treatments likely are most effective with minimal
residual tumor burden, which should be associated with
decreased levels of soluble immunosuppressive factors and
KEY POINTS
Immunotherapy has proven efficacy for other solid

tumors and is now a focus for glioblastoma.
An important component of glioblastoma-mediated
immunosuppression stems from soluble factors such as
TGF-b.
Checkpoint inhibition probably confers a nonspecific
state of immune activation that is a promising approach
to glioblastoma, given the success of checkpoint
inhibitors in melanoma, lung cancer, and renal cancers.
Current strategies for biomarkers have focused on
protein expression, immune cell characterization, and
mutational burden.
Tumor-selective viruses and viral-based vectors have
shown increasingly promising results as easy-to-use
immunostimulants.
decreased potential for immune cell re-education by bulky

tumor. Accordingly, most recent immunotherapy trials have
been and are now being conducted in highly selected patient
populations in the newly diagnosed setting and among
patients with little or no residual tumor after concomitant temozolomide chemoradiotherapy. Given these developments, it can be assumed that neurosurgery will be
attributed a larger role not only in newly diagnosed patients
but also in the setting of recurrent disease, if the first definitively positive immunotherapy data become available.
Although removing the source of immunosuppression surgically seems to be the most straightforward approach to
overcome immunosuppression, additional strategies include
neutralizing TGF-b, at least transiently, or employing immune adjuvants, such as granulocyte macrophage colonystimulating factor or toll-like receptor 2 agonists. The most
powerful combinatorial strategy for tumor-specific vaccines,
however, is likely to be the checkpoint inhibitors, as outlined
above. Such additional efforts at boosting immune responses may be essential for a benefit from upcoming
strategies of immunotherapy in broader populations of
patients, including patients with bulky disease, heavily
pretreated patients, and older patients who are likely to
exhibit impaired immune responsiveness.
CHECKPOINT INHIBITORS
An important component of tumor-induced immunosuppression involves the costimulatory interaction. Normally,
when a T cell encounters an antigen-presenting cell (APC)
expressing the appropriate antigen, a second interaction
with a checkpoint molecule is required to either activate or
suppress the T cell (Fig. 2).7,8 This second interaction plays
an important role in modulating an immune response.
Furthermore, there are multiple costimulatory molecules,
which suggest that a hierarchy of activation status exists. In
addition, this interaction is not unique to APCs and T cells;
other immune cells, such as natural killer cells and regulatory
T cells (Tregs), also have costimulatory molecules.8 Hence,
this suggests a hierarchy of immune cell activation as well as
an ability to attenuate an immune response.
Checkpoint inhibitors are a class of antibodies that are
designed to interrupt or activate these costimulatory molecules. Intense investigation is underway for the utilization
of checkpoint inhibitors for the treatment of solid tumors.
CTLA-4 was one of the first molecules to be studied. Leach
et al9 found that they could induce an antitumor immune
response within a murine model for melanoma by using an
antiCTLA-4 antibody.9 The second checkpoint inhibitor that
has been intensely studied is antiPD-1. AntiPD-1 has been
shown to induce an antitumor immune response in multiple
solid tumors, including glioblastoma.10,11
Proven Efficacy of Checkpoint Inhibitors

Several large clinical trials in humans verified the observed
efficacy of checkpoint inhibitors in the preclinical models. In
2010, Hodi et al12 ran a large phase III trial that demonstrated
e133
improved survival among patients with melanoma who were

treated with antiCTLA-4. The investigators also found that a
subset of patients were long-term survivors.12 Topalian et al13
then published their experience in treating multiple solid
tumors with antiPD-1, in which they found that antiPD-1
improved survival for patients with melanoma, renal cell
carcinoma, and nonsmall cell lung cancer. These and other
important studies resulted in U.S. Food and Drug Administration (FDA) approval of checkpoint inhibitors. Checkpoint
inhibitors first gained FDA approval for use in patients with
melanoma in 2011, with the approval of antiCTLA-4. The FDA
approved antiPD-1 in September 2014 for melanoma.
Shortly after, antiPD-1 was approved for lung cancer (both
adenocarcinoma and squamous cell carcinoma) and renal cell
cancer. Hence, the enthusiasm for the use of checkpoint
inhibitors in glioblastoma is high.
Checkpoint Inhibitors for Glioblastoma

There are promising preclinical data suggesting that
checkpoint inhibitors may promote an antitumor immune
response. Fecci et al14 demonstrated improved survival
mediated by a CD4+ T-cell immune response in a murine
glioma model treated with antiCTLA-4. Zeng et al and
others have shown that antiPD-1 monotherapy improved
survival in a murine model for glioblastoma. Interestingly,
antiPD-L1 alone did not result in much survival improvement. With antiPD-1 use, the CD8+ T cells appear to
be responsible for the antitumor immune response.10,11
CURRENT TRIALS FOR GLIOBLASTOMA

Because we observed improved survival with the use of
checkpoint inhibitors in the previously mentioned solid
tumors, these compounds are now being applied to glioblastoma. Several clinical trials are currently underway, with
encouraging preclinical results.
NCT02017717
NCT02017717 is a large, randomized, phase III open-label
trial sponsored by Bristol Myers Squibb for patients with a
first-time recurrence of glioblastoma that used antiPD-1 as
the treatment backbone to assess its safety and, ultimately,
its efficacy. The study began with a small safety run-in,
during which patients were treated with antiPD-1 alone
or antiPD-1 with antiCTLA-4. Interim safety data that were
presented at the 2015 American Society of Clinical Oncology
Annual Meeting showed that the rates of severe adverse
events were significantly higher among patients who received the combination of antiPD-1 and antiCTLA-4; 40%
of patients had to discontinue therapy (10 patients per arm).
Thus, a decision was made to expand the antiPD-1 cohort
and to use bevacizumab as the comparator arm. Although
this was a study of safety, a partial response in one patient
who was treated with anti-PD-1 alone, and a few cases of
pseudoprogression, was observed. The trial has finished
accrual.
e134
NCT02337491
NCT02337491 is phase II trial based on the premise that antiVEGF therapy could be synergistic with immunotherapy.15
This trial is a single-institution study sponsored by Merck using
Mercks antiPD-1 drug. The study will measure progressionfree survival at 6 months. The trial has two arms: antiPD-1
alone and antiPD-1 with bevacizumab.
NCT02336166
NCT02336166 is a phase II trial sponsored by the Ludwig
Foundation to study the antiPD-1 therapy developed by
AstraZeneca. The trial has multiple objectives. The first
cohort (cohort A) will study the efficacy of antiPD-L1 for
patients with newly diagnosed glioblastoma that has
unmethylated O(6)-methylguanine DNA methyltransferase.
The remarkable factor in this arm is that temozolomide
(TMZ) will be withheld; patients will receive only antiPD-L1
and radiation. The rationale behind this design is that radiation and antiPD-L1 are synergistic, so systemic chemotherapy may be counterproductive. The second cohort
(cohort B) will assess the efficacy of antiPD-L1 alone for
patients who have recurrent glioblastoma. The third cohort
(cohort C) will administer antiPD-L1 to patients who have
recurrent glioblastoma that is progressing with bevacizumab
treatment.
NCT02617589
NCT02617589 is a phase III trial sponsored by Bristol Myers
Squibb to assess the efficacy of antiPD-1 with radiation
among patients who have newly diagnosed glioblastoma.
The comparator arm will enroll patients for treatment with
radiation and temozolomide (standard of care).
NCT02313272
NCT02313272 is a phase I trial to assess the safety of an
antiPD-1 drug developed by Merck with bevacizumab and
hypofractionated stereotactic irradiation for patients who
have recurrent high-grade gliomas.
NCT02530502
NCT02530502 is a phase I/II trial for patients with newly
diagnosed glioblastoma. Phase I will assess the safety of
combining an antiPD-1 drug developed by Merck with
radiation and temozolomide for patients with newly diagnosed glioblastoma. Phase II will compare the efficacy of
adding antiPD-1 to radiation and temozolomide for patients treated with radiation and temolozomide.
NCT02311582
NCT02311582 is a phase I, open-label, randomized safety
study to assess the addition of MK-3475, an antiPD-1
drug developed by Merck, for treatment among patients
who are being treated with laser ablation for recurrent
glioblastoma.
BIOMARKERS
What we have learned from checkpoint inhibitor trials in
other tumors is that the overall response rate is between
20% and 30%. Therefore, it is important to identify patients
who would not respond and to minimize toxicities and treat
them with other therapies. Current strategies for biomarkers
have focused on protein expression, immune cell characterization, and mutational burden.
The melanoma and lung cancer trials have focused on the
expression of PD-L1 on tumor cells as a biomarker to predict
response. Investigators have found that tumors that express
PD-L1 were more likely than PD-L1negative tumors to
respond (overall response rate) to antiPD-1 therapy.16
Interestingly, the expression of PD-L1 was not important
in the trial of patients with melanoma who received the
concurrent combination of antiPD-1 and antiCTLA-4.
Some theorized that an adaptive immune response occurred with the combination therapy and that combination
caused tumor cells to express PD-L1 as a defense mechanism.17 In an interesting twist, when the therapies were
sequencedpatients received antiCTLA-4 first, followed by
antiPD-1the overall response rate again correlated to the
PD-L1 status of the tumor.18
Another approach has been to assess the activation status
of various immune cells as a predictor of response. As an
example, investigators assessed the activation status of CD8+
T cells by measuring eomesodermin for patients with melanoma after they received treatment with antiCTLA-4 and
found that the eomesodermin status predicted relapse-free
survival. Other markers of interest are interferon gamma
expression, Helios expression, and various other checkpoint
molecules.19
Last, investigators have assessed mutational burden as a
predictor of response. Chan et al20 correlated mutational
burden among patients with lung cancer to response rates
among patients who received antiPD-1 therapy. They
found that patients with lung cancer who had a history of
smoking had higher mutational burdens than patients with
lung cancer who did not smoke and that the higher number
of mutations correlated to improved survival.20 Le et al21
also studied patients with colon cancer and found that
patients who had a defective DNA repair gene had a higher
number of mutations in their tumor, and this again correlated to improved survival. Some theorize that the reason for
the observed improved antitumor immune responses in
patients with tumors that had more mutations was that
the tumors expressed a higher number of target antigens for
the immune system.21,22
TOXICITY
Immune-related toxicities are an important issue for patients who receive immunotherapy. Most of the toxicities
are related to autoimmune reactions. The most common
toxicities include colitis, pneumonitis, hepatitis, pancreatitis,
dermatitis, hypophysitis, and thyroiditis. If the toxicities are
not recognized early, these reactions could become life
threatening. Treatments often require stopping the treatment, starting high-dose corticosteroids, and possibly administering infliximab.12,13,23
IMAGING
Determining response to immunotherapy with imaging has
become an area of intense interest. In the trial reported by
Hodi et al,12 a large number of patients with melanoma who
received antiCTLA-4 experienced pseudoprogression.
Furthermore, the researchers found that pseudoprogression
could take months to resolve.24,25 As a result, many trials
have built in lag times for imaging to allow patients to
continue therapy and to avoid considering the treatment a
failure. In glioblastoma, Okada et al26 have developed an
iRANO protocol specifically tailored for immunotherapy.
COMBINATION THERAPY
As previously mentioned, the response rates for patients who
receive immunotherapy has ranged from 20% to 30%. Studies
are investigating ways to combine checkpoint inhibitors with
other modalities, such as radiation, bevacizumab, and devices. Zeng et al10 demonstrated in a preclinical glioblastoma
model that the combination of focused radiation with
antiPD-1 is synergistic. Clinical trials (e.g., NCT02313272) are
looking at stereotactic radiation use in combination with
checkpoint inhibitors. Bevacizumab also may work synergistically with immunotherapy.20 Preclinical data suggest
that this approach could be effective, and a phase II trial
(NCT02337491) is looking at the combination of bevacizumab
and antiPD-1 for patients with glioblastoma. Finally, another
interesting approach is the combination of laser ablation with
antiPD-1. A phase I trial (NCT02313272) is looking at the
combination of antiPD-1 with laser ablation.
In conclusion, checkpoint inhibitors have shown great
promise in other solid tumors. There is much excitement
about checkpoint inhibitors in the setting of glioblastoma.
Several large trials are underway to assess the efficacy of
checkpoint inhibitors in glioblastoma.
VIRAL- AND GENE-MEDIATED

IMMUNOTHERAPIES FOR GLIOBLASTOMA
We have discussed the various types of immunotherapy for
glioblastoma. Now, we discuss one additional mode that
involves the use of genetically engineered viruses to deliver
cytotoxic/immunostimulatory genes into tumors.27 Two
main types of viruses are used in this technology: replicationdefective vectors, in which viral genes have been removed so
there is no expression of viral genes or generation of progeny
viruses, but there is expression of an immunostimulatory
and/or cytotoxic gene, and tumor replicationselective viruses (oncolytic viruses), in which a viral pathogen is engineered so that its pathogenicity is now targeted to tumor
cells and not normal cells.28 It is recognized now that the
presence of viral genes and viral proteins in both of these
technologies can elicit powerful anticancer immune responses, which are a major component of efficacy.29
e135
FIGURE 1. Immunosuppressive Properties of Glioma Stem Cells
Abbreviation: TGF-b, transforming growth factor-beta.

Modified from Hatiboglu et al.42
Characteristics that differentiate this immunotherapy

from others are the following: this immunotherapy involves a neurosurgical component of direct injection into
glioblastoma, either by stereotaxy or free-hand injection
during a craniotomy; is off-the-shelf, in that manipulation of
cells from the patient before or after treatment is not involved; and does not require a priori knowledge of the
identity of tumor-specific antigens that require targeting
and, at least in theory, the cytotoxic action of the viral
vector-delivered cytotoxic gene or of the replicating oncolytic viruses will expose the repertoire of all tumor-specific
antigens to the immune system.
There are more than 2 decades of clinical trial experiences
that used various types of replication-defective vectors to
deliver various types of cytotoxic/immunostimulatory genes,
but none have resulted in an FDA-approved clinical product,
and two phase III clinical trialsa retroviral vector to deliver the
cytotoxic/immunostimulatory herpes thymidine kinase (TK)
gene for recurrent glioblastoma30 and an adenoviral vector to
deliver TK in newly diagnosed glioblastoma (ASPECT)have
failed.31 However, we recently reported the results of a phase II
trial in which aglatimagene besadenovec (AdV-tk), a nonreplicating adenovirus expressing the herpesvirus TK gene,
was free-hand injected in a resected, newly diagnosed, malignant glioma cavity.32,33 The patient then received the oral
antiherpetic prodrug (valacyclovir) while undergoing standardof-care radiochemotherapy (gene-mediated cytotoxic immunotherapy [GMCI]). There are multiple modes of cytotoxic and
immunostimulatory anticancer action (Fig. 3):
e136
1. The delivered TK gene product phosphorylates the administered valacyclovir drug, which becomes incorporated
at sites where DNA is becoming repaired or where DNA is
replicating. This leads to termination of DNA repair and/or
replication, leading to immunogenic cell death.
2. Standard of care also leads to cytotoxicity, and
radiation-induced DNA damage leads to additional
unsuccessful and cytotoxic attempts at DNA repair
using the phosphorylated valacyclovir pools.
3. The delivered TK antigen and viral vector proteins act
as superantigens, providing an immunostimulatory
stimulus in the glioblastoma microenvironment.
4. The cytotoxic death of glioblastoma cells is also immunogenic, releasing and exposing multiple glioma
antigens to the immune system.
Our published mature phase II data appear to show encouraging, albeit not definite, results in terms of possible
efficacy on the basis of the extent of residual tumor burden:
the median overall survival (OS) durations for patients who
underwent gross total resection were 25.0 and 16.9 months
(a difference of 8.1 months) for GMCI/standard of care
and standard of care, respectively (hazard ratio 0.59; 95%
CI, 0.350.998; p = .0492); for patients who underwent
subtotal resections, the difference was only 1 month (13.5
vs. 12.5 months for GMCI/standard of care vs. standard of
care; p = .4584).33 To further improve this therapy, we have
returned to the laboratory. The current theory is that to
be effective, an anticancer immune response by cytotoxic
FIGURE 2. Examples of Activating (CD28) and Inhibiting (PD-1) Immune Checkpoints
T cells requires removal of immune checkpoint signaling

mediated by PD-1/PD-L1,34-36 the CTLA-4/B7 family,37,38 and
other molecules. We thus hypothesize that a combination of
GMCI and checkpoint inhibitors may lead to more effective
immunotherapy.8 In fact, we are finding that the application
of GMCI in mouse gliomas does lead to increase signaling of
immune checkpoint networks and that inhibition of these
networks does lead to even more encouraging antigliomal

effects of GMCI.
For the second type of viral-mediated therapy, multiple
types of oncolytic viruses have been tested in clinical trials of
glioblastoma, all in the recurrent setting. No trial for glioblastoma has progressed to phase III, but there are two
oncolytic viruses (an oncolytic adenovirus from DNAtrix
FIGURE 3. Schematics of the Published Clinical Trials and Different Modes of Anticancer Action of Gene-Mediated
Cytotoxic Immunotherapy
(Top panel) Schematic of the published clinical trials of gene-mediated cytotoxic immunotherapy (GMCI). On the day of surgery, the adenoviral vector that delivers TK (AdVtk) is injected in the resected newly diagnosed glioblastoma tumor cavity. The oral agent (valacyclovir) is administered to the patient on days 1 to 14. Standard-of-care
radiation and temozolomide are also administered as per the Stupp regimen. (Lower panel) Schematic of the different modes of anticancer action of GMCI.
e137
and an oncolytic retrovirus from Tocagen) that have

completed phase II evaluation and planning for advanced
phase III trials. Commercial interest in this area of
oncolytic virusbased immunotherapy has been resurrected recently by the FDA approval of a herpes simplex
virus (HSV) oncolytic virus for melanoma.29,39 Our group
has been involved in the preclinical and clinical development of an oncolytic virus, based on HSV1, that
has been engineered to selectively replicate and destroy gliomas on the basis of tumor deregulation of the
p16 tumor suppressor pathway and expression of the
glioma stem-cell marker, nestin.40,41 This oncolytic virus
(rQNestin34.5v.2) shows potency in animal models of
gliomas compared with older, clinical trialtested versions
of herpes oncolytic viruses. Extensive preclinical data
have been obtained to justify filing of an investigational

new drug application with the FDA for a planned firstin-human clinical trial among patients with recurrent
glioblastoma.
In summary, the use of tumor-selective viruses and viralbased vectors is increasingly delivering promising results as
an easy to use immunostimulant approach, but only the
successful completion of phase III trials for several of these
products will show if the treatments have achieved their
promise as anticancer agents or if additional laboratory
development is required.
ACKNOWLEDGMENT
We thank Eileen Kim for her help with Fig. 2.
References
1. Roszman T, Elliott L, Brooks W. Modulation of T-cell function by gliomas.
Immunol Today. 1991;12:370-374.
2. Frei K, Gramatzki D, Tritschler I, et al. Transforming growth factor-b
pathway activity in glioblastoma. Oncotarget. 2015;6:5963-5977.
3. Nduom EK, Weller M, Heimberger AB. Immunosuppressive mechanisms
in glioblastoma. Neuro Oncol. 2015;17:vii9-vii14.
4. Weiss T, Weller M, Roth P. Immunotherapy for glioblastoma: concepts
and challenges. Curr Opin Neurol. 2015;28:639-646.
5. Preusser M, Lim M, Hafler DA, et al. Prospects of immune checkpoint modulators in the treatment of glioblastoma. Nat Rev Neurol. 2015;11:504-514.
6. Preusser M, Berghoff AS, Wick W, et al. Clinical neuropathology minireview 6-2015: PD-L1emerging biomarker in glioblastoma? Clin
Neuropathol. 2015;34:313-321.
7. Sharma P, Allison JP. Immune checkpoint targeting in cancer therapy: toward
combination strategies with curative potential. Cell. 2015;161:205-214.
8. Pardoll DM. The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer. 2012;12:252-264.
9. Leach DR, Krummel MF, Allison JP. Enhancement of antitumor immunity by CTLA-4 blockade. Science. 1996;271:1734-1736.
10. Zeng J, See AP, Phallen J, et al. Anti-PD-1 blockade and stereotactic
radiation produce long-term survival in mice with intracranial gliomas.
Int J Radiat Oncol Biol Phys. 2013;86:343-349.
11. Reardon DA, Freeman G, Wu C, et al. Immunotherapy advances for
glioblastoma. Neuro Oncol. 2014;16:1441-1458.
12. Hodi FS, ODay SJ, McDermott DF, et al. Improved survival with ipilimumab in patients with metastatic melanoma. N Engl J Med. 2010;363:
711-723.
13. Topalian SL, Hodi FS, Brahmer JR, et al. Safety, activity, and immune
correlates of anti-PD-1 antibody in cancer. N Engl J Med. 2012;366:
2443-2454.
14. Fecci PE, Ochiai H, Mitchell DA, et al. Systemic CTLA-4 blockade
ameliorates glioma-induced changes to the CD4+ T cell compartment
without affecting regulatory T-cell function. Clin Cancer Res. 2007;13:
2158-2167.
15. Hodi FS, Lawrence D, Lezcano C, et al. Bevacizumab plus ipilimumab in
patients with metastatic melanoma. Cancer Immunol Res. 2014;2:
632-642.
16. Taube JM, Klein A, Brahmer JR, et al. Association of PD-1, PD-1 ligands,
and other features of the tumor immune microenvironment with response to anti-PD-1 therapy. Clin Cancer Res. 2014;20:5064-5074.
e138
17. Grosso J, Horak CE, Inzunza D, et al. Association of tumor PD-L1 expression and immune biomarkers with clinical activity in patients (pts)
with advanced solid tumors treated with nivolumab (anti-PD-1;BMS936558;ONO-4538). J Clin Oncol. 2013;31(suppl; abstr 3016).
18. Callahan MK, Postow MA, Wolchok JD. CTLA-4 and PD-1 pathway
blockade: combinations in the clinic. Front Oncol. 2014;4:385.
19. Wang W, Yu D, Sarnaik AA, et al. Biomarkers on melanoma patient
T cells associated with ipilimumab treatment. J Transl Med. 2012;10:
146.
20. Rizvi NA, Hellmann MD, Snyder A, et al. Cancer immunology: mutational
landscape determines sensitivity to PD-1 blockade in nonsmall-cell
lung cancer. Science. 2015;348:124-128.
21. Le DT, Uram JN, Wang H, et al. PD-1 blockade in tumors with mismatchrepair deficiency. N Engl J Med. 2015;372:2509-2520.
22. Diaz LA Jr, Le DT. PD-1 blockade in tumors with mismatch-repair deficiency. N Engl J Med. 2015;373:1979.
23. Weber JS, Kahler KC, Hauschild A. Management of immune-related
adverse events and kinetics of response with ipilimumab. J Clin Oncol.
2012;30:2691-2697.
24. Hodi FS, Ribas A, Daud A, et al. Patterns of response in patients with
advanced melanoma treated with pembrolizumab (MK-3475) and
evaluation of immune-related response criteria (irRC). J Immunother
Cancer. 2014;2:P103.
25. Topalian SL, Sznol M, McDermott DF, et al. Survival, durable tumor
remission, and long-term safety in patients with advanced melanoma
receiving nivolumab. J Clin Oncol. 2014;32:1020-1030.
26. Okada H, Weller M, Huang R, et al. Immunotherapy response assessment in neuro-oncology: a report of the RANO working group. Lancet
Oncol. 2015;16:e534-e542.
27. Kaufmann JK, Chiocca EA. Glioma virus therapies between bench and
bedside. Neuro Oncol. 2014;16:334-351.
28. Chiocca EA, Rabkin SD. Oncolytic viruses and their application to cancer
immunotherapy. Cancer Immunol Res. 2014;2:295-300.
29. Lawler SE, Chiocca EA. Oncolytic virus-mediated immunotherapy:
a combinatorial approach for cancer treatment. J Clin Oncol. 2015;33:
2812-2814.
30. Rainov NG. A phase III clinical evaluation of herpes simplex virus type 1
thymidine kinase and ganciclovir gene therapy as an adjuvant to surgical
resection and radiation in adults with previously untreated glioblastoma multiforme. Hum Gene Ther. 2000;11:2389-2401.
31. Westphal M, Yla-Herttuala S, Martin J, et al; ASPECT Study Group.

Adenovirus-mediated gene therapy with sitimagene ceradenovec followed by intravenous ganciclovir for patients with operable high-grade
glioma (ASPECT): a randomised, open-label, phase 3 trial. Lancet Oncol.
2013;14:823-833.
32. Chiocca EA, Aguilar LK, Bell SD, et al. Phase IB study of gene-mediated
cytotoxic immunotherapy adjuvant to up-front surgery and intensive
timing radiation for malignant glioma. J Clin Oncol. 2011;29:3611-3619.
33. Wheeler LA, Manzanera AG, Bell SD, et al. Phase 2 multicenter study of
gene-mediated cytotoxic immunotherapy as adjuvant to surgical resection
for newly diagnosed malignant glioma. Neuro Oncol. Epub 2016 Feb 2.
34. Barber DL, Wherry EJ, Masopust D, et al. Restoring function in exhausted
CD8 T cells during chronic viral infection. Nature. 2006;439:682-687.
35. Freeman GJ, Long AJ, Iwai Y, et al. Engagement of the PD-1 immunoinhibitory receptor by a novel B7 family member leads to negative
regulation of lymphocyte activation. J Exp Med. 2000;192:1027-1034.
36. Mahoney KM, Rennert PD, Freeman GJ. Combination cancer immunotherapy and new immunomodulatory targets. Nat Rev Drug Discov.
2015;14:561-584.
37. Lane P. Regulation of T and B cell responses by modulating interactions

between CD28/CTLA4 and their ligands, CD80 and CD86. Ann N Y Acad
Sci. 1997;815:392-400.
38. Korman A, Yellin M, Keler T. Tumor immunotherapy: preclinical and
clinical activity of anti-CTLA4 antibodies. Curr Opin Investig Drugs. 2005;
6:582-591.
39. Andtbacka RH, Kaufman HL, Collichio F, et al. Talimogene laherparepvec
improves durable response rate in patients with advanced melanoma.
J Clin Oncol. 2015;33:2780-2788.
40. Aghi M, Visted T, Depinho RA, et al. Oncolytic herpes virus
with defective ICP6 specifically replicates in quiescent cells with
homozygous genetic mutations in p16. Oncogene. 2008;27:
4249-4254.
41. Kambara H, Okano H, Chiocca EA, et al. An oncolytic HSV-1 mutant
expressing ICP34.5 under control of a nestin promoter increases survival of animals even when symptomatic from a brain tumor. Cancer
Res. 2005;65:2832-2839.
42. Hatiboglu MA, Wei J, Wu AS, et al. Immune therapeutic targeting of
glioma cancer stem cells. Target Oncol. 2010;5:217-227.
e139
Treatment of Low-Grade Gliomas

in the Era of Genomic Medicine
CHAIR
Minesh P. Mehta, MD, MBChB, FASTRO
University of Maryland Medical Center
Baltimore, MD
SPEAKERS
Kenneth D. Aldape, MD
Houston, TX
Susan M. Chang, MD
San Francisco, CA
LOWER-GRADE GLIOMAS, MOLECULAR MARKERS, AND THERAPEUTIC CHOICES
Treatment of Adult Lower-Grade Glioma in the Era of Genomic

Medicine
Susan M. Chang, MD, Daniel P. Cahill, MD, PhD, Kenneth D. Aldape, MD, and Minesh P. Mehta, MBChB, FASTRO
OVERVIEW
By convention, gliomas are histopathologically classified into four grades by the World Health Organization (WHO) legacy
criteria, in which increasing grade is associated with worse prognosis and grades also are subtyped by presumed cell of origin.
This classification has prognostic value but is limited by wide variability of outcome within each grade, so the classification is
rapidly undergoing dramatic re-evaluation in the context of a superior understanding of the biologic heterogeneity and
molecular make-up of these tumors, such that we now recognize that some low-grade gliomas behave almost like malignant
glioblastoma, whereas other anaplastic gliomas have outcomes comparable to favorable low-grade gliomas. This clinical
spectrum is partly accounted for by the dispersion of several molecular genetic alterations inherent to clinical tumor
behavior. These molecular biomarkers have become important not only as prognostic factors but also, more critically, as
predictive markers to drive therapeutic decision making. Some of these, in the near future, will likely also serve as potential
therapeutic targets. In this article, we summarize the key molecular features of clinical significance for WHO grades II and III
gliomas and underscore how the therapeutic landscape is changing.
everal challenges limit the efficacy of surgery, radiotherapy (RT), chemotherapy, and targeted agents for
WHO grades II and III glioma (categorized as lower-grade
glioma, or LrGG). These include the diffuse and imprecisely
defined boundaries of extension, proximity to, and growth
into critical cortical and subcortical structures, inability of
therapeutic agents to adequately cross the blood-brain
barrier, active transport mechanisms of drug efflux, high
plasma protein binding of most conventional agents, and
multiple and redundant drug and radiation resistance
mechanisms. Biologic tumor heterogeneity, with respect to
response to treatment and resistance mechanisms, despite
similar grade or histology, is another important challenge.
The era of genomic medicine holds the promise of generating personalized medicine treatment paradigms to improve outcomes. The ability of isocitrate dehydrogenase
(IDH) mutations, loss of heterozygosity of 1p and 19q, and
methylation of methylguanine methyl transferase (MGMT)
status to serve as predictive biomarkers is of specific interest
in the treatment of LrGG.
MOLECULAR CHARACTERIZATION OF LOWERGRADE GLIOMA

Clinicopathologic assessment, valuable to distinguish
grades of glioma, does not address the distinct genetic
and biologically different subgroups that are nested within

each grade category. 1 In addition, the interevaluator
variability that has been demonstrated, especially for
mixed histologies, is a major limitation in provision of
uniform and reproducible pathologic diagnoses.2 Recent
molecular characterization has identified potential biomarkers that can serve to define the framework of different
subtypes of these tumors with more uniform prognosis
and also potentially can serve as indicators for treatment
selection.3,4
Somatic mutations in the isocitrate dehydrogenase genes
IDH1 and IDH2 are of important prognostic value. These
mutations are present in 50% to 80% of WHO grades II and III
astrocytic and oligodendroglial tumors in adults and also
occur with similar frequency in secondary glioblastoma.
These mutant enzymes lead to conversion of a-ketoglutarate
into D-2-hydroxyglutarate, an oncometabolite that drives
the oncogenic activity of IDH mutations.5 Irrespective of
treatment choice, patients with IDH-mutated tumors experience superior survival compared with patients who
have IDH wild-type gliomas of the same histologic grade; this
difference clearly underscores the prognostic significance of
this marker. In tumors of oligodendroglial lineage, the most
common cytogenetic alteration consists of an unbalanced
t1,19(q10;p10) translocation that results in the fusion of the
From the University of California, San Francisco, San Francisco, CA; Harvard Medical School, Boston, MA; Toronto General Hospital/Research Institute, Toronto, Canada; University of
Maryland, Baltimore, MD.
Corresponding author: Minesh P. Mehta, MD, FASTRO, Department of Radiation Oncology, University of Maryland, 22 South Greene St., Baltimore, MD 21201; email: mmehta@umm.edu.
75
CHANG ET AL
chromosomal arms of 1p and 19q, accompanied by the loss

of one hybrid chromosome, which thus results in the loss of
heterozygosity of 1p and 19q. These tumors have better
prognoses than histologically identical tumors of the same
grade that do not harbor this codeletion. Furthermore, this
biomarker has been proven to be a critical predictive biomarker of chemotherapy response, and improved survival
after chemoradiotherapy, compared with RT alone.
Recent reports also have reinforced the redefinition of subgroups of LrGG by incorporating additional
biomarkersfor example, activating mutations of the telomerase reverse transcription (TERT) promotor and mutations and losses of alpha thalassemia/mental retardation
syndrome X linked (ATRX) and tumor protein P53 (TP53). By
using these genetic alterations, the vast majority of adult
diffuse gliomas fall into one of three distinct categories that
more reproducibly predict clinical outcome and guide
treatment by creating more homogeneous subsets.3,4 These
consist of (1) molecular glioblastomas, which are highly
aggressive tumors characterized by the TERT promoter
mutation and concomitant deletions of chromosome 10 and
gain of chromosome 7 with EGFR amplification, and they are
notable for the absence of IDH1/2 mutations (i.e., they
are IDH wild type); (2) molecular astrocytomas, which are
characterized by mutations in IDH1/2, TP53, and ATRX, and
the absence of promoter mutations or codeletion of 1p19q;
and (3) molecular oligodendrogliomas, which are characterized by IDH mutation and TERT promoter mutation,
KEY POINTS
76
WHO grades II and III gliomas, conventionally

substratified as astrocytomas or oligodendrogliomas (or
mixed), are highly heterogeneous tumors with complex,
overlapping, and varied biological behavior.
In the past decade, significant new therapeutic strides
have been made to improve overall survival in almost all
of these entities, in large measure, as a consequence of
robust, international clinical trial collaborations with
parallel tissue collection for molecular analysis.
This tissue collection effort, with well-annotated
matching clinical outcomes, has secondarily identified
several key molecular markers that have crucial
prognostic significance, such as IDH1/2 mutations,
1p19q co-deletions, MGMT methylation, and more.
These markers are also increasingly identified as crucial
predictive markers, which thereby influence therapeutic
decision making.
Some of these molecular markers, such as the IDH1/2
mutation, are now also emerging as future therapeutic
targets. Multivariate analysis of clinical and molecular
markers is identifying subgroups that likely need to be
treated with greater therapeutic intensity and others for
which therapeutic de-intensification could potentially
be contemplated, posing novel challenges for clinical
trial design and conduct.
1p19q codeletion, and mutations in CIC and FUBP1, with an

absence of mutations in TP53 or ATRX.
WHO GRADE III ANAPLASTIC

OLIGODENDROGLIOMA: THE CASE FOR A
PREDICTIVE BIOMARKER FOR CHEMOTHERAPY
On the basis of several retrospective series and phase II
trials that demonstrated the chemosensitivity of oligodendrogliomas, two prospective randomized phase III trials
evaluated the role of postoperative chemoradiotherapy
compared with RT alone.6,7 In both studies, the experimental
arm was RT combined with procarbazine, lomustine, and
vincristine (PCV). In the Radiation Therapy Oncology Group
(RTOG) trial RTOG-9402, patients were randomly assigned
either to four cycles of intensified PCV followed by RT or to
immediate RT without chemotherapy. Mature data from this
study showed that the median survival of those with
codeleted tumors who were treated with PCV plus RT was
twice that of patients who received only RT (14.7 vs. 7.3
years). The survival of patients with codeleted tumors was
better than that of those with non-codeleted tumors regardless of treatment, which emphasizes the prognostic
value of the 1p19q codeletion. The survival was not statistically different for patients with tumors that lacked the 1p
and 19q deletion, irrespective of treatment (median survival,
2.6 vs. 2.7 years), which suggests that the codeletion also has
predictive value. In another analysis of patients enrolled in
this study, IDH mutation status had predictive value as well;
although patients with codeleted tumors lived longest,
patients with non-codeleted IDH-mutated tumors also lived
longer after chemoradiotherapy compared to RT alone.8 The
implication here is that, for WHO grade III tumors, conventionally labeled as oligodendroglioma, treatment with
chemoradiotherapy imparts a substantial survival advantage
when the tumors are either codeleted for 1p19q or are noncodeleted tumors only with expression of an IDH mutation.
Non-codeleted tumors without IDH mutations have the
worst prognosis and do not benefit from the addition of
chemotherapy.
In the European Organization for Research and Treatment of Cancer (EORTC) 26951 trial, 368 patients who had
predominantly grade III oligodendroglial tumors received
immediate RT only or RT followed by six cycles of PCV.
Patients with tumors that had the 1p and 19q codeletion
had better outcomes regardless of therapy (prognostic).
In addition, MGMT promoter methylation was of prognostic value in this cohort. Similar to the results of RTOG9402, long-term follow-up showed that the addition of
PCV to RT significantly increased progression-free survival
(PFS; median, 157 vs. 50 months) in patients with the
1p19q codeletion, and there was a trend toward increased
overall survival (OS; OS not reached in the RT/PCV group
vs. OS of 112 months in the RT-alone group; hazard
ratio [HR] 0.56; 95% CI, 0.311.03). These two studies
confirmed the value of the 1p19q codeletion as an important prognostic and predictive marker and influenced
the design of studies in anaplastic glioma; in ongoing
international studies, patient arms are being stratified on

the basis of this biomarker. An international intergroup
phase III trial is being conducted in patients who have
newly diagnosed grade 3 glioma with the 1p19q status
codeletion.7 Patients are randomly assigned to two arms,
RT with concomitant and adjuvant temozolomide or RT
with adjuvant PCV.
WHO GRADE III ANAPLASTIC ASTROCYTOMA

The role of chemotherapy in anaplastic astrocytoma is not
well established. Most phase III trials have demonstrated
no benefit of chemotherapy compared with radiation
alone in this tumor. Carmustine and PCV are associated
with minimal improvement in survival. The Glioma MetaAnalysis Trialists group showed a 6% increase in 1- and
2-year survival rates for patients who received chemotherapy (2-year survival, 37% vs. 31%), whereas a large
randomized trial of adjuvant PCV compared with RT
alone did not show any benefit from PCV.9,10 RTOG-9813
was a phase III study to compare radiation with bischloroethylnitrosourea (BCNU) or lomustine to radiation with
temozolomide, and the results of this study demonstrated
no difference in survival between the two chemotherapy
arms. Molecular characterization in a subset of patients
showed that IDH mutation was a strong prognostic factor.
The number of patients was too small to determine any
differential benefit of the type of chemotherapy on the
basis of IDH mutation status.11
The NOA-04 phase III trial compared the efficacy of RT
followed by chemotherapy at progression to initial chemotherapy followed by RT at progression in newly diagnosed
anaplastic gliomas and did not show any difference in the
median time to failure, PFS, or OS among the arms. The study
demonstrated the prognostic value of IDH1 mutations in
anaplastic gliomas, with a favorable impact that was more
distinct than that of 1p19q codeletion or MGMT promoter
methylation.12
A large international trial in which patients with newly
diagnosed non1p19q-codeleted grade III glioma were
randomly assigned to radiation with or without concomitant and with or without adjuvant temozolomide after
RT recently completed accrual, and initial interim analysis
showed a significant benefit with adjuvant temozolomide
(p = .0084). Results for concurrent temozolomide and for the
overall treatment regimen are not yet available. Tissue
acquisition was mandatory for this study and will allow
evaluation of the role of concurrent or adjuvant temozolomide on the basis of molecular subgroups.
WHO GRADE II GLIOMA

Data and conclusions pertaining to WHO grade II gliomas
have largely been derived from patient subsets entered onto
therapeutic trials. As a consequence, these recommendations do not apply to the patients with lowest-risk disease, that is, younger patients with completed resected
tumors, especially oligodendroglioma.
Because of the relatively low proliferative index of grade II

glioma, these tumors were previously considered chemotherapy resistant. Reports of objective responses in patients
with LrGG, however, raised interest in the use of chemotherapy in the adjuvant setting. In a small Southwest Oncology Group trial, there was no survival difference among
patients with incompletely excised grade II gliomas randomly assigned to RT alone or to the combination of RT and
lomustine.13 The RTOG-9802 study examined the role of
adjuvant PCV chemotherapy for high-risk adults (i.e., those
of any age with tumors that had less than total resection, or
those older than age 40 with any resection) who had grade II
glioma. Two hundred fifty-one patients were randomly
assigned to RT alone or RT followed by six cycles of PCV. At a
median follow-up of 12 years, a significant improvement in
OS in the RT/PCV group (13.3 years) compared with the RTalone group (7.8 years) was reported.14
The significance of 1p19q, MGMT, and IDH was unknown at
the time of study initiation, and several molecular analyses
were post hoc, performed on non-mandatorily collected
tumor blocks, slides, or cores submitted for 117 patients; of
these, 113 (57 of 126 [45%] and 56 of 125 [45%] in the RT
alone and RT/PCV arms, respectively) were suitable for IDHR132H immunohistochemistry interpretation. IDH1-R132H
mutations were detected in 35 (61.4%) of 57 patients and 36
(64.3%) of 56 patients in the RT alone and RT/PCV arms,
respectively, and in 77.6%, 53.8%, and 48.0% of patients with
oligodendroglioma, oligoastrocytoma, and astrocytoma, respectively. Tissue sufficient for determination of other IDH
mutations or of the 1p19q codeletion was available for
analysis in 29 and 34 patients in the RT and in the RT/PCV
arms, respectively. In a subset analysis of PFS by IDH1-R132H
mutation status, patients with mutations experienced significantly longer PFS than those without (p , .001), regardless
of treatment, which established the mutation as prognostically significant. In addition, patients with tumoral IDH1R132H mutations had longer PFS if they received RT/PCV than
if they received RT alone (p = .003), which established the
mutation status as a predictive variable. The number of events
in patients without the IDH1-R132H mutation was too small to
determine the association of treatment effect in this subset.
In an exploratory subset analysis of OS by IDH1-R132H mutation status, patients with mutations experienced significantly longer OS than those without the mutation, regardless
of treatment (log rank p = .02). Median survival times with and
without the mutation were 13.1 and 5.1 years, respectively.
Patients with tumoral IDH1-R132H mutations also experienced longer OS if they received RT/PCV than if they received
RT alone (log rank p = .02). The number of events in patients
without the IDH1-R132H mutation was too small to determine the association of treatment effect in this subset.15
RTOG-0424 was a study of concurrent and adjuvant
temozolomide and RT in a high-risk low-grade glioma
population, and the 3-year OS rate was 73.1% (95% CI,
65.3%80.8%), which was significantly improved compared
with the prespecified historical control reported by Pignatti
et al (p , .0001).16 Molecular analysis on this trial is pending.
77
CHANG ET AL
An ongoing intergroup phase III trial is attempting to address

the role of adjunctive temozolomide in LrGG.
ROLE OF RADIOTHERAPY
Historically, RT trials have established that earlier intervention
with RT prolongs PFS and reduces symptom burden, especially
related to seizures, but does not improve survival compared
with deferred RT.17 In addition, a categorical dose effect on
survival is not observed with RT in the 45- to 64-Gy range.18
Grade III tumors were historically labeled malignant gliomas
and were conglomerated in trials that included grade IV
glioblastomas, which made it difficult to discern the precise
survival advantage from RT. More contemporary trials in grade
II astrocytoma and oligodendroglioma, as well as in grade III
oligodendroglioma, have confimed that chemoradiotherapy
yields a survival advantage compared with RT alone. A common theme to emerge from these trials is that, for the favorable tumors, especially those with IDH mutations, a
dramatic survival advantage emerges with combination chemoradiotherapy; in the case of grade III tumors, even those in
the 1p19q non-codeleted subset, a survival advantage with
chemoradiotherapy is observed if the tumor has IDH mutant
status. An important caveat here is that these patients have
prolonged survival, so concerns have been expressed about
possible cognitive decline after RT; in practice, some clinicians
simply treat with chemotherapy alone and reserve RT for
salvage therapy. Level-1 data to support this practice do not
exist, and, until it is categorically established that RT dose
reduction or dose elimination is safe, these patients should
continue to be treated with combination chemoradiotherapy.
The IDH wild-type tumors do not exhibit this survival advantage
with chemoradiotherapy, so it would be reasonable to consider
treating patients who have these tumors with RT alone.
The EORTC has completed a large, international, randomized trial in 477 patients with WHO grade II tumors
(22033-26033), stratified by 1p deletion, and randomized to
RT alone or temozolomide alone. A difference between the
two treatment strategies was not observed for PFS, although
RT was numerically favored. However, molecular tumor
characterization may allow the treatment approach to be
personalized and one or the other treatment modality to be
selected. A post hoc analysis of molecular markers was done
in 407 patients and was reported recently. IDH1 or IDH2
mutations were detected in 85.8% of patients. Codeletions
of 1p19q were identified in 33%. MGMT was methylated in
90%, of which the majority, 86%, was IDH mutant. Mutation
of IDH1 or IDH2, regardless of 1p19q codeletion, was a
positive prognostic factor. Patients with IDH-mutated, noncodeleted tumors had shorter PFS after treatment with
temozolomide than after RT (HR 1.86; 95% CI, 1.212.87; log
rank p = .0043), whereas no difference was observed between these treatments for patients with IDH wild-type
tumors and IDH-mutated, codeleted tumors.19
One additional study, the phase III German NOA-04 trial,
deserves mention, relative to the role of RT in LrGG. The
NOA-04 trial compared the efficacy of RT followed by
chemotherapy at progression with the reverse sequence in
78
patients with newly diagnosed anaplastic gliomas. Three

hundred eighteen patients, most of whom had grade 3
anaplastic astrocytoma (with very few oligodendroglioma),
were randomly assigned 2:1:1 (arms A:B1:B2) to receive
conventional RT (arm A), PCV (arm B1), or temozolomide
(arm B2) at diagnosis. At progression or unacceptable toxicity, patients in arm A were treated with PCV or temozolomide (1:1 random assignment), whereas patients in arms
B1 or B2 received RT. The median PFS (HR 1.0; 95% CI,
0.71.3) and OS (HR 1.2; 95% CI, 0.81.9) were similar for all
three arms. Hypermethylation of the MGMT promoter (HR
0.59; 95% CI, 0.361.0) and mutations of IDH1 gene (HR 0.48;
95% CI, 0.290.77) reduced the risk of progression, and IDH
mutations had the largest impact. MGMT promoter methylation also was associated with prolonged PFS in the
chemotherapy and RT arm. Therefore, initial RT or chemotherapy achieved comparable results in patients with
anaplastic gliomas.12 This trial has sometimes been used to
support the logic of initiating chemotherapy first, followed
by RT at progression, for grade II gliomas (both astrocytic and
oligodendroglial) and grade III oligodendroglioma, but it
must be borne in mind that the results may not necessarily
extrapolate to these entities.
SURGERY IN LOWER-GRADE GLIOMA

The role of surgical resection in adult diffuse gliomas is
twofold: diagnostic and therapeutic. Historically, the diagnostic grading of a glioma has been dependent upon the
type of procedure performed and the volume of tissue
specimen provided to the pathologist for review. Thus, a
less-extensive biopsy procedure can undersample a glioma,
which can result in diagnostic histologic undergrading. For
example, Glantz et al20 demonstrated that, in 262 patients
undergoing resection, 214 (82%) were ultimately rendered a
final diagnosis of glioblastoma (WHO grade IV) and 48 (18%)
were diagnosed with anaplastic astrocytoma (WHO grade
III). In contrast, of 67 patients undergoing stereotactic biopsy, 33 (49%) were diagnosed with glioblastoma, whereas
34 (51%) were diagnosed with anaplastic astrocytoma.
These observations led to the conclusion that it was likely
that some of the anaplastic astrocytomas diagnosed by
stereotactic biopsy were undersampled and actually represented glioblastomas.20
A similar theme was noted by Jackson et al,21 in a study
based on biopsy specimens and the subsequent resection
specimens from the same patient (paired samples), without
intervening delay or therapeutic intervention. In this study,
the authors noted that the diagnoses rendered from the
biopsy specimen differed from the diagnoses achieved by
using the more extensive surgical resection specimen in 50
(49%) of 82 instances. They concluded that stereotactic
biopsy is frequently inaccurate in provision of a correct
diagnosis and thus recommended surgical resection, even
purely for diagnostic purposes.21
In a more recent analysis, Kim et al22 noted that molecular
testing can be used to resolve these diagnostic discrepancies
that develop because of limited pathologic sampling. Similar
to results by Glantz et al,20 these investigators noted a

volume-based discrepancy between diagnoses of grade IV
and grade III tumors, whereby grade IV diagnoses were more
common in larger resection specimens. Notably, however,
the frequency of IDH1 mutation was constant between the
two cohorts, which demonstrated that molecular status was
volume independent. Importantly, the diagnosis of anaplastic astrocytoma from a less-extensive resection was
associated with significantly poorer survival (glioblastomalike). Thus, the investigators concluded that, for smaller
surgical specimens, underdiagnosis of glioblastoma may
occur when the analysis is restricted to histopathology
alone.22 Moving forward, the WHO will now integrate
molecular scoring into the diagnostic categorization of these
lesions.23
With regard to the therapeutic benefit of resection, recent
studies have focused on the benefit of removal of the
nonenhancing tumor that makes up the majority (or entirety) of an LrGG. Most studies of surgery for glioma are
retrospective and nonrandomized and, therefore, are potentially biased because of patient selection. In a molecular
study, Beiko et al24 found that IDH1 mutation was independently associated with complete resection of enhancing disease in grade III and IV astrocytomas and that
additional resection of nonenhancing disease was associated with an additional survival benefit only in the IDH1mutant cohort.24 For LrGG, Shaw et al25 observed 111
patients who were assigned to observation after surgeondetermined gross-total resection in the RTOG-9802 study of
low-risk, low-grade (WHO grade II) glioma.25 The most
favorable-risk cohort shared three factors: tumor diameter
less than 4 cm on preoperative MRI scan, oligodendroglioma
histology, and less than 1 cm of residual tumor on postoperative MRI scan, indicative of the survival benefit
associated with more extensive resection. In a wellcharacterized cohort, Smith et al26 demonstrated an association between gross-total resection of nonenhancing
tumor and survival in low-grade (WHO grade II) gliomas. The
majority of patients in these studies likely had IDH1-mutant
tumors.
Thus, the emerging molecular classification forms a
practical guide for surgical decision making in patients with
adult diffuse gliomas. For molecular glioblastomas, the
evidence supports a surgical goal of complete resection of
enhancing disease. There is no evidence yet in this population that surgical resection of nonenhancing disease
offers a survival benefit; surgical goals therefore should be
curtailed for lesions in eloquent cortex with a focus on
safety. For molecular astrocytomas, evidence supports a
surgical goal of maximal resection for both enhancing and
nonenhancing disease. The standard of care for subsequent
adjuvant treatment of patients with molecular astrocytoma
awaits guidance from ongoing randomized studies. Finally,
for molecular oligodendrogliomas, a surgical goal of maximal
safe resection of enhancing and nonenhancing disease again
is warranted but tempered with the knowledge that, for
these patients, well-established adjuvant therapies have
been proven highly effective in the randomized studies

discussed in this article.
TARGETED THERAPY IN LOWER-GRADE GLIOMA

Aberrant signaling in pathways, including the PI3K/Akt/
mTOR network, have been identified in grade II glioma,
and clinical trials are ongoing to target this pathway as a
therapeutic approach.27 In addition, ongoing studies are
evaluating inhibitors of IDH.28 The ability to image levels of
the oncometabolite 2-hydroxyglutarate is an exciting area
of research to develop noninvasive robust biomarkers of
treatment response and clinical outcome in IDH-mutated
tumors.29 These approaches mandate the selection of patients on the basis of specific molecular/cytogenetic characteristics and highlight the future of medical treatment of
LrGG in the genomic era.
CONCLUSION
There is increasing evidence that multiple molecular markers
are prognostic, and possibly even predictive, in patients with
LrGG. The 1p19q codeletion historically has conferred improved prognosis and recently has been identified as a
predictive marker. IDH mutations, similarly, have also been
identified as prognostic and, possibly, predictive.5,8,11 By
using a combination of IDH1/2, 1p19q, and TERT mutations,
it is possible to subdivide these tumors into three working
subgroups: molecular glioblastoma, molecular astrocytoma,
and molecular oligodendroglioma. Effectively, this stratifies
LrGGs from the worst prognostic category that has outcomes
comparable to GBM to the best with very long survival rates
measured in decades and includes an in between category
as well; all categories likely differ in response to therapy.
Such stratification now calls into question both the existing
grade-based conventional therapeutic strategies and the
value of ongoing clinical trials that do not include these as
stratification variables.
For the worst prognostic group of LrGG, more effective
therapies might be necessary, because conventional therapies are relatively ineffective; patients with these tumors
could be enrolled in clinical trials of glioblastoma, perhaps
as a separate subgroup. Among those with the best prognosis, median survival exceeds a decade, and this subgroup
might include patients who can actually be observed in lieu
of immediate intervention. These patients might be suited
for treatment de-intensification, but this possibility raises
the conundrum of how to risk treatment de-intensification,
because even a modest loss in survival would take longer
than a decade to detect. One possible approach would be
to de-intensify only those components that have greater
toxicity; for example, perhaps the indirect data about the
efficacy of temozolomide in LrGGs would be adequate to
consider its use instead of PCV?11,30,31 From an RT perspective, perhaps these are the tumors better suited for
proton therapy, with its inherently lower integral dose.
How do we move this genomic information directly to the
therapeutic realm, specifically in terms of tailored therapies?
79
CHANG ET AL
IDH-directed therapies have entered phase I clinical testing,

but, with prolonged median survival rates, how exactly does
one establish the therapeutic potential of such an agent? Of
course, testing in the recurrence situation with response
rates or PFS as endpoints is one initial step, but making the
leap to a survival-based trial in the newly diagnosed context
will likely prove challenging. For example, one recent report
suggests that IDH inhibitors might in fact be radioprotective,
which potentially negates their clinical value. 32 In the
meantime, data are emerging that these IDH-mutated tumors might be amenable to more complete resections and,
maybe, that these are the patients who, after a gross total
resection, could in fact receive observation only.24,33 Are
these perhaps also the patients best suited for and driving the lengthy survival rates associated with supratotal
resection, a surgical concept promulgated by Duffau?33

Will metabolic imaging that is based on the IDH-associated
oncometabolite 2-hydroxyglutartate provide an avenue to
monitor residual disease or the effects of various therapies
or will it perhaps become a tool for contouring the target
for RT?34
IDH mutations represent the emergence of a new paradigm; recently described molecular classifications categorically demonstrate a number of less common, but
potentially targetable, alterations, all of which individually
are sufficiently rare that it would be near impossible to
mount classic randomized phase III trials for each and
every one. Perhaps the concept of the National Cancer
Institutesponsored MATCH trial could be expanded to
these LrGGs.35
References
1. Louis DN, Ohgaki H, Wiestler OD, et al. The 2007 WHO classification of
tumours of the central nervous system. Acta Neuropathol. 2007;114:
97-109.
2. van den Bent MJ. Interobserver variation of the histopathological diagnosis in clinical trials on glioma: a clinicians perspective. Acta
Neuropathol. 2010;120:297-304.
3. Brat DJ, Verhaak RG, Aldape KD, et al; Cancer Genome Atlas Research
Network. Comprehensive, integrative genomic analysis of diffuse
lower-grade gliomas. N Engl J Med. 2015;372:2481-2498.
4. Eckel-Passow JE, Lachance DH, Molinaro AM, et al. Glioma groups based
on 1p/19q, IDH, and TERT promoter mutations in tumors. N Engl J Med.
2015;372:2499-2508.
5. Yan H, Parsons DW, Jin G, et al. IDH1 and IDH2 mutations in gliomas.
N Engl J Med. 2009;360:765-773.
6. Cairncross G, Wang M, Shaw E, et al. Phase III trial of chemoradiotherapy for anaplastic oligodendroglioma: long-term results of
RTOG 9402. J Clin Oncol. 2013;31:337-343.
7. van den Bent MJ, Brandes AA, Taphoorn MJ, et al. Adjuvant procarbazine, lomustine, and vincristine chemotherapy in newly diagnosed
anaplastic oligodendroglioma: long-term follow-up of EORTC brain
tumor group study 26951. J Clin Oncol. 2013;31:344-350.
8. Cairncross JG, Wang M, Jenkins RB, et al. Benefit from procarbazine,
lomustine, and vincristine in oligodendroglial tumors is associated with
mutation of IDH. J Clin Oncol. 2014;32:783-790.
9. Stewart LA. Chemotherapy in adult high-grade glioma: a systematic
review and meta-analysis of individual patient data from 12 randomised
trials. Lancet. 2002;359:1011-1018.
10. Medical Research Council Brain Tumor Working Party. Randomized trial
of procarbazine, lomustine, and vincristine in the adjuvant treatment of
high-grade astrocytoma: a Medical Research Council trial. J Clin Oncol.
2001;19:509-518.
11. Chang SM, Zhang P, Cairncross JG, et al. Results of NRG oncology/RTOG
9813: A phase III randomized study of radiation therapy (RT) and
temozolomide (TMZ) versus RT and nitrosourea (NU) therapy for anaplastic astrocytoma (AA). J Clin Oncol. 2015;33 (suppl; abstr 2002).
12. Wick W, Hartmann C, Engel C, et al. NOA-04 randomized phase III trial of
sequential radiochemotherapy of anaplastic glioma with procarbazine,
lomustine, and vincristine or temozolomide. J Clin Oncol. 2009;27:
5874-5880.
13. Shaw EG, Wisoff JH. Prospective clinical trials of intracranial low-grade
glioma in adults and children. Neuro Oncol. 2003;5:153-160.
80
14. Buckner JC, Pugh SL, Shaw EG, et al. Phase III study of radiation therapy
(RT) with or without procarbazine, CCNU, and vincristine (PCV) in lowgrade glioma: RTOG 9802 with Alliance, ECOG, and SWOG. J Clin Oncol.
2014;32:5s (suppl;abstr 2000).
15. Buckner JC, Shaw EG, Pugh SL, et al. Long-term results of R9802: a Phase
III study of radiation therapy (RT) with or without procarbazine, CCNU,
and vincristine (PCV) in low-grade gliomaRTOG with NCCTG, ECOG,
and SWOG. N Engl J Med. In press.
16. Pignatti F, van den Bent M, Curran D, et al. Prognostic factors for survival
in adult patients with cerebral low-grade glioma. J Clin Oncol. 2002;20:
2076-2084.
17. Karim AB, Afra D, Cornu P, et al. Randomized trial on the efficacy of
radiotherapy for cerebral low-grade glioma in the adult: European
Organization for Research and Treatment of Cancer Study 22845 with
the Medical Research Council study BRO4an interim analysis. Int J
18. Shaw E, Arusell R, Scheithauer B, et al. Prospective randomized trial of
low- versus high-dose radiation therapy in adults with supratentorial
low-grade glioma: initial report of a North Central Cancer Treatment
Group/Radiation Therapy Oncology Group/Eastern Cooperative Oncology Group study. J Clin Oncol. 2002;20:2267-2276.
19. Baumert BG, Hegi ME, Mason WP, et al. Radiotherapy in relation to
temozolomide: subgroup analysis of molecular markers of the randomized phase III study by the EORTC/NCIC-CTG/TROG/MRC-CTU
(EORTC 22033-26033) in patients with a high-risk low-grade glioma.
20. Glantz MJ, Burger PC, Herndon JE II, et al. Influence of the type of
surgery on the histologic diagnosis in patients with anaplastic gliomas.
Neurology. 1991;41:1741-1744.
21. Jackson RJ, Fuller GN, Abi-Said D, et al. Limitations of stereotactic biopsy
in the initial management of gliomas. Neuro Oncol. 2001;3:193-200.
22. Kim BY, Jiang W, Beiko J, et al. Diagnostic discrepancies in malignant
astrocytoma due to limited small pathological tumor sample can be
overcome by IDH1 testing. J Neurooncol. 2014;118:405-412.
23. Louis DN, Perry A, Burger P, et al; International Society Of
NeuropathologyHaarlem. International Society Of Neuropathology
Haarlem consensus guidelines for nervous system tumor classification
and grading. Brain Pathol. 2014;24:429-435.
24. Beiko J, Suki D, Hess KR, et al. IDH1 mutant malignant astrocytomas are
more amenable to surgical resection and have a survival benefit associated with maximal surgical resection. Neuro Oncol. 2014;16:81-91.
25. Shaw EG, Berkey B, Coons SW, et al. Recurrence following neurosurgeondetermined gross-total resection of adult supratentorial low-grade
glioma: results of a prospective clinical trial. J Neurosurg. 2008;109:
835-841.
26. Smith JS, Chang EF, Lamborn KR, et al. Role of extent of resection in the
long-term outcome of low-grade hemispheric gliomas. J Clin Oncol.
2008;26:1338-1345.
27. McBride SM, Perez DA, Polley MY, et al. Activation of PI3K/mTOR
pathway occurs in most adult low-grade gliomas and predicts patient
survival. J Neurooncol. 2010;97:33-40.
28. Rohle D, Popovici-Muller J, Palaskas N, et al. An inhibitor of mutant IDH1
delays growth and promotes differentiation of glioma cells. Science.
2013;340:626-630.
29. Andronesi OC, Loebel F, Bogner W, et al. Treatment response assessment in IDH-mutant glioma patients by non-invasive 3D functional
spectroscopic mapping of 2-hydroxyglutarate. Clin Cancer Res. 2016;22:
1632-1641.
30. Vogelbaum MA, Hu C, Peereboom DM, et al. Phase II trial of preirradiation and concurrent temozolomide in patients with newly
31.
32.
33.
34.
35.
diagnosed anaplastic oligodendrogliomas and mixed anaplastic oligoastrocytomas: long-term results of RTOG BR0131. J Neurooncol.
2015;124:413-420.
Fisher BJ, Hu C, Macdonald DR, et al. Phase 2 study of temozolomidebased chemoradiation therapy for high-risk low-grade gliomas: preliminary results of Radiation Therapy Oncology Group 0424. Int J Radiat
Oncol Biol Phys. 2015;91:497-504.
Molenaar RJ, Botman D, Smits MA, et al. Radioprotection of IDH1mutated cancer cells by the IDH1-mutant inhibitor AGI-5198. Cancer
Res. 2015;75:4790-4802.
Duffau H. Long-term outcomes after supratotal resection of diffuse lowgrade gliomas: a consecutive series with 11-year follow-up. Acta
Neurochir (Wien). 2015;158:51-58.
Dang L, White DW, Gross S, et al. Cancer-associated IDH1 mutations
produce 2-hydroxyglutarate. Nature. 2009;462:739-744.
Clinical Trial NCT02465060. NCI-MATCH: targeted therapy directed by
genetic testing in treating patients with advanced refractory solid tumors or lymphomas. https://clinicaltrials.gov/ct2/show/NCT02465060.
Accessed November 11, 2015.
81
DEVELOPMENTAL THERAPEUTICS AND

TRANSLATIONAL RESEARCH
Biomarkers, Blood-Based Testing,

and the Heterogeneous Tumor
CHAIR
Charles Swanton, MBPhD, FRCP, FMedSci
The Francis Crick Institute
London, United Kingdom
SPEAKERS
Julia A. Beaver, MD
Silver Spring, MD
Melissa Lynne Johnson, MD
Nashville, TN
MANAGEMENT OF INTRATUMOR HETEROGENEITY
Tumor Evolutionary Principles: How Intratumor

Heterogeneity Influences Cancer Treatment and Outcome
Subramanian Venkatesan, BSc, and Charles Swanton, MBPhD, FRCP, FMedSci
OVERVIEW
Recent studies have shown that intratumor heterogeneity contributes to drug resistance in advanced disease. Intratumor
heterogeneity may foster the selection of a resistant subclone, sometimes detectable prior to treatment. Next-generation
sequencing is enabling the phylogenetic reconstruction of a cancers life history and has revealed different modes of cancer
evolution. These studies have shown that cancer evolution is not always stochastic and has certain constraints. Consideration of cancer evolution may enable the better design of clinical trials and cancer therapeutics. In this review, we
summarize the different modes of cancer evolution and how this might impact clinical outcomes. Furthermore, we will
discuss several therapeutic strategies for managing emergent intratumor heterogeneity.
he selection of fitter subclones in the context of tumor

evolution was postulated by Nowell in the 1970s.1 The
existence of multiple phenotypes within a single tumor was
determined 3 decades ago2; however, the emergence of
next-generation sequencing has enabled the more detailed
study of cancer evolution and intratumor heterogeneity.
Genome instability is one of the proposed hallmarks of
cancer and is believed to be acquired early in tumorigenesis,
promoting the acquisition of other cancer hallmarks.3,4 It
is a collective term, referring to a spectrum of different
genetic aberrations ranging from point mutations to chromosomal rearrangements, gains, and losses.5,6 Genome
instability equips cancer with the potential to create new
genetic aberrations in daughter cells and to adapt to new
environments.3 Once one of these genetic aberrations
confers a fitness advantage, genetically distinct subclones
may arise within a single cancer. Many cancers show hallmarks of genome instability that support the development
of intratumor heterogeneity, evidenced by elevated rates
of point mutations,7-9 chromosomal rearrangements,10 and
somatic copy-number aberrations.10,11 Intratumor heterogeneity and cancer evolution contribute to resistance to
therapy and therefore demand investigation.
As the number of studies reporting on the extent of
intratumor heterogeneity in different cancer types is rising
sharply, it is becoming clear that cancer evolution is not
always stochastic. Recurring commonalities observed within
each cancer type suggests there are certain rules and
constraints dictating how a tumor can evolve. Identifying
rule books of cancer evolution might have important

implications for designing trials and cancer treatments. In
this review, we will touch upon how intratumor heterogeneity may affect clinical outcome and discuss the different
modes of cancer evolution. We will also discuss different
therapeutic strategies for the management of intratumor
heterogeneity.
THE RELATIONSHIP BETWEEN INTRATUMOR

HETEROGENEITY AND CLINICAL OUTCOME
Genome instability and intratumor heterogeneity are closely
associated, as genome instability increases the rate of
mutation (in the broadest sense referring to nucleotide and
to whole chromosome level changes) and the subsequent
chance of subclonal (i.e., in a fraction of cancer cells) diversification (reviewed by Burrell et al12). The influence of
genome instability and intratumor heterogeneity upon clinical outcome is becoming clearer. Chromosomal instability
(CIN) is a specific form of genome instability and refers to the
rate of acquiring whole-chromosome or segmental chromosomal aneuploidies.13 CIN may contribute to extensive
intratumor heterogeneity in clear cell renal cell carcinomas
(ccRCC)14; therefore, the identification of CIN-driving events
may inform how to manage intratumor heterogeneity. In
some cases, genome-doubled tumors are thought to precede the acquisition of a CIN phenotype.15 The ability of a
diploid cell to tolerate a genome-doubling event subsequently fosters the propagation of aneuploid cells and
the progressive evolution of chromosomally unstable tumor
From the UCL Cancer Institute, CRUK Lung Cancer Centre of Excellence, London, United Kingdom; The Francis Crick Institute, London, United Kingdom.
Corresponding author: Charles Swanton, MBPhD, FRCP, FMedSci, Translational Cancer Therapeutics Laboratory, The Francis Crick Institute, 44 Lincolns Inn Fields, London WC2A 3LY;
email: charles.swanton@crick.org.uk.
e141
VENKATESAN AND SWANTON
cells.15 Emerging evidence suggests that genome doubling

itself is a predictor of a poorer relapse-free survival in earlystage colorectal15 and ovarian cancers.16
Interestingly, a mouse model with a heterozygous knockout
of the gene centrosome-associated protein E (Cenp-e) led
to increased aneuploidy and CIN.17 The Cenp-e+/2 mice developed more spontaneous spleen and lung tumors compared
with the Cenp-e+/+ mice. In contrast, genetically (through
homozygous p19ARF deletion) and chemically (through 7,12dimethylbenz[a]anthracene exposure) induced tumorigenesis was inhibited in the Cenp-e+/2 mice compared with
wild-type mice. These results suggest that CIN, in some
cases, can both promote and inhibit tumorigenesis.17 This
study supports the mutational meltdown theory postulated
by Lynch et al regarding deleterious mutations in asexual
populations.18 Here, the accumulation of deleterious mutations combined with the inability of sexual recombination
contributes to the extinction process, which they termed
mutational meltdown. Cancer cells also divide and reproduce in an asexual manner; therefore, the observations
by Weaver et al17 suggested that the mutational meltdown
theory18 may also apply to human cancers. Our group investigated the association between the level of CIN and
clinical outcome.19,20 We found, in different cancer types,
that patients with tumors containing intermediate levels
of CIN had the worst prognosis, whereas tumors with
high levels of CIN were associated with improved prognosis
(Fig. 1).19,20 These observations have been recapitulated in
recent studies of high-grade serous ovarian cancer,21 estrogen receptornegative breast cancer,19,22 and a pancancer study across 12 cancer types.23
Although the number of patients with ovarian cancer
studied by Schwarz et al and patients with esophageal cancer
studied by Murugaesu et al were probably not large enough
to detect the nonlinear relationship between the extent of
intratumor heterogeneity and response to therapy, they still
demonstrated that patients with more heterogeneous
cancers had poorer response.21,24 In glioblastoma, Kim et al
report that TP53 mutations coincide with higher frequencies
of subclonal mutations and better clinical outcome, which
KEY POINTS
e142
Cancer evolution is not always stochastic and is

subjected to several constraints evidenced by parallel
and convergent evolution.
There is a trade-off between tumor fitness and genome
instability.
Different modes of cancer evolution may be operative
over the lifetime of a cancer.
Multiregion sequencing and/or longitudinal monitoring
of cancer subclones enables a better understanding of
evolutionary life histories.
Understanding cancer evolution may inform the better
design of clinical trials and cancer treatments.
FIGURE 1. The Trade-off Between Genome

Instability/Intratumor Heterogeneity and Clinical
Outcome
might be related to the trade-off between tumor fitness and

genome instability as previously discussed (Fig. 1).25 Collectively, these studies suggest a range of optimal genome
instability favoring the tumors fitness, which at the right
threshold, may negatively affect clinical outcome (Fig. 1).
However, beyond a certain threshold, excessive genome
instability may result in diminished cancer cell survival and
is paradoxically associated with an improvement in cancer
survival outcomes (Fig. 1).
The influence of specific somatic alterations on the rate of
evolution has been subject to investigation. Genomic aberrations in cancer may accumulate either gradually such as
through point mutations (microevolution), or in a punctuated manner through processes such as chromoplexy,26 or
in a single catastrophic event such as through chromothripsis (macroevolution).27 The distinction between these
different magnitudes of genomic changes might be of importance for the rate of disease progression. For example,
chromothripsis was identified in a small subset of patients in
multiple myeloma and was often related to rapid relapse and
death within 1 to 2 years.28
It is of clinical importance to elucidate how cancers tolerate such somatic events and propagate diversity and
branched evolution, as this may inform how cancer progresses and support new therapeutic approaches to limit
these processes. Through deep, multiregion, and longitudinal sequencing studies of the primary tumor and metastases, it might be possible to order somatic events more
effectively and identify genetic events enabling the tolerance and propagation of genome instability. For example,
prostate cancer metastases were enriched for TP53 mutations, whereas this mutation was subclonal in the primary
tumor.29 Conceivably, metastasis may be enhanced by lossof-function of TP53, possibly allowing for greater tolerance
of genome instability and the subsequent gain of a
metastasis-driving event.
Unfortunately, the clinical effect of targeted therapeutic
drugs has been limited by the rapid acquisition of drug
resistance in the metastatic setting. There are now multiple
reports suggesting that drug-resistant subclones reside
within the tumor at low frequency prior to therapy in
melanoma,30 prostate cancer,31 colorectal cancer,32,33 ovarian

cancer,21 and medulloblastoma.34 We recently reported in a
study of nine cancer types that mutations in many driver
genes can be subclonal across multiple different tumor types.
These data suggest that targeting founder events present in
every tumor cell may facilitate a more effective drug development process.35
MODES OF CANCER EVOLUTION

In the past decade, different modes of cancer evolution have
been discovered. Historically, cancer was believed to follow
the multistep process of oncogene activation, tumor suppressor gene loss, and subsequent clonal sweeps by the
fittest clone.1,36 With current advances in next-generation
sequencing technology, the initial hypothesis has been refined and additional models of macroevolution26,27,37 and
neutral evolution38-40 have been postulated. In this section,
we will summarize the current state of knowledge of cancer
evolution (Table 1 and Fig. 2).
Darwinian Cancer Evolution

Deep sequencing analysis of hematologic and solid tumors
indicate that cancer is usually of clonal origin and often
follows a branched tumor evolutionary pattern.41 Using
tumor purity estimates, local copy number aberrations, and
mutation variant allele frequency, it is possible to calculate
the cancer cell fraction of a mutation and phylogenetically
reconstruct the life history of cancers.42-45 The trunk of the
phylogenetic tree describes the clonal events where the
founder driver events are present. Trunk driver events are
present in all cancer cells derived from the cell of origin.
As the cancer genome evolves during cancer progression, subclonal events are introduced into a subset of the
progeny, termed branched events (Fig. 2A). A clonal sweep
occurs if a branch driver event increases the fitness of a
subclone to the extent that it out-competes all other subclones in the tumor. This mode of cancer evolution has been
termed linear evolution (Fig. 2B). Reports suggest that
branched evolution is common in cancer, as there may
be one or more subclonal/branched driver events present within distinct subclones of a tumor. Nevertheless,
in several cancer types, cases of linear evolution have been
documented such as in acute myeloid leukemia, 46,47
chronic lymphocytic leukemia, 48 and VHL-associated
ccRCC. 49
We have studied four clonally independent primary kidney
cancers in the context of a germline VHL mutation.49 Several
observations were made from this patient: (1) the tumors
were polyclonal, (2) the tumors shared no other mutations apart from the germline VHL mutation highlighting
independent evolutionary trajectories despite similar microenvironments, and (3) the tumors showed signs of
evolutionary constraints shown by the absence of branched
evolution and the presence of convergent evolution toward
the PI3K-Akt-mTOR pathway (Fig. 2C).49 Convergent evolution refers to the process of independent tumors within
the same patient acquiring (epi)genetic alterations in similar

genes, protein complexes, or signaling pathways (Fig. 2C).49
Conversely, parallel evolution refers to subclones derived
from the same parental clone acquiring (epi)genetic alterations in similar genes, protein complexes, or signaling
pathways (Fig. 2D).50,51 Although many tumors follow a
branched evolutionary pattern, presumably many of these
tumors experience common selective pressures from the
microenvironment or constraints inherited from the parental founder cell, which force the different subclones to
independently converge on similar cellular processes. The
level of convergent and parallel evolution is of emerging
interest because of developments in the use of multiregion
and single cell sequencing (Table 1).
Up to this point, we have described cancer as co-evolving,
mutually independent subclones; however, reality is more
complex as these subclones also interact. In an environment
with limited space and nutritional resources, there are
complex dynamics of clonal cooperation and clonal interference (Fig. 2E), which are only now beginning to be
understood. In vitro52-55 and in vivo54-57 data show that
subclonal populations can protect and/or drive cancer as a
whole. Marusyk et al elegantly demonstrated in a mouse
model that a minor subclone with inferior fitness drove
tumor growth in a noncell-autonomous fashion through
the secretion of a specific ligand.56 This enabled the tumor to
maintain growth, clonal diversity, and even metastatic potential. These tumor phenotypes could be abrogated once
the driving subclones were omitted from the tumor, highlighting the importance of examining minor subclones in
addition to dominant subclones. Similar clonal dynamics
are likely also operative in patients. Clonal interference
between a dominant subclone and a minor subclone may
prevent the outgrowth of the minor subclone during tumor
progression (Fig. 2E). However, the harsh selective pressure
imposed by the tumor microenvironment, the cancer
therapeutics directed toward the dominant clone, or the
process of metastasis in general may permit a resistant
minor subclone residing in the tumor to emerge by competitive release.58 Gatenby et al hypothesized that a tumor
may consist of chemotherapy-sensitive cells that have a
higher fitness in the absence of chemotherapeutic pressure
and suppress the outgrowth of less fit chemotherapyresistant subpopulation. They demonstrated that adapting
the chemotherapy dose according to tumor dynamicsthat
is, higher doses upon tumor growth and lower doses upon
tumor reductionwas able to stabilize the tumor and
prevent the resistant subpopulation from reaching lethal
proportions.58 This therapeutic strategy was appropriately
called adaptive therapy, and, related to this, it was recently
shown to be potentially relevant in the treatment of
BRAFV600E-mutant melanomas (Fig. 2F).59
A different approach to address an emerging treatmentresistant subclone may include altering the drug instead
of the drug doses with the aim to eradicate the cancer
(i.e., sequential therapy; Fig. 2G). In sequential therapy,
the treatment is sequentially adapted according to the
e143
evolution of new treatment-resistant events. The potential of sequential therapy has been demonstrated in the
treatment of drug-sensitive EGFR-driven (often EGFRL585R
or EGFR exon 19 deletion) lung cancer with first-generation
EGFR inhibitors (erlotinib or gefitinib). Tumors often
progress as a result of the emergence of an EGFRT790M
resistanceconferring mutation.60 However, this mutation
renders cancer cells sensitive to the third-generation EGFR
inhibitors, AZD929161 and rociletinib.62 Nevertheless, sequential therapy may eventually encounter a resistant
subclone that is not yet amenable to treatment because
treating tumors that harbor the EGFRT790M variant with
either AZD929163 or rociletinib 64 appears to select for the
re-emergence of EGFRT790M-negative cells, EGFRC797S
mutant cells, or EGFRT790M-positive cells with a yet unidentified resistance event (Fig. 2G). Clonal interactions are
not limited to the primary tumor but also extend to the
metastatic sites. Phylogenetic reconstruction of the primary
tumor and the accompanying metastases is enabling a

clearer picture of the metastatic process. Studies by
Gundem et al and Hong et al in prostate cancer show that
the following patterns of metastasis occur frequently
(Fig. 2H): (1) multiple subclones can seed the same
metastatic site (termed polyclonal seeding), (2) combinations of shared subclones between these polyclonal
seeds are recurring (hinting toward clonal cooperation),
(3) subclones within a metastasis can spread to another
metastatic site (termed metastasis-to-metastasis spread
or metastatic cross-seeding), and (4) tumor cells from
the metastasis may return back to the primary tumor
(termed self-seeding).65 The formation of metastatic subclones were long thought to occur through the clonal selection of the fittest and latest evolving subclone. However,
recent studies show that metastatic progression can also
occur through the early diverged clonal expansion of the
trunk clone.21,29
TABLE 1. Different Modes of Cancer Evolution Reported in a Selection of Different Cancer Types
Cancer Type
Mode of Evolution
AML
Branched evolution80,81
Parallel evolution81: FLT3-ITD
Linear evolution46,47,82
Breast Cancer
Branched evolution83
Parallel evolution83: PTEN, FGFR2, TP53, RUNX1
Punctuated and catastrophic evolution7,83: chromoplexy, chromothripsis, kataegis
Colorectal Cancer
Catastrophic evolution27: chromothripsis

Neutral evolution39,40
Esophageal
Cancer
Branched evolution24,84,85
Parallel evolution24: NOTCH1, GNPTAB
Polyclonal tumorigenesis85
Follicular
Lymphoma
Parallel evolution86,87: chromatin-modifying genes (KMT2D, KMT2C, EZH2, CREBBP, MEF2B, EP300, HIST1H1E,
HIST1H1C, SWI/SNF complex [ARID1A, SMARCA4])
Punctuated evolution7: kataegis
Branched evolution88,89 in distant relapse90
Glioma

Linear evolution in local relapse25,90
Parallel evolution25,90,91: EGFR, TP53, PTEN, RB1, NF1, CDKN2A/B
Melanoma
Metastasic cross-seeding93
Parallel evolution30,92,93: KRAS, BRAF, MEK1, PIK3R2, PHLPP1, CTNNB1
Multiple
Myeloma
Branched evolution94-96
Linear evolution94-97
Parallel evolution96,97: NRAS, KRAS, BRAF, NF1, RASA2, FAM46C
Punctuated and catastrophic evolution28,96: kataegis, chromothripsis
Continued
e144
TABLE 1. Different Modes of Cancer Evolution Reported in a Selection of Different Cancer Types (contd)
Cancer Type
Mode of Evolution
NSCLC
Punctuated and catastrophic evolution7,27,98,100: chromoplexy, chromothripsis, kataegis
Ovarian Cancer
Branched evolution21
Prostate Cancer
Punctuated and catastrophic evolution26: chromoplexy
Parallel evolution102: MYC, methylation sites
Polyclonal seeding103
Sporadic ccRCC
Parallel evolution45,51,68: mTOR pathway, SETD2, PTEN, KDM5C, PIK3CA, BAP1, PBRM1, SWI/SNF complex
VHL-associated
ccRCC
Linear evolution49
Convergent evolution49: mTOR pathway (TSC1, mTOR)
Abbreviations: AML, acute myeloid leukemia; NSCLC, nonsmall cell lung cancer; ccRCC, clear cell renal cell carcinoma.
Neutral Evolution
In addition to the model of Darwinian selection of the
fittest subclone, a different model was postulated by
Sottoriva et al termed the big bang model, or neutral
cancer evolution (Fig. 2I).39 In this study in colorectal
cancer, individual tumor glands and two bulk fragments
originating from the left and right side of the resected
tumors were subjected to different genomic techniques
including whole-exome sequencing and single nucleotide
polymorphism arrays. As expected, they found a proportion of the aberrations to be clonal, but unexpectedly,
the same set of subclonal (or private) mutations could be
found in different tumor glands on opposite sides of the
resection. These observations suggest that once the initial
cell has transformed into a cancer cell, subclonal mutations
rapidly accumulate in the primordial tumor. Subsequently,
the daughter cells of specific subclones may migrate small
distances within the primordial tumor mass. Indeed, this
short-range cellular migration may be a basic feature of
aggressive tumor growth.66 The distance between these
daughter cells is presumably amplified by the subsequent
(big banglike) growth, possibly explaining why tumor
glands of the same subclone are scattered throughout the
tumor. Other explanations for the finding that exactly the
same subclonal mutations are found in different parts of the
colorectal cancer could include that these subclonal mutations are acquired late in tumor evolution with the
subsequent migration of these cancer cells. However, as
the authors stressed, this is unlikely because these private
mutations were clonal in individual tumor glands and it is

improbable that a whole gland would migrate.
This model of neutral cancer evolution states that after
malignant transformation, individual subclones grow at similar
rates, despite showing distinct mutational patterns. In this
context, the timing of a new mutation occurring is the major
determinant of its prevalence within the tumor, rather than
clonal selection for that mutation. In follow-up studies, the
role of neutral evolution across different cancer types was
investigated.38,40 These studies found that neutral evolution
may play an important role in tumor development of cervical
squamous cell carcinoma and endocervical adenocarcinoma,
colorectal, stomach, lung, prostate, bladder, and urothelial
cancers.38,40 However, once treatment has been initiated,
it is likely that other forms of Darwinian cancer evolution
take over and eventually force the selection of treatmentresistant subclones.
APPROACHES TO TRANSLATE KNOWLEDGE OF

INTRATUMOR HETEROGENEITY INTO CANCER
TREATMENT
Over the past few years, the cancer research community has
hunted for substantially mutated genes or driver genes. The
underlying premise was that cancer cells have become
addicted to the driver gene and, consequently, inhibition of
these genes may halt tumor progression. However, the
presence of multiple subclones with different driver and
resistance mutations in a single tumor is a major hurdle for
e145
FIGURE 2. Different Modes of Cancer Evolution Contribute to Different Patterns of Intratumor Heterogeneity
e146
targeted therapy. Cancer drug development is undoubtedly

made more costly by the ability of a resistant subclone to be
selected during treatment in a matter of weeks or months.
There is a need to implement the knowledge gained by
cancer evolutionary models into designing better anticancer
therapies. In this section, we will provide a brief summary
of different strategies incorporating knowledge of cancer
evolutionary medicine.
A therapeutic strategy that has been advocated by us and
others to maximize tumor response is to target trunk driver
events present in every tumor cell.67 In this case, it is important to accurately determine the clonal driver events.
Parallel evolution is widespread in different cancer types and
represents the constraints imposed on cancer evolution
(Table 1). Therefore, targeting parallel evolutionary events and
exploiting these cancer dependencies may provide a compelling approach to therapy.49,68,69 Indeed, these observations
were initially made in ccRCC, in which rapalogs worked exceptionally well in tumors that harbored multiple mutations
that converged upon the mTOR pathway,68 suggesting that
parallel evolution could pose as an Achilles heel of cancer.
As cancer evolves during treatment, dynamic therapy
strategies may allow the clinician to maintain a stable
population of treatment-sensitive cells (adaptive therapy),58
or even force the tumor down a particular evolutionary
route resulting in acquired sensitivity to a different drug
(sequential therapy). The utility of sequential therapy has
recently been demonstrated in a patient with an ALKrearranged nonsmall cell lung cancer, who sequentially
responded to different generations of ALK inhibitors.70 To
inform the next step in adaptive and sequential therapy,
longitudinal monitoring of the cancer is essential. Liquid
biopsy analysis of circulating tumor DNA or circulating tumor
cells is a convenient way to monitor cancer genetic changes.
Furthermore, liquid biopsy has the potential to detect
subclonal events,71-74 minimal residual disease, and the

acquisition of new somatic alterations associated with drug
resistance,31,32,75 and it can be used to monitor response73,76 (reviewed in Crowley et al77 and Krebs et al78).
In addition to adapting treatment to the adapting cancer, it
may be attractive to interfere with the causes of genome
instability. Reducing genome instability may reduce the
adaptability of cancer cells to treatments. Conversely, increasing genome instability may tip the cancer cell beyond
the levels of tolerance, forcing cell-autonomous lethality. As
there are many causes of genome instability (reviewed in
Aguilera et al6 and Burrell et al12), identifying the mechanistic basis of micro- and macroevolutionary events may
enable the therapeutic intervention and prevention of major
resistance-forming events from occurring.
CONCLUSION
The number of studies investigating cancer evolution is rising
sharply and has contributed to early insights into cancer evolutionary rule books. Sequencing of bulk tumor samples in
projects such as The Cancer Genome Atlas have set the foundations for precision medicine; however, intratumor heterogeneity is still largely underestimated in these datasets. Multiregion
sequencing and longitudinal monitoring via liquid biopsy together
with postmortem studies is necessary to attain a more accurate
picture of cancer evolution. TRACERx (TRAcking nonsmall cell
lung Cancer Evolution through therapy [Rx]) is such a prospective
study, which aims to define the evolutionary trajectories
of nonsmall cell lung cancer.79
Although the complexities of cancer evolution are just
beginning to be understood, much has been learned over the
past decade. We anticipate that cancer evolutionary models
will inform clinicians how to design clinical trials and cancer
treatments.
References
1.
2.
3.
4.
5.
6.
7.
8.
Nowell PC. The clonal evolution of tumor cell populations. Science.

1976;194:23-28.
Heppner GH. Tumor heterogeneity. Cancer Res. 1984;44:2259-2265.
Negrini S, Gorgoulis VG, Halazonetis TD. Genomic instabilityan
evolving hallmark of cancer. Nat Rev Mol Cell Biol. 2010;11:220-228.
Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation.
Cell. 2011;144:646-674.
Aguilera A, Gomez-Gonz
alez B. Genome instability: a mechanistic view

of its causes and consequences. Nat Rev Genet. 2008;9:204-217.
Aguilera A, Garca-Muse T. Causes of genome instability. Annu Rev
Genet. 2013;47:1-32.
Alexandrov LB, Nik-Zainal S, Wedge DC, et al; Australian Pancreatic
Cancer Genome Initiative; ICGC Breast Cancer Consortium; ICGC
MMML-Seq Consortium; ICGC PedBrain. Signatures of mutational
processes in human cancer. Nature. 2013;500:415-421.
Lawrence MS, Stojanov P, Mermel CH, et al. Discovery and saturation
analysis of cancer genes across 21 tumour types. Nature. 2014;505:
495-501.
9.
10.
11.
12.
13.
14.
15.
Kandoth C, McLellan MD, Vandin F, et al. Mutational landscape and

significance across 12 major cancer types. Nature. 2013;502:333-339.
Yang L, Luquette LJ, Gehlenborg N, et al. Diverse mechanisms of
somatic structural variations in human cancer genomes. Cell. 2013;
153:919-929.
Zack TI, Schumacher SE, Carter SL, et al. Pan-cancer patterns of somatic copy number alteration. Nat Genet. 2013;45:1134-1140.
Burrell RA, McGranahan N, Bartek J, et al. The causes and consequences of genetic heterogeneity in cancer evolution. Nature. 2013;
501:338-345.
Geigl JB, Obenauf AC, Schwarzbraun T, et al. Defining chromosomal
instability. Trends Genet. 2008;24:64-69.
Martinez P, Birkbak NJ, Gerlinger M, et al. Parallel evolution of tumour
subclones mimics diversity between tumours. J Pathol. 2013;230:
356-364.
Dewhurst SM, McGranahan N, Burrell RA, et al. Tolerance of wholegenome doubling propagates chromosomal instability and accelerates cancer genome evolution. Cancer Discov. 2014;4:175-185.
e147
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
e148
Carter SL, Cibulskis K, Helman E, et al. Absolute quantification of somatic

DNA alterations in human cancer. Nat Biotechnol. 2012;30:413-421.
Weaver BA, Silk AD, Montagna C, et al. Aneuploidy acts both oncogenically and as a tumor suppressor. Cancer Cell. 2007;11:25-36.
Lynch M, Burger R, Butcher D, et al. The mutational meltdown in
asexual populations. J Hered. 1993;84:339-344.
Roylance R, Endesfelder D, Gorman P, et al. Relationship of extreme
chromosomal instability with long-term survival in a retrospective
analysis of primary breast cancer. Cancer Epidemiol Biomarkers Prev.
2011;20:2183-2194.
Birkbak NJ, Eklund AC, Li Q, et al. Paradoxical relationship between
chromosomal instability and survival outcome in cancer. Cancer Res.
2011;71:3447-3452.
Schwarz RF, Ng CK, Cooke SL, et al. Spatial and temporal heterogeneity
in high-grade serous ovarian cancer: a phylogenetic analysis. PLoS
Med. 2015;12:e1001789.
Jamal-Hanjani M, AHern R, Birkbak NJ, et al. Extreme chromosomal
instability forecasts improved outcome in ER-negative breast cancer:
a prospective validation cohort study from the TACT trial. Ann Oncol.
2015;26:1340-1346.
Andor N, Graham TA, Jansen M, et al. Pan-cancer analysis of the extent
and consequences of intratumor heterogeneity. Nat Med. 2016;22:
105-113.
Murugaesu N, Wilson GA, Birkbak NJ, et al. Tracking the genomic
evolution of esophageal adenocarcinoma through neoadjuvant
chemotherapy. Cancer Discov. 2015;5:821-831.
Kim H, Zheng S, Amini SS, et al. Whole-genome and multisector exome
sequencing of primary and post-treatment glioblastoma reveals
patterns of tumor evolution. Genome Res. 2015;25:316-327.
Baca SC, Prandi D, Lawrence MS, et al. Punctuated evolution of
prostate cancer genomes. Cell. 2013;153:666-677.
Stephens PJ, Greenman CD, Fu B, et al. Massive genomic rearrangement acquired in a single catastrophic event during cancer
development. Cell. 2011;144:27-40.
Magrangeas F, Avet-Loiseau H, Munshi NC, et al. Chromothripsis
identifies a rare and aggressive entity among newly diagnosed
multiple myeloma patients. Blood. 2011;118:675-678.
Hong MK, Macintyre G, Wedge DC, et al. Tracking the origins and
drivers of subclonal metastatic expansion in prostate cancer. Nat
Commun. 2015;6:6605.
Kemper K, Krijgsman O, Cornelissen-Steijger P, et al. Intra- and intertumor heterogeneity in a vemurafenib-resistant melanoma patient
and derived xenografts. EMBO Mol Med. 2015;7:1104-1118.
Romanel A, Tandefelt DG, Conteduca V, et al. Plasma AR and
abiraterone-resistant prostate cancer. Sci Transl Med. 2015;7:312re10.
Diaz LA Jr, Williams RT, Wu J, et al. The molecular evolution of acquired
resistance to targeted EGFR blockade in colorectal cancers. Nature.
2012;486:537-540.
Laurent-Puig P, Pekin D, Normand C, et al. Clinical relevance of KRASmutated subclones detected with picodroplet digital PCR in advanced
colorectal cancer treated with anti-EGFR therapy. Clin Cancer Res.
2015;21:1087-1097.
Morrissy AS, Garzia L, Shih DJ, et al. Divergent clonal selection dominates
medulloblastoma at recurrence. Nature. 2016;529:351-357.
McGranahan N, Favero F, de Bruin EC, et al. Clonal status of actionable
driver events and the timing of mutational processes in cancer
evolution. Sci Transl Med. 2015;7:283ra54.
Fearon ER, Vogelstein B. A genetic model for colorectal tumorigenesis.
Cell. 1990;61:759-767.
Gerlinger M, McGranahan N, Dewhurst SM, et al. Cancer: evolution
within a lifetime. Annu Rev Genet. 2014;48:215-236.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
Sottoriva A, Graham T. A pan-cancer signature of neutral tumor

evolution. bioRxiv. 2015. http://biorxiv.org/content/early/2015/02/
11/014894. Accessed February 12, 2016.
Sottoriva A, Kang H, Ma Z, et al. A Big Bang model of human colorectal
tumor growth. Nat Genet. 2015;47:209-216.
Williams MJ, Werner B, Barnes CP, et al. Identification of neutral
tumor evolution across cancer types. Nat Genet. Epub 2016 Jan 18.
Greaves M. Evolutionary determinants of cancer. Cancer Discov. 2015;
5:806-820.
Campbell PJ, Pleasance ED, Stephens PJ, et al. Subclonal phylogenetic
structures in cancer revealed by ultra-deep sequencing. Proc Natl
Acad Sci USA. 2008;105:13081-13086.
Miller CA, White BS, Dees ND, et al. SciClone: inferring clonal architecture and tracking the spatial and temporal patterns of tumor
evolution. PLOS Comput Biol. 2014;10:e1003665.
Roth A, Khattra J, Yap D, et al. PyClone: statistical inference of clonal
population structure in cancer. Nat Methods. 2014;11:396-398.
Gerlinger M, Rowan AJ, Horswell S, et al. Intratumor heterogeneity
and branched evolution revealed by multiregion sequencing. N Engl J
Med. 2012;366:883-892.
Klco JM, Spencer DH, Miller CA, et al. Functional heterogeneity of
genetically defined subclones in acute myeloid leukemia. Cancer Cell.
2014;25:379-392.
Garg M, Nagata Y, Kanojia D, et al. Profiling of somatic mutations in
acute myeloid leukemia with FLT3-ITD at diagnosis and relapse. Blood.
2015;126:2491-2501.
Landau DA, Tausch E, Taylor-Weiner AN, et al. Mutations driving CLL and
their evolution in progression and relapse. Nature. 2015;526:525-530.
Fisher R, Horswell S, Rowan A, et al. Development of synchronous VHL
syndrome tumors reveals contingencies and constraints to tumor
evolution. Genome Biol. 2014;15:433.
Juric D, Castel P, Griffith M, et al. Convergent loss of PTEN leads to
clinical resistance to a PI(3)Ka inhibitor. Nature. 2015;518:240-244.
Gerlinger M, Horswell S, Larkin J, et al. Genomic architecture and
evolution of clear cell renal cell carcinomas defined by multiregion
sequencing. Nat Genet. 2014;46:225-233.
Hobor S, Van Emburgh BO, Crowley E, et al. TGFa and amphiregulin
paracrine network promotes resistance to EGFR blockade in colorectal
cancer cells. Clin Cancer Res. 2014;20:6429-6438.
Liu J, Joha S, Idziorek T, et al. BCR-ABL mutants spread resistance to
non-mutated cells through a paracrine mechanism. Leukemia. 2008;
22:791-799.
Zhang M, Tsimelzon A, Chang CH, et al. Intratumoral heterogeneity
in a Trp53-null mouse model of human breast cancer. Cancer Discov.
2015;5:520-533.
Inda MM, Bonavia R, Mukasa A, et al. Tumor heterogeneity is an active
process maintained by a mutant EGFR-induced cytokine circuit in
glioblastoma. Genes Dev. 2010;24:1731-1745.
Marusyk A, Tabassum DP, Altrock PM, et al. Non-cell-autonomous
driving of tumour growth supports sub-clonal heterogeneity. Nature.
2014;514:54-58.
Cleary AS, Leonard TL, Gestl SA, et al. Tumour cell heterogeneity
maintained by cooperating subclones in Wnt-driven mammary cancers. Nature. 2014;508:113-117.
Gatenby RA, Silva AS, Gillies RJ, et al. Adaptive therapy. Cancer Res.
2009;69:4894-4903.
Sun C, Wang L, Huang S, et al. Reversible and adaptive resistance to
BRAF(V600E) inhibition in melanoma. Nature. 2014;508:118-122.
Pao W, Miller VA, Politi KA, et al. Acquired resistance of lung adenocarcinomas to gefitinib or erlotinib is associated with a second
mutation in the EGFR kinase domain. PLoS Med. 2005;2:e73.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
Cross DA, Ashton SE, Ghiorghiu S, et al. AZD9291, an irreversible EGFR

TKI, overcomes T790M-mediated resistance to EGFR inhibitors in lung
cancer. Cancer Discov. 2014;4:1046-1061.
Walter AO, Sjin RT, Haringsma HJ, et al. Discovery of a mutantselective covalent inhibitor of EGFR that overcomes T790Mmediated resistance in NSCLC. Cancer Discov. 2013;3:1404-1415.
Thress KS, Paweletz CP, Felip E, et al. Acquired EGFR C797S mutation
mediates resistance to AZD9291 in non-small cell lung cancer harboring EGFR T790M. Nat Med. 2015;21:560-562.
Piotrowska Z, Niederst MJ, Karlovich CA, et al. Heterogeneity underlies
the emergence of EGFRT790 wild-type clones following treatment of
T790M-positive cancers with a third-generation EGFR inhibitor.
Cancer Discov. 2015;5:713-722.
Kim MY, Oskarsson T, Acharyya S, et al. Tumor self-seeding by circulating cancer cells. Cell. 2009;139:1315-1326.
Waclaw B, Bozic I, Pittman ME, et al. A spatial model predicts that
dispersal and cell turnover limit intratumour heterogeneity. Nature.
2015;525:261-264.
Yap TA, Gerlinger M, Futreal PA, et al. Intratumor heterogeneity:
seeing the wood for the trees. Sci Transl Med. 2012;4:127ps10.
Voss MH, Hakimi AA, Pham CG, et al. Tumor genetic analyses of
patients with metastatic renal cell carcinoma and extended benefit
from mTOR inhibitor therapy. Clin Cancer Res. 2014;20:1955-1964.
Wei EY, Hsieh JJ. A river model to map convergent cancer evolution
and guide therapy in RCC. Nat Rev Urol. 2015;12:706-712.
Shaw AT, Friboulet L, Leshchiner I, et al. Resensitization to Crizotinib by the
Lorlatinib ALK Resistance Mutation L1198F. N Engl J Med. 2016;374:54-61.
Murtaza M, Dawson SJ, Pogrebniak K, et al. Multifocal clonal evolution
characterized using circulating tumour DNA in a case of metastatic
breast cancer. Nat Commun. 2015;6:8760.
Siravegna G, Mussolin B, Buscarino M, et al. Clonal evolution and
resistance to EGFR blockade in the blood of colorectal cancer patients.
Nat Med. 2015;21:827.
Russo M, Siravegna G, Blaszkowsky LS, et al. Tumor heterogeneity and
lesion-specific response to targeted therapy in colorectal cancer.
Cancer Discov. 2016;6:147-153.
Lohr JG, Adalsteinsson VA, Cibulskis K, et al. Whole-exome sequencing
of circulating tumor cells provides a window into metastatic prostate
cancer. Nat Biotechnol. 2014;32:479-484.
Murtaza M, Dawson SJ, Tsui DW, et al. Non-invasive analysis of acquired resistance to cancer therapy by sequencing of plasma DNA.
Nature. 2013;497:108-112.
Dawson SJ, Tsui DW, Murtaza M, et al. Analysis of circulating tumor DNA
to monitor metastatic breast cancer. N Engl J Med. 2013;368:1199-1209.
Crowley E, Di Nicolantonio F, Loupakis F, et al. Liquid biopsy: monitoring
cancer-genetics in the blood. Nat Rev Clin Oncol. 2013;10:472-484.
Krebs MG, Metcalf RL, Carter L, et al. Molecular analysis of circulating
tumour cells-biology and biomarkers. Nat Rev Clin Oncol. 2014;11:129-144.
Jamal-Hanjani M, Hackshaw A, Ngai Y, et al. Tracking genomic cancer
evolution for precision medicine: the lung TRACERx study. PLoS Biol.
2014;12:e1001906.
Welch JS, Ley TJ, Link DC, et al. The origin and evolution of mutations in
acute myeloid leukemia. Cell. 2012;150:264-278.
Paguirigan AL, Smith J, Meshinchi S, et al. Single-cell genotyping
demonstrates complex clonal diversity in acute myeloid leukemia. Sci
Transl Med. 2015;7:281re2.
Hughes AE, Magrini V, Demeter R, et al. Clonal architecture of secondary acute myeloid leukemia defined by single-cell sequencing.
PLoS Genet. 2014;10:e1004462.
Yates LR, Gerstung M, Knappskog S, et al. Subclonal diversification of
primary breast cancer revealed by multiregion sequencing. Nat Med.
2015;21:751-759.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
Ross-Innes CS, Becq J, Warren A, et al; Oesophageal Cancer Clinical

and Molecular Stratification (OCCAMS) Study Group; Oesophageal
Cancer Clinical and Molecular Stratification OCCAMS Study Group.
Whole-genome sequencing provides new insights into the clonal
architecture of Barretts esophagus and esophageal adenocarcinoma.
Nat Genet. 2015;47:1038-1046.
Stachler MD, Taylor-Weiner A, Peng S, et al. Paired exome analysis of
Barretts esophagus and adenocarcinoma. Nat Genet. 2015;47:
1047-1055.
Green MR, Kihira S, Liu CL, et al. Mutations in early follicular lymphoma progenitors are associated with suppressed antigen presentation. Proc Natl Acad Sci USA. 2015;112:E1116-E1125.
Okosun J, Bodor C, Wang J, et al. Integrated genomic analysis
identifies recurrent mutations and evolution patterns driving the
initiation and progression of follicular lymphoma. Nat Genet. 2014;
46:176-181.
Sottoriva A, Spiteri I, Piccirillo SG, et al. Intratumor heterogeneity in
human glioblastoma reflects cancer evolutionary dynamics. Proc Natl
Acad Sci USA. 2013;110:4009-4014.
Johnson BE, Mazor T, Hong C, et al. Mutational analysis reveals the
origin and therapy-driven evolution of recurrent glioma. Science.
2014;343:189-193.
Kim J, Lee IH, Cho HJ, et al. Spatiotemporal evolution of the primary
glioblastoma genome. Cancer Cell. 2015;28:318-328.
Francis JM, Zhang CZ, Maire CL, et al. EGFR variant heterogeneity in
glioblastoma resolved through single-nucleus sequencing. Cancer
Discov. 2014;4:956-971.
Shi H, Hugo W, Kong X, et al. Acquired resistance and clonal evolution
in melanoma during BRAF inhibitor therapy. Cancer Discov. 2014;4:
80-93.
Sanborn JZ, Chung J, Purdom E, et al. Phylogenetic analyses of melanoma reveal complex patterns of metastatic dissemination. Proc Natl
Acad Sci USA. 2015;112:10995-11000.
Keats JJ, Chesi M, Egan JB, et al. Clonal competition with alternating
dominance in multiple myeloma. Blood. 2012;120:1067-1076.
Magrangeas F, Avet-Loiseau H, Gouraud W, et al. Minor clone
provides a reservoir for relapse in multiple myeloma. Leukemia. 2013;
27:473-481.
Bolli N, Avet-Loiseau H, Wedge DC, et al. Heterogeneity of genomic
evolution and mutational profiles in multiple myeloma. Nat Commun.
2014;5:2997.
Melchor L, Brioli A, Wardell CP, et al. Single-cell genetic analysis
reveals the composition of initiating clones and phylogenetic patterns
of branching and parallel evolution in myeloma. Leukemia. 2014;28:
1705-1715.
de Bruin EC, McGranahan N, Mitter R, et al. Spatial and temporal
diversity in genomic instability processes defines lung cancer evolution. Science. 2014;346:251-256.
Zhang J, Fujimoto J, Zhang J, et al. Intratumor heterogeneity in localized lung adenocarcinomas delineated by multiregion sequencing.
Science. 2014;346:256-259.
George J, Lim JS, Jang SJ, et al. Comprehensive genomic profiles of
small cell lung cancer. Nature. 2015;524:47-53.
Haffner MC, Mosbruger T, Esopi DM, et al. Tracking the clonal origin of
lethal prostate cancer. J Clin Invest. 2013;123:4918-4922.
Brocks D, Assenov Y, Minner S, et al; ICGC Early Onset Prostate
Cancer Project. Intratumor DNA methylation heterogeneity reflects
clonal evolution in aggressive prostate cancer. Cell Reports. 2014;8:
798-806.
Gundem G, Van Loo P, Kremeyer B, et al; ICGC Prostate UK Group. The
evolutionary history of lethal metastatic prostate cancer. Nature.
2015;520:353-357.
e149


CHAIR
Steven J. Lemery, MD, MHS
Silver Spring, MD
SPEAKERS
Francisco J. Esteva, MD, PhD
New York University Clinical Cancer Center
New York, NY
Martina Weise, MD
Federal Institute for Drugs and Medical Devices
Bonn, Germany
BIOSIMILARS: HERE AND NOW

Steven J. Lemery, MD, MHS, Francisco J. Esteva, MD, PhD, and Martina Weise, MD
OVERVIEW
Congress passed the Biologics Price Competition and Innovation Act (BPCI Act) as part of the Affordable Care Act on March
23, 2010. The BPCI Act authorized an approval pathway for biosimilar and interchangeable products. It defines biosimilarity
to mean that the biological product is highly similar to the reference product notwithstanding minor differences in clinically
inactive components and that there are no clinically meaningful differences between the biological product and the
reference product in terms of safety, purity, and potency of the product. The biosimilar pathway has the potential to
facilitate access to biologic products through increased competition, in the same manner as biosimilars have done for almost
10 years in Europe. The goal of a biosimilar program is not to independently establish safety and effectiveness for each
condition of use. Rather, the goal is to demonstrate biosimilarity through an extensive analytical characterization and a
targeted clinical program designed to assess for clinically meaningful differences, if they exist. The regulatory approaches in
both the United States and Europe involve a totality-of-the-evidence approach to demonstrate biosimilarity. Importantly,
the biosimilar pathway allows for extrapolation of data across indications so that a sponsor, with adequate scientific
justification, need not conduct clinical studies in each intended condition of use. Without extrapolation, development may
not be feasible for many products, and patients and resources could be diverted from clinical studies of newer agents for
cancer.
ongress passed the BPCI Act on March 23, 2010, as part

of the Affordable Care Act. The BPCI Act created a new
abbreviated licensure pathway for biologic products that are
shown to be biosimilar to, or interchangeable with, a U.S.
Food and Drug Administration (FDA)-licensed reference
product. According to the BPCI Act, a reference product is
the single biologic product licensed (i.e., approved) by FDA
that the proposed biosimilar is being compared with in an
application.1 The statutory definition requires the reference
product to be licensed under Section 351(a) of the Public
Health Service Act. The reference product must be an originator drug approved based on a full dossier of preclinical and
clinical data, and not a biosimilar, which is licensed under
Section 351(k) and not 351(a).1
The abbreviated licensure pathway for biosimilar and interchangeable products is possible because the proposed
product may rely for licensure on, among other things, publicly
available information on FDAs prior determination that the
reference product is safe, pure, and potent. This article will
discuss scientific considerations as understood by the authors
regarding the development and approval of biosimilar therapeutic biologic proteins (e.g., monoclonal antibodies or
certain cytokines) in the United States. It also highlights the
experience of biosimilars in Europe. As such, this article
should not be considered formal or binding FDA or European

Medicine Agencies (EMA) advice on regulatory or policy
matters related to biosimilar drugs.
The goal of a biosimilar development program differs from
traditional drug development. The goal of a traditional drug
development program is to demonstrate that the biologic
product is safe and effective for one or more indications. The
objective of a biosimilar development program is to demonstrate through a totality-of-the-evidence approach that
the biosimilar product is highly similar to the U.S.-licensed
reference product, notwithstanding minor differences in
clinically inactive components and that there are no clinically
meaningful differences between the biosimilar product and
the reference product in terms of safety, purity, and potency.2 A biosimilar sponsor does not need to independently
establish the safety and effectiveness of its product; it establishes this through a demonstration of biosimilarity. The
biosimilar pathway also allows for extrapolation of data
across indications. With adequate scientific justification, a
sponsor would not need to conduct a clinical study to
support approval of a biosimilar for each indication in the
reference products label.
Although the biosimilar pathway is an abbreviated approval pathway, the data package required for approval of a
From the U.S. Food and Drug Administration, Silver Spring, MD; New York University Clinical Cancer Center, New York, NY; Federal Institute for Drugs and Medical Devices (BfArM), Bonn,
Germany.
Corresponding author: Steven J. Lemery, U.S. Food and Drug Administration, 10903 New Hampshire Ave., Silver Spring, MD 20993; email: steven.lemery@fda.hhs.gov.
e151
LEMERY, ESTEVA, AND WEISE
biosimilar product can be quite extensive. Therefore, the

development of these products by sponsors and the review
of these products by regulatory authorities, including FDA
and EMA, is quite thorough to ensure that the products are
highly similar and will be safe and effective. Ultimately, FDA
uses a totality-of-the-evidence approach to determine that a
product is analytically highly similar to the approved reference product and that there are no clinically meaningful
differences between the biosimilar product and the reference product.2 Likewise, EMA considers a totality-of-theevidence approach when evaluating biosimilars. Under the
provisions of the BPCI Act, FDA licensed the first biosimilar
product, Zarxio (filgrastim-sndz), on March 6, 2015.3-5
Because oncologists prescribe biologic products, they need
to familiarize themselves with biosimilar productshow they
are approved and how they will be incorporated into clinical
practice. This article describes biosimilar products and their
approval process. Aspects on incorporation of the products
into clinical practice may differ by state and are not covered in
this article.
Biosimilar products are not generic drug products. Furthermore, the data package required to support the approval
of a biosimilar differs from a generic drug. Generic drugs are
copies of brand-name drugs. They contain the same active
ingredient as the reference listed drug (e.g., the brand-name
drug) and are the same in terms of dosage form, strength, and
KEY POINTS
e152
Approval in the United States requires a biosimilar

product to be highly similar to the reference product
notwithstanding minor differences in clinically inactive
components and that there are no clinically
meaningful differences between the biological product
and the reference product in terms of safety, purity, and
potency of the product.
Although biosimilar products are not identical to the
reference product, the reference product also may differ
from batch to batch and may change over time. As such,
the biologic product a patient receives in clinical practice
is unlikely to be identical to the reference product
administered to patients in the clinical study or studies
that supported approval.
Comparative clinical studies in biosimilar development
programs should be directed toward an assessment of
clinically meaningful differences, if they exist, and not
directed toward independently establishing the safety
and effectiveness of the biosimilar product.
The BPCI Act allows for the potential, with adequate
scientific justification, to extrapolate data to support a
demonstration of biosimilarity in more than one
indication.
Both the United States and the EMA use a totality-ofthe-evidence approach when assessing whether a
product meets the standards to be designated as
biosimilar.
route of administration. They also must be bioequivalent to

the reference listed drug.6
Unlike small molecule drugs, whose structure usually can be
completely defined and reproduced, biologic products typically
are more complex and difficult to characterize. In addition, all
biologics display a certain degree of microheterogeneity because of the inherent variability of the biologic systems from
which they are derived as well as their manufacturing processes.7 The reference product manufactured by an originator
may differ from batch to batch and may change over time
because of modifications of the manufacturing process after
licensing. As such, the biologic product that a patient receives
in clinical practice is unlikely to be identical to the reference
product administered to patients in the clinical study or studies
that supported approval. However, product quality is tightly
controlled within predefined acceptance ranges, and changes
in the manufacturing processes are evaluated and approved by
the competent regulatory authority to ensure unchanged
clinical performance of the product. Since 1996, FDA has
approved many manufacturing process changes for reference biologic products based on demonstration of product
comparabilitybefore and after the process change(s)
often without the need for additional nonclinical or clinical
data. The EMA also has approved many such changes to
approved biologic products. These process changes can include changes in suppliers of cell culture media, new purification methods, or new manufacturing sites. A 2013 report
from Dr. Schneider from the Danish Health and Medicines
Authority indicated that Remicade (infliximab) has undergone
over 35 manufacturing process changes since approval and
MabThera (rituximab) has undergone at least five process
changes.8
Both the biosimilar and reference product are essentially
different versions of the same active drug substance.19 As
such, scientific principles that underlie the comparability
exercise for manufacturing process changes of the originator
product are also applicable to the analytical similarity exercise
to demonstrate biosimilarity.9 Although the scientific principles are applicable to both exercises, additional steps to
justify similarity are usually necessary in biosimilar development, either analytically or clinically, because of differences in
cell lines and manufacturing processes, among others. Ultimately, the sponsors, including the originators, and regulators
need to understand how any differences in the structure of
the product may affect the function of the product. They also
need to ensure that these differences are evaluated thoroughly to ensure safety and effectiveness of the product.
APPROACHES TO ASSESS ANALYTICAL

SIMILARITY
In the United States, a determination of biosimilarity requires data demonstrating that the biologic product is highly
similar to the reference product, notwithstanding minor
differences in clinically inactive components. The FDA generally expects that the biosimilar product will be composed of
the same amino acid sequence as the reference product.2
Because of the complexity of biologic products, FDA expects

sponsors to characterize both their product and the reference
product. FDA also expects that this characterization will serve
as the foundation for the demonstration of biosimilarity.2
Such characterization generally will include analyses of multiple attributes including, but not limited to, primary structures (e.g., amino acid sequences), higher order structures,
enzymatic post-translational modifications (e.g., glycosylation
patterns), other variants (e.g., deamidation or charge variants), and any other intentional modifications (e.g., addition
of PEGylation).2
Comparative functional assays provide additional data
to support a determination that the biosimilar product
is highly similar to the reference product. For example,
comparative functional assays may include binding assays,
antibody-dependent cell-mediated cytotoxicity assays, and
cell-based bioactivity assays to assess the potential for cytokine release.2 Furthermore, a sponsor should consider other
factors such as product expression systems, manufacturing
processes, impurities, and stability.2
The growing number of analytical tools allows sponsors
and regulatory agencies to better understand how attributes
of a protein or biologic product can affect the function of that
protein. As such, the comparative analytical characterization
in a biosimilar program is quite extensive.
A sponsor can submit analytical data to regulatory
agencies before clinical development. This allows regulatory
authorities to assess residual uncertainty regarding the determination of whether the proposed biosimilar product is
biosimilar to the reference product, and what additional
studies are needed to support the determination that the
product is biosimilar. Factors considered in the assessment of
residual uncertainty may include: (1) which specific attributes
were tested, understanding that it is necessary to assess for
differences in any critical quality attributes; (2) the number of
attributes tested (in a theoretical example, a more extensive
characterization with a fingerprint-like analysis10 could reduce uncertainty); (3) the number of lots tested for both the
proposed biosimilar product and the reference product; and
(4) what differences, if any, were observed between products
and what impact the differences could have on safety and
efficacy.
CLINICAL STUDIES
As stated, the foundation of a biosimilar development
program is data demonstrating that the biosimilar product is
analytically highly similar to the reference product. In addition, the BPCI Act states that an application for a biosimilar
product must include, among other things, information from
animal studies and a clinical study or studies (including the
assessment of immunogenicity and pharmacokinetics or
pharmacodynamics) that are sufficient to demonstrate
safety, purity, and potency in one or more conditions of use
for which the reference product is licensed and for which
licensure is sought for the biosimilar product.1 However,
the BPCI Act provides FDA discretion to determine whether a
particular element is unnecessary in an application to

support a demonstration of biosimilarity.
Comparative clinical study results (e.g., a clinical study
powered to demonstrate similar pharmacokinetic profiles
or a clinical study to demonstrate a lack of clinically meaningful differences in safety and efficacy) cannot be used to
support that a proposed biosimilar product is biosimilar to the
reference product when the proposed biosimilar product has
otherwise not been demonstrated to be highly similar
to the reference product. Highly similar and no clinically
meaningful differences are two separate standards, each of
which needs to be met to support a demonstration of biosimilarity. Clinical data cannot be used to overcome relevant
differences in quality attributes.2,11
In general, FDA expects an application for a biosimilar
product to include data from a clinical pharmacology study
demonstrating pharmacokinetic similarity between the
biosimilar product and the reference product.12 If one or
more relevant pharmacodynamic markers exist, demonstration of pharmacodynamic similarity may reduce residual
uncertainty and permit a more selective and targeted approach to additional clinical studies. Important elements
that FDA considers in the pharmacokinetic and/or pharmacodynamic study include: (1) the appropriate (i.e., sensitive)
population to detect differences, if they exist; (2) the appropriate dose, which may differ from the approved dose and
ought to be on the steep part of the dose response curve; (3)
route of administration; and (4) and the assays used to measure
pharmacokinetic and/or pharmacodynamic markers.12
In some cases, a healthy volunteer population, if ethically
appropriate, may be the most appropriate and sensitive
population rather than a patient population for the pharmacokinetic and/or pharmacodynamic similarity assessment
because tumor burden may increase the variability in pharmacokinetic measurements of certain biologic products. With
justification, the dose and schedule of the biologic products
used in the study could be lower than the approved dose of
the reference product to improve the safety margin and/or
sensitivity of such a study.
The FDA recommends sponsors to pursue a step-wise
approach to biosimilar development. For example, FDA
can review the analytical data and clinical pharmacokinetic
and/or pharmacodynamic data and determine what residual
uncertainly exists before determining whether additional
clinical studies are necessary to address the uncertainty and
support a demonstration of biosimilarity. If sponsors request
FDA advice on the design of clinical studies before FDAs
review of analytical data, which does not follow the FDArecommended step-wise approach, FDA may assume residual
uncertainty exists. The FDA may recommend a more extensive
clinical program than otherwise might have been necessary.
Furthermore, the design of additional studies should be
influenced by the type of residual uncertainty. For example,
as a scientific matter, an adequate assessment of immunogenicity is expected in a biosimilar development program.2
The nature of the data and the design of a study to generate
comparative immunogenicity data may depend on specific
e153
analytical characteristics of the biosimilar compared with the

reference product that could change the risk of an immune
response (e.g., different excipients or other formulation
changes) as well as the publicly known risk of immunogenicity
(and risk of adverse events if immunogenicity occurs) associated with the reference product.
The objective of a comparative clinical study in a biosimilar
development program is to support a demonstration of no
clinically meaningful differences between the biosimilar
product and the reference product. In these comparative
clinical studies, this determination is not made by independently establishing the safety and effectiveness of the
biosimilar product. As such, some endpoints used for the
approval of the reference product in the oncology setting
(e.g., overall survival or progression-free survival) may be less
informative. This is because they are not only influenced by
the performance of the medicinal product but also by factors
such as the general health status of the patient. Therefore,
they are insufficiently sensitive to detect product-specific
differences between the reference product and the biosimilar product, if they exist. In some cases, if scientifically
relevant, pharmacodynamic endpoints, such as objective response rate, may be more appropriate. These endpoints
measure the pharmacological action of the biologic in a less
confounded way and may be more sensitive to any potential differences in the biologic products.
If multiple indications are approved for the FDA-licensed
reference product, and the biosimilar sponsor intends to
seek licensure for more than one of these indications, FDA
may provide advice regarding which indication is more
sensitive in the assessment of clinically meaningful differences. Selection of a sensitive indication may allow a sponsor
to request extrapolation of data to other indications without
conducting additional clinical studies.
In general, FDA recommends that sponsors design comparative clinical studies to demonstrate that the proposed
biosimilar product has neither decreased nor increased activity (e.g., efficacy) or has safety concerns (e.g., toxicity) that
differ from the reference product.2 Unless there is justification for a one-sided test, or other design, such a comparative
clinical study generally uses a two-sided test based on a
prespecified similarity margin.2 The FDA will consider whether
the proposed study duration will provide sufficient exposure
to the proposed biosimilar and the reference product to
assess for clinically meaningful differences, and whether the
study is adequately designed to assess for potential differences in immunogenicity.2
Because FDA uses a totality-of-the-evidence approach to
support a demonstration of biosimilarity, it will consider
whether observed differences in a clinical study between the
proposed biosimilar and the reference product constitute
clinically meaningful differences. A comparative evaluation
of multiple data points in a comparative clinical study (e.g.,
efficacy and cardinal-safety findings), combined with pharmacokinetic similarity and an extensive analytical comparison
demonstrating that the products are highly similar, will provide solid ground to conclude that the biosimilar product will
e154
be safe and effective for the proposed indications. The FDA

understands that because of chance, there may be differences
between the study arms in the incidence of certain adverse
events in a comparative clinical study. Therefore, sponsors
may discuss proposals with FDA to prioritize the collection and
analysis of certain cardinal or key adverse events of interest.
For example, anti-VEGF pathway antibodies increase the
risk of hypertension, epistaxis, proteinuria, and dysponia.
An analysis of these adverse events may be more relevant to
an assessment of no clinically meaningful differences for
anti-VEGF pathway antibodies than analyses of fatigue or
pain in a patient population with advanced cancer.
EXTRAPOLATION OF CLINICAL DATA ACROSS

INDICATIONS
The FDA has essentially allowed extrapolation of data for
approved biologic products without additional clinical studies
for years following the introduction of certain manufacturing
process changes. Many biologic drugs used in oncology are
approved for multiple indications, including nononcology
indications. Based on the BPCI Act, if the proposed product
meets the statutory requirements for licensure as a biosimilar
product based on, among other things, data derived from a
clinical study sufficient to demonstrate safety, purity, and
potency in an appropriate condition of use, the potential
exists for the biosimilar product to be licensed for one or more
additional conditions of use for which the reference product
is licensed.13 A conclusion of biosimilarity in one or more
indications potentially allows for extrapolation of data to
support a conclusion of biosimilarity for other indications.13
With adequate scientific justification, a sponsor would not
need to conduct a clinical study to support licensure of a
biosimilar for each condition of use for which the reference
product is licensed.
Sponsors need to provide scientific justification for such
extrapolation. Justification should include a discussion of
how the mechanism of action of the drug relates to each
condition of use; whether pharmacokinetics and/or pharmacodynamics may differ in different patient populations;
differences in expected toxicities in different patient populations; and any other factor that may affect the safety or
efficacy in the different patient populations.2,13 It is important
to note that differences among indications or conditions of
use do not preclude extrapolation or necessitate additional
clinical data. However, any differences need to be adequately
addressed with data and information to support extrapolation.
Ultimately, sponsors should consider conducting clinical
studies that are adequately sensitive to detect clinically
meaningful differences between the biosimilar product and
the comparator product and that would best reduce any
residual uncertainty regarding extrapolation based on the
factors described above. For example, choosing to conduct
the comparative clinical study in a relevant indication with a
consistent and/or large treatment effect may have greater
sensitivity to detect differences between the products, if
they exist. Also, choosing to conduct a comparative clinical
study in a relevant indication whose mechanism of action

overlaps with other indications being sought could reduce
the residual uncertainty regarding extrapolation.
Extrapolation can reduce the time and cost of development by abrogating the need to conduct clinical studies for
each indication previously approved for the reference product.
Extrapolation is one of the tenets of the abbreviated approval
pathway for biosimilar products. When FDA approves a biosimilar product, it has determined that the product meets
the requirements for approval and has been demonstrated to
have no clinically meaningful differences from the reference
product in terms of safety, purity, and potency, including any
indications that were supported by extrapolation.
In summary, extrapolation can be supported for a product
that is highly similar to the U.S.-licensed reference product
based on the totality of the evidence in which the biosimilar
product would be expected to be safe and effective for each
indication of use, even without conducting a separate clinical
study in each indication. Without extrapolation, for products
with multiple indications, sponsors would likely need to
conduct comparative clinical studies with thousands to tens
of thousands, or more, patients depending on the treatment
effect and the sensitivity of the population(s). Such a requirement could have multiple consequences. These could
include keeping development costs high and rendering the
pathway infeasible and potentially diverting patients and
resources from clinical studies of newer therapies for cancer.
Despite these considerations, however, licensing of safe and
effective drugs is the highest priority for regulators and
extrapolation is only allowed if a sponsor provides adequate
scientific justification.
INTERCHANGEABILITY
The statutory provisions for interchangeability are unique to
the United States. According to the BPCI Act, for FDA to
determine that a product is interchangeable with a reference product, the product must be biosimilar to the reference product and can be expected to produce the same
clinical result in any given patient.14 Additionally, FDA must
determine that for a biological product that is administered
more than once to an individual, the risk in terms of safety or
diminished efficacy of alternating or switching between use
of the proposed interchangeable product and the reference
product is not greater than the risk of using the reference
product without such alternation or switching.14 Similar to
U.S. generic drugs, an interchangeable biologic product may
be substituted for the reference product without the intervention of the health care provider who prescribed the
product.15 According to the BPCI Act, only a product designated as interchangeable may be substituted. Therefore,
the treating health care providerand not the pharmacist
can decide whether to prescribe a biosimilar product not
designated as interchangeable at any time, including deciding
whether to change their patient from one product to another.
Although most biosimilars in Europe are approved centrally by EMA, individual member states make substitution
policies.16 Currently, none of the European Union (EU) countries have authorized unrestricted automatic substitution.
Some countries (e.g., Belgium, Italy, Norway, and the United
Kingdom) prohibit pharmacy-level substitution, whereas other
countries (e.g., France, Germany, and the Netherlands) allow
substitution in certain instances. France permits substitution
of a biosimilar for the prescribed (reference) biologic in patients
who are treatment nave.17 The Dutch Medicines Evaluation
Board accepts substitution of biosimilars under the condition
that both the treating physician and the pharmacist are involved.18 In Germany, the pharmacist may only substitute a
bioidentical product (i.e., a biosimilar from the same
manufacturer).
Importantly, lack of designation as interchangeable does
not imply that a biosimilar product is an inferior product
to a biosimilar that has also been determined to be interchangeable with the reference product. By being biosimilar, the biosimilar product has been determined to be
highly similar and has no clinically meaningful differences
from the reference product for the labeled conditions of use.
Nevertheless, a demonstration of interchangeability might
require additional data (e.g., to evaluate the risk of switching
or alternating) to address the additional standards, and
sponsors need to make a business decision about whether to
pursue a claim of interchangeability.
DISCUSSION
Biosimilar products are an additional treatment option for
health care providers and patients. They may help facilitate
access to treatment and decrease costs for patients and the
health care system in the United States, in the same manner
as these products have done for almost 10 years in Europe.
The FDA licensed the first biosimilar product, Zarxio (filgrastimsndz), on March 6, 2015.3
Health care providers in Europe have already gained substantial experience with safely prescribing biosimilar products. The legal pathway for authorizing biosimilar medicines
was established in 2003, and the first biosimilar product,
somatotropin, was approved in 2006.16,19,20 Approved biosimilar products in Europe include hematopoietic growth
factors, such as filgrastim and epoetin, as well as insulin,
somatotropin, follitropin, and the monoclonal antibody
infliximab.9,20 The uptake of biosimilars in the EU has increased steadily but at different rates in different countries.
Germany has shown the highest uptake of approximately
50% volume.21 In certain European countries, adoption of
such products has been widespread and thousands of patients have been safely treated with biosimilar drugs.16,22-26
Concerns about safety and efficacy have been voiced for
biosimilar filgrastim and epoetins, especially in extrapolated
indications for which no clinical data with the biosimilar have
been generated.9 Epoetins were among the first biosimilars
approved in the EU. The major concern with epoetins is the
rare development of neutralizing, cross-reacting anti-epoetin
antibodies that can cause pure red cell aplasia (PRCA).
Product-related factors, such as aggregation and contaminants
e155
(e.g., leachables from rubber stoppers or tungsten from syringes), have been implicated in the development of antiepoetin antibodies.27,28
Although regulators and sponsors now better understand
the factors that contribute to immunogenicity, as a scientific
matter, both FDA and EMA expect an adequate assessment
of immunogenicity as part of biosimilar development programs. The EMA guideline on the development of biosimilar
epoetins requests clinical immunogenicity data with subcutaneous use in patients with renal anemia and, therefore,
with the route of administration and in the patient population
at highest risk for developing PRCA.29 As such, immunogenicity
data from these patients can then be extrapolated to lower-risk
clinical settings, such as intravenous use or to immunocompromised patients who have chemotherapy-induced anemia.
The immunogenicity assessment of approved biosimilar
epoetins in Europe includes both prelicensing studies and,
because of the rarity of PRCA, extensive postmarketing
surveillance programs to gain a more precise estimate of its
frequency. Reassuringly, postmarketing studies and pharmacovigilance surveillance in Europe confirm efficacy and
safety of licensed biosimilars, lending further support to the
adequacy of the respective scientific guidelines and approval
processes.22-25
Biosimilar products are not generics. Requirements for
approval of a biosimilar product differ in many ways from
generic drug products. Analytical testing for biosimilar drug
products is much more extensive and encompasses numerous
structural and functional assays using state-of-the art science.

In the United States, such comparative analytical testing must
show that the biosimilar product is analytically highly similar
to the reference product. Although the specific language of
the regulations differs in Europe, the same scientific principles
apply with regard to the requirements for demonstrating
analytical similarity.
In addition to analytical testing, biosimilar applications in
the United States generally will contain comparative data
from one or more clinical studies showing pharmacokinetic
similarity and, if necessary, an additional comparative clinical
study or studies to support a demonstration of no clinically
meaningful differences. Data in humans further serve to
reduce the residual uncertainty on the clinical impact arising
from any observed analytical differences and support a
demonstration of biosimilarity. Ultimately, FDA and EMA
determinations that a product is biosimilar based on a totalityof-the-evidence approach to ensure that these products are
safe and effective for the labeled indications, including for the
treatment of patients with cancer. Such access will help to
facilitate treatment of patients with cancer in both the United
States and Europe.
ACKNOWLEDGMENT
The authors would like to thank Leah Christl, PhD, Associate
Director for Therapeutic Biologics, FDA, for providing
comments and edits to the manuscript.
References
1. Patient Protection and Affordable Care Act, Pub. L No. 111-148, Sections
7002(a)(2) and (b)(3); Public Health Service Act, Pub. L No. 78-410,
Sections 351(k), 351(i)(2), and 351(i)(4).
2. U.S. Food and Drug Administration. Scientific Considerations in
Demonstrating Biosimilarity to a Reference Product: Guidance for Industry. http://www.fda.gov/downloads/DrugsGuidanceCompliance
RegulatoryInformation/Guidances/UCM291128.pdf. Accessed March 17,
2016.
3. U.S. Food and Drug Administration. BLA Approval Letter. http://www.
accessdata.fda.gov/drugsatfda_docs/appletter/2015/
125553Orig1s000ltr.pdf. Accessed January 7, 2016.
4. U.S. Food and Drug Administration. FDA Approves First Biosimilar Product
Zarxio. http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/
ucm436648.htm. Accessed January 29, 2016.
5. Christl LD, Deisseroth A. Development and Approval of Biosimilar
Products. The ASCO Post. March 25, 2015;6.
6. U.S. Food and Drug Administration. Abbreviated New Drug Application (ANDA): Generics. http://www.fda.gov/Drugs/DevelopmentApproval
Process/HowDrugsareDevelopedandApproved/ApprovalApplications/
AbbreviatedNewDrugApplicationANDAGenerics/. Accessed January 8, 2016.
7. Weise M, Bielsky MC, De Smet K, et al. Biosimilars: what clinicians
should know. Blood. 2012;120:5111-5117.
8. Schneider CK. Biosimilars in rheumatology: the wind of change. Ann
Rheum Dis. 2013;72:315-318.
9. Weise M, Kurki P, Wolff-Holz E, et al. Biosimilars: the science of extrapolation. Blood. 2014;124:3191-3196.
e156
10. U.S. Food and Drug Administration. Quality Considerations in Demonstrating Biosimilarity of a Therapeutic Protein Product to a Reference
Product: Guidance for Industry. http://www.fda.gov/downloads/
drugs/guidancecomplianceregulatoryinformation/guidances/
ucm291134.pdf. Accessed March 17, 2016.
11. European Medicines Agency. Guideline on Similar Biological Medicinal
Products. http://www.ema.europa.eu/docs/en_GB/document_library/
Scientific_guideline/2014/10/WC500176768.pdf. Accessed January 29,
2016.
12. U.S. Food and Drug Administration. Guidance for Industry: Clinical
Pharmacology Data to Support a Demonstration of Biosimilarity to a
Reference Product, Draft Guidance. http://www.fda.gov/downloads/
drugs/guidancecomplianceregulatoryinformation/guidances/
ucm397017.pdf.
13. U.S. Food and Drug Administration. Guidance for Industry on Biosimilars: Q & As Regarding Implementation of the BPCI Act of 2009:
Questions and Answers Part I. http://www.fda.gov/Drugs/Guidance
ComplianceRegulatoryInformation/Guidances/ucm259809.htm.
14. U.S. Food and Drug Administration. Information for Industry (Biosimilars). http://www.fda.gov/Drugs/DevelopmentApprovalProcess/
HowDrugsareDevelopedandApproved/ApprovalApplications/Therapeutic
BiologicApplications/Biosimilars/ucm241720.htm. Accessed January 8,
2016.
15. U.S. Food and Drug Administration. Information for Consumers (Biosimilars). http://www.fda.gov/Drugs/DevelopmentApprovalProcess/
16.
17.
18.
19.
20.
21.
22.
HowDrugsareDevelopedandApproved/ApprovalApplications/Therapeutic
BiologicApplications/Biosimilars/ucm241718.htm. Accessed January 8,
2016.
Renwick MJ SK, Smolina K, Gladstone EJ, et al. Postmarket policy
considerations for biosimilar oncology drugs. Lancet Oncol. 2016;17:
e31-e38.
Generics and Biosimilars Initiative. France to Allow Biosimilars Substitution. http://gabionline.net/Policies-Legislation/France-to-allowbiosimilars-substitution. Accessed January 29, 2016.
Medicines Evaluation Board. MEB Stance on Prescribing Biosimilars. http://
english.cbg-meb.nl/latest/news/2015/03/31/meb-stance-on-prescribing-%
E2%80%9Cbiosimilars%E2%80%9D. Accessed January 29, 2016.
European Medicines Agency. Questions and Answers on Biosimilar
Medicines (Similar Biological Medicinal Products). http://www.ema.
europa.eu/docs/en_GB/document_library/Medicine_QA/2009/12/
WC500020062.pdf. Accessed January 8, 2016.
European Medicines Agency. European Public Assessment Reports.
http://www.ema.europa.eu/ema/index.jsp?curl=pages%2Fmedicines%2
Flanding%2Fepar_search.jsp&mid=WC0b01ac058001d124&searchTab=
searchByAuthType&alreadyLoaded=true&isNewQuery=true&status=
Authorised&keyword=Enter+keywords&searchType=name&taxonomy
Path=&treeNumber=&searchGenericType=biosimilars&generics
KeywordSearch=Submit. Accessed January 8, 2016.
Generics and Biosimilars Initiative. Use of Biosimilars in Europe Differs
Across Countries. http://gabi-journal.net/news/use-of-biosimilars-ineurope-differs-across-countries. Accessed January 29, 2016.
Michallet M, Luporsi E, Soubeyran P, et al; ORHEO study group. BiOsimilaRs in the management of anaemia secondary to chemotherapy in
23.
24.
25.
26.
27.
28.
29.
HaEmatology and Oncology: results of the ORHEO observational study.

BMC Cancer. 2014;14:503.
Dellanna F, Fluck RJ, Lonnemann G, et al. Results from a safety cohort of
patients with renal anemia receiving the biosimilar epoetinzeta: The
PASCO I study. Clin Nephrol. 2015;84:280-288.
Horbrand F, Bramlage P, Fischaleck J, et al. A population-based study
comparing biosimilar versus originator erythropoiesis-stimulating agent
consumption in 6,117 patients with renal anaemia. Eur J Clin Pharmacol.
2013;69:929-936.
Schmitt M, Publicover A, Orchard KH, et al. Biosimilar G-CSF based
mobilization of peripheral blood hematopoietic stem cells for autologous
and allogeneic stem cell transplantation. Theranostics. 2014;4:280-289.
Gascon
P, Aapro M, Ludwig H, et al. Treatment patterns and outcomes
in the prophylaxis of chemotherapy-induced (febrile) neutropenia with
biosimilar filgrastim (the MONITOR-GCSF study). Support Care Cancer.
2016;24:911-925.
Boven K, Stryker S, Knight J, et al. The increased incidence of pure red
cell aplasia with an Eprex formulation in uncoated rubber stopper
syringes. Kidney Int. 2005;67:2346-2353.
Seidl A, Hainzl O, Richter M, et al. Tungsten-induced denaturation and
aggregation of epoetin alfa during primary packaging as a cause of
immunogenicity. Pharm Res. 2012;29:1454-1467.
European Medicines Agency. Committee for Medicinal Products for
Human Use: Guideline on non-clinical and clinical development of
similar biological medicinal products containing recombinant erythropoietins (revision). http://www.ema.europa.eu/docs/en_GB/document_
library/Scientific_guideline/2010/04/WC500089474.pdf. Accessed January
29, 2016.
e157

Clinical Trial Eligibility Criteria:

Tradition Versus Reality
CHAIR
Edward S. Kim, MD, FACP
Levine Cancer Institute, Carolinas HealthCare System
Charlotte, NC
SPEAKERS
Gwynn Ison, MD
College Park, MD
Jeffrey R. Infante, MD
Nashville, TN
PRACTICAL VERSUS TRADITIONAL ELIGIBILITY CRITERIA
Transforming Clinical Trial Eligibility Criteria to Reflect

Practical Clinical Application
Edward S. Kim, MD, FACP, Jennifer Atlas, MD, Gwynn Ison, MD, and Jennifer L. Ersek, MSPH
OVERVIEW
Historically, oncology clinical trials have focused on comparing a new drugs efficacy to the standard of care. However, as our
understanding of molecular pathways in oncology has evolved, so has our ability to predict how patients will respond to a
particular drug, and thus comparison with a standard therapy has become less important. Biomarkers and corresponding
diagnostic testing are becoming more and more important to drug development but also limit the type of patient who may
benefit from the therapy. Newer clinical trial designs have been developed to assess clinically meaningful endpoints in
biomarker-enriched populations, and the number of modern, molecularly driven clinical trials are steadily increasing. At the
same time, barriers to clinical trial enrollment have also grown. Many barriers contribute to nonenrollment in clinical trials,
including patient, physician, institution, protocol, and regulatory barriers. At the protocol level, eligibility criteria have
become a large roadblock to clinical trial accrual. Over time, eligibility criteria have become more and more restrictive. To
accrue an adequate number of patients to molecularly driven trials, we should consider eligibility criteria carefully and
attempt to reduce restrictive criteria. Reducing restrictive eligibility criteria will allow more patients to be eligible for clinical
trial participation, will likely increase the speed of drug approvals, and will result in clinical trial results that more accurately
reflect treatment of the population in the clinical setting.
linical trials and their availability for treatment of patients serve as a longstanding foundation in the practice of medicine and drug development. Through all phases
of study (I-IV), the risk-benefit ratio has always encompassed
assessing safety as well as offering opportunity for treatment
benefit. In the past, the primary focus of early-phase studies
was defining safe doses, and late-phase studies served as the
final hurdle to regulatory approval. Initially, with only limited treatments available, therapies were tested against best
supportive care. As more therapies were approved, combination therapies were compared with historical standards,
leading to the desire for homogenous patient populations.
Although homogenous populations allow for direct comparison of study results, they also reduce the number of patients
eligible to participate in the study. From the perspective of
study investigators and sponsors, there may be an interest in
keeping study populations tight and homogenous to allow for
comparison with standard therapies; however, increasingly
strict eligibility criteria lead to increased difficulty in accrual.
With modern clinical trials incorporating biomarkers and
targeted therapies, there is less need to compare the results
of these trials to standard therapies, and thus trial eligibility
could be relaxed. Clinical models and structures that support

therapeutic development are not keeping pace with scientific
innovations.1 We must carefully evaluate the evolution of
clinical trial design and the corresponding eligibility criteria
based on molecular targets and drugs and, in turn, implement
these into the conduct of future clinical trials to successfully
incorporate the use of novel therapeutics into oncologic care
with results that are generalizable to the greater population.
IMPACT OF BIOMARKERS AND SURROGATE

ENDPOINTS ON CLINICAL TRIALS
Molecular medicine is quickly changing the landscape of
medical oncology. The era of molecular medicine is leading to drug approval at an unprecedented rate in oncology,
with more than 40 drug approvals based on specific tumor
biomarkers in the past decade.2 These drugs add to the
growing armamentarium of treatment options available to
patients with cancer. However, it is becoming apparent
that phenotypic characteristics are limited in their ability to
predict response. More recently, biomarkers are included as
critical components in modern clinical trials involving targeted agents, shifting the priority to identifying targets
From the Department of Medicine, Hematology and Oncology, Wake Forest School of Medicine, Winston-Salem, NC; Office of Hematology and Oncology Products, Center for Drug
Evaluation and Research, U.S. Food and Drug Administration, Silver Spring, MD; Levine Cancer Institute, Carolinas HealthCare System, Charlotte, NC.
Corresponding author: Edward S. Kim, MD, FACP, Solid Tumor Oncology and Investigational Therapeutics, Levine Cancer Institute, Carolinas HealthCare Stystem, 1021 Morehead
Medical Dr., Suite 3100, Charlotte, NC 28204; email: edward.kim@carolinashealthcare.org.
83
KIM ET AL
for therapeutic development alongside drug development.

Consideration also must be given to appropriate choice of
primary endpoints. Although overall survival remains the
gold standard, progression-free survival, tumor response,
and duration of tumor response may serve as surrogate
endpoints in early-phase studies and may allow for early
approval. Interpretation of emerging data from trials is
hindered by a lack of specific reporting standards for biomarkers.3 Stronger evidence-based outcomes incorporating
biomarkers could be achieved by implementing standardized biomarker-reporting systems with molecular selection
criteria guided by the mutation gene at the variant level.3
As an example, listing specific variant mutations (e.g., del19,
T790) instead of listing a mutation generically as an EGFR
mutation would be more descriptive.
GOAL OF CLINICAL TRIALS

Clinical trials are fundamental to understanding the risks and
benefits of treatment in a specific intended population.
Clinical trials provide a formalized process of integrating
scientific findings into clinical practice in a safe and systematic manner. With increasing response rates noted in many
molecularly driven trials, small incremental gains may no
longer be adequate for drug approval. There needs to be a
focus on utilizing molecular medicine to achieve clinically
meaningful endpoints.
From a regulatory perspective, the U.S. Food and Drug
Administrations (FDA) primary objective in evaluating trial
eligibility criteria, from a first-in-human trial to a trial intended to support a marketing application, is the safety of
the patient.4 Eligibility criteria that encourage clinical trial
accrual, permit patient access to investigational drugs, and
create clinical trial results that are generalizable to the
broader U.S. population are directly in line with FDAs
mission.
TRADITIONAL CLINICAL TRIAL ELIGIBILITY

CRITERIA AND DESIGN
Traditional clinical trials have been categorized on the basis
of phases of drug development. In phase I clinical trials, a
small number of patients are enrolled, and the focus is on
safety. The goal of phase I clinical trials is primarily to establish the maximum tolerated dose that would have
KEY POINTS
84
Discuss the impact of molecular medicine and

biomarkers in oncology clinical trials.
Review traditional versus modern clinical trial design.
Review of barriers to protocol enrollment.
Discuss issues related to traditional versus modern
clinical trial eligibility.
Discuss regulatory issues related to clinical trial
eligibility.
acceptable toxicity if the drug is found to have clinical efficacy. The most important endpoint traditionally in phase I
trials has been identifying dose-limiting toxicity (DLT). Historically, phase I trials have not restricted patient eligibility
on the basis of diagnosis or molecular phenotype; however,
this is changing.
From a regulatory standpoint, many deficiencies sent to
sponsors of investigational new drug applications (INDs) do
not pertain to the eligibility criteria, provided that the firstin-human study is conducted in an appropriately advanced
cancer patient population, but pertain to the definition of
DLTs. The stringency applied to the DLT definition stems
from the interpretability of the study results, as potential
signals of toxicity should be captured as early in development as possible. Unfortunately, the inflexible nature of the
commonly used 3+3 design of first-in-human trials may lead
to the premature discontinuation of dose escalation because
of toxicities that are either unrelated to the drug or are not
dose-dependent. To minimize the potential for premature
discontinuation of dose escalation, strict eligibility criteria
regarding organ function, life expectancy, comorbidities,
performance status, HIV status, and laboratory parameters
are instituted.
In phase II clinical trials, the primary objective is to determine whether the therapy is effective. The primary
endpoints in these studies have historically been tumor
response and survival. Because of inconsistent factors, including eligibility criteria and protocol design, phase II trials
can be difficult to interpret in comparison with one another.
Historically, phase III clinical trials have evaluated an experimental therapy in comparison with standard-of-care
therapy. Randomized clinical trials remain the gold standard. Following trials in these phases, sponsors present data
from their studies and file applications for approval. Federal
regulations govern aspects of IND applications and New
Drug Applications and Biologic License Applications. 5,6
Although there are no specific regulations regarding eligibility criteria, deciding whom to include or exclude from a
trial may determine the difference between a successful trial
and a failed one.
Phase IV clinical trials, often referred to as postmarketing
surveillance trials, are completed after a drug has received
approval. Common reasons for conducting these trials include monitoring a drugs long-term efficacy, observing
the long-term toxicities and impact on quality of life, and
comparisons between a novel drug and other therapies that
are available on the market. The strict criteria developed
in the early-phase setting are not relaxed for later-phase
studies, as there is a legitimate interest to avoid confounding
of the interpretation of the results of trials by factors related
to the disease condition, rather than the therapeutic product.
The unintended consequence of this approach is that the
results of these studies may not be applicable to the patients
who eventually will receive the drug in the postmarketing
setting. Although postmarketing studies may capture some
of this information, they are not systematically performed, resulting in off-label use without appropriate safety
information. Studying subgroups with broader eligibility

earlier in development may provide information that is more
useful to prescribers and patients in the postapproval setting, where comorbidities are more frequent, and may aid in
making the benefit-risk calculation for a new agent with
more relevant data.
MODERN/PRACTICAL CLINICAL TRIALS

Modern trial designs, such as adaptive, umbrella, and basket
trials, have allowed the FDA to develop and implement
accelerated approval programs in some cases, based on
surrogate endpoints. A growing number of targeted therapies have been adopted into the FDA programs that assist
with the accelerated development and approval of promising drugs (Fast Track designation, Breakthrough Therapy
designation, and Priority Review designation).7 The FDA
granted accelerated approval to 35 oncology products for
47 new indications from December 11, 1992, to July 1, 2010,
with clinical benefit confirmed in postmarketing trials for
26 of the 47 new indications, with subsequent conversion to
regular approval.8 The FDA has given breakthrough therapy
designation and accelerated approval to several targeted
drugs to be used in cancer therapy in an effort to speed
up the availability of promising drugs based on surrogate
endpoints felt to be reasonably likely to predict clinical
benefit in diseases with unmet medical needs or drugs that
have demonstrated a substantial improvement over current
available therapy. The median time between accelerated
approval and regular approval of oncology products was
3.9 years (range, 0.8-12.6 years), and the mean time was
4.7 years.8 Reduction in approval time clearly shows the
benefit of these FDA accelerated approval programs, with
the implication that appropriate postmarketing trials should
be completed to confirm efficacy results. Examples of targeted therapies developed in conjunction with biomarkers
that received approval under expedited programs include
combination trametinib and dabrafenib for unresectable
or metastatic melanoma with BRAF V600E or V600K mutations,9 osimertinib for metastatic EGFR T790M mutationpositive nonsmall cell lung cancer (NSCLC),10 and erlotinib
and gefitinib for NSCLC with EGFR exon 19 deletions or
exon 21 (L858R) substitution mutations.11,12 There continues
to be a growing interest in developing tumor biomarkers to
assess the effectiveness of therapy, especially in relation to
targeted therapy, and in using surrogate endpoints as a basis
for accelerated approval by the FDA. We must continue to
strive to overcome the design and regulatory barriers that
exist so as not to impede approval of promising drugs.
Adaptive Trials
In 2004, the FDA published a critical path initiative document highlighting the need for new, innovative clinical trial
designs as well as integration of biomarkers.13 Adaptive
designs, as defined by Vladimir Dragalin,14 are a multistage
study design that uses accumulating data to decide how
to modify aspects of the study without undermining the
validity and integrity of the trial. Adaptive clinical trial

designs are becoming more and more popular, as statisticians can preplan adjustments to the study design on the
basis of data from the study itself. Multiple variations of
the adaptive design exist, and combinations can be used
simultaneously (multiple adaptive design). In the phase I
setting, adaptive dose-finding designs are used to establish
minimum and maximum dose levels and may be used in
conjunction with the group sequential design. The group
sequential design allows preplanned examination of the
study data to determine if the trial can be stopped early
because of efficacy or futility. In clinical trials involving the
use of biomarkers, biomarker-adaptive designs are used to
allow for changes to the study design on the basis of biomarker response.15,16
Phase I trials have increasingly been restricted to specific
tumor types known to harbor molecular targets or to tumors that demonstrate the presence of a specific molecular
target.17 The use of contemporary designs in dose escalation studies that allow for more flexibility with regard to
dose selection could permit the broadening of the eligibility criteria. An adaptive design that allows for dose escalation, dose de-escalation, and dose re-escalation may
filter out potential signals of toxicity stemming from the drug
product, as opposed to the disease process or underlying
comorbidities.
Umbrella Trials
Umbrella trials involve the testing of a variety of drugs targeting different mutations in a single cancer subtype. They
allow a large number of patients to be screened for multiple
biomarkers, which is particularly beneficial for low-prevalence
markers. These trials involve a group of two or more enrichment designs, connected through a central infrastructure
that oversees screening and identification of patients. The
rationale for the umbrella trial design is to facilitate screening
and accrual to trials, with the goal of identifying a patient
population with the best chance for improved outcomes
with therapy. The BATTLE trial (Biomarker-Based Approaches
of Targeted Therapy for Lung Cancer Elimination) in 2005 is
an example of an umbrella trial. In the study, patients with
previously diagnosed and treated lung cancer underwent
repeat biopsies to assess biomarkers, which were used to
select treatment recommendations with one of several biologic agents.16 Other examples of umbrella trials include
ALCHEMIST, Lung-MAP, and I-SPY 2.18-20
Basket Trials
Basket trials are genotype-focused designs involving the
testing of a single drug on a specific mutation or mutations
in a variety of cancer types. Therefore, these trials have the
ability to include rare cancers that would be difficult to
study in randomized controlled trials.21 The basket trial
approach allows flexibility to continually open and close
treatment arms, which makes it possible to screen the
impact of many drugs over several different tumors under
85
KIM ET AL
one study. Patients are matched to the drug based solely on

the genetic abnormality, not on the type of cancer. One
example of a basket trial is vemurafenib in V600-mutated
nonmelanomas.22 In this trial, patients were enrolled into
prespecified cohorts of multiple tumor types. Cohorts were
closed or combined for tumor types with low accrual. Response rate and progression-free survival were evaluated,
and the authors concluded that vemurafenib is active in
some, but not all, tumor types. The National Cancer Institutes (NCI) MATCH is a basket/master trial, meaning that
new drugs can be added to the trial at any time. In 2015 the
NCI launched the NCI-MATCH trial (Molecular Analysis for
Therapy Choice), with plans to screen up to 3,000 patients,
with an enrollment of at least 1,000 individuals to a targeted
drug combination and independent of tumor histology in
patients with advanced solid tumors (gastrointestinal stromal tumor, NSCLC, breast, gastric, melanoma, and thyroid)
and lymphomas.23 The primary objective of this trial is to
determine if the efficacy of targeted therapy is better than
that of standard therapy. Another example of the basket trial
approach includes the NCI M-PACT trial (Molecular ProfilingBased Assignment of Cancer Therapy), which will randomly
assign patients with a known mutation in a specific genetic
pathway to either pathway-driven targeted therapies or a
treatment not known to be pathway-specific.24 These types
of trials will help answer questions about what types of
patients, tumors, or aberrations have the best outcomes
with molecularly targeted therapy compared with standard
chemotherapy.25
Registry Trials
Clinical studies are investigational in nature, whereas the
focus of registry studies is on observation. Registry studies
allow observational conclusions to be drawn from the
evaluation of multiple drugs directed at multiple biologic
targets in several cancers. As an example, the American
Society of Clinical Oncologys (ASCOs) TAPUR (Targeted
Agent and Profiling Utilization Registry Study) will enroll
patients with advanced solid tumors, B-cell non-Hodgkin
lymphoma, and multiple myeloma from three health care
systems, including the Carolinas HealthCare Systems Levine
Cancer Institute, with the goal of matching tumor profiles
with available agents. A molecular tumor board will review
the proposed drug-target match and report to the physician
on potential treatments either on or off the study. Safety and
efficacy outcomes will be recorded; however, the primary
endpoint is objective response rate.26
BARRIERS TO CLINICAL TRIAL ENROLLMENT

Many barriers to clinical trial enrollment exist and are
noted in the literature. These can be broken into several
broad categories: patient-level barriers (including sociodemographic and disease-related barriers), physician-level
barriers, institutional-level barriers, and regulatory and
protocol-level barriers, including restrictive eligibility
criteria.
86
There are several obstacles to trial enrollment associated

with patient characteristics. Because of the more extensive
need for tissue and blood samples in molecularly driven
studies, some patients decline clinical trial enrollment because
of the fear that additional testing will delay the initiation of
critical antineoplastic therapy or out of fear of undergoing
additional procedures. Socioeconomic issues, including distance from centers with trials and frequency of follow-up
visits, may influence the populations enrolled on protocols.
Patients who lack insurance may present with more advanced
stages of cancer and symptomatic disease, precluding trial
enrollment. Because of socioeconomic differences, there are
subsets of patients with poor enrollment on clinical trials due
to lack of understanding of the rationale for the clinical studies
and/or mistrust of physicians. Christian et al27 reported that
although the enrollment of minority patients on clinical trials in
the United States has remained relatively stable, because the
overall number of individuals enrolled in trials has increased
over time, the minority percentage has therefore decreased. A
decreased number of minority patients represented in clinical
trials could lead to potentially important biologic differences
being overlooked.
Physician-driven barriers to protocol enrollment exist as
well. These include the complexity of protocol design and
eligibility criteria, lack of awareness of an available protocol,
absence of an available protocol for an individual patient, or
lack of consideration for clinical trial enrollment of patients
with decreased performance status. Sohal et al28 reported
that the lack of clinical trial access and a decline in functional
status or death were the most commonly cited reasons
patients were not enrolled on molecularly driven protocols
by their physicians. The inability of physicians to keep up to
date with the research literature of molecular compounds
and markers is overwhelming and will limit discussions with
patients about clinical trials.
Additionally, there are funding and regulatory barriers to
clinical trial enrollment that directly affect the protocollevel barriers. With the evolution of trial design and safety
regulations, there is a growing number of mandated requirements in an effort to standardize protocols. Typically,
criteria mandated for exclusion of certain patients perceived
as being potentially at higher risk to receive investigational
therapy. These patients at higher risk may be defined as
having a prior history of malignancy, active brain metastases, suboptimal hepatic or renal function, or HIV positivity.
However, as the field becomes more comfortable with early
integration of available interdisciplinary data during the
course of drug development,1 the FDAs Office of Hematology
and Oncology Products would encourage sponsors to broaden
these criteria, allowing for the best examination of real world
experience as well as providing patients with life-threatening
diseases access to investigational agents. From a funding
perspective, increasing complexity of trials and eligibility
criteria has led to increased costs to conduct such studies.
With decreasing funding available, studies must be carefully
designed, as meaningful clinical benefits will likely be examined more closely in the future.
IMPORTANCE OF CLINICAL TRIAL ELIGIBILITY

Although eligibility criteria are a necessary aspect of clinical
protocols, their purpose is to define the characteristics of
the intended patient population to receive the study drug.
However, as trial designs adapt to incorporate molecular
medicine, so must eligibility criteria. From a design perspective, restrictions on eligibility are effective only if the
restrictions lead to increased statistical power of the study.29
In clinical practice, eligibility criteria employed on clinical
trials are rarely adhered to.
Traditionally, we have relied on the decision-making capacity of the individual physician, with consent of the patient, to determine eligibility for trial enrollment; however,
the paradigm has shifted over time so that all aspects of
the trial, including determination of patient eligibility, dose
modifications, and allowed supportive measures, are explicitly included in the protocol.29 Reevaluation of clinical
trial eligibility criteria, especially those related to the scientific objective and generalizability of the results to
other populations, throughout the entire drug development
process should be considered, while keeping the safety of
the patient at the forefront.
RESTRICTIVE ELIGIBILITY CRITERIA: A

ROADBLOCK TO TRIAL ENROLLMENT
Although eligibility criteria remain an essential component
of clinical trials, the growing number of eligibility criteria and
complexity of trial designs has increasingly become a barrier
to trial enrollment. Eligibility criteria have generally been
categorized as inclusion (e.g., must have a specific molecular
aberration) versus exclusion criteria (e.g., evidence of brain
metastases). We assessed barriers to enrollment on trials,
including patient-related barriers, physician-driven barriers,
and funding and regulatory barriers.17 Protocol eligibility
criteria represent a modifiable barrier to increasing clinical
trial availability to a larger subset of patients. Table 1 lists a
comparison of traditional versus potential modern clinical
trial eligibility criteria.
Certain diagnostic restrictions are required to ensure that
patients enrolled on clinical protocols involving targeted
drugs would have a reasonable chance of clinical benefit.
Protocol barriers, such as tissue availability and specific
laboratory testing, are increasingly frequent requirements of
trials involving molecularly targeted agents. George29 described the types of problems restrictive eligibility criteria
create in phase III clinical trials in cancer with regard to
scientific interpretation, medical applicability, complexity,
costs, and patient accrual. In molecularly driven trials, eligibility criteria related to patient characteristics, including
restrictions on age, performance status, laboratory findings
addressing organ function, and history of previous cancer
diagnosis, previous antineoplastic therapy, or evidence of
metastatic disease to the brain, have become cumbersome.
Many restrictive criteria could be reevaluated when
considering the effects of traditional cytotoxic therapy
versus modern targeted therapy because these drug side-
effect profiles are quite different. As more clinical information about the toxicity profile of a particular drug is
obtained, the eligibility criteria for subsequent trials may be
tailored to account for this information. For example, if it is
observed that an agent causes hepatic toxicity during early
development, then the eligibility criteria for the subsequent
trials may be amended either to include more intensive
evaluation of hepatic injury and function and/or to exclude
patients with a certain degree of baseline hepatic impairment. Dedicated studies in patients with hepatic dysfunction would also be informative, especially in diseases that
have a propensity to metastasize to the liver. However,
other criteria, such as a minimum age requirement or comorbidities (e.g., underlying HIV positivity), could be relaxed or
amended to allow for less stringent requirements as the development progresses into later phases.17 Safety concerns are a
commonly cited cause for restrictive eligibility criteria. If a drug
has a toxicity profile known to be harmful in a particular subset
of patients, then these patients should be excluded. One example of this would be giving bevacizumab to patients with lung
cancer with squamous cell carcinoma histology. Ideally, safety
criteria should be limited to those in which no medical judgment
is allowed. Increased complexity of protocol design and instructions on dose modifications do not replace a clinicians
judgment concerning an individual patients suitability for
clinical trial enrollment.
Patients with brain metastases are a patient subset
typically excluded from clinical trials. Over time, the
number of patients with brain metastases are likely to
increase in frequency, based on the natural history of
many tumor types and the sites of frequent brain metastases. In addition, as adjuvant chemotherapy continues
to improve outcomes, it is likely that we will continue to
see more disease relapses in the brain given the limited
number of antineoplastic therapies that can penetrate the
blood-brain barrier. Carden et al30 advocate that we
should rethink exclusion of patients with asymptomatic
brain metastases from clinical trials. Because of the development of more-sensitive brain-imaging techniques,
metastatic brain lesions are often detected earlier and
may be asymptomatic at the time of detection. In many
trials, brain imaging is a part of the screening workup for
trial eligibility. Traditionally, patients with brain metastases have been excluded from trial participation because
of the shortened life expectancy associated with symptomatic brain metastases and a concern that the patient
would not receive a sufficient duration of the study drug;
therefore, clinical benefit would not be expected and may
confound study results. Although safety is a priority,
universally precluding patients with brain metastases
from clinical trials limits their access to potentially effective therapy, making it difficult to extrapolate the use
of these drugs to the general population.
The more restrictive the eligibility criteria, the more
limited the heterogeneity of the study population, limiting
the ability to generalize trial results to the population who
will be treated off-study. We must be concerned that the
87
KIM ET AL
TABLE 1. Comparison of Traditional Versus Potential Modern Clinical Trial Eligibility Criteria
Traditional
Modern
Trial Methodology
Preclinical Data
Less requirements, usually some diagnostic restrictions
More diagnostic restrictions; usually require biopsy/tissue and

specific laboratory testing to identify actionable target
Clinical Trial Design Dose escalation

(Early Phase)
Dose escalation with expansion cohorts
Clinical Trial Design Randomized registries

(Late Phase)
Randomized in biomarker enriched population
Clinical Trial
Endpoints
(Early Phase)
Maximum tolerated dose
Response rate; however, FDA policies in place for Breakthrough

Therapies and Accelerated Approval. Greater use of surrogate
endpoints.
Clinical Trial
Endpoints
(Late Phase)
OS, PFS
OS
Target Population
Unselected
Molecularly targeted subset
Age
Typically restricted
Typically restricted, but could advocate for broadening age restrictions given different side-effect profiles of these drugs
Performance
Status
Typically limited to ECOG PS 0-1
Could include EGOG PS 2
Life Expectancy
Typically, patients must be expected to live several months
Could possibly be relaxed given different side-effect profiles
Organ Function/
Laboratory
Abnormalities
Required; some patients may be excluded because of abnormal Required; some patients may be excluded because of abnormal
laboratory values
laboratory values. May require reevaluation as a result of
different side-effect profiles.
Medical History
Required; some conditions are excluded
Required; some conditions are excluded
Previous Therapy
May exclude from trials looking at front-line therapy
May exclude from trials looking at front-line therapy
Medications/
Steroid Use
Less likely to exclude
Increased concern for drug-drug interactions
Comorbidities
More stringent exclusion based on comorbidities (e.g., cardiac

comorbidities)
Some comorbidities traditionally restricting enrollment could be

relaxed
QOL/Depression
Sometimes required that patients self-reported QOL/depression Could be relaxed; including patients with low self-reported QOL
scores are high enough for inclusion
could be important for trials where efficacy is similar, but one
drug is better tolerated
Patient Characteristics
Disease Characteristics
Confirmed
Required
Diagnosis/Tumor
Histology
Somewhat important, but druggable target is main driver
Target Lesion
Required
Required
Nontarget Lesion
Usually not required
May be required
Brain Metastases
Traditionally excluded
Traditionally excluded; possibly needs to be modified, especially

for patients with asymptomatic brain metastases
Metastatic Disease Usually allowable

Other Than Brain
Usually allowable
Prior/Concurrent
Malignancy
Usually excluded
Usually excluded
Prior Progression
Sometimes restricted in traditional trials
Could be relaxed in trials of molecularly targeted drug
Prior Clinical
Response
Usually allowable
Usually allowable
Molecular Analysis
Biopsy/Specimen
Not necessarily required
Required
Study Therapy
Treated Previously Traditionally excluded

With Study Drug
Could be included if subsequent biopsies have target abnormality
Known Sensitivity
Not necessarily required
Usually required
Other Investigational Drug
Traditionally excluded
Usually excluded
Continued
88
TABLE 1. Comparison of Traditional Versus Potential Modern Clinical Trial Eligibility Criteria (contd)
Traditional
Modern
Toxicity
More likely to exclude from trial or lead to patient coming off trial; Different side-effect profile than traditional cytotoxic therapy;
likely more grade 3 or 4 toxicity seen with traditional cytotoxic
generally less grade 3 or 4 toxicity seen with targeted therapy
therapy
Investigator
Opinion
Traditionally more investigator judgment incorporated; drug or Less investigator opinion, shift to explicit instructions on all
combination therapy compared with physician preference
protocol aspects
Regulatory
Consent/
Compliance
Required
Additional
Not often required
Optional Consent
Required
Often required for specimen collection/molecular analysis
Abbreviations: FDA, U.S. Food and Drug Administration; OS, overall survival; PFS, progression-free survival; ECOG, Eastern Cooperative Oncology Group; PS, performance status; QOL,
quality of life.
study population is not accurately reflecting associated

toxicities seen with these novel agents if only the youngest and healthiest patients qualify for study enrollment.
Unger et al31 found that clinical trial participation was
associated with better survival in the first year of enrollment, which was postulated to be related to exclusion
of patients having a poor prognosis with comorbidities
and decreased functional status.
Restrictive, complex eligibility criteria likely result in
decreased accrual rates. Slow accrual to clinical trials
means delays in answers to important clinical questions.
The negative impact of low accrual rates extends to prolonged trial duration, early closure of studies with poor
accrual, and delays in analysis of trial results.32 If eligibility
criteria are broadened, this may result in a faster accrual
rate and observation of a wider array of clinical toxicities
that more accurately reflect the general intended population. However, expanding clinical trial eligibility would
perhaps allow for trial conclusions to more accurately
mimic general clinical practice and allow for extrapolation
to patients who were not treated on-study.
FUTURE DIRECTIONS
Personalized, precision medicine will require the incorporation of clinical exams, molecular testing, and
utilization of information obtained from ongoing adaptive

clinical trials to determine the appropriate treatment for
each patient. There must be a concentrated effort to
limit restrictive eligibility criteria while allowing continued
modification of eligibility criteria as a drug moves through
the research process. ASCO initiatives support the evolution
of more practical, modernized, clinical trial design.1,17 At
the center of this movement for eligibility criteria reform
is the implementation of trials that involve targeted
therapies with molecular biomarkers. The higher the
strength of a drug and a specific molecular target measured on the tumor, the higher the priority of identifying
this patient population for treatment. In these settings,
an active drug-to-target (e.g., high response rate) should
supersede other potentially restrictive eligibility criteria.
Because many markers have low incidence rates, greater
collaboration will be needed to effectively complete
meaningful studies. We must continue to prioritize the
development of modernized, practical clinical protocols and encourage enrollment in these innovative trials.
We must increase access and awareness of molecularly
driven protocols to include more patients as we answer
these important clinical questions and continue to integrate new targeted drugs and biomarkers into clinical
practice.
References
1. Meropol NJ, Kris MG, Winer EP. The American Society of Clinical
Oncologys blueprint for transforming clinical and translational cancer
research. J Clin Oncol. 2012;30:690-691.
2. Masters GA, Krilov L, Bailey HH, et al. Clinical cancer advances 2015:
annual report on progress against cancer from the American Society of
Clinical Oncology. J Clin Oncol. 2015;33:786-809.
3. Mockus SM, Patterson SE, Statz C, et al. Clinical trials in precision
oncology. Clin Chem. 2016;62(3):442-448. Epub 2015 Nov 27.
4. Code of Federal Regulations. General principles of the IND submission,
21 CFR Sect. 312. 22; 2014.
5. Code of Federal Regulations. IND content and format, 21 CFR Sect. 312.
23; 2014.
6. Code of Federal Regulations. Applications for FDA approval to market a
new drug, 21 CFR Sect. 314; 2014.
7. U.S. Department of Health and Human Services. U.S. Food and Drug
Administration. Guidance for Industry Expedited Programs for Serious
ConditionsDrugs and Biologics. http://www.fda.gov/downloads/Drugs/
GuidanceComplianceRegulatoryInformation/Guidances/UCM358301.pdf.
8. Johnson JR, Ning YM, Farrell A, et al. Accelerated approval of oncology
products: the food and drug administration experience. J Natl Cancer
Inst. 2011;103:636-644.
9. U.S. Food and Drug Administration. Trametinib/Dabrafenib. http://
www.fda.gov/Drugs/InformationOnDrugs/ApprovedDrugs/ucm476837.
htm. Accessed January 28, 2016.
10. U.S. Food and Drug Administration. Osimertinib. http://www.fda.gov/
Drugs/InformationOnDrugs/ApprovedDrugs/ucm472565.htm. Accessed
January 28, 2016.
89
KIM ET AL
11. U.S. Food and Drug Administration. Gefitinib. http://www.fda.gov/

January 28, 2016.
12. U.S. Food and Drug Administration. Erlotinib. http://www.fda.gov/
January 28, 2016.
13. U.S. Food and Drug Administration. Innovation or stagnation: challenge
and opportunity on the critical path to new medical products. Washington, DC. 2004.
14. Dragalin V. Adaptive designs: terminology and classification. Drug Inf J.
2006;40(4):425-435.
15. Bowalekar S. Adaptive designs in clinical trials. Perspect Clin Res. 2011;2:
23-27.
16. Kim ES, Herbst RS, Wistuba II, et al. The BATTLE trial: personalizing
therapy for lung cancer. Cancer Discov. 2011;1:44-53.
17. Kim ES, Bernstein D, Hilsenbeck SG, et al. Modernizing eligibility criteria
for molecularly driven trials. J Clin Oncol. 2015;33:2815-2820.
18. National Cancer Institute. ALCHEMIST (The Adjuvant Lung Cancer
Enrichment Marker Identifcation and Sequencing Trials): Question
and Answers. http://www.cancer.gov/newscenter/newsfromnci/2014/
ALCHEMISTlaunchQandA. Accessed February 2, 2015.
19. Lung-MAP. http://www.lung-map.org. Accessed January 28, 2016.
20. I-SPY 2 Trial. http://www.ispy2trial.org. Accessed January 28, 2016.
21. Siu LL, Conley BA, Boerner S, et al. Next-generation sequencing to guide
clinical trials. Clin Cancer Res. 2015;21:4536-4544.
22. Hyman DM, Puzanov I, Subbiah V, et al. Vemurafenib in multiple
nonmelanoma cancers with BRAF V600 mutations. N Engl J Med. 2015;
373:726-736.
90
23. National Cancer Institute. NCI Molecular Analysis for Therapy

Choice Program (MATCH) & Pediatric MATCH. http://www.cancer.gov/
clinicaltrials/noteworthy-trials/match. Accessed February 2, 2015.
24. Do K, OSullivan Coyne G, Chen AP. An overview of the NCI precision
medicine trials-NCI MATCH and MPACT. Chin Clin Oncol. 2015;4:
31-38.
25. Kim ES. The future of molecular medicine: biomarkers, BATTLEs, and big
data. Am Soc Clin Oncol Educ Book. 2015;22-27.
26. Andrews A. ASCO and NCI launch largest precision medicine trials using
real-world evidence. Am Health Drug Benefits. 2015;8:37.
27. Christian MC, Trimble EL. Increasing participation of physicians and
patients from underrepresented racial and ethnic groups in National
Cancer Institute-sponsored clinical trials. Cancer Epidemiol Biomarkers
Prev. 2003;12:277s-283s.
28. Sohal DP, Rini BI, Khorana AA, et al. Prospective clinical study of
precision oncology in solid tumors. J Natl Cancer Inst. 2015;108:
djv332.
29. George SL. Reducing patient eligibility criteria in cancer clinical trials.
J Clin Oncol. 1996;14:1364-1370.
30. Carden CP, Agarwal R, Saran F, et al. Eligibility of patients with brain
metastases for phase I trials: time for a rethink? Lancet Oncol. 2008;9:
1012-1017.
31. Unger JM, Barlow WE, Martin DP, et al. Comparison of survival outcomes among cancer patients treated in and out of clinical trials. J Natl
Cancer Inst. 2014;106:dju002.
32. Lara PN Jr, Higdon R, Lim N, et al. Prospective evaluation of cancer

Immunotherapy: Beyond
Checkpoint Inhibitors
CHAIR
Gregory L. Beatty, MD, PhD
Hospital of the University of Pennsylvania
Philadelphia, PA
SPEAKERS
George E. Peoples, MD
Cancer Vaccine Development Program
San Antonio, TX
Marcela Valderrama Maus, MD, PhD
Abramson Cancer Center of the University of Pennsylvania
Philadelphia, PA
ARMING THE IMMUNE SYSTEM TO PREVENT CANCER RECURRENCE
Arming the Immune System Through Vaccination to Prevent

Cancer Recurrence
Diane F. Hale, MD, Timothy J. Vreeland, MD, and George E. Peoples, MD
OVERVIEW
Cancer vaccines have the potential to provide a nontoxic treatment for the prevention of cancer recurrence in the adjuvant
setting. Many cancer vaccines have been tested in multiple phase III trials with minimal success. However, through these
failed clinical trials, we have learned that the ideal setting for vaccine therapy is the adjuvant setting. Also, we have learned
important lessons about patient selection to maximize the probability of success. This article will highlight some of the
successes, our trial results in the adjuvant setting, and future directions.
he creation of a vaccine to prevent or cure cancer is a very

alluring goal, but the pursuit of this goal has been fraught
with difficulty. To create a cancer vaccine, the first key is to
find the perfect target: a tumor-associated antigen (TAA) or
antigens expressed on a high percentage of tumor cells but
not on normal cells. Once the target is identified, the goal is
to use an efficient, safe, and easily reproducible method to
stimulate the immune system to eliminate targeted tumor
cells. These goals have been approached in a number of ways
in the saga of cancer vaccine creation, but with few clinically
meaningful successes. To date, the majority of cancer vaccines
have been tested in the metastatic setting with disappointing
results. The high disease burden, immune escape mechanisms,
and immune suppression of the tumor micro-environment
encountered in the setting of metastatic or aggressive disease
is more than the immune system can overcome. This may not
be the ideal setting for cancer vaccines to become clinically
relevant, at least not as monotherapy.
There are, however, exceptions to the overall discouraging
results of vaccines in the metastatic setting, and important
lessons can be learned from these exceptions. The most
notable exception, and the only vaccine granted U.S. Food
and Drug Administrationapproval, is sipuleucel-T, an autologous dendritic cell vaccine directed against the TAA prostatic
acid phosphate. This vaccine is approved for use in metastatic,
asymptomatic or minimally symptomatic castration-resistant
(hormone-refractory) prostate cancer. The success of sipuleucelT in the metastatic setting is likely due to the indolent nature of
prostate cancer, which makes it a more realistic target for a
cancer vaccine, even in advanced disease. Similarly, subset
analyses from multiple trials of vaccines given to patients

with advanced cancers have shown that, although vaccines
produced disappointing results overall, they perform much
better in patients with minimal or no residual disease. Taken
as a whole, these results indicate that cancer vaccines perform better when used in patients with less aggressive disease, whether by biology or by disease burden.
Building on these lessons, cancer vaccines have been used
to treat patients with the absolute minimal disease burden
after they are rendered disease-free by standard-of-care
therapies. Priming the immune system to create an active
immune response to prevent recurrence rather than trying
to eradicate advanced, established, and aggressive disease is
very appealing. First, after patients complete and recover
from standard-of-care therapy (generally immunosuppressive chemotherapy), they have improved immune function,
allowing a full development of antitumor immunity.1 Second,
in a disease-free state, there is theoretically less tumor-related
immune suppression, fewer immune escape mechanisms, and
no tumor microenvironment.1 With the bulk of tumor removed
or eradicated, the immune system can then be used to eliminate
the small number of tumor cells left after therapy, which is a
more realistic goal for this mechanism. Finally, the long-term
immune memory created by a successful cancer vaccine offers
continued protection against dormant circulating or disseminated tumor cells that may still be present or may re-emerge in a
disease-free host; this characteristic and distinct advantage
distinguishes active specific immunotherapy from traditional
cancer therapies. Consequently, cancer vaccines offer the unique
promise of long-term protection against cancer recurrence.
From the Department of Surgery, San Antonio Military Medical Center, Fort Sam Houston, TX; Department of Surgery, Womack Army Medical Center, Fort Bragg, NC; Cancer Vaccine
Development Program, San Antonio, TX; Department of Surgery, Uniformed Services University of the Health Sciences, Bethesda, MD.
Corresponding author: George E. Peoples, MD, Cancer Vaccine Development Program, c/o Metis Foundation, 300 Convent St., San Antonio, TX 78205; email: georgepeoples2@
hotmail.com.
e159
HALE, VREELAND, AND PEOPLES
Although using the vaccine platform to prevent recurrence

in the adjuvant setting is quite attractive, this setting has
unique challenges. There is less tolerance for toxicity when
patients are disease-free; therefore, any therapy being
added in the adjuvant setting must come with a very low
toxicity profile or else compliance is expected to be poor.
Also, proving a clinically substantial decrease in recurrence
can be quite challenging when evaluating patients with less
aggressive disease. The attractive aspect of studying a new
therapy in patients with advanced disease is the relatively
short time period needed to show a reduction in tumor
burden or improvement in progression-free survival. Conversely, in the adjuvant setting, the time period required to
reach a noteworthy number of recurrences may be quite
lengthy. Depending on the tumor biology and documented
recurrence rates, this may necessitate a trial spanning
multiple years and with large treatment groupsultimately
resulting in a very expensive trial. If the vaccine strategy is
truly more suitable in the adjuvant setting, these limitations
lead to a difficult first step to prove efficacy and likely explain
why this strategy has been around for so long but has so few
successes. As a result, much of the current evidence showing
benefit of vaccines comes from subset analyses of larger
trials focusing on patients with minimal disease burden.
We will review the evidence of vaccine success in patients
with minimal disease burden and current immunologic
vaccine therapies being studied in the adjuvant setting, and
we will lay out the best available data supporting this theory
with current successes. We will also discuss future plans for
combination therapies.
PROOF OF CONCEPT IN PHASE III TRIALS

Although developing and funding an adjuvant study may be
difficult, it has been attempted in several cancer types with a
variety of vaccine platforms. Table 1 summarizes the previous and current phase III trials of vaccines given in the
adjuvant setting (with no or minimal evidence of disease),
KEY POINTS
e160
Cancer vaccines represent a nontoxic therapeutic

modality with great antitumor specificity.
Peptide vaccines effectively stimulate a specific immune
response and are simple, easily produced, and
efficiently exported to the community.
The adjuvant setting is an appropriate clinical setting to
test cancer vaccines as it optimizes the chances of an
effective immune response when the tumor burden is
the lowest.
Combinations of other immunotherapies with vaccines
may be even more clinically effective.
In the metastatic setting, combining vaccines as
T-celleliciting agents with checkpoint inhibitors may
synergistically enhance the benefits of each modality
and improve long-term clinical results.
for the prevention of solid tumor recurrence. There have

been many vaccines that found clinical improvement in
phase II evaluation but failed to find clinical significance in
phase III. The lessons learned from previous trials have
assisted in creating improved trial designs and, most importantly, selecting patients who will derive the most benefit
from a cancer vaccine. Highlighted below are a few key
phase III trials that demonstrate important points for future
trial design.
OncoVax is an autologous tumor cell vaccine combined
with bacillus Calmette-Guerin that has been studied
extensively in colon cancer. There have been three phase
III trials using this vaccine in the adjuvant setting after
resection of stage II/III colon and rectal cancer, enrolling a
total of 764 patients.2-4 The initial phase III trial enrolled
patients with both colon and rectal cancer. Overall, there
was no significant differences in the treatment groups, but
when breaking down the results by tumor type (colon vs.
rectal) and focusing only on those patients with colon
cancer, there was a significant improvement in both overall
survival (OS) and disease-free survival (DFS) for the vaccinated versus control patients (OS 83% vs. 52%, hazard ratio
[HR] 3.97; 95% CI, 1.2412.72; p = .020; DFS 67% vs. 44%, HR
2.67; 95% CI, 1.056.76; p = .039).2 This led to the second
phase III trial, which focused on patients with surgically
resected stage II/III colon cancer. Again, the overall results of
this trial failed to show a significant difference with treatment,
but within the subset of stage II disease, vaccinated patients did
show an improved recurrence free interval (p = .011) and 4-year
recurrence-free survival (88% vs. 74%, p = .032).3 The largest of
the phase III OncoVax trials enrolled 412 patients with surgically
resected stage II/III colon cancer and had a 7.6-year median
follow-up.4 This trial did not show a significant improvement in
OS, DFS, or time-to-recurrence with vaccination.4 Based on the
combined results of these trials, the next phase III OncoVax trial
is focused on patients with minimal disease, where subset
analysis has shown the most promise, enrolling only patients
with resected stage II colon cancer (NCT02448173). Overall, the
work with this vaccine nicely demonstrates identification of
patients with no evidence of disease with low disease burden
as the most suitable population for cancer vaccines.
A similar finding of improved outcomes in vaccinated
patients with lower disease burden was found in a trial using
vitespen for renal cell carcinoma. Vitespen (HSPPC-96) is an
autologous, tumor-derived, heat shock protein vaccine. In
the phase III evaluation, it was given to patients in the
adjuvant setting that were stage I-IV, surgically resected and
with no evidence of disease; with 1.9-year median followup, there were no overall differences in recurrence events
(p = .506). However, on subset analysis of patients with
stage I/II disease, there was a decrease in recurrence for
patients who were treated with the vaccine versus patients
in the control group (15% vs. 27%, p = .056).5
Patient selection in trials using an adjuvant vaccine to
prevent recurrence is a key to demonstrate clinical efficacy.
Beyond a low disease burden, it is also important to understand the specific mechanism of action of each vaccine to
Vaccine
Tyrosinase, gp100,
MART-1
GM2-KLH
MAGE-A3
MART-1, NA17-A,
gp100,
tyrosinase
Melanoma
Melanoma
Melanoma
Melanoma
(Ocular)
Algenpantucel-L
(hyperacute)
Belagenpumatecel- Tumor cell

L (lucanix)
Polyvalent cultured Tumor cell

allogenic melanoma vaccine
(canvaxin)
Pancreas
NSCLC
Melanoma
TUMOR CELL VACCINES
Tumor cell
Protein
RCC
HSPPC-96
(vitespen)
Peptide
Protein
Multiple
peptides
Protein
Protein
Multiple
Peptides
Multiple
peptides
Peptide
Glioblastoma Rindopepimut
(CDX-110)
MAGE-A3
POL-103A
Melanoma
NSCLC
E75
Breast
PEPTIDE VACCINES
DC
Glioblastoma ICT-107
DC
DC
DCaT-RNA
Strategy
Glioblastoma DC Vax-L
Melanoma
(Uveal)
DENDRTIC CELL VACCINES
Solid Tumor
Type
GM-CSF
Alum
GM-CSF
Allogeneic melanoma vaccine
Allogenic tumor cell with antisense plasmid
Alpha-1,3-glactosyltransferase
expressing allogeneic pancreatic tumor cell vaccine
Autologous, tumor-derived, heat

shock protein
14-mer peptide of EGFRvIII with

KLH
recMAGE-A3
recMAGE-A3
BCG
GM-CSF
AS15
AS15
200*
No. Patients
700*
13
1,351*
1,314
722*
Adjuvant, NED, stage III/IV
1,656
Adjuvant (maintenance after first-line che- 532

motherapy), evidence of disease allowed
to have SD (CR or PR)
Adjuvant, NED, resected pancreatic cancer

stage I/II
Adjuvant, NED, surgically resected, stage I-IV 728
Adjuvant, must have attempted surgical

700*
resection and chemoradiation, may have
evidence of disease, must have stable
disease from postoperative to
postchemoradiation
Adjuvant, NED, MAGE-A3 positive, resected, 2,272

stage IB, II, IIIA
Adjuvant, NED
Adjuvant, NED, stage IIIB-C
Adjuvant, NED, stage II
Adjuvant, NED, surgically resected, stage IV 815
Adjuvant, NED, surgically resected, stage IIb, 1,059*

IIc, III
Adjuvant, NED, high-risk for recurrence
Adjuvant, evidence of disease after standard 414*

of care , 1 cm3 of disease may be present
Adjuvant, NED, new diagnosis in addition to 348*

standard of care, must be resected
disease
Adjuvant, NED, resected, monosomy 3
Immunoadjuvant Study Type
GM2 antigen combined with KLH QS-21

to improve immunogenicity
Shed peptides from 3 allogenic

melanoma cell lines
HER2/neu peptide vaccine
Autologous DC pulsed with 6

peptides (MAGE-1, HER2, AIM2, TRP-2, GP100, IL-13Ra2)
Autologous DC pulsed with autologous whole tumor lysate
Autologous tumor RNA loaded

onto autologous DC
Vaccine Type
TABLE 1. Adjuvant Phase III Solid Tumor Vaccine Trials
NCT01983748
Reference or
NCT Number
NCT02546102
OS (p = .528), RFS (p = .131)
Currently enrolling, study start date April 2012, estimated primary completion date July 2016
NCT01989572
NCT01546571
Active (not enrolling), study start date November 2011, NCT01479244

estimated primary completion date April 2018
Currently enrolling, study start date December 2015,

estimated primary completion December 2019
Active (not enrolling), study start date December 2006, NCT00045968

estimated primary completion September 2016
Currently enrolling, study start date June 2014, estimated primary completion June 2020
Results
MMAIT
STOP Trial
IMPRESS
ACT IV
MAGRIT
DERMA
NCT00796445
Vansteenkiste
et al20
NCT00036816
NCT01072981
Continued
Terminated after interim analysis for futility: 5-year

Morton et al22
estimated survival 42.3% for stage IV and 63.4% for
stage III
No difference OS (p = .594), no difference PFS (p = .947); Giaccone

regression analysis found improved survival with
et al21
randomization within 12 weeks of completion of
chemotherapy (p = .002) and prior radiation improved survival (p = .032)
Active (not enrolling), study start date April 2010,

estimated primary completion date June 2016
1.9-year median follow-up, no difference in overall

Wood et al5
recurrence events (37.7% vaccine vs. 39.8% control; p
= .506), slightly improved in stage I/II (15.2% vaccine
vs. 27% control; p = .056)
Active (not enrolling), study start date November 2011, NCT01480479

estimated primary completion date November 2016
38.8-month median follow-up, median DFS 60.5

months vaccine vs. 57.9 months control (p = .74)
Terminated for low accrual
Terminated after showing lack of efficacy
EORTC 18961 Terminated after second interim analysis: 1.8 years

Eggermont
median follow-up, detrimental OS (HR 1.66; p = .02)
et al19
and futile RFS (HR 1.00; p = .99). After 4-year followup, vaccinated patients RFS 1.2% (HR 1.03) and OS
rate 2.1% (HR 1.16)
MAVIS
PRESENT
Study Name
e161
e162
Active specific
immunotherapy
(OncoVax)
Active specific
immunotherapy
(OncoVax)
Active specific
immunotherapy
(OncoVax)
Active specific
immunotherapy
(OncoVax)
Newcastle disease,
virus-infected
autologous tumor cell vaccine
(ATV-; NDV)
Colon and
Rectal
Colon
Colon
Colon
Colon and
Rectal
Tumor cell
Tumor cell
Tumor cell
Tumor cell
Tumor cell
Tumor cell
Strategy
Viral vector of autologous tumor

cells from resected liver
metastasis
Autologous tumor cells
Allogeneic melanoma vaccine
Vaccine Type
BCG
BCG
BCG
BCG
DETOX (detoxified Freund

adjuvant)
Adjuvant, NED, after resection (to include

liver metastasis resection)
Adjuvant, NED, stage II colon
Adjuvant, NED, stage II/III colon
Adjuvant, NED, stage II/III colon
Adjuvant, NED, stage II/III colon or rectal
Adjuvant, NED, stage II/III
Immunoadjuvant Study Type
51
550*
412
254
98
689
No. Patients
ECOG E5283
8701
S9035
Study Name
Reference or
NCT Number
116.1-month follow-up for vaccine, 112.4-month

follow-up for control, no difference in OS, DFS, or
metastases-free survival. Subgroup analysis: advantage for vaccinated colon OR (p = .042) and
metastases-free survival (p = .047)
Currently enrolling, study start date May 2015, estimated primary completion date July 2020
Schulze et al23
NCT02448173
7.6-year median follow-up, ITT no difference OS, DFS, or Harris et al4

time to recurrence, cohort analysis vaccinated patient with treatment compliance signaled a trend in
DFS (p = .078) and OS (p = .12)
5.3-year median follow-up, recurrence free interval

Vermorken et
(p = .023) improved in vaccine with larger impact on
al3
RFS (p = .032) in stage II vaccine
93-month median follow-up, OS trend toward favoring Hoover et al2

vaccine (p = .088), cohort analysis: colon cancer
significant OS (p = .02) and DFS (p = .039) for vaccine,
no benefit in rectal
12.1-year median follow-up, ITT no difference in RFS Carson et al6

(p = .58) or OS (p = .61), subset analysis vaccinated
and HLA-A2/Cw3+ 10-year RFS 66% vs. controls 54%
(p = .02)
Results
*Anticipated number.
Abbreviations: DC, dendritic cell; NED, no evidence of disease; GM-CSF, granulocyte macrophage colony-stimulating factor; OS, overall survival; DFS, disease-free survival; NSCLC, nonsmall cell lung cancer; RCC, renal cell carcinoma; SD, stable
disease; CR, complete response; PR, partial response; BCG, bacillus Calmette-Guerin; ITT, intention-to-treat.
Polyvalent tumorcell lysate

(melacine)
Vaccine
Melanoma
Solid Tumor
Type
TABLE 1. Adjuvant Phase III Solid Tumor Vaccine Trials (cont'd)
identify the patients who will have the best clinical benefit.
The Southwest Oncology Groups phase III trial of melacine
demonstrates the importance of human leukocyte antigen
(HLA) characteristics in patient selection and understanding
how the vaccine stimulates the immune system. Melacine
is a vaccine created from an allogeneic tumor cell line that
contains numerous melanoma-associated TAAs. The phase III
trial randomly selected 689 patients with stage II/III resected
melanoma to receive the vaccine or to be in the control group.
At a median follow-up of 12.1 years, the trial had an overall
negative result, but subset analysis demonstrated a significant
recurrence-free survival (RFS) and OS advantage in the patients who received the vaccine compared with the control
group when limiting analysis to patients with HLA-A2+ and/or
HLA-Cw3+ expression (5-year RFS, 78% vs. 65%; 10-year RFS,
66% vs. 54%; p = .02; 5-year OS, 90% vs. 76%; 10-year OS, 75%
vs. 63%; p = .01).6 These findings demonstrate the importance
of identifying the appropriate target population for a vaccine
and selecting patients with the appropriate biologic features.
PROOF OF EFFICACY
As previously discussed, demonstrating efficacy of a new
therapy in the adjuvant setting is inherently difficult, but it has
been done. We will highlight the work our group has completed
with cancer vaccines in the adjuvant setting; these trials are
summarized in Table 2. We have extensively researched three
HER2-based peptide vaccines (Fig. 1). These peptides (E75, GP2,
and AE37) were paired with an immunoadjuvant, granulocyte
macrophage colony-stimulating factor (GM-CSF) and given in
the adjuvant setting to disease-free patients with breast cancer
at high risk of recurrence. Two of the vaccines are HLA-A2 and
HLA-A3restricted (E75 and GP2) and directly stimulate
CD8+ T cells to generate immunity. The third, AE37, is a longer
peptide vaccine, is not HLA-restricted, and stimulates CD4+
T cells to create HER2-directed immunity. Similar to the HER2based peptide vaccines, we have also targeted folate binding
protein (FBP)-expressing tumors with a peptide vaccine (E39
and GM-CSF) in ovarian, endometrial, and breast cancers.
Lastly, we have recently expanded into an autologous dendritic
cell vaccine platform using autologous tumor lysate particleloaded dendritic cells in the adjuvant setting, targeting patients
with surgically resected melanoma.
minimal toxicity with the majority of local and systemic

toxicities being grade 1. It was also able to produce effective
immunity to HER2. With a median follow-up of 60 months,
the vaccinated patients in this trial showed a trend toward
improved DFS with a 48% reduction in relative risk of recurrence (RRR; p = .08).7 Given the dose escalation nature of
the first stage of the trial, some patients did not receive the
optimal dose of the vaccine. When we examined patients
based on dosing, we found a significant increase in DFS in
patients who received the optimal dosing regimen compared with suboptimal dosed and observational controls
(94.6% vs. 87.1% vs. 80.2%, respectively; p = .05).7 During
this trial, it was discovered that, after completing the primary vaccination series, patients demonstrated waning
immunity to the targeted peptide. Therefore, a voluntary
booster program was instituted and patients received additional booster inoculations every 6 months. When examining patients who were optimally dosed and boosted, we
found an improvement numerically in DFS versus all other
vaccinated patients versus the control arm (95.2% vs. 88.4%
vs. 80.2%; p = .11).7 Another important patient characteristic, which had been previously demonstrated and was also
identified here, was the effect of HER2 expression on response
to the vaccine.8 For patients with non-overexpression of HER2
(immunohistochemistry [IHC] 1+ or 2+), there was a trend
toward improvement in 5-year DFS in vaccinated patients
versus the control (88% vs. 78%; p = .16).7 There were also
important limitations of this trial; there was no blinded,
placebo-treated control group receiving GM-CSF, which led to
questions about the effects of this immunoadjuvant therapy.
Thus, this feature was incorporated into our future trials.
Integrating the lessons learned from this initial trial, a phase
III trial was designed that included a GM-CSFtreated control
arm and optimal dosing of the vaccine to include a booster
series, and it enrolled only patients with HER2 1-2+ expression.
The phase III PRESENT (Prevention of Recurrence in Early Stage
node-positive breast cancer with low to intermediate HER2
Expression with NeuVax Treatment) trial was initiated in the
adjuvant setting for patients with high-risk HLA-A2+ or HLA-A3+
breast cancers who were deemed disease free after standardof-care therapy. The patients were randomly selected to receive E75 with GM-CSF or GM-CSF only. This trial has completed enrollment and is monitoring for the primary endpoint
of 3-year DFS.
E75 (Nelipepimut-S, NeuVax)

E75 (nelipepimut-S or NeuVax) is a peptide (KIFGSLAFL, HER2
aa: 369-377) derived from the extracellular domain of HER2
(Fig. 1) and given in combination with GM-CSF. Its immunologic activity is mainly restricted to specific HLA types HLAA2 and HLA-A3, but with the possibility of binding several
other less common alleles HLA-A24 and HLA-A26. E75 was
evaluated in one of the largest adjuvant breast cancer
vaccine phase I/II trials to date, with 195 enrolled and 187
evaluable patients.7 Patients who were positive for HLA-A2/
A3 were placed in the treatment arm, and patients who were
negative for HLA-A2/A3 were followed prospectively in an
observational control group. This vaccine was found to have
AE37 and GP2

Our second large phase II peptide vaccine trial in the adjuvant setting, which is nearing completion, is a dual trial
testing GP2 and AE37. GP2 (IISAVVGIL, HER2 aa: 654-675) is a
peptide derived from the transmembrane domain of HER2
(Fig. 1). GP2 is a major histocompatibility complex (MHC)
class I peptide vaccine that stimulates antigen-specific CD8+
T cells in patients positive for HLA-A2 or HLA-A3. AE37
(GVGSPYVSRLLGICL, HER2 aa: 776-790) is the Ii-Key (LRMK)
hybrid of the HER2-derived MHC class II peptide, AE36, from
the intracellular domain of HER2 (Fig. 1) and is not HLArestricted. The trial design selected patients by HLA status,
e163
TABLE 2. Current and Future Trials in the Adjuvant NED Setting to Prevent Recurrence
Solid
Tumor
Type
Immunoadjuvant Phase Study Type
No.
Patients Result
HER2 (aa:
369-377)
GM-CSF
II
Adjuvant, NED,
high-risk
recurrence
195
Median follow-up 60 months; 5-year

Mittendorf
DFS ITT: 90% vaccinated vs. 80%
et al7
control (p = .08); subset populations
NCT00841399
with most benefit optimally dose 95%
vaccine vs. 80% control (p = .05) and NCT00854789
both optimally dosed and boosted:
95% vaccinated vs. 80% control (p =
.11)
Peptide
HER2 (aa:
369-377)
GM-CSF
II
Adjuvant, NED,
DCIS
108*
Anticipated start date March 2016, estimated primary completion September 2019
AE37
Peptide
HER2 (aa:
776-790
+LRMK)
GM-CSF
II
Adjuvant, NED,
high-risk
recurrence
298
Median follow-up 25 months; 5-year DFS Mittendorf

et al10
IIT: 81% vaccine vs. 80% controls (p =
.70); subset populations with improved
NCT00524277
benefit HER2 1-2+ expressing tumors:
77% vaccine vs. 66% controls (p = .21)
and TNBC 78% vaccine vs. 49% controls
(p = .12)
Breast
GP2
Peptide
HER2 (aa:
654-675)
GM-CSF
II
Adjuvant, NED,
high-risk
recurrence
180
Schneble
et al9
DFS IIT: 88% vaccine vs. 81% controls
(p = .43); subset population with imNCT00524277
proved benefit HER3+ expressing tumors: 94% vaccine vs. 89% controls (p
= .86)
Prostate
E75 (NeuVax,
nelipepimut-S)
Peptide
HER2 (aa:
369-377)
GM-CSF
I/IIA
Adjuvant, NED
s/p resection,
high-risk
recurrence
40
Median follow-up 58 months; PSA recurrence rate: 29% vaccine vs. 26%
control (p = .9); DFS: 41.3 vs. 37.9
months (p = .6)
Ovarian
E39
Peptide
Folate binding protein (aa:

191-199)
GM-CSF
I/IIA
Adjuvant, NED
51
Greene
et al15
DFS ITT: 43% vaccine vs. 34% control
(p = .36); subset population with imNCT01580696
proved benefit optimally dosed
group: 86% vaccine vs. 34% control (p
= .03)
Melanoma Dendritoma
DC
Autologous
DC fused
with autologous
tumor
lysate
IL-2
I/IIA
Adjuvant, NED
and ED, stage
IV patients
enrolled
25
Median OS 16 months; overall 5-year

DFS: 29%; OS improvements in NED
patients: 80% vs. ED 14% (p = .004)
Greene et al17
Melanoma TLPLDC
DC
GM-CSF
Autologous
DC loaded
with autologous
tumor
lysate
IIB
Adjuvant, NED
120*
Actively enrolling, estimated primary

completion June 2018
NCT02301611

completion December 2016
NCT01570036

completion October 2016
NCT02297698
Vaccine
Strategy
Breast
E75 (NeuVax,
nelipepimut-S)
Peptide
Breast
(DCIS)
E75 (NeuVax,
nelipepimut-S)
Breast
Vaccine
Type
Reference or
NCT Number
PEPTIDE
NCT02636582
Gates et al24
DENDRITIC CELL
COMBINATION
300*
Breast
E75 (NeuVax,
Nelipepimut-S) and
trastuzumab
(herceptin)
Peptide and HER2 (aa:

369-377)
monocloand HER2
nal
monocloantibody
nal
antibody
GM-CSF
II
Adjuvant, NED,
high-risk recurrence with
HER2 1-2+
expressing
tumors
Breast
E75 (NeuVax,
Nelipepimut-S) and
trastuzumab
(herceptin)
Peptide and HER2 (aa:

369-377)
monocloand HER2
nal
monocloantibody
nal
antibody
GM-CSF
II
100*
Adjuvant, NED,
high-risk recurrence with
HER2 3+
expressing
tumors
*Anticipated number.
Abbreviations: DC, dendritic cell; NED, no evidence of disease; GM-CSF, granulocyte macrophage colony-stimulating factor; OS, overall survival; DFS, disease-free survival; NSCLC,
nonsmall cell lung cancer; RCC, renal cell carcinoma; SD, stable disease; CR, complete response; PR, partial response; BCG, bacillus Calmette-Guerin; ITT, intention-to-treat; DCIS,
ductal carcinoma in situ; TNBC, triple-negative breast cancer; TLPLDC, tumor lysate particle-loaded dendritic cell.
e164
FIGURE 1. HER2 Peptide Vaccines (E75, GP2, AE37)
Abbreviations: MHC, major histocompatibility complex; HLA, human leukocyte antigen.
and placed the patients who were positive for HLA-A2/A3

into the GP2 trial and patients who were negative for HLAA2/A3 into the AE37 trial. Patients in both trials were then
randomly selected to receive the respective vaccine or the
control, GM-CSF alone. This dual trial has also completed
enrollment, with 298 patients enrolled in the AE37 trial and
180 patients in the GP2 trial.9,10 From the analysis of these
trials, we have established that the minimal toxicity in all
patients was primarily related to the GM-CSF given similar
toxicity levels between the vaccine and control arms.
The primary analysis of the GP2 trial revealed a 37% reduction
in RRR for the vaccinated arm with a 34-month median followup. Vaccinated patients showed a nonsignificant increase in
estimated 5-year DFS (88% vs. 81%; p = .43). Interestingly,
subset analysis by HER2 expression in this trial showed that
the best DFS was found in patients with HER2 overexpression
(IHC 3+ or FISH+) who received the vaccine after completing
trastuzumab. Although intention-to-treat (ITT) analysis in this
subgroup showed a 94% estimated 5-year DFS in vaccinated
patients versus 89% in the control arms (p = .86), the pertreatment analysis (excluding patients who experienced recurrence during the primary vaccination series and therefore
did not complete the series) of HER2 overexpression showed
improved 5-year DFS in the vaccinated patients versus the
control (100% vs. 89%; p = .08). Although it is difficult to show a
statistically significant improvement over the already impressive DFS of patients receiving standard-of-care trastuzumab,
based on these results, the GP2 vaccine appears to have
synergism with the HER2-directed trastuzumab.9
The primary analysis of the AE37 arms revealed no significant difference in 5-year DFS for vaccinated versus
control patients (83.5% vs. 82.1%; p = .7). Again, planned
subset analysis revealed the patient populations with the
greatest benefit. First, patients with HER2 1-2+ expression
showed a trend of separation in the Kaplan Meier plots, with
5-year DFS enhanced in patients who were vaccinated versus
controls (77% vs. 66%; p = .21) and a 32% reduction in RRR.

This was even more pronounced in patients with triplenegative breast cancer, who experienced a 56% reduction in
RRR and had a trend toward improved 5-year DFS with the
vaccination (78% vs. 50%; p = .12). Patients with triplenegative breast cancer currently have limited options for
long-term recurrence prevention, making the AE37 vaccine
an interesting option for future study.10
Overall, through these phase II studies, a potential benefit
of peptide vaccine therapy in the adjuvant setting was
identified in patient populations depending on biologic characteristics of the tumor and synergy with current standardof-care therapy, specifically passive immunotherapy.
E39
Similar to HER2 expression in breast cancer, FBP has also
been found to be overexpressed in breast, lung, endometrial, and ovarian cancers and has been linked to a poor
prognosis.11-14 E39 (EIWTHSYKV, FBP aa: 191-199,) is an HLAA2restricted FBP peptide vaccine, also given with GM-CSF.
Our group has completed an initial phase I/IIA trial of the E39
with GM-CSF vaccine given in the adjuvant setting to patients with ovarian and endometrial cancers after they received standard-of-care therapies. Similar to previous
peptide-based vaccine trials, there was minimal toxicity
with the treatment. Interim analysis after median follow-up
of 12 months revealed overall ITT 5-year estimated DFS was
43% for vaccinated patients versus 34% for controls (p = .36).
Like the initial E75 trial, given the phase I/IIA design of this
trial, not all patients received what was deemed to be the
optimal dose of the vaccine. In subset analysis based on
dosing, optimally dosed patients had a statistically significant increase in survival over suboptimally dosed and control
patients (86% vs. 34% vs. 21%, respectively; p = .03). Because
of the aggressive nature of ovarian and endometrial cancers
as opposed to breast cancer, this trial was able to show a
e165
survival difference between groups in a relatively short

time.15
Tumor Lysate Particle-Loaded Dendritic Cell Vaccine

We are currently evaluating a novel dendritic cell vaccine
technology. The new technology revolutionizes the ability to
efficiently produce a personalized dendritic cell vaccine. The
vaccine is produced by loading autologous tumor lysate into
prepared yeast cell wall particles, which are naturally and
efficiently phagocytized by isolated and immature dendritic
cells. Along with the immunoadjuvant CpG, the loading
process matures the dendritic cells resulting in the final
product referred to as the tumor lysate particle-loaded
dendritic cell (TLPLDC) vaccine. The use of autologous tumor lystate allows for the inclusion of the full range of
tumor antigens from a given patients tumor, making it
widely applicable to any patient and any tumor type.
TLPLDC can be created in 48 hours and requires only a small
amount of tumor (1 cm3). In comparison with previous
dendritic cell vaccine technologies, TLPLDC costs less and
requires fewer dendritic cells. This technology affords the
ability to create a personalized vaccine for any solid tumor
in an expeditious manner.
The TLPLDC vaccine technology builds from the previous
technology of dendritoma (fusion of dendritic cells with
autologous tumor), which is a relatively cumbersome
technique.16 We have completed a phase I/IIA trial of the
dendritoma vaccine in 25 patients with stage IV melanoma.
With a median follow-up of 16 months, the overall 5-year
DFS was 29%. Following the theme of previous trials, the
patients with no evidence of disease prior to initiation of
therapy experienced a significant benefit compared with
those with evidence of disease (80% vs. 14%; p = .004).17
Although these results are encouraging, the dendritoma
technology is too cumbersome and expensive to be evaluated
in a larger trial. The TLPLDC vaccine technology accomplishes
the same goal as the dendritoma technologythe introduction
of the entire antigenic repertoire from the autologous tumor
into the cytoplasm of the dendritic cell. The efficacy of the
TLPLDC vaccine has been confirmed in an ongoing basket trial of
multiple solid tumors.18 Current results from this ongoing phase
I/IIA trial indicate that the autologous TLPLDC vaccination is
well-tolerated and safe. The TLPLDC vaccine had clinical benefit
in 60% of patients, with 30% objective response including
complete response in metastatic melanoma.18 Therefore,
building on the results of the dendritoma trial in melanoma and
utilizing the improved TLPLDC technology (confirmed to be
effective in melanoma18), we have initiated a prospective,
randomized, blinded, placebo-controlled phase IIB trial in patients with fully resected stage III and IV melanoma to prevent
recurrence (NCT02301611), with estimated completion of the
primary endpoint in 2018.
breast cancer with HER2 overexpression changed based on

the approval of trastuzumab. During these trials, we were
able to compare patients who received a combination of
peptide vaccine and trastuzumab with patients in the control
arms, who received trastuzumab without a vaccine. From
these trials, there have been 89 patients with HER2 overexpression who received trastuzumab, with 55 of these
patients also receiving one of our peptide vaccines. With a
median follow-up of 36 months, patients receiving the
combination of vaccine and trastuzumab had a 100% DFS
versus only 84% in those receiving trastuzumab alone
(p = .012; unpublished data from project narrative). To
further elucidate the potential for synergy and verify the
possible improved clinical benefit, we are currently enrolling two separate trials combining E75 and trastuzumab
in the adjuvant setting. The first trial is focused on the
patients with HER2 overexpression (HER2 3+ or FISH+;
NCT01570036), and the second trial is enrolling patients
with HER2 non-overexpression (HER2 1-2+; NCT02297698).
In both trials, patients are randomly selected to receive
trastuzumab with E75 and GM-CSF versus trastuzumab and
GM-CSF alone. Patients with HER2 1-2+ expression currently have limited options in the adjuvant setting after
completion of chemotherapy, as they do not qualify to
receive trastuzumab. Our trial, along with the NSABP-47
trial, seeks to address the use of trastuzumab in this
population. Additionally, we seek to assess the vaccine in
this population as well as the toxicity associated with
trastuzumab alone or in combination with the vaccine in
these patients in the adjuvant setting.
Though beyond the scope of this article, vaccines may
now, largely for the first time, be successfully applied to the
nonadjuvant setting. Given the successes of checkpoint
inhibitors (anti-CTLA4 and PD-1) to protect and promote
endogenous antitumor T cells, there is now rationale for
combining tumor-reactive T-celleliciting vaccines with
clinically proven checkpoint inhibitors for patients with
larger tumor burdens. Furthermore, the knowledge gained
from the checkpoint inhibitorrelated studies underscores
why so many vaccines have previously failed to demonstrate benefit in patients who have metastatic disease
with established tumors capable of suppressing/evading
the vaccine-induced immune response. To assess the likely
synergism between vaccines and checkpoint inhibitors,
we have initiated a new trial for patients with metastatic
melanoma receiving standard-of-care checkpoint inhibitors
but who have stable or slowly progressive disease. For
these patients, we will add a personalized TLPLDC vaccine
in combination with their checkpoint inhibitors. The endpoints of the trial are safety and tumor response.
CONCLUSION
Combination Immunotherapy
Exploring combination therapy of passive and active immunotherapies is a growing area of interest. During the
enrollment of the E75 and GP2 trials, the standard of care for
e166
Cancer vaccines represent a nontoxic therapeutic modality

with great antitumor specificity to prime the immune system, but they have yet to be proven and/or gain acceptance
in clinical practice. Based on lessons learned from previous
trials, the most promising use of this therapy is to prevent

recurrence in disease-free patients. Our group has made
strides toward showing the efficacy of this approach with
multiple vaccines/strategies applied to several tumor types.
Additionally, there is potential synergistic clinical benefit of
combination therapy with monoclonal antibodies, which
needs to be further evaluated. Finally, with a better understanding of the tumor microenvironment and the tools to
protect tumor-reactive T cells in this hostile setting, cancer
vaccines may now, largely for the first time, be effective in
patients with metastatic disease if given in combination with

checkpoint inhibitors.
DISCLAIMER
The view(s) expressed herein are those of the author(s) and
do not reflect the official policy or position of San Antonio
Military Medical Center, Womack Army Medical Center, the
U.S. Army Medical Department, the U.S. Army Office of the
Surgeon General, the Department of the Army, Department
of Defense, or the U.S. Government.
References
1. Hale DF, Clifton GT, Sears AK, et al. Cancer vaccines: should we be
targeting patients with less aggressive disease? Expert Rev Vaccines.
2012;11:721-731.
2. Hoover HC Jr, Brandhorst JS, Peters LC, et al. Adjuvant active specific immunotherapy for human colorectal cancer: 6.5-year median follow-up of a
phase III prospectively randomized trial. J Clin Oncol. 1993;11:390-399.
3. Vermorken JB, Claessen AM, van Tinteren H, et al. Active specific
immunotherapy for stage II and stage III human colon cancer: a randomised trial. Lancet. 1999;353:345-350.
4. Harris JE, Ryan L, Hoover HC Jr, et al. Adjuvant active specific immunotherapy for stage II and III colon cancer with an autologous tumor cell
vaccine: Eastern Cooperative Oncology Group Study E5283. J Clin Oncol.
2000;18:148-157.
5. Wood C, Srivastava P, Bukowski R, et al; C-100-12 RCC Study Group. An
adjuvant autologous therapeutic vaccine (HSPPC-96; vitespen) versus
observation alone for patients at high risk of recurrence after nephrectomy for renal cell carcinoma: a multicentre, open-label, randomised phase III trial. Lancet. 2008;372:145-154.
6. Carson WE, III, Unger JM, Sosman JA, et al. Adjuvant vaccine immunotherapy of resected, clinically node-negative melanoma: long-term
outcome and impact of HLA class I antigen expression on overall
survival. Cancer Immunol Res. 2014;2:981-987.
7. Mittendorf EA, Clifton GT, Holmes JP, et al. Final report of the phase I/II
clinical trial of the E75 (nelipepimut-S) vaccine with booster inoculations to prevent disease recurrence in high-risk breast cancer
patients. Ann Oncol. 2014;25(9):1735-1742.
8. Benavides LC, Gates JD, Carmichael MG, et al. The impact of HER2/neu
expression level on response to the E75 vaccine: from U.S. Military
Cancer Institute Clinical Trials Group Study I-01 and I-02. Clin Cancer
Res. 2009;15:2895-2904.
9. Schneble E, Perez S, Murray J, et al. Primary analysis of the prospective,
randomized, phase II trial of GP2+GM-CSF vaccine versus GM-CSF alone
administered in the adjuvant setting to high-risk breast cancer patients.
J Clin Oncol. 2014;32 (suppl 26; abstr 134).
10. Mittendorf EA, Schneble EJ, Perez SA, et al. Primary analysis of the
prospective, randomized, single-blinded phase II trial of AE37 vaccine
versus GM-CSF alone administered in the adjuvant setting to high-risk
breast cancer patients. J Clin Oncol. 2014;32:5s (suppl; abstr 638).
11. Peoples GE, Anderson BW, Lee TV, et al. Vaccine implications of folate
binding protein, a novel cytotoxic T lymphocyte-recognized antigen
system in epithelial cancers. Clin Cancer Res. 1999;5:4214-4223.
12. Peoples GE, Anderson BW, Fisk B, et al. Ovarian cancer-associated
lymphocyte recognition of folate binding protein peptides. Ann Surg
Oncol. 1998;5:743-750.
13. Garin-Chesa P, Campbell I, Saigo PE, et al. Trophoblast and ovarian
cancer antigen LK26. Sensitivity and specificity in immunopathology and
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
molecular identification as a folate-binding protein. Am J Pathol. 1993;

142:557-567.
Li PY, Del Vecchio S, Fonti R, et al. Local concentration of folate binding protein GP38 in sections of human ovarian carcinoma by in vitro
quantitative autoradiography. J Nucl Med. 1996;37:665-672.
Greene J, Schneble E, Berry J, et al. Preliminary results of the phase I/IIa
dose finding trial of a folate binding protein vaccine (E39+GM-CSF) in
ovarian and endometrial cancer patients to prevent recurrence. J Clin
Oncol. 2015;33 (suppl; abstr e14031).
Holmes LM, Li J, Sticca RP, et al. A rapid, novel strategy to induce tumor
cell-specific cytotoxic T lymphocyte responses using instant dentritomas. J Immunother. 2001;24:122-129.
Greene JM, Schneble EJ, Jackson DO, et al. A phase I/IIa clinical trial in
stage IV melanoma of an autologous tumor-dendritic cell fusion
(dendritoma) vaccine with low dose interleukin-2. Cancer Immunol
Immunother. Epub 2016 Feb 19.
Greene J, Hale D, Schneble E, et al. Initial phase I/IIa trial results of an
autologous tumor lysate + yeast cell wall particles + dendritic cells
vaccine (TLPLDC) in patients with solid tumors. Cancer Immunol Res.
2016;4(suppl, abstr A044).
Eggermont AM, Suciu S, Rutkowski P, et al. Adjuvant ganglioside GM2-KLH/
QS-21 vaccination versus observation after resection of primary tumor
. 1.5 mm in patients with stage II melanoma: results of the EORTC 18961
randomized phase III trial. J Clin Oncol. 2013;31:3831-3837.
Vansteenkiste J, Cho B, Vanakesa T, et al. MAGRIT, a double-blinded,
randomized, placebo-controlled phase III study to assess the efficacy
of the recMAGE-A3 + AS15 cancer immunotherapeutic as adjuvant
therapy in patients with resected MAGE-A3-positive non-small cell lung
cancer (NSCLC). Ann Oncol. 2014;25(suppl 4):iv409-iv16.
Giaccone G, Bazhenova LA, Nemunaitis J, et al. A phase III study of
belagenpumatucel-L, an allogeneic tumour cell vaccine, as maintenance therapy for non-small cell lung cancer. Eur J Cancer. 2015;51:
2321-2329.
Morton D, Mozzillo N, Thompson J, et al. An international, randomized,
phase III trial of bacillus Calmette-Guerin (BCG) plus allogeneic melanoma vaccine (MCV) or placebo after complete resection of melanoma
metastatic to regional or distant site. J Clin Oncol. 2007;25:8508.
Schulze T, Kemmner W, Weitz J, et al. Efficiency of adjuvant active
specific immunization with Newcastle disease virus modified tumor
cells in colorectal cancer patients following resection of liver metastases: results of a prospective randomized trial. Cancer Immunol
Immunother. 2009;58:61-69.
Gates JD, Carmichael MG, Benavides LC, et al. Longterm followup assessment of a HER2/neu peptide (E75) vaccine for prevention of recurrence in high-risk prostate cancer patients. J Am Coll Surg. 2009;208:
193-201.
e167
GREGORY L. BEATTY
Overcoming Therapeutic Resistance by Targeting Cancer

Inflammation
Gregory L. Beatty, MD, PhD
OVERVIEW
Tumor-infiltrating myeloid cells are a prominent feature of most solid malignancies. This inflammatory immune response,
driven by tumor-intrinsic signaling pathways, is a major checkpoint to therapeutic efficacy achieved with immunotherapy
and standard cytotoxic therapies. To overcome therapeutic resistance mediated by cancer inflammation, ongoing clinical
trials are evaluating strategies that (1) deplete myeloid cells from tumors, (2) inhibit tumor-promoting properties of myeloid
cells, and (3) redirect myeloid cells with tumor-inhibitory activity.
he microenvironment that surrounds solid tumors is

fundamental to tumor biology and a critical barrier to
therapeutic efficacy. A defining feature of this microenvironment is a leukocyte infiltrate that can vary widely in
its complexity.1 For example, some tumors demonstrate a
robust infiltration of T lymphocytes, whereas others show a
near absence or a spatial restriction of T lymphocytes to the
tumor margin.2-4 In contrast, innate immune cells composed
of macrophages, granulocytes, and immature myeloid cells
are universally seen within tumors, although to varying
degrees.5 In some tumors, this innate immune inflammatory
response can be robust, even representing the majority of
the cellular mass of a tumor lesion.
Although inflammation is a well-recognized proponent of
cancer, the phenotype of tumor-infiltrating inflammatory
cells is pliable and strongly dependent on cues received
from the surrounding microenvironment.6 As such, tumorinfiltrating leukocytes can acquire either pro- or antitumor
properties.7 With a rapidly advancing understanding of this
biology regulating the phenotype and recruitment of inflammatory cells to tumors, novel therapeutic targets have
been identified and are being investigated in early-phase
clinical trials for their potential to enhance current standard
therapies (e.g., radiation and chemotherapy) and to derail
resistance mechanisms to the success of T celldependent
immunotherapies.
INFLAMMATION: A HALLMARK OF CANCER

Although the immunosurveillance hypothesis proposed
by Thomas and Burnet in 1957 postulated a critical role for
the immune system in controlling the development and
outgrowth of nascent transformed cells,8,9 it has become

increasingly clear that the immune system, in addition to its
potential to be harnessed for antitumor activity, can be a
major proponent of tumor development and a barrier to
therapeutic efficacy.5,10 This paradoxic role of the immune
system in cancer can be explained by the plasticity of the
immune system, which can acquire either pro- or antitumor
properties. For productive immunosurveillance to eliminate
cancer, components of innate and adaptive immunity must
be aligned or else peripheral tolerance will ensue.11 In many
settings, alignment is not achieved because leukocytes
recruited to tumors most often orchestrate an immunosuppressive, rather than an immunostimulatory, microenvironment that acts to spoil the antitumor potential of
the immune system.12 In preclinical models, elimination
of this myeloid response by inhibiting ICAM-1mediated
macrophage recruitment to tumors early during cancer
initiation can block tumor development.13 Similarly,
blockade of colony-stimulating factor 1/receptor colonystimulating factor 1 (CSF-1/CSF-1R) signaling, a key signaling
pathway for macrophages, can improve the response to
immunotherapy with PD-1 and CTLA-4 antagonists.14
Tumor-infiltrating inflammatory cells, including macrophages, dendritic cells, granulocytes, and immature myeloid cells, are commonly seen across a vast majority of solid
tumor types. These myeloid cell subsets are recruited to
tumors by a variety of chemoattractants released by both
malignant cells and nonmalignant cells residing within the
tumor microenvironment. Myeloid cells that infiltrate tumors are primarily thought to originate from progenitors in
the bone marrow and are recruited to tumor tissue via the
From the Abramson Cancer Center of the University of Pennsylvania, Philadelphia, PA; Division of Hematology-Oncology, Department of Medicine, Perelman School of Medicine,
University of Pennsylvania, Philadelphia, PA.
Corresponding author: Gregory L. Beatty, MD, PhD, Abramson Cancer Center of the University of Pennsylvania, Smilow Center for Translational Research, Room 8-112, Bldg 421, 3400
Civic Center Blvd., Philadelphia, PA 19104-5156; email: gregory.beatty@uphs.upenn.edu.
e168
OVERCOMING THERAPEUTIC RESISTANCE BY TARGETING CANCER INFLAMMATION
bloodstream.15 However, the spleen may also be an abundant

source of tumor-infiltrating myeloid cells.16 In contrast, the
contribution to the tumor microenvironment by tissueresident myeloid cells, which are derived from yolk-sac
progenitors, remains ill defined.17,18
The recruitment of myeloid cells to the tumor microenvironment is dependent on chemoattractants, such as
chemokine:chemokine receptor interactions.19 For example,
the chemokine CCL2 regulates myeloid trafficking through
interaction with chemokine receptor CCR2, which is
expressed at high levels on a subset of monocytes, commonly referred to as inflammatory or classical monocytes. In some cases, a single chemokine may bind to
multiple chemokine receptors. For example, CCL5 has
binding affinity for chemokine receptors CCR1, CCR3, and
CCR5. Similarly, a single chemokine receptor may also bind
multiple chemokines. For example, chemokine receptor
CCR5 can bind both CCL3 and CCL5. Together, this complexity associated with expression of chemokine receptors
on distinct myeloid subsets and the interaction of chemokine receptors with select chemokines determines, at least
in part, the heterogeneity of the myeloid infiltration into
tumors.
Both malignant and nonmalignant cells can release chemokines within the tumor microenvironment that act to
regulate the infiltration of myeloid cells. For example, in
preclinical models of breast cancer, CCL2 released by
both malignant and nonmalignant cells is required for
recruitment of CCR2+ monocytes to promote cancer
metastasis.20 However, the importance of tumor genetics
in defining the chemokine milieu within tumors is only
KEY POINTS
Tumor-infiltrating myeloid cells, including macrophages,

dendritic cells, granulocytes, and immature myeloid
cells, are a hallmark of solid malignancies.
Tumor-intrinsic oncogenic signaling pathways (e.g.,
Kras, PTEN/Akt/PI3K, WNT/b-catenin) shape the
composition, phenotype, and function of tumorinfiltrating myeloid cells with implications for defining
tumor immunogenicity.
The pro- versus antitumor biology of myeloid cell
subsets within solid tumors is pliable and determined by
microenvironmental signals released by malignant and
nonmalignant cells.
Strategies to intervene on the myeloid response to
cancer include therapeutic approaches designed to (1)
deplete myeloid cells from the tumor
microenvironment, (2) inhibit the tumor-promoting
properties of myeloid cells, and (3) redirect tumorinfiltrating myeloid cells with tumor-inhibitory
properties.
Therapeutic manipulation of tumor-infiltrating myeloid
cells holds promise for enhancing the efficacy of T-cell
immunotherapy and standard cytotoxic therapies.
recently being unraveled. For example, activation of Kras in

pancreatic cancer induces the secretion of chemokines, such
as interleukin (IL)-8,21 and myeloid growth factors, including
granulocyte-macrophage colony stimulating factor,22 which
can promote the recruitment and differentiation of myeloid
cells with protumor activities.23 PTEN loss in preclinical
models of pancreatic cancer has also been associated
with a robust infiltration of inflammatory cells.24 In addition, PTEN loss in melanoma has been found to correlate
with T-cell exclusiona finding that was associated with an
increase in proinflammatory cytokines including CCL2 and
vascular endothelial growth factor.25 In this study, neutralization of vascular endothelial growth factor enhanced
T-cell infiltration into tumors, and pharmacologic inhibition
of the PI3K pathway improved the effectiveness of antiPD1
immunotherapy. Similar findings have been reported with
activation of the b-catenin pathway, which was shown to
regulate the expression of tumor-derived chemokines,
specifically CCL4.26 Here, CCL4 was important for the recruitment of dendritic cells involved in T-cell priming. This
emerging role for tumor-intrinsic signaling pathways in
regulating the recruitment of inflammatory cells has also
been observed in models of prostate adenocarcinoma,
where loss of Smad4 leads to hyperactivation of Hippo-YAP
signaling, which induces CXCL5 upregulation in cancer cells
and subsequent recruitment of immunosuppressive CXCR2+
CD11b+ Gr-1+ myeloid cells.27 Together, these studies suggest the importance of tumor genetics in shaping the inflammatory immune response to cancer.
The phenotype of myeloid cells within tumors can be quite
variable and probably best described as a spectrum ranging from pro- to antitumor based on functional properties
(e.g., T cellsuppressive activity, cytokine production, tumoricidal activity, pro- versus antiangiogenic properties, etc.).7
However, a diverse array of factors (e.g., extracellular matrix,
hypoxia, cytokines, and other soluble factors) can influence
the phenotype of a myeloid cell. As a result, the complexity
of the myeloid response to cancer has remained a challenge
to describe in vivo. Many attempts have been made to
categorize tumor-infiltrating myeloid cells based on protein
and molecular signatures determined under in vitro culture
conditions. For example, macrophages studied under distinct in vitro culture conditions with interferon gamma and
lipopolysaccharide or IL-4 and IL-13 can be classified as M1
or M2 macrophages, respectively.28 However, almost invariably when macrophages are obtained from in vivo
settings, they more commonly display a mixture of an M1
and M2 phenotype, thereby diluting the potential relevance
of this simplistic approach.29
In cancer, the myeloid cell-differentiation program becomes altered from the earliest stages of myelopoiesis,30
occurring within the bone marrow, through myeloid cell
transit in the peripheral blood to infiltration into the tumor
microenvironment.31 At each step along this lifecycle of a
tumor-infiltrating myeloid cell, soluble factors released in
response to cancer have been identified that shape myeloid
cell biology. For example, in pancreatic cancer, mobilization
e169
GREGORY L. BEATTY
of CCR2+ monocytes from the bone marrow into the peripheral blood is enhanced and correlates with elevated
serum levels of CCL2.32 Within the peripheral blood, human
monocytes are then exposed to serum-soluble factors that
can modulate monocyte biology. For example, in patients
with renal cell carcinoma, human blood monocytes were
found to be induced with pro-angiogenic properties by the
IL-1IL-1R cytokine axis even before entry into tumor tissue.33 Once in the tumor microenvironment, tumor-derived
soluble factors then further shape the ultimate fate of
tumor-infiltrating myeloid cells. However, because the
secretome of malignant cells can vary between tumor-type,
the cellular fate of myeloid cells is tumor-specific and likely
differentially skewed between cancers and perhaps even
between lesions within the same patient.34,35
TARGETING INFLAMMATION FOR THERAPEUTIC

BENEFIT IN CANCER
Changes in inflammatory cell biology that occur in cancer can
affect the efficacy of standard treatments such as chemotherapy. In response to cytotoxic stress, tumor cells secrete
chemokines that can recruit myeloid cells, which may then
suppress the efficacy of chemotherapy and/or inhibit the
development of antitumor T celldependent immunity.10,36
Tumor-infiltrating inflammatory cells can also directly suppress T-cell antitumor immunity by producing immunosuppressive cytokines and depleting critical amino acids,
including tryptophan and arginine, which are necessary for
normal T-cell activity.37-39 For these reasons, strategies to
intervene on cancer-induced inflammation are actively
being investigated in early-phase clinical trials.
There are three potential approaches (Table 1) to altering
the inflammatory response to cancer including:
1. blocking inflammatory cell recruitment to tumors,
2. inhibiting tumor-promoting activities of myeloid cells,
and
3. redirecting tumor-infiltrating myeloid cells with antitumor activity.
Blocking Recruitment
Strategies to block inflammatory cell recruitment to tumors
involve the neutralization of chemokines with antibodies,
the delivery of small-molecule inhibitors to block chemokine
receptor signaling, and selective depletion of myeloid
subsets. For example, neutralizing antibodies against CCL2
(e.g., carlumab) and antagonists of CCR2 (e.g., PF-04136309
and CCX872) are being evaluated in early phase 1 studies in patients with advanced solid malignancies to block
monocyte recruitment to tumors. Similarly, antagonists of
CXCR1/2 (e.g., AZD5069 and reparixin) are being studied in
early-phase clinical trials to inhibit granulocyte recruitment to
tumors. To deplete myeloid cell populations, ongoing clinical
investigations are testing antagonists of CSF1R (PLX3397,
ARRY-382, IMC-CS4, FPA008, AMG820, and RO5509554)
and a chemotherapeutic agent, trabectedin, which was
found to selectively induce capsase-8dependent apoptosis
in monocytes via signaling through tumor necrosis factor
related apoptosis-inducing ligand receptor 2.40
Inhibiting Activity
Tumor-infiltrating myeloid cells are active components of
the tumor microenvironment. Within tumors, myeloid cells
shape the supply of nutrients important for T-cell activity
(e.g., tryptophan and arginine) and release soluble factors
(e.g., IL-6, IL-10, and transforming growth factor beta) that
can be immunosuppressive and supportive of tumor
survival.37,41-43 Ongoing studies are evaluating strategies to
inhibit indoleamine 2,3-dioxygenase (IDO), a key enzyme
expressed by myeloid cells and important for tryptophan
metabolism in tumors. In preclinical models, IDO inhibition
was shown to promote antitumor T-cell activity,37 and in
clinical trials, IDO inhibition has demonstrated safety and
tolerability.44,45 Similarly, early-phase clinical trials are underway to evaluate the safety and preliminary efficacy of
inhibiting distinct signaling pathways in myeloid cells, such
as JAK/STAT signaling, which is important for the biologic
activity of cytokines, including IL-6 and IL-10 (Table 1).
TABLE 1. Therapeutic Strategies to Cancer Inflammation in Solid Malignancies

Approach to Cancer Inflammation
Target
Agents
Block Infiltration or Deplete Myeloid Subsets
CCL2/CCR2
Carlumab, PF-04136309, CCX872
CXCR1/CXCR2
AZD5069, reparixin
CSF-1/CSF-1R
PLX3397, ARRY-382, IMC-CS4, FPA008, AMG820, RO5509554
TRAIL-R2
Trabectedin
IDO
Epacadostat/INCB024360, indoximod, NLG919, 1-methyl-D-tryptophan
Inhibit Protumor Activity

Redirect With Antitumor Activity
JAK
Ruxolitinib, pacritinib, momelotinib, INCB039110
CD40
RO7007789/CP-870,893, APX005M, ADC-1013, Chi Lob 7/4, SEA-CD40
CSF-1R
PLX3397, ARRY-382
CD47-Sirpa
CC-90002, TTI-621
Abbreviations: CCL2, chemokine (C-C motif) ligand 2; CCR2, chemokine (C-C motif) receptor 2; CXCR1, chemokine (C-X-C motif) receptor 1; CXCR2, chemokine (C-X-C motif) receptor 2;
CSF-1, colony-stimulating factor 1; CSF-1R, colony-stimulating factor 1 receptor; TRAIL-R2, tumor necrosis factorrelated apoptosis-inducing ligand receptor 2; IDO, indoleamine 2,3
dioxygenase; JAK, Janus kinase; Sirpa, signal regulatory protein alpha.
e170
TABLE 2. Targeting Cancer Inflammation in Combination With Checkpoint Immunotherapy

Therapeutic Approach
Targets
Agents
Block Infiltration or Deplete Myeloid Subsets in Combination With Checkpoint Inhibition
CXCR2, PD-L1
AZD5069, MEDI4736
CSF-1R, PD-1
FPA008, nivolumab
Inhibit Inflammatory Signaling Pathways in Combination With Checkpoint Blockade
Redirect Inflammation in Combination With Checkpoint Blockade
CSF-1R, PD-L1
RO5509554, MPDL3280A
IDO, PD-1
INCB024360, nivolumab
IDO, PD-1
INCB024360, pembrolizumab
IDO, PD-L1
INCB024360, MEDI4736
IDO, CTLA4
INCB024360, ipilimumab
JAK, PD-1
INCB039110, pembolizumab
CD40, PD-L1
RO7007789, MPDL3280A
CSF1R, PD-1
PLX3397, pembrolizumab
Abbreviations: CXCR2, chemokine (C-X-C motif) receptor 2; CSF-1, colony-stimulating factor 1; CSF-1R, colony-stimulating factor 1 receptor; IDO, indoleamine 2,3 dioxygenase; CTLA4,
cytotoxic T-lymphocyteassociated protein 4; JAK, Janus kinase.
Redirecting Activity
The biology of tumor-infiltrating myeloid cells is pliable
and therefore may potentially be redirected from pro- to
antitumor or, alternatively, from immunosuppressive to
immunostimulatory. This premise forms the rationale for
therapeutic approaches designed to unleash the antitumor
potential of tumor-infiltrating myeloid cells. For example,
ongoing clinical studies are evaluating antagonists of the
CD47-Sirpa pathway, which delivers inhibitory signals to
macrophages and blocks macrophage phagocytosis of tumor
cells. In preclinical studies, CD47 antagonists have demonstrated the capacity to induce macrophage-dependent antitumor activity46 and promote dendritic cellmediated
induction of T-cell antitumor immunity. 47 An alternative
approach to redirecting myeloid cells with antitumor
properties is to block signaling pathways (e.g., CSF-1), which
can instill macrophages with protumor functions.48 For
example, although antibody-based strategies that target
CSF-1R (e.g., IMC-CS4, FPA008, AMG820, RO5509554) may
deplete myeloid cells via antibody-dependent cellular
cytotoxicity,49 small-molecule inhibitors of CSF-1R (e.g.,
PLX3397, ARRY-382) have the potential for shifting the
polarity of myeloid cells from pro- to antitumor.48 Finally,
CD40 agonists have demonstrated the capacity to redirect
tumor-infiltrating myeloid cells with both tumoricidal and
antifibrotic activity.50,51 In recent years, several CD40 agonists have been developed and are now undergoing clinical
investigation (Table 1).
Each of the aforementioned strategies for modulating cancer
inflammation is being evaluated in clinical trials. Although
merely manipulating the myeloid response to cancer is unlikely
by itself, to produce long-lasting and major tumor regressions,
these approaches hold promise in combination with cytotoxic

therapies and immunotherapy. To this end, multiple clinical
trials are underway investigating modulation of cancer inflammation in combination with immunotherapy (e.g., PD-1/
PD-L1 blockade; Table 2), chemotherapy, and radiation. These
studies will help to inform whether distinct strategies designed
to manipulate the myeloid cell response to cancer can complement existing therapeutic approaches.
CONCLUSION
Tumor-infiltrating myeloid cells are a prominent feature
of most solid malignancies and most often portend a poor
prognosis. However, the impact of myeloid cells on tumor
development, progression, and growth can vary substantially, which is explained by their inherent cellular
plasticity and dependence on local microenvironmental
cues, which determine their ultimate cellular fate. Tumorintrinsic oncogenic signaling pathways have emerged as a
major determinant of the complexity of the myeloid cell
infiltrate in tumorsa finding that has strong implications
in understanding resistance mechanisms to the efficacy of
cytotoxic and immunotherapeutic strategies. Based on
strong preclinical rationale, ongoing clinical trials are
investigating strategies designed to deplete, inhibit, and
redirect myeloid cells in cancer for therapeutic benefit.
ACKNOWLEDGMENT
This work was supported by a National Institutes of Health
grant K08 CA138907 and by grant 2013107 from the Doris
Duke Charitable Foundation.
References
1. Coussens LM, Zitvogel L, Palucka AK. Neutralizing tumor-promoting
chronic inflammation: a magic bullet? Science. 2013;339:286-291.
2. Llosa NJ, Cruise M, Tam A, et al. The vigorous immune microenvironment of microsatellite instable colon cancer is balanced by multiple
counter-inhibitory checkpoints. Cancer Discov. 2015;5:43-51.
3. Galon J, Costes A, Sanchez-Cabo F, et al. Type, density, and location of

immune cells within human colorectal tumors predict clinical outcome.
Science. 2006;313:1960-1964.
4. Gajewski TF, Schreiber H, Fu YX. Innate and adaptive immune cells in the
tumor microenvironment. Nat Immunol. 2013;14:1014-1022.
e171
GREGORY L. BEATTY
5. Qian BZ, Pollard JW. Macrophage diversity enhances tumor progression

and metastasis. Cell. 2010;141:39-51.
6. Lavin Y, Winter D, Blecher-Gonen R, et al. Tissue-resident macrophage
enhancer landscapes are shaped by the local microenvironment. Cell.
2014;159:1312-1326.
7. Long KB, Beatty GL. Harnessing the antitumor potential of macrophages
for cancer immunotherapy. OncoImmunology. 2013;2:e26860.
8. Burnet FM. The concept of immunological surveillance. Prog Exp Tumor
Res. 1970;13:1-27.
9. Thomas L. On immunosurveillance in human cancer. Yale J Biol Med.
1982;55:329-333.
10. DeNardo DG, Brennan DJ, Rexhepaj E, et al. Leukocyte complexity
predicts breast cancer survival and functionally regulates response to
chemotherapy. Cancer Discov. 2011;1:54-67.
11. Mellor AL, Munn DH. Creating immune privilege: active local suppression that benefits friends, but protects foes. Nat Rev Immunol.
2008;8:74-80.
12. Beatty GL, Gladney WL. Immune escape mechanisms as a guide for
cancer immunotherapy. Clin Cancer Res. 2015;21:687-692.
13. Liou GY, Doppler H, Necela B, et al. Mutant KRAS-induced expression of
ICAM-1 in pancreatic acinar cells causes attraction of macrophages to
expedite the formation of precancerous lesions. Cancer Discov. 2015;5:
52-63.
14. Zhu Y, Knolhoff BL, Meyer MA, et al. CSF1/CSF1R blockade reprograms
tumor-infiltrating macrophages and improves response to T-cell
checkpoint immunotherapy in pancreatic cancer models. Cancer Res.
2014;74:5057-5069.
15. Franklin RA, Liao W, Sarkar A, et al. The cellular and molecular origin of
tumor-associated macrophages. Science. 2014;344:921-925.
16. Cortez-Retamozo V, Etzrodt M, Newton A, et al. Origins of tumorassociated macrophages and neutrophils. Proc Natl Acad Sci USA. 2012;
109:2491-2496.
17. Ostuni R, Kratochvill F, Murray PJ, et al. Macrophages and cancer: from
mechanisms to therapeutic implications. Trends Immunol. 2015;36:
229-239.
18. Gomez Perdiguero E, Klapproth K, Schulz C, et al. Tissue-resident
macrophages originate from yolk-sac-derived erythro-myeloid progenitors. Nature. 2015;518:547-551.
19. Bonecchi R, Locati M, Mantovani A. Chemokines and cancer: a fatal
attraction. Cancer Cell. 2011;19:434-435.
20. Qian BZ, Li J, Zhang H, et al. CCL2 recruits inflammatory monocytes to
facilitate breast-tumour metastasis. Nature. 2011;475:222-225.
21. Sparmann A, Bar-Sagi D. Ras-induced interleukin-8 expression plays a
critical role in tumor growth and angiogenesis. Cancer Cell. 2004;6:
447-458.
22. Pylayeva-Gupta Y, Lee KE, Hajdu CH, et al. Oncogenic Kras-induced GMCSF production promotes the development of pancreatic neoplasia.
Cancer Cell. 2012;21:836-847.
23. Bayne LJ, Beatty GL, Jhala N, et al. Tumor-derived granulocytemacrophage colony-stimulating factor regulates myeloid inflammation and T cell immunity in pancreatic cancer. Cancer Cell. 2012;
21:822-835.
24. Ying H, Elpek KG, Vinjamoori A, et al. PTEN is a major tumor suppressor
in pancreatic ductal adenocarcinoma and regulates an NF-kB-cytokine
network. Cancer Discov. 2011;1:158-169.
25. Peng W, Chen JQ, Liu C, et al. Loss of PTEN Promotes Resistance to T CellMediated Immunotherapy. Cancer Discov. 2016;6:202-216.
26. Spranger S, Bao R, Gajewski TF. Melanoma-intrinsic b-catenin
signalling prevents anti-tumour immunity. Nature. 2015;523:
231-235.
27. Wang G, Lu X, Dey P, et al. Targeting YAP-Dependent MDSC Infiltration
Impairs Tumor Progression. Cancer Discov. 2016;6:80-95.
e172
28. Mantovani A, Sica A, Locati M. Macrophage polarization comes of age.

Immunity. 2005;23:344-346.
29. Murray PJ, Allen JE, Biswas SK, et al. Macrophage activation and polarization: nomenclature and experimental guidelines. Immunity. 2014;
41:14-20.
30. Redente EF, Orlicky DJ, Bouchard RJ, et al. Tumor signaling to the bone
marrow changes the phenotype of monocytes and pulmonary macrophages during urethane-induced primary lung tumorigenesis in A/J
mice. Am J Pathol. 2007;170:693-708.
31. Sica A, Bronte V. Altered macrophage differentiation and immune
dysfunction in tumor development. J Clin Invest. 2007;117:1155-1166.
32. Sanford DE, Belt BA, Panni RZ. M, et al. Inflammatory monocyte mobilization decreases patient survival in pancreatic cancer: a role for
targeting the CCL2/CCR2 axis. Clin Cancer Res. 2013;19:3404-3415.
33. Chittezhath M, Dhillon MK, Lim JY, et al. Molecular profiling reveals a
tumor-promoting phenotype of monocytes and macrophages in human
cancer progression. Immunity. 2014;41:815-829.
34. Lundholm M, Hagglof C, Wikberg ML, et al. Secreted Factors from
Colorectal and Prostate Cancer Cells Skew the Immune Response in
Opposite Directions. Sci Rep. 2015;5:15651.
35. Bogels M, Braster R, Nijland PG, et al. Carcinoma origin dictates differential skewing of monocyte function. OncoImmunology. 2012;1:
798-809.
36. Qian DZ, Rademacher BL, Pittsenbarger J, et al. CCL2 is induced by
chemotherapy and protects prostate cancer cells from docetaxelinduced cytotoxicity. Prostate. 2010;70:433-442.
37. Liu X, Shin N, Koblish HK, et al. Selective inhibition of IDO1 effectively
regulates mediators of antitumor immunity. Blood. 2010;115:
3520-3530.
38. Bronte V, Zanovello P. Regulation of immune responses by L-arginine
metabolism. Nat Rev Immunol. 2005;5:641-654.
39. Kortylewski M, Kujawski M, Wang T, et al. Inhibiting Stat3 signaling in
the hematopoietic system elicits multicomponent antitumor immunity.
Nat Med. 2005;11:1314-1321.
40. Germano G, Frapolli R, Belgiovine C, et al. Role of macrophage targeting
in the antitumor activity of trabectedin. Cancer Cell. 2013;23:249-262.
41. Lesina M, Kurkowski MU, Ludes K, et al. Stat3/Socs3 activation by IL-6
transsignaling promotes progression of pancreatic intraepithelial
neoplasia and development of pancreatic cancer. Cancer Cell. 2011;19:
456-469.
42. Spranger S, Spaapen RM, Zha Y, et al. Up-regulation of PD-L1, IDO, and T
(regs) in the melanoma tumor microenvironment is driven by CD8(+)
T cells. Sci Transl Med. 2013;5:200ra116.
43. Uyttenhove C, Pilotte L, Theate I, et al. Evidence for a tumoral immune
resistance mechanism based on tryptophan degradation by indoleamine 2,3-dioxygenase. Nat Med. 2003;9:1269-1274.
44. Soliman HH, Jackson E, Neuger T, et al. A first in man phase I trial of the
oral immunomodulator, indoximod, combined with docetaxel in patients with metastatic solid tumors. Oncotarget. 2014;5:8136-8146.
45. Beatty GL, ODwyer PJ, Clark J, et al. Phase I study of the safety,
pharmacokinetics (PK), and pharmacodynamics (PD) of the oral inhibitor of indoleamine 2,3-dioxygenase (IDO1) INCB024360 in patients
(pts) with advanced malignancies. J Clin Oncol. 2013;31(suppl; abstr
3025).
46. Weiskopf K, Ring AM, Ho CC, et al. Engineered SIRPa variants as immunotherapeutic adjuvants to anticancer antibodies. Science. 2013;
341:88-91.
47. Liu X, Pu Y, Cron K, et al. CD47 blockade triggers T cell-mediated destruction of immunogenic tumors. Nat Med. 2015;21:1209-1215.
48. Pyonteck SM, Akkari L, Schuhmacher AJ, et al. CSF-1R inhibition alters
macrophage polarization and blocks glioma progression. Nat Med.
2013;19:1264-1272.
49. Ries CH, Cannarile MA, Hoves S, et al. Targeting tumor-associated

macrophages with anti-CSF-1R antibody reveals a strategy for cancer
therapy. Cancer Cell. 2014;25:846-859.
50. Beatty GL, Torigian DA, Chiorean EG, et al. A phase I study of an agonist
CD40 monoclonal antibody (CP-870,893) in combination with gemcitabine
in patients with advanced pancreatic ductal adenocarcinoma. Clin Cancer

Res. 2013;19:6286-6295.
51. Beatty GL, Chiorean EG, Fishman MP, et al. CD40 agonists alter tumor
stroma and show efficacy against pancreatic carcinoma in mice and
humans. Science. 2011;331:1612-1616.
e173

Pharmacokinetics, Dynamics,
and Genomics in the Era of
Immunotherapy and Small
Molecules
CHAIR
Emiliano Calvo, MD, PhD
START Madrid, Centro Integral Oncologico
Clara Campal
Madrid, Spain
SPEAKERS
Christine Walko, PharmD, BCOP
Tampa, FL
E. Claire Dees, MD
Chapel Hill, NC
PHARMACOGENOMICS, PHARMACOKINETICS, AND PHARMACODYNAMICS OF TARGETED THERAPIES
Pharmacogenomics, Pharmacokinetics, and

Pharmacodynamics in the Era of Targeted Therapies
Emiliano Calvo, MD, PhD, Christine Walko, PharmD, FCCP, BCOP, E. Claire Dees, MD, and Belen Valenzuela,
PharmD, PhD
OVERVIEW
The complex nature of the pharmacologic aspects of cancer therapeutics has become more apparent in the past several years
with the arrival of a cascade of target-based agents and the difficult challenge of bringing individualized precision medicine to
oncology. Interpatient variability in drug action, singularly in novel agents, is in part caused by pharmacogenomic (PG),
pharmacokinetic, and pharmacodynamic (PD) factors, and drug selection and dosing should take this into consideration to
optimize the benefit for our patients in terms of antitumor activity and treatment tolerance. In this regard, somatic genetic
evaluation of tumors is useful in not only predicting response to initial targeted therapies but also in anticipating and guiding
therapy after the development of acquired resistance; therapeutic drug monitoring of novel small molecules and
monoclonal antibodies must be incorporated in our day-to-day practice to minimize the negative effect on clinical outcome
of interindividual variability on pharmacokinetic processes of these drugs for all patients, but especially for fragile patient
populations and those with organ dysfunction or comorbidities. For these populations, incorporating frailty assessment
tools into trials of newer agents and validating frailty-based dose adjustment should be an important part of further drug
development.
ver the past several years, we have witnessed a dramatic and encouraging burst of newly approved anticancer therapeutics. The treatment of cancer has continued
to shift from traditional cytotoxic therapies aimed at inhibiting
the growth of fast-dividing cells toward targeting a particular
cellular pathway that appears to be activated in a patients
particular cancer. The concept of individualized therapy
embraces the idea that each patient and his or her cancer
both have unique properties, including how the drug will be
handled in the body and whether it will be effective against
the patients cancer. Many of these new drugs are targeted
agents, small molecules, and monoclonal antibodies with
unique biologic effects and challenges in optimizing their clinical
use include variability in pharmacogenomics, pharmacokinetics,
and pharmacodynamics among patients with cancer.
Pharmacogenomic analysis of tumor tissue is an important
tool in identifying therapies that will be effective against
certain cancers while also preventing patients for whom
these therapies are unlikely to be effective from suffering
the physical and economic toxicities inherent to each
treatment. Tyrosine kinase inhibitors (TKIs) and monoclonal
antibodies are administered at fixed-flat or weight-based
doses, and frequently, they are not the right dose for the
majority of patients. In fact, most failures in phase III1 trials

are related to a lack of efficacy that can be attributed to
underdosing situations, because interindividual variability
in pharmacokinetic processes plays a critical role in drug
exposure variability and, consequently, in the clinical outcome
observed. In addition, dosing anticancer chemotherapy in
elderly individuals and frail patients has long been a problem in
oncology. Because the pivotal clinical trials of most anticancer
chemotherapeutics included only patients with normal organ
function and excellent performance status, we have limited
information in dosing a more real-world population.
Therefore, we review some of the data available on the variability in drug action, particularly in newer agents, caused by
genomic, kinetic, and dynamic factors and how dosing should be
approached in the context of individualized precision medicine.
ROLE OF PHARMACOGENOMICS IN THE

MOLECULAR ERA
The process of dissecting the heterogeneity of an individual
patients cancer begins as soon as a sample of the malignancy
is able to be assessed and subjected to a gamut of staining,
immunohistochemistry, fluorescence in situ hybridization,
From the DeBartolo Family Personalized Medicine Institute, H. Lee Moffitt Cancer Center, Tampa, FL; UNC Lineberger Comprehensive Cancer Center, Chapel Hill, NC; Platform of
Oncology, Hospital Quiron,
Torrevieja, Alicante, Spain; START Madrid, Early Clinical Drug Development Program, Centro Integral Oncologico
Clara Campal, Madrid, Spain.

~a, 10, 28050 Madrid,
Corresponding author: Emiliano Calvo, MD, PhD, START Madrid, Early Clinical Drug Development Program, Centro Integral Oncologico
Clara Campal, Calle On

Spain; email: emiliano.calvo@start.stoh.com.
e175
CALVO ET AL
and cytogenetics, as well as genetic and methylation assays in some instances. All of these provide both prognostic and predictive information used to begin to narrow
down specific treatment options and expected outcomes.
The discovery of specific biomarkers has been essential for
drug development. For example, the initial trial of panitumumab and best supportive care (BSC) compared with BSC
alone among patients with metastatic colorectal cancer that
had progressed after standard therapy demonstrated
an improvement in median progression-free survival (mPFS)
from 7.3 weeks to 8 weeks in the BSC and panitumumab
groups, respectively. In comparison, mean progression-free
survival (PFS) was 8.5 weeks and 13.8 weeks for the BSC
and panitumumab groups, respectively, suggesting the existence of a subset of patients who responded better to the
panitumumab.2 Unlike the story of the HER2-inhibitor trastuzumab, overexpression of EGFR was not the predictive
biomarker for the EGFR inhibitor. Rather, mutations in the
Kirsten rat sarcoma (KRAS) oncogene downstream of EGFR
were predictive of responses to these inhibitors in colorectal
cancer, as evidenced by a secondary analysis of the same trial
showing KRAS mutations in 43% of the patients and correlation with mPFS. For patients with wild-type KRAS, panitumumab resulted in mPFS of 12.3 weeks compared with
7.3 weeks with BSC, whereas patients with KRAS mutations
had similar mPFS regardless of treatment (7.4 weeks compared with 7.3 weeks in the panitumumab and BSC groups,
respectively).3 Based on these and additional findings, a newly
diagnosed metastatic colorectal cancer is standardly assessed
for the alterations in KRAS and NRAS and this helps to determine whether EGFR-directed therapies such as cetuximab
KEY POINTS
e176
Precision medicine in oncology must take into account

interpatient variability in drug action, particularly in
newer agents, caused by genomic, kinetic, and dynamic
factors.
Pharmacogenomics of somatic tumors, through genetic
analyses of solid samples and liquid biopsies of the
disease, relates not only to initial selection of therapy
but also to sequencing of therapy options.
Interindividual variability on pharmacokinetic processes
is a definite source of imprecision in the use of targeted
drugs, with profound clinical consequences, especially in
frail patients, that can be decreased through therapeutic
drug monitoring.
Measuring frailty might become an important focus of
oncology research, because some of these tools
accurately predict toxicity and may prompt
management change.
To continue to establish genetic biomarkers and
therapeutic drug monitoring of novel therapies, we
must evolve the designs of our clinical trials, make them
as inclusive as possible, and facilitate access to patients
willing to enroll in them.
or panitumumab will be useful in the management of the

tumor.4 This emphasizes not only finding the right biomarker
for each agent but also ensuring that the selected patient
sample will assess the value of this biomarker. Ideal predictive
biomarker trials should include two comparison treatment
groups and should demonstrate that the outcome is different
for patients with biomarker-positive results compared with
those with biomarker-negative results.5
The role of somatic mutation testing has continued to
evolve past the initial diagnosis of tumors to help prioritize
and guide additional treatment options. Improved understanding of cancer biology and drug pharmacology has
allowed for expanded investigation into novel therapy
options. Greater understanding of the role of a mutated protein in a process essential to cancer survival, such as DNA
repair, has supported the interrogation of alterations in genes
involved in this process in terms of response to DNA-repair
inhibitors like the PARP inhibitor, olaparib. PARP inhibitors
were initially developed with the hope of taking advantage
of synthetic lethality for patients deficient in the tumor suppressor genes BRCA1 and BRCA2. PARP is required for baseexcision repair of damaged DNA, whereas BRCA is involved
in homologous repair. In a BRCA1- or BRCA2-deficient tumor,
PARP is able to compensate for this loss of homologous repair;
thus, inhibition of PARP would ultimately result in cell death.6
The PARP inhibitor olaparib was initially approved by the U.S.
Food and Drug Administration (FDA) in December 2014 for
patients with ovarian cancer with defective BRCA genes. Numerous other mutations can occur in genes involved in DNA
repair and are common in cancers, including prostate cancer.
The clinical efficacy of olaparib in metastatic castrationresistant prostate cancer was assessed among 16 patients
with homologous deletions, deleterious mutations, or both in
DNA-repair genes. These genes included ATM, FANCA, CHEK2,
PALB2, and others, including BRCA1 and BRCA2. A response
was seen for 88% of patients, including four of five patients
with ATM alterations.7 Ongoing trials with olaparib and investigational PARP inhibitors are underway for patients with a
variety of tumor types with DNA-repair gene alterations.
As our knowledge about cancer biology grows, so too
does the cancers ability to overcome these novel therapies.
Somatic genetic evaluation of tumors is useful in not only
predicting response to initial targeted therapies but also
in anticipating and directing therapy after the development
of acquired resistance. EGFR-activating mutations were
identified in 2004 and are found for approximately 10%
of patients with nonsmall cell lung cancer (NSCLC) with
adenocarcinoma histology. The majority of these EGFRactivating mutations occur in exons 18 to 21 and predict
response to EGFR TKIs including erlotinib, gefitinib, and
afatinib. Median PFS with these agents is approximately
9 to 14 months when resistance develops. The EGFR
T790M mutation is responsible for approximately 60% of the
acquired resistance and prevents TKI binding through steric
hindrance and increased EGFR affinity for ATP.8,9 Novel antiEGFR TKIs are being developed to overcome this specific
mutation and include the recently approved osimertinib,
which has affinity for both activating mutations in EGFR and

the T790M. The use of serial sequencing is necessary for
depiction of this mutational landscape. The value of this
sequential genetic analysis and integration with correlative
science is illustrated by a case report of a patient with NSCLC
harboring an anaplastic lymphoma kinase (ALK) rearrangement. The patient was initially started on crizotinib and
developed resistance to the agent after 18 months. Genetic
analysis at this time showed an ALK C1156Y mutation known
to confer resistance to crizotinib. The patient was then treated
with ceritinib and other non-ALK directed therapies until she
was ultimately enrolled in a clinical trial with a novel ALK inhibitor, lorlatinib. The patient initially had a 41% decrease in
her tumor and a response that lasted for 8 months. After
progressive disease, a subsequent genetic analysis of the tumor showed the C1156Y mutation but also a new L1198F
mutation. This later mutation results in resistance to lorlatinib
but paradoxically enhances response to the original crizotinib,
ultimately negating the C1156Y mutation.10 This was elegantly
demonstrated with correlative in vitro studies and crystal
structures of the mutations and drug therapies, underscoring
the importance of team science approaches to interrogating
extraordinary responding patients. Exploring and determining
the clinical effect a mutation can have on a particular protein
can help develop the library of known mutations and further
explore a variant of nearly known significance. These variants are found in clinically relevant genes in areas of the gene
that are known to related to clinical function but the exact
variant itself has not been functionally characterized. Encouraging collaborative science exploration of these variants
using functional assays and pharmacologic insight will help to
elucidate the clinical value of these types of variants.
Finally, it may not be the specific mutation itself but rather
a number of mutations that may help to predict benefit from
therapy. The search for a reproducible biomarker to predict
the response to novel immunotherapies, including the PD-1
inhibitors nivolumab and pembrolizumab as well as the CTLA-4
inhibitor ipilimumab, has yielded limited results. A promising
finding demonstrated across tumor types and immunotherapies is the correlation between the number of mutations and
clinical response to treatment. Whole-exome sequencing of
34 patients with NSCLC treated with pembrolizumab demonstrated longer PFS in tumors with a higher number of
nonsynonymous mutations (hazard ratio 0.19; 95% CI,
0.080.47, p = .0004). The median number of nonsynonymous
mutations among patients with a durable clinical response
and those without was 302 and 148, respectively.11 This was
also shown for patients with melanoma who were receiving
ipilimumab, in which those with more than 100 nonsynonymous mutations as well as a specific neoantigen landscape demonstrated improved survival.12 Finally, tumors
with mismatch repair deficiency, which is associated with
a higher number of mutations, were also associated with
improved mPFS from pembrolizumab.13 The specific
mutation cut point is dependent on the analysis and will
need further prospective validation trials to confirm this
biomarker.
PHARMACOKINETICS VARIABILITY AND

THERAPEUTIC DRUG MONITORING OF
NEWER AGENTS
Oral Targeted Therapies
TKIs are the main oral targeted therapies and comprise a
growing group of drugs for the personalized treatment of
both hematologic and solid tumors. To date, more than 15
TKIs have been approved and many others are at different
stages of their clinical development.
Regarding the absorption process, limited solubility of TKIs
in the gastrointestinal tract, gastrointestinal surgery, and
first-pass metabolism by enterocytes and the liver are the
main factors limiting their oral absorption. Moreover, concomitant administration with food significantly modifies
plasma levels for some TKIs. For example, lapatinib and nilotinib exposure is increased more than 50% with high-fat meals,
whereas no influence of food has been reported for imatinib,
dasatinib, sunitinib, or sorafenib absorption. In the opposite
case, drug exposure of axitinib and pazopanib is increased by
49% and twofold, respectively, under fasting conditions.1,14,15
Finally, adherence to treatment is another factor that determines variability in drug exposure. Thus, poor imatinib compliance results in very low trough plasma concentrations, reduces
cytogenetic responses, and contributes to treatment failure.16,17
There are a number of important host-related variables
that are correlated with variability in toxic and therapeutic
drug response. For example, TKIs bind to plasma proteins,
mainly albumin and alpha-1-acid glycoprotein (AGP). Because the only unbound (free) drugs are pharmacologically
active, hypoalbuminemia secondary to cachexia or liver
metastases can increase the amount of free drug, leading to
increased toxicity. In fact, AGP levels have been correlated
with imatinib and erlotinib clearance. Sarcopenia and low
body mass index have been also shown to be predictors of
dose-limiting toxicity of sunitinib and sorafenib.18,19 The
major route of elimination of TKIs is metabolism by CYP3A4
enzymes. Hepatic expression on these enzymes may vary
more than 20-fold among patients and may lead to variability in plasma levels. Moreover, genetic polymorphisms
influence TKI metabolism, as has been determined for
CYP2D6 and gefitinib or UGT1A9 and sorafenib.16 It is also
important to take into account the potential inhibition or
induction of metabolic enzymes as a result of drugdrug
interactions because patients with cancer are frequently
receiving more than one drug treatment. Finally, chronic administration may also induce drug elimination, as occurred
with imatinib among patients with a gastrointestinal stromal tumor, in which clearance was increased by 33% after
3 months of treatment.20
There is accumulating evidence for potential benefits of
therapeutic drug monitoring (TDM) in the treatment of
TKIs.21 Relationships between exposure and response
(efficacy/toxicity) have been established for most TKIs. Area
under the plasma concentration-time curve (AUC) and
trough plasma concentration (Cmin) are the pharmacokinetic
parameters often correlated with clinical outcomes and/or
toxicity effects of TKIs in several types of cancer.22 For
e177
CALVO ET AL
imatinib, the pharmacokinetic target has been retrospectively established in both patients with chronic myeloid
leukemia (CML) and those with gastrointestinal stromal
tumor. For imatinib, a Cmin value below 1,000 ng/mL is
associated with lower clinical benefit and shorter time to
progression. To validate this pharmacokinetic target and
include TDM as a clinical routine tool for patients receiving
imatinib treatment, two prospective randomized controlled
trials were performed.22,23 The final results of the randomized OPTIM imatinib study of 133 patients with chronic
phase CML were recently communicated,24 which showed
that only one-third of patients were correctly treated with
the standard fixed dose (i.e., these patients achieved the
Cmin threshold of 1,000 ng/mL) and two-thirds of patients
were not exposed enough to imatinib. The imatinib dose was
increased guided by TDM to obtain the Cmin target and
resulted in a higher major molecular response rate at
12 months (63% vs. 37%). Moreover, among 493 patients
with imatinib-resistant or imatinib-intolerant CML treated
with nilotinib, it was recently demonstrated that those who
had a nilotinib Cmin below 500 ng/mL had a significantly
longer time to achieve complete cytogenetic response and
molecular response. In addition, nilotinib Cmin was also
correlated with elevations in total bilirubin and lipase
levels.24 For dasatinib, results of the prospective OPTIM
dasatinib trial among patients newly diagnosed with
chronic-phase-CML demonstrated that a dasatinib maximum serum concentration (Cmax) above 50 ng/mL was associated with clinical response and a Cmin below 2.5 ng/mL
prevented fluid retention and pleural effusion.25 These results provide a strong rationale to use TDM as a new strategy
for optimizing the drug dosage to maximize the clinical
benefit (faster and more pronounced clinical responses) for
patients with CML and to promote another prospective
clinical trials to clarify the pharmacokinetic targets for more
TKIs.
Monoclonal Antibodies
The pharmacokinetics of monoclonal antibodies are complex
and different from others anticancer drugs but their interindividual variability on pharmacokinetic processes is
similar to that observed for TKIs (for example, trastuzumab
clearance varies up to 40%). Monoclonal antibodies are
dosed by body weight or at fixed doses but high variability in
mAb exposure has been observed after their administration
at the labeled dose, supporting the need to individualize
dosing to account for variability and ensure efficacy.26,27
Absorption and elimination are the main sources of pharmacokinetic variability. Absorption after subcutaneous or
intramuscular administrations is slow, incomplete, and
variable. The Cmax is reached between 2 and 8 days after
administration. Factors that have been identified as potential sources to limit mAb absorption are loss through
presystemic catabolism, flow of blood and lymph, age, body
weight (especially in obese patients), and injection site.28
The elimination routes of monoclonal antibodies have not
e178
been totally elucidated and several mechanisms have been

proposed, including proteolysis by the liver and reticuloendothelial system, target-mediated elimination, and
nonspecific endocytosis.29 Because the target-mediated
route of elimination depends on the amount of the target
expression and the affinity to binding, it leads to nonlinear
and time-dependent changes in clearance.
Although TDM studies for monoclonal antibodies are
limited, exposure-response relationships have been described. Strong exposure response has determined for
trastuzumab. Thus, the results of a case-matched control
comparison study showed that patients with the lowest
quartile of trastuzumab exposure at the end of the cycle 1
(Cmin , 11.8 mg/L on day 21) had a median overall survival
8 months shorter than in the other quartiles, whereas no
relationship was found between exposure and toxicity.30
Based on these results, the FDA review team recommended
performing a prospective trial of trastuzumab TDM-dose
intensification among patients who were underexposed.28
An exposure-response relationship for trastuzumab-emtansine
was also found among patients with HER2-positive metastatic breast cancer. After adjusting for baseline risk factors,
a higher trough concentration at day 21 (the end of cycle 1)
was associated with improved efficacy (measured as overall
survival, PFS, and objective response rates).31 These results
justify that trastuzumab-emtansine TDM-guided dosing
is a necessary tool to improve patient exposure and avoid
inadequate exposure.
For patients with metastatic colorectal cancer treated with
cetuximab, a relationship between trough serum concentration and clinical benefit was also found. Thereby, patients
with a Cmin below the median value of 40 mg/L at day 14 had
significantly shorter mPFS than patients with a Cmin above
this value (3.3 vs. 7.8 months).32
Similar results to those seen in solid tumors had been
reported as well for hematologic malignancies. Thus, Cmin
values of alemtuzumab,27 rituximab,27 and obinutuzumab33
were higher in responders versus nonresponders and were
related to better clinical outcomes.
DOSING PATIENTS WHO ARE ELDERLY OR FRAIL:

PHARMACODYNAMIC CHALLENGES
Although people older than age 65 comprise approximately
60% of all patients with newly diagnosed cancer, only 36% of
patients enrolled in clinical trials are age 65 or older.34 This
becomes a more important disparity as the population ages
and as the population of patients with cancer ages. Many
oncologists empirically dose reduce older patients, often
without much data to guide them. There are growing efforts
among several collaborative and cooperative groups to
evaluate anticancer therapeutics specifically in elderly
populations. In addition, techniques for measuring frailty,
independent of age, have been developed, are being incorporated into clinical trials, and may help guide dosing.
Because marrow reserve and organ function often decline
with age, efforts to evaluate therapies specifically for patients
with organ dysfunction are particularly relevant to the appropriate treatment of older adults with cancer.
Several commonly used cytotoxic chemotherapy regimens
have been evaluated for older patients and those with
impaired organ function. Yet there is variability in regard to
which traditional anticancer chemotherapy agents need
dose reductions in these populations. If we look at breast
cancer chemotherapy as an example, there appears to be no
age-related difference in toxicity in the common breast
cancer adjuvant therapy regimens, doxorubicin and cyclophosphamide (AC) or cyclophosphamide, epirubicin, and
fluorouracil (CEF). One small study evaluated AC across a
range of ages and found no age-related differences in
neutropenia complications, cardiac dysfunction, or quality of
life.35 Another small study evaluated CEF and similarly found
no age-related difference in leukopenia or other toxicity
among patients older than age 70 compared with those
younger than age 70.36 In a large cooperative group trial
evaluating doses of AC, there was no difference in grade 4
toxicity across age strata.37 By contrast, several investigators
have found that the adjuvant regimen cyclophosphamide,
methotrexate, and fluorouracil (CMF) is less well tolerated
by older patients than by their younger counterparts.38,39
However, this disparity in toxicity may be partly explained by
age-related renal dysfunction. At least one study has shown
that apparent age-related differences in toxicity with CMF
were abrogated by dosing according to creatinine clearance
(CrCl).40
The Cancer and Leukemia Group B (CALGB; now ALLIANCE)
evaluated age-related toxicity in three of their trials of
adjuvant chemotherapy in node-positive breast cancer. The
regimens were three schedules of CAF, AC with paclitaxel, or
AC without paclitaxel and two schedules of AC followed by
paclitaxel. Interestingly, even in these relatively modern
trials (accrued from 1985 to 1999), only 7% of patients
enrolled were age 65 or older. Nevertheless, this pooled
retrospective analysis was able to show that patients age 65
or older are 66% more likely to have grade 4 hematologic
toxicity than younger patients, discontinue therapy early
more commonly, and have a higher incidence of treatmentrelated death.41
More recently, several large randomized trials tested alternate adjuvant breast cancer regimens in older populations. For example, in the Italian ELDA trial, women age 65
to 79 were randomly assigned to receive adjuvant CMF or
weekly docetaxel. No difference in disease-free or overall
survival was seen. The pattern of toxicity was notable for
worse hematologic toxicity in the CMF arm and worse
nonhematologic toxicity in the docetaxel arm. Furthermore,
docetaxel resulted in worse quality of life for participants.
Thus, docetaxel was not recommended as an acceptable
alternate for elderly patients.42 CALGB (now ALLIANCE)
conducted a randomized trial of adjuvant breast cancer
therapy for older patients comparing oral capecitabine to
either CMF or AC. This randomized study enrolled 633
patients, 65% of whom were older than age 70. Although
capecitabine was less toxic, it was found to be less
effective than the standard regimens. Thus, it also was not

recommended as an acceptable alternative for elderly
patients.43
The taxanes are one of the most widely used classes of
chemotherapy for breast cancer as well as for other tumor
types. The toxicity of this class of drugs has been specifically
evaluated in elderly patients in several trials. Docetaxel was
evaluated in the above-described trial.42 The relationship
between age and paclitaxel pharmacokinetics and toxicity
was evaluated in a CALGB trial, which enrolled 155 patients
ranging in age from 55 to 86. This study found age-related
decline in paclitaxel clearance and in severity of neutropenia,
which did not appear to result in adverse clinical sequelae
such as hospitalization or infection.44 In a subsequent pooled
analysis of two other cooperative group trials, these same
investigators again demonstrated an age-related increase in
leukopenia as well as anorexia, hyperbilirubinemia, and
neuropathy. However, no age-related difference in paclitaxel efficacy was noted.45 Finally, Hurria et al46 recently
evaluated nab-paclitaxel among older patients across an age
range. Although these investigators found an age-related
increase in AUC (borderline significance), there was no relationship between age and dose reduction, dose omissions,
or grade 3 toxicity. However, the chemotherapy toxicity risk
score did predict increased toxicity. This score (which included
age and variables such as CrCl, anemia, hearing impairment,
falls, and need for assistance with instrumental activities of
daily living) is an attempt to capture frailty, which may be
more important than chronological age in toxicity prediction
and treatment and dosing decisions.46
Measuring frailty has become an important focus of geriatric oncology research. Several tools have been developed, including the chemotherapy toxicity prediction tool
described above and used by the Cancer and Aging Research
Group (CARG)47 and the Chemotherapy Risk Assessment
Scale for High-Age Patients (CRASH) score.48 The types and
utility of geriatric assessments have recently been the
subject of a consensus paper from the International Society
of Geriatric Oncology.49An intriguing new imaging tool for
measuring frailty by assessing sarcopenia (or skeletal muscle
wasting) on CT scans has also been shown to predict chemotherapy toxicity and outcomes (reviewed in KazemiBajestani et al50).
Although it is clear from multiple studies that some of these
tools measuring frailty accurately predict toxicity and may
prompt management change, studies validating that dosing
based on risk scores yield better outcomes still are needed.
Similar to our conundrum with the more traditional
chemotherapy drugs, we have little data to guide us in
treating elderly individuals, frail patients, or patients with
organ dysfunction with novel targeted agents. Table 1 reviews
many of the agents approved in 2014 to 2015, the age range
of patients enrolled in the pivotal trials, and data in the
package insert regarding dosing in geriatric patients and
those with organ dysfunction.
It is encouraging to note that the majority of the pivotal
trials for these newer targeted therapies enrolled a higher
e179
CALVO ET AL
TABLE 1. Recently Approved Anticancer Therapeutics
Mechanism
of Action
Drug Name
Indication
Alectinib
ALK plus NSCLC ALK inhibitor
Median
Patient Age in
Pivotal Trial,
Years (Range)
Package Insert Guidelines

Geriatric
Use
54 (2979);
No data
18% age
65 and older
Renal
Impairment
Hepatic
Impairment
No data in severe No data in

moderate
to severe
Reference
Shaw et al55
Ou et al56
52 (2279);
10% age
65 and older
Belinostat
Peripheral
T cell
lymphoma
HDAC inhibitor
64 (2881);
No need
48% age
for dose
65 and older
reduction
No data in CrCl , No data in

39 mL/min
moderate
to severe
OConnor et al57
Blinatumomab
Ph- preBcell ALL
Bispecific Ab
to CD19/CD3
32 (1877);
Patients older
13% age
than age 65
65 and older
had higher
neuro- and
infectious
toxicity
No data
No data
Topp et al58
Ceritinib
ALK plus
NSCLC
ALK inhibitor
53 (2280);
No difference in
16% age
toxicity or
65 and older
efficacy
Dose reduce for

CrCl , 30
mL/min
No data in
moderate
to severe
Shaw et al59
Cobimetinib
BRAF
mutation
melanoma
MEK inhibitor
55 (2388)
No data in severe No data in

moderate
to severe
Larkin et al60
Daratumumab
MM
Anti-CD38
64 (4476);
No differences
45% age
in safety or
65 and older
efficacy
No dose
adjustment
needed
No data in
moderate
to severe
Lokhorst et al61
Elotuzumab
MM
Anti-SLAMF7
67 (3788);
No comment
57% age
65 and older
No data
No data
Lonial et al62
Ibrutinib
Mantle
cell, CLL
BTK inhibitor
67 (5678);
Toxicity more
62% age
frequent
65 and older
among
patients
older than
age 65
No dose
adjustment
for CrCl . 25
mL/min. No
data for CrCl
, 25 mL/min
Dose reduce in
Dreyling et al63
mild. Avoid in
moderate
to severe
No data
Byrd et al64
67 (3086)
61% older
than age 65
Idelalisib
CLL, B-cell
NHL, SLL
PI3K-delta
inhibitor
71 (4890);
Higher
78% age
incidence
65 and older
of SAEs
and death
No dose
modification
for CrCl . 15
mL/min
Increased
AUC if
LFTs . ULN
Furman et al65
Ixazomib
MM
Proteasome
inhibitor
55% age 65
and older
No difference
in safety or
efficacy
Dose reduce for

CrCl , 30
mL/min
Dose reduce
for total
bilirubin .
1.53 ULN
Moureau et al66
Levatinib
Thyroid
cancer
Kinase inhibitor
64 (NR);
No difference
45% age
in safety or
65 and older
efficacy
Dose reduce for

CrCl , 30
mL/min
Dose reduce
for ChildPugh C
Schlumberger
et al67
Necitumumab
Squamous
NSCLC
EGFR antagonist
62 (3284);
Higher incidence
39% age
of VTE among
65 and older
patients older
than age 70
No data
No data
Thatcher et al68
Nivolumab
Melanoma
squamous
NSCLC
AntiPD-1
59 (2388);
No difference
35% age
in safety or
65 and older
efficacy
No need for
dose reduction
No dose
reduction
in mild. No
data in
moderate
to severe
Weber et al69
Continued
e180
TABLE 1. Recently Approved Anticancer Therapeutics (Contd)
Drug Name
Indication
Mechanism
of Action
Median
Patient Age in
Pivotal Trial,
Years (Range)
Package Insert Guidelines

Geriatric
Use
Renal
Impairment
Hepatic
Impairment
Reference
Borghaei et al70
61 (3784);
37% age
65 and older
Olaparib
BRCA
mutation
ovarian
cancer
PARP inhibitor
57 (2979);
Grade 3 AEs
20% age
more
65 and older
frequent
among
patients
older than
age 65
No data if CrCl
, 50 mL/min
No data
Kaufman et al71
Osimertinib
EGFR
mutation
NSCLC
EGFR inhibitor
45% older
than age 65
No data if
CrCl , 30
mL/min
No data in
moderate
to severe
Ramalingam
et al72
Palbociclib
ER+ HER22
breast
cancer
CDK4/6 inhibitor 63 (5471);

No difference
44% age
in safety or
65 and older
efficacy
No dose
reduction
required for
CrCl . 30
mL/min. No
data in severe
No dose
reduction
in mild; no
data in
moderate
to severe
Finn et al73
Panobinostat
MM
Pan-deacetylase
inhibitor
63 (5659)
More GI,
42% age
cardiac, and
65 and older
hematologic
AEs in older
patients
No dose
reduction
needed
Dose
San-Miguel
modification
et al74
for mild
and moderate
Pembrolizumab
Melanoma
AntiPD-1
60 (2594)
No difference
39% age
in safety or
65 and older
efficacy
No dose
adjustment
needed
No dose
adjustment
needed in
mild. No
data in
moderate
to severe
Hamid et al75
Sonidegib
Basal cell
carcinoma
Smoothened
receptor
antagonist
65 (2493)
No difference in
No dose
52% age
effectiveness;
adjustment
65 and older
higher incidence
needed
of grade 3 or
4 AEs in older
patients
No dose
adjustment
needed in
mild. No data
in moderate
to severe
Migden et al76
Oncolytic virus
63 (2294)
No difference
48% age
in safety or
65 and older
efficacy
No data
Andtbacka et al77
Talimogene
Melanoma
Laherparepvec
Grade 3 and
4 AEs more
frequent
among
patients
older than
age 65
No data
Abbreviations: ALK, anaplastic lymphoma kinase; NSCLC, nonsmall cell lung cancer; ALL, acute lymphoblastic leukemia; MM, multiple myeloma; CLL, chronic lymphocytic
leukemia; NHL, non-Hodgkin lymphoma; SLL, small lymphocytic lymphoma; HDAC, histone deacetylase; Ab, antibody; MEK, mitogen-activated protein kinase; BTK, Brutons
tyrosine kinase; PI3K, phosphoinositide 3-kinase; CDK, cyclin-dependent kinase; SAE, serious adverse event; VTE, venous thromboembolism; AE, adverse event; GI,
gastrointestinal; CrCl, creatinine clearance; AUC, area under the plasma concentration-time curve; LFT, liver function test; ULN, upper limit of normal; Ph, Philadelphia
chromosome.
proportion of older patients than the historical trials summarized above. More than 40% patients were age 65 or older
in 12 of the 20 trials highlighted. The relationship between
age and toxicity appears to vary between drugs. Some
studies have noted increased toxicity in older patients
(ibrutinib, idelalisib, necitumumab, osimertinib, panobinostat, sonidegib, blinatumumab, olaparib). Others report no
age-related increase in toxicity (belinostat, ceritinib, daratumumab, ixazomib, lenvatinib, nivolumab, palbociclib),
although most comment that sample sizes are small enough
to lessen confidence. These new targeted therapies and

immunotherapies also vary in need for dose reduction for
renal or hepatic dysfunction (Table 1).
Few targeted therapies have been specifically evaluated to
date in populations who have organ dysfunction or are elderly or frail. Several examples of treatments that have been
assessed in special populations include the CARG study of
bevacizumab among elderly patients51 and clinical trials of
pazopanib52 and sorafenib53 among patients with organ
dysfunction.
e181
CALVO ET AL
Some of the newer drugs shown in Table 1 have in fact

been evaluated in populations who are older than age 65
and are more representative of the disease population. For
example, in the pivotal study of idelalisib and rituximab in
relapsed chronic lymphocytic leukemia,54 78% of patients enrolled were age 65 or older, 40% had a CrCl below 60 mL/min,
and 85% had a Cumulative Illness Rating Scale score greater
than 6. Although patients older than age 65 benefited from
idelalisib similarly to their younger counterparts, the package
insert reveals that there were higher rates or serious adverse
events, treatment discontinuation, and death among patients
older than age 65 in this trial as well as the pivotal trials in
non-Hodgkin lymphoma and small lymphocytic lymphoma.
In conclusion, pivotal studies of recently approved drugs
generally have had higher representation of patients older
than age 65. This yields more information on toxicity in the
older population with these therapies. Unfortunately, it
appears that toxicity of many of these new targeted drugs is
higher in older patients. However, package inserts do not
give guidance on alternate dosing or schedules in older and
frail patients. In this special population, TDM or validated
frailty assessments could be useful tools to characterize the
real exposure and manage possible dose reduction accordingly, minimizing the risk of greater reductions than those
that are really needed.
FUTURE DIRECTIONS OF PERSONALIZED

THERAPY ACCORDING TO PHARMACOGENOMIC,
PHARMACOKINETIC, AND PHARMACODYNAMIC
PARAMETERS
The role of somatic tumor genetics relates not only to initial
selection of therapy but also to sequencing of therapy options, and it must be optimized in standard clinical practice.
To continue to establish genetic biomarkers of novel therapies, we need clinical trials and patients willing to enroll in
them. Enthusiasm behind basket trials must continue to fuel
histology agnostic, biomarker-driven trials that are as inclusive as possible for the tumor types in which they will be
most useful. In addition, novel methods of obtaining samples for genetic analysis are needed to address challenges of
tumor acquisition as well as heterogeneity of metastatic

cancers. Cell-free DNA and liquid biopsies are currently being
assessed as predictors of response and more experience
with these assays will help to determine their value and
place in standard clinical management. Pharmacogenomic
analysis of tumor tissue has evolved from focused interrogation of select genes to guide targeted therapies to
multigene panels being used in conjunction with molecular
tumor boards to help direct optimal therapy for patients.
The encouraging results for pharmacokinetic optimization
of TKIs and monoclonal antibodies justify the need to increase knowledge about the exposure-response relationship
for other anticancer targeted agents to maximize the
therapeutic benefit of these expensive therapies. As Gao
et al14 said, Perhaps the time is ripe for us to quit guessing
and start measuring. The implementation of TDM in routine
clinical practice would require (1) simple, rapid, and reproducible analytical methods to quantify plasma or serum
drug concentrations; (2) a pharmacokinetic target related to
efficacy or toxicity; (3) population pharmacokinetic models
previously developed and a Bayesian approach to determine
the individual pharmacokinetic parameters taking into account the interindividual variability in all of the pharmacokinetic processes; and (4) highly qualified professionals
(within multidisciplinary teams) trained to translate the TDM
results and to propose new doses adjusted to individual
needs of each patient. This is a feasible way of optimizing our
current and future therapeutic armamentarium and it
should be encouraged.
Finally, considering that with these agents, as with the
older drugs, chronological age is not as important as
physiologic age/frailty, incorporating frailty assessment
tools into trials of newer agents and validating frailty-based
dose adjustment will be an important part of further drug
development. Ongoing efforts to assess approved anticancer therapeutics for older patient populations and those
with organ dysfunction are an important effort for the
oncology clinical research community.
All together, these efforts will allow us to push precision
medicine in oncology to the next level for our patients.
References
1. Klumpen HJ, Samer CF, Mathijssen RH, et al. Moving towards dose
individualization of tyrosine kinase inhibitors. Cancer Treat Rev. 2011;
37:251-260.
2. Van Cutsem E, Peeters M, Siena S, et al. Open-label phase III trial of
panitumumab plus best supportive care compared with best supportive
care alone in patients with chemotherapy-refractory metastatic colorectal cancer. J Clin Oncol. 2007;25:1658-1664.
3. Amado RG, Wolf M, Peeters M, et al. Wild-type KRAS is required for
panitumumab efficacy in patients with metastatic colorectal cancer.
J Clin Oncol. 2008;26:1626-1634.
4. Allegra CJ, Rumble RB, Hamilton SR, et al. Extended RAS gene mutation
testing in metastatic colorectal carcinoma to predict response to antiepidermal growth factor receptor monoclonal antibody therapy:
e182
5.
6.
7.
8.
9.
American Society of Clinical Oncology Provisional Clinical Opinion

Update 2015. J Clin Oncol. 2016;34:179-185.
Ballman KV. Biomarker: predictive or prognostic? J Clin Oncol. 2015;33:
3968-3971.
Iglehart JD, Silver DP. Synthetic lethalitya new direction in cancer-drug
development. N Engl J Med. 2009;361:189-191.
Mateo J, Carreira S, Sandhu S, et al. DNA-repair defects and olaparib in
metastatic prostate cancer. N Engl J Med. 2015;373:1697-1708.
Thomas A, Liu SV, Subramaniam DS, et al. Refining the treatment of
NSCLC according to histological and molecular subtypes. Nat Rev Clin
Oncol. 2015;12:511-526.
Oxnard GR, Arcila ME, Sima CS, et al. Acquired resistance to EGFR
tyrosine kinase inhibitors in EGFR-mutant lung cancer: distinct natural
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
history of patients with tumors harboring the T790M mutation. Clin

Cancer Res. 2011;17:1616-1622.
Shaw AT, Friboulet L, Leshchiner I, et al. Resensitization to crizotinib by
the lorlatinib ALK resistance mutation L1198F. N Engl J Med. 2016;374:
54-61.
Rizvi NA, Hellmann MD, Snyder A, et al. Cancer immunology. Mutational
landscape determines sensitivity to PD-1 blockade in non-small cell lung
cancer. Science. 2015;348:124-128.
Snyder A, Makarov V, Merghoub T, et al. Genetic basis for clinical
response to CTLA-4 blockade in melanoma. N Engl J Med. 2014;371:
2189-2199.
Le DT, Uram JN, Wang H, et al. PD-1 blockade in tumors with mismatchrepair deficiency. N Engl J Med. 2015;372:2509-2520.
Gao B, Yeap S, Clements A, et al. Evidence for therapeutic drug
monitoring of targeted anticancer therapies. J Clin Oncol. 2012;30:
4017-4025.
Widmer N, Bardin C, Chatelut E, et al. Review of therapeutic drug
monitoring of anticancer drugs part twotargeted therapies. Eur J
Cancer. 2014;50:2020-2036.
Ibrahim AR, Eliasson L, Apperley JF, et al. Poor adherence is the main
reason for loss of CCyR and imatinib failure for chronic myeloid leukemia patients on long-term therapy. Blood. 2011;117:3733-3736.
Bui BN, Italiano A, Miranova A, et al. Trough imatinib plasma levels in
patients treated for advanced gastrointestinal stromal tumors evidence
of large interpatient variations under treatment with standard doses.
J Clin Oncol. 2008;26:15s (suppl; abstr 10564).
Huillard O, Mir O, Peyromaure M, et al. Sarcopenia and body mass index
predict sunitinib-induced early dose-limiting toxicities in renal cancer
patients. Br J Cancer. 2013;108:1034-1041.
Mir O, Coriat R, Blanchet B, et al. Sarcopenia predicts early dose-limiting
toxicities and pharmacokinetics of sorafenib in patients with hepatocellular carcinoma. PLoS One. 2012;7:e37563.
Eechoute K, Fransson MN, Reyners AK, et al. A long-term prospective
population pharmacokinetic study on imatinib plasma concentrations
in GIST patients. Clin Cancer Res. 2012;18:5780-5787.
Yu H, Steeghs N, Nijenhuis CM, et al. Practical guidelines for therapeutic
drug monitoring of anticancer tyrosine kinase inhibitors: focus on the
pharmacokinetic targets. Clin Pharmacokinet. 2014;53:305-325.
Gotta V, Widmer N, Decosterd LA, et al. Clinical usefulness of therapeutic concentration monitoring for imatinib dosage individualization:
results from a randomized controlled trial. Cancer Chemother Pharmacol. 2014;74:1307-1319.
Rousselot P, Johnson-Ansah H, Huguet F, et al. 133 personalized daily
doses of imatinib by therapeutic drug monitoring increase the rates of
molecular responses in patients with chronic myeloid leukemia. Final
results of the randomized OPTIM imatinib study. Paper presented at:
57th Annual Meeting of the American Society of Hematology; December 2015; Orlando, FL.
Giles FJ, Yin OQ, Sallas WM, et al. Nilotinib population pharmacokinetics
and exposure-response analysis in patients with imatinib-resistant or
-intolerant chronic myeloid leukemia. Eur J Clin Pharmacol. 2013;69:
813-823.
Rousselot P, Mollica L, Guerci-Bresler A, et al. Dasatinib daily dose
optimization based on residual drug levels resulted in reduced risk of
pleural effusions and high molecular response rates: final results of the
randomized OPTIM dasatinib trial. Paper presented at: 19th Congress of
the European Hematology Association; June 2014; Milan, Italy.
Oude Munnink TH, Henstra MJ, Segerink LI, et al. Therapeutic drug
monitoring of monoclonal antibodies in inflammatory and malignant
diseasetranslating TNF-a experience to oncology. Clin Pharmacol
Ther. Epub 2015 Aug 11.
27. Mould DR, Dubinsky MC. Dashboard systems: pharmacokinetic/

pharmacodynamic mediated dose optimization for monoclonal antibodies. J Clin Pharmacol. 2015;55(Suppl 3):S51-S59.
28. Wang W, Chen N, Shen X, et al. Lymphatic transport and catabolism of
therapeutic proteins after subcutaneous administration to rats and
dogs. Drug Metab Dispos. 2012;40:952-962.
29. Keizer RJ, Huitema AD, Schellens JH, et al. Clinical pharmacokinetics of
therapeutic monoclonal antibodies. Clin Pharmacokinet. 2010;49:493-507.
30. Yang J, Zhao H, Garnett C, et al. The combination of exposure-response
and case-control analyses in regulatory decision making. J Clin Pharmacol. 2013;53:160-166.
31. Wang J, Song P, Schrieber S, et al. Exposure-response relationship of TDM1: insight into dose optimization for patients with HER2-positive
metastatic breast cancer. Clin Pharmacol Ther. 2014;95:558-564.
32. Azzopardi N, Lecomte T, Ternant D, et al. Cetuximab pharmacokinetics
influences progression-free survival of metastatic colorectal cancer
patients. Clin Cancer Res. 2011;17:6329-6337.
33. Gibiansky E, Gibiansky L, Carlile DJ, et al. Population pharmacokinetics
of obinutuzumab (GA101) in chronic lymphocytic leukemia (CLL) and
non-Hodgkins lymphoma and exposure-response in CLL. CPT Pharmacometrics Syst Pharmacol. 2014;3:e144.
34. Talarico L, Chen G, Pazdur R. Enrollment of elderly patients in clinical
trials for cancer drug registration: a 7-year experience by the US Food
and Drug Administration. J Clin Oncol. 2004;22:4626-4631.
35. Dees EC, OReilly S, Goodman SN, et al. A prospective pharmacologic
evaluation of age-related toxicity of adjuvant chemotherapy in women
with breast cancer. Cancer Invest. 2000;18:521-529.
36. Cascinu S, Del Ferro E, Catalano G. Toxicity and therapeutic response to
chemotherapy in patients aged 70 years or older with advanced cancer.
Am J Clin Oncol. 1996;19:371-374.
37. Fisher B, Anderson S, Wickerham DL, et al. Increased intensification and
total dose of cyclophosphamide in a doxorubicin-cyclophosphamide
regimen for the treatment of primary breast cancer: findings from
National Surgical Adjuvant Breast and Bowel Project B-22. J Clin Oncol.
1997;15:1858-1869.
38. Crivellari D, Bonetti M, Castiglione-Gertsch M, et al. Burdens and
benefits of adjuvant cyclophosphamide, methotrexate, and fluorouracil
and tamoxifen for elderly patients with breast cancer: the International
Breast Cancer Study Group Trial VII. J Clin Oncol. 2000;18:1412-1422.
39. Hurria A, Brogan K, Panageas KS, et al. Patterns of toxicity in older
patients with breast cancer receiving adjuvant chemotherapy. Breast
40. Gelman RS, Taylor SG 4th. Cyclophosphamide, methotrexate, and 5fluorouracil chemotherapy in women more than 65 years old with advanced breast cancer: the elimination of age trends in toxicity by using
doses based on creatinine clearance. J Clin Oncol. 1984;2:1404-1413.
41. Muss HB, Berry DA, Cirrincione C, et al; Cancer and Leukemia Group B
Experience. Toxicity of older and younger patients treated with adjuvant chemotherapy for node-positive breast cancer: the Cancer and
Leukemia Group B Experience. J Clin Oncol. 2007;25:3699-3704.
42. Perrone F, Nuzzo F, Di Rella F, et al. Weekly docetaxel versus CMF as
adjuvant chemotherapy for older women with early breast cancer: final
results of the randomized phase III ELDA trial. Ann Oncol. 2015;26:
675-682.
43. Muss HB, Berry DA, Cirrincione CT, et al; CALGB Investigators. Adjuvant
chemotherapy in older women with early-stage breast cancer. N Engl J
Med. 2009;360:2055-2065.
44. Lichtman SM, Hollis D, Miller AA, et al; Cancer and Leukemia Group B
(CALGB 9762). Prospective evaluation of the relationship of patient age
and paclitaxel clinical pharmacology: Cancer and Leukemia Group B
(CALGB 9762). J Clin Oncol. 2006;24:1846-1851.
e183
CALVO ET AL
45. Lichtman SM, Hurria A, Cirrincione CT, et al; Cancer and Leukemia Group B.
Paclitaxel efficacy and toxicity in older women with metastatic breast cancer:
combined analysis of CALGB 9342 and 9840. Ann Oncol. 2012;23:632-638.
46. Hurria A, Blanchard MS, Synold TW, et al. Age-related changes in
nanoparticle albumin-bound paclitaxel pharmacokinetics and pharmacodynamics: influence of chronological versus functional age. Oncologist. 2015;20:37-44.
47. Hurria A, Togawa K, Mohile SG, et al. Predicting chemotherapy toxicity
in older adults with cancer: a prospective multicenter study. J Clin
Oncol. 2011;29:3457-3465.
48. Extermann M, Boler I, Reich RR, et al. Predicting the risk of chemotherapy
toxicity in older patients: the Chemotherapy Risk Assessment Scale for
High-Age Patients (CRASH) score. Cancer. 2012;118:3377-3386.
49. Wildiers H, Heeren P, Puts M, et al. International Society of Geriatric
Oncology consensus on geriatric assessment in older patients with
50. Kazemi-Bajestani SM, Mazurak VC, Baracos V. Computed tomographydefined muscle and fat wasting are associated with cancer clinical
outcomes. Semin Cell Dev Biol. Epub 2015 Sep 3.
51. Mohile SG, Hardt M, Tew W, et al; Cancer and Aging Research Group.
Toxicity of bevacizumab in combination with chemotherapy in older
patients. Oncologist. 2013;18:408-414.
52. Shibata SI, Chung V, Synold TW, et al. Phase I study of pazopanib in
patients with advanced solid tumors and hepatic dysfunction: a National Cancer Institute Organ Dysfunction Working Group study. Clin
Cancer Res. 2013;19:3631-3639.
53. Miller AA, Murry DJ, Owzar K, et al. Phase I and pharmacokinetic study
of sorafenib in patients with hepatic or renal dysfunction: CALGB 60301.
J Clin Oncol. 2009;27:1800-1805.
54. Furman RR, Sharman JP, Coutre SE, et al. Idelalisib and rituximab in
relapsed chronic lymphocytic leukemia. N Engl J Med. 2014;370:
997-1007.
55. Shaw AT, Gandhi L, Gadgeel S, et al. Alectinib in ALK-positive, crizotinibresistant, non-small-cell lung cancer: a single-group, multicentre, phase
2 trial. Lancet Oncol. 2016;17:234-242.
56. Ou SI, Ahn JS, De Petris L, et al. Alectinib in crizotinib-refractory ALKrearranged non-small-cell lung cancer: a phase II global study. J Clin
Oncol. Epub 2015 Nov 23.
57. OConnor OA, Horwitz S, Masszi T, et al. Belinostat in patients with
relapsed or refractory peripheral T-cell lymphoma: results of the pivotal
phase II BELIEF (CLN-19) study. J Clin Oncol. Epub 2015 Jun 22.
58. Topp MS, Gokbuget N, Stein AS, et al. Safety and activity of blinatumomab
for adult patients with relapsed or refractory B-precursor acute lymphoblastic leukaemia: a multicentre, single-arm, phase 2 study. Lancet
Oncol. 2015;16:57-66.
59. Shaw AT, Ou SH, Bang YJ, et al. Crizotinib in ROS1-rearranged non-smallcell lung cancer. N Engl J Med. 2014;371:1963-1971.
60. Larkin J, Ascierto PA, Dreno B, et al. Combined vemurafenib and
cobimetinib in BRAF-mutated melanoma. N Engl J Med. 2014;371:
1867-1876.
61. Lokhorst HM, Plesner T, Laubach JP, et al. Targeting CD38 with daratumumab monotherapy in multiple myeloma. N Engl J Med. 2015;373:
1207-1219.
62. Lonial S, Weiss BM, Usmani SZ, et al. Daratumumab monotherapy in
patients with treatment-refractory multiple myeloma (SIRIUS): an
open-label, randomised, phase 2 trial. Lancet. Epub 2016 Jan 6.
e184
63. Dreyling M, Jurczak W, Jerkeman M, et al. Ibrutinib versus temsirolimus

in patients with relapsed or refractory mantle-cell lymphoma: an international, randomised, open-label, phase 3 study. Lancet. Epub 2015
Dec 4.
64. Byrd JC, Brown JR, OBrien S, et al. Ibrutinib versus ofatumumab in previously
treated chronic lymphoid leukemia. N Engl J Med. 2014;371:213-223.
65. Furman RR, Sharman JP, Coutre SE, et al. Idelalisib and rituximab in
relapsed chronic lymphocytic leukemia. N Engl J Med. 2014;370:
997-1007.
66. Moreau P, Masszi T, Grzasko N, et al. Ixazomib, an investigational oral
proteasome inhibitor (PI), in combination with lenalidomide and
dexamethasone (IRd), significantly extends progression-free survival
(PFS) for patients (Pts) with relapsed and/or refractory multiple myeloma (RRMM): the phase 3 Tourmaline-MM1 Study (NCT01564537).
Blood. 2015;126 (abstr 727).
67. Schlumberger M, Tahara M, Wirth LJ, et al. Lenvatinib versus placebo in
radioiodine-refractory thyroid cancer. N Engl J Med. 2015;372:621-630.
68. Thatcher N, Hirsch FR, Luft AV, et al. Necitumumab plus gemcitabine
and cisplatin versus gemcitabine and cisplatin alone as first-line therapy
in patients with stage IV squamous non-small-cell lung cancer (SQUIRE):
an open-label, randomised, controlled phase 3 trial. Lancet Oncol. 2015;
16:763-774.
69. Weber JS, DAngelo SP, Minor D, et al. Nivolumab versus chemotherapy
in patients with advanced melanoma who progressed after anti-CTLA-4
treatment (CheckMate 037): a randomised, controlled, open-label,
70. Borghaei H, Paz-Ares L, Horn L, et al. Nivolumab versus docetaxel in
advanced nonsquamous non-small-cell lung cancer. N Engl J Med. 2015;
373:1627-1639.
71. Kaufman B, Shapira-Frommer R, Schmutzler RK, et al. Olaparib monotherapy in patients with advanced cancer and a germline BRCA1/2
mutation. J Clin Oncol. 2015;33:244-250.
72. cSS, Chih-Hsin Yang J, Lee CK, et al. AZD9291, a mutant-selective EGFR
inhibitor, as first-line treatment for EGFR mutation-positive advanced
non-small cell lung cancer (NSCLC): results from a phase 1 expansion
cohort. J Clin Oncol. 2015;33 (suppl; abstr 8000).
73. Finn RS, Crown JP, Lang I, et al. The cyclin-dependent kinase 4/6 inhibitor palbociclib in combination with letrozole versus letrozole alone
as first-line treatment of oestrogen receptor-positive, HER2-negative,
advanced breast cancer (PALOMA-1/TRIO-18): a randomised phase 2
74. San-Miguel JF, Hungria VT, Yoon SS, et al. Panobinostat plus bortezomib
and dexamethasone versus placebo plus bortezomib and dexamethasone in patients with relapsed or relapsed and refractory multiple
myeloma: a multicentre, randomised, double-blind phase 3 trial. Lancet
Oncol. 2014;15:1195-1206.
75. Hamid O, Robert C, Daud A, et al. Safety and tumor responses with
lambrolizumab (anti-PD-1) in melanoma. N Engl J Med. 2013;369:
134-144.
76. Migden MR, Guminski A, Gutzmer R, et al. Treatment with two different
doses of sonidegib in patients with locally advanced or metastatic basal
cell carcinoma (BOLT): a multicentre, randomised, double-blind phase 2
77. Andtbacka RH, Kaufman HL, Collichio F, et al. Talimogene laherparepvec
improves durable response rate in patients with advanced melanoma.
J Clin Oncol. 2015;33:2780-2788.
Potentially Resectable Metastatic

Colorectal Cancer: A
Multidisciplinary Discussion
CHAIR
Gunnar Folprecht, MD
University Hospital Carl Gustav Carus
Dresden, Germany
SPEAKERS
David M. Liu, MD
University of British Columbia
Vancouver, BC, Canada
Laura A. Dawson, MD
Toronto, ON, Canada
Steven J. Nurkin, MD
Roswell Park Cancer Institute
Buffalo, NY
GUNNAR FOLPRECT
Liver Metastases in Colorectal Cancer

Gunnar Folprecht, MD
OVERVIEW
Resection of colorectal liver metastases is a treatment standard because patients experience long-term disease-free survival
or are even cured after undergoing this procedure. Improved surgical techniques for liver resection in combination with
downsizing liver metastases by chemotherapy, interventions to induce liver hypertrophy before resection, and the use of
ablative techniques have allowed us to expand the indications for liver surgery and local treatment in situations with limited
metastatic colorectal cancer. Resectability and identification of patients who might benefit from liver surgery and local
ablative techniques are key factors for the treatment of patients with colorectal cancer. Despite the wide acceptance of liver
surgery and ablative techniques, there are many open questions on the management of limited metastatic disease, such as
which patients benefit from an aggressive surgical approach, what the indications for ablative and other local techniques are,
and what the role of chemotherapy is for patients with resectable or resected disease. Unfortunately, results of randomized
trials are only available for a limited number of these questions.
esection of colorectal liver metastases is a treatment

standard because patients experience long-term diseasefree survival or are even cured after undergoing this procedure. Improved surgical techniques for liver resection and
downsizing liver metastases by chemotherapy, interventions
to induce liver hypertrophy before resection, and the use of
ablative techniques have allowed us to widen the indications
for liver surgery and local treatment in situations with limited
metastatic colorectal cancer.
Resectability and identification of patients who might benefit
from liver surgery and local ablative techniques are key factors
for the treatment of patients with colorectal cancer. Despite the
wide acceptance of liver surgery and ablative techniques, there
are many open questions on the management of limited
metastatic disease, such as which patients benefit from an
aggressive surgical approach, what the indications for ablative
and other local techniques are, and what the role of chemotherapy is for patients with resectable or resected disease.
Unfortunately, results of randomized trials are only available
for a limited number of these questions.
RESECTABILITY
Technical resectability is a necessary condition for the resection of metastases. For liver metastases, resectability is
determined by 25% to 30% of functional liver tissue with
sufficient inflow (portal vein/liver artery) and outflow (liver
veins). Conversion chemotherapy1,2 and several surgical
techniques can improve technical resectability (Fig. 1).
In daily clinical practice, technical resectability and prognostic factors contribute to the decision making on whether
to resect liver metastases. A higher number of metastases, a
short disease-free interval before diagnosis of the metastases,
the presence of extrahepatic metastases, larger metastases, an advanced primary tumor stage, and elevated tumor
marker levels (carcinoembryonic antigen [CEA] or CA 19-9)
are important risk factors for the prognosis of patients.3 The
Memorial Sloan Kettering Cancer Center score4 is one of the
most widely used scores and is easily calculated. There is no
fixed upper limit for the number of liver metastases defining
nonresectability or a lower limit for the disease-free interval
before the (re)occurrence of liver metastases, however,
these two factors have often have an important influence in
daily treatment decisions. The lack of clear definitions on
prognostic factors limiting the value of resection contributes
to the high subjectivity of decisions and might explain the
high variability of treatment decisions observed, i.e., in surgical
reviews of CT scans the CELIM trial.2 Multidisciplinary care
team discussions can only partly overcome the lack of studies
defining limits for prognostic resectability.
In addition to technical resectability and tumor-related
prognostic factors, treatment decisions are driven by patientrelated comorbidity, which limits surgical treatment options
and chemotherapy tolerance (Fig. 1A). Systematic research
for comorbidity and multimodal treatment of liver metastases
is rare and is therefore not included in this review.
From the University Hospital Carl Gustav Carus, University Cancer Center, Medical Department I, Dresden, Germany.
Corresponding author: Gunnar Folprecht, MD, Universitatsklinikum Carl Gustav Carus Dresden, Fetscherstrae 74, 01307, Dresden, Germany; email: gunnar.folprecht@
uniklinikum-dresden.de.
e186
LIVER METASTASES IN COLORECTAL CANCER
Retrospective analyses have demonstrated that patients

who underwent resection for their metastases experienced
improved survival compared with patients with medical
treatment alone. This knowledge is based on retrospective
cohorts1,4 and patients resected for liver and lung metastases in prospective trials.5 Unfortunately, these reports are
influenced by several biases.6 Resections are offered to
patients whose disease responded to chemotherapy and
who have a better prognosis, and the resection cohort
excludes patients who have an early disease progression, a
worsened performance status, or died before the decision
for resection (i.e., after conversion chemotherapy) could be
made.
The EORTC CLOCC trial7 was the first prospective randomized trial to demonstrate a survival benefit of resection/
ablation versus systemic treatment alone. Patients with up
to 10 liver metastases were selected to receive medical
treatment alone or medical treatment with radiofrequency
ablation. For the vast majority of patients, radiofrequency
ablation was performed as an open procedure and combined with resection. In the combined treatment arm, the
overall survival was significantly improved compared
with systemic treatment alone (8-year survival, 36% vs. 9%,
respectively; hazard ratio [HR] 0.58; 95% CI, 0.380.88;
p , .01).8 This trial demonstrated that resection/ablation
has a major impact on overall survival. Among patients with
up to 10 liver metastases without extrahepatic spread,
approximately 30% were cured by the combined treatment. Furthermore, progression-free survival increased
from 9.9 months (chemotherapy alone) to 16.8 months
(combined modality treatment),8 allowing treatment-free
intervals.
Unfortunately, overall and disease-free survival rates after
resection decrease for patients with poor prognostic factors.
In the CELIM study, the median disease-free survival after
conversion therapy and resection was 16.8 months among
patients with less than five metastases, 8.2 months among
patients with 5 to 10 metastases, and 2.5 months among patients with more than 10 metastases.9 Similarly, patients with
KEY POINTS
Resection of (liver) metastases is part of standard

treatment in metastatic colorectal cancer.
Further trials are needed to better define limits of
resectability, especially among patients with a poor
prognosis.
Resectability can be improved by conversion
chemotherapy and surgical/interventional methods
requiring close multidisciplinary cooperation.
FOLFOXIRI-based regimens and EGFR combinations
(for RAS wild type disease) have demonstrated higher
response and resection rates in randomized trials
Perioperative FOLFOX therapy should be considered for
patients with higher risk factors and resectable disease.
a Basingstoke risk score higher than 20 experience a median

survival of less than 1.5 years following liver resection.10 Although patients were treated in the era of fluorouracil (5-FU),
survival for the highest risk group is probably not markedly
longer than in a general patient population with systemic
treatment alone.11,12 It is important to note that current
prognostic scores are not prospectively evaluated to be predictive for the benefit of resections. In addition, re-resections of
further metastases can be beneficial, and patients with observed indications for re-resection have similar prognoses as
those with initial resections.13-15 Furthermore, surgery/ablative
techniques have the potential to contribute to therapy as a
further line of palliative treatment, but prospective trials among
patients with technically resectable, poor prognostic disease
are needed to determine which patients benefit most from
resections.
IMPROVING RESECTABILITY
Among patients whose disease is not regarded as suitable for
up-front resection due to the location of the metastases
(technical resectability) or poor prognostic factors, chemotherapy is mostly the first treatment step, but other
interventions also can improve resectability.
Preoperative portal vein embolization of the liver segments planned for resection, which induces a hypertrophy of
the not-embolized liver segments, enlarges the remaining
functional liver tissue.16 This procedure is most frequently
applied before an extended right hemihepatectomy if the
left segments (2/3) are relatively small and often combined
with a two-step resection, i.e., resection of metastases in the
left segments before the right hemihepatectomy.17,18 The
relatively aggressive concept of a combination of a portal
vein ligation with an intraoperative in-situ splitting of the
liver (associating liver partition and portal vein ligation for
staged hepatectomy; ALPPS) induces a more rapid liver
hypertrophy and is followed by a second resection.19 In a
recently published register of 320 patients, second resection was performed 14 days after the first intervention (median). During those 2 weeks, the functional
liver remnant increased from 21% to 40%.20 This technique increases the options for resection of liver metastases and primary hepatobiliary cancers but was
associated with a 90-day mortality of 8.8% among patients with colorectal liver metastases of 5%.20 The combination of ablation with resection of metastases provides a
more parenchymal-sparing approach for extensive liver
metastases.21
These techniques increase the level of treatment complexity, and additional treatment steps need additional time
for procedures and/or patient recovery, especially in multistep resections for liver metastases or sequential resections, i.e., of liver and lung metastases plus resection of
the primary. In the COIN, AIO 0207, and CAIRO3 trials investigating treatment-free intervals versus maintenance
therapy, the median time from end of chemotherapy
treatment until progression without chemotherapy was 3.0
e187
GUNNAR FOLPRECT
FIGURE 1. (A) Dimensions of Resectability and (B) Heatmap Combining Prognostic and Technical Factors
to 4.1 months.22-24 This interval is shorter than the time

needed for two or three resections. In several centers,
chemotherapy is administered between treatment cycles,
but the number of required interventions may limit the
feasibility of resection at all metastatic sites.
CHEMOTHERAPY IN NONRESECTABLE DISEASE

There is a strong correlation between the response rate with
chemotherapy and the resection rate when different studies
are compared.25 Conceptually, the effect of shrinking larger,
technically nonresectable metastases is more obvious than
e188
when prognostic factors, such as a high number of metastases, limit resection (Fig. 2). In the latter situation, chemotherapy is often used to observe the course of disease
and to resect disease with less aggressive tumor growth.
With current regimens achieving response rates of 60%
65%, liver resections were reported in approximately 40% of
patients enrolled in trials focusing on liver-limited disease.26
In randomized trials, FOLFOXIRI (5-FU, leucovorin, oxaliplatin, and irinotecan) improved response27,28 and resection
rates28 compared with infusion chemotherapy doublets, as
did cetuximab/FOLFIRI (5-FU, leucovorin, and irinotecan)
compared with FOLFIRI alone.29 Bevacizumab improved
response rates in studies with IFL (irinotecan, 5-FU, leucovorin) or fluoropyrimidine monotherapy30,31 but not when
combined with FOLFOX in first-line treatment.32
When EGFR antibodies and bevacizumab are directly
compared, higher response rates were reported in the
CALGB 80405 study,33 and nonsignificant trends were reported in the FIRE3 and PEAK trials.34,35 Furthermore, differences were observed for the non-RECIST evaluation of the
deepness of response favoring anti-EGFR antibodies.36
A literature-based meta-analysis of all three trials directly comparing anti-EGFR antibodies with bevacizumab
described a significantly higher response rate with anti-EGFR
antibodies (odds ratio 1.3; 95% CI, 1.11.6).37 Similarly, in
the CALGB study, 82 patients (14.2%) underwent resection
after treatment with cetuximab, and 50 patients (8.9%)
underwent resection after receiving bevacizumab-based
therapy.38 This difference is statistically significant (p , .01).
No differences in resection rates were reported in the FIRE3
and PEAK trials.34,35 Based on these results, FOLFOXIRI (or
EGFR-based therapies among patients with RAS wild-type
disease) might be optimal if conversion therapy is the primary
treatment aim.
During conversion chemotherapy, patients will have resectable disease after a median treatment duration of
4 months. If resectability is not achieved within 6 months, it
is very unlikely that further chemotherapy will facilitate later
resections.2 Therefore, multidisciplinary discussion of the
treatment strategy might be scheduled every 2 months (or
after 3 and 6 months) during conversion chemotherapy. No
trial has been performed to determine the optimal duration
of neoadjuvant therapy when resectability is achieved.
When resectability was achieved within 3 to 6 months after
preoperative treatment, it might be reasonable to orientate

on the treatment durations for patients with resectable
disease and to continue the preoperative therapy to
3 months and total chemotherapy duration to 6 months,
although on treatment durations are missing.
RESECTABLE LIVER METASTASES:

PERIOPERATIVE TREATMENT
Several studies on systemic or liver-directed chemotherapy
have investigated adjuvant or perioperative treatment
of resectable liver metastases. A meta-analysis of two
studiesboth closed early due to poor recruitment
demonstrated a strong trend towards improved overall
survival with adjuvant bolus 5-FU compared with surgery
alone (HR 1.32; 95% CI, 0.951.82; p = .095).39 The EORTC
trial 40983 (EPOC) randomly selected patients with up to
four metastases to receive perioperative treatment with
FOLFOX or liver surgery alone and showed a trend toward
improved disease-free survival with perioperative chemotherapy (HR 0.81; 95% CI, 0.641.02; p = .068). The overall
survival was not significantly different between both arms,
and the absolute difference in overall survival was relatively
small (3.4%; 95% CI, 7.113.8).40 A subgroup analysis of the
EPOC trial has shown interaction between CEA level, performance status, and obesity on the chemotherapy effect;
the HR for patients with normal CEA, patients with a performance status of 1 or higher, and obese patients (body
mass index $ 30) was greater than 1.0, suggesting harm
from perioperative treatment. The HRs for the opposite
groups (elevated CEA, performance status 0, and not obese)
were 0.60 or lower.41 Based on these results, it can be
FIGURE 2. (A) Conversion Chemotherapy in Technical Nonresectable Metastases (B) Effect of Chemotherapy for
Patients With a High Number of Metastases
e189
GUNNAR FOLPRECT
speculated that perioperative therapy might be indicated for

patients with higher risk for recurrence and excellent performance status.
This is especially important because the patients in the
trials mentioned above had relatively favorable risk
factors: In the EORTC EPOC trial,40 the majority of patients had a single liver metastasis and had metachronous metastases. In resectable metastases, there is no
evidence for adding any antibodies to perioperative or
adjuvant chemotherapy.
The British New EPOC trial randomly selected patients
with liver metastases to receive perioperative chemotherapy
with or without the EGFR-antibody cetuximab. Progressionfree and overall survival were worse with the addition of
cetuximab.42 The background for these findings remained
unclear and was the subject of long discussions, i.e., whether
less extensive surgery and higher rates of positive margins or
unknown biologic aspects explain the unexpected outcome.43
With the missing benefit of adding any antibody in adjuvant
treatment in stage III colon cancer44,45 and the lack of positive
results in perioperative treatment of resectable metastatic
disease, FOLFOX remains the regimen with the best evidence
of potential benefit for patients with resectable disease and
might be indicated especially in the presence of higher risk
factors.
CHEMOTHERAPY AND LIVER TOXICITY

Liver-related toxicity was described for irinotecan (steatohepatitis, 20%) and oxaliplatin (sinusoidal obstruction,
19%).46 Patients with steatohepatitis experience increased
mortality. In the EORTC 40983 study (investigating patients
with resectable disease), one of the 160 (0.6%) patients did
not undergo resection because of oxaliplatin-related liver
toxicity.47 Although chemotherapy-induced liver toxicity
was intensively discussed after these publications, this risk
has to be balanced in the context of patients with metastatic, nonresectable (or poorly resectable) disease. Because operative morbidity increases with the number of
chemotherapy cycles,48 it is recommended to not further
extend preoperative chemotherapy if metastases become
resectable. Furthermore, surgical strategies for patients who
have been heavily pretreated must be adapted, i.e., by
parenchymal-saving resections.21
INTRA-ARTERIAL THERAPIES
High response rates in phase II trials using intra-arterial
chemotherapy have revived the discussion on this treatment. In the French OPTILIV study, pretreated patients were
treated with systemic cetuximab and intra-arterial FOLFOXIRI.
The response rate was 41%, and 31% of patients were able to
undergo resection for their liver metastases.49 A protocol at
the Memorial Sloan Kettering Cancer Center used systemic
oxaliplatin and irinotecan combined with intra-arterial FUDR
in 49 pretreated patients, achieving response rate of 76%. In
total, 47% of patients were able to undergo resection in this
trial.50 These intra-arterial strategies are options if resection is
the treatment aim in pretreated patients without extrahepatic disease.
Selective internal radiation therapy (SIRT) adds another
treatment modality. In the SIRFLOX trial, patients with liverlimited or liver-dominant metastases were randomly selected to FOLFOX plus SIRT or FOLFOX alone. Progressionfree survival (primary endpoint) was not different between
the treatment arms (HR 0.93; 95% CI, 0.771.12; p = .43), and
the overall response rates were similar (76% vs. 68%; p = .11).
Progression-free survival for the liver lesions was longer in
the SIRT group (HR 0.69; 95% CI, 0.550.90), but overall
survival data are not yet available.51 Currently, it remains
unclear whether the later progression in the liver translates
to a longer overall survival.
CONCLUSION
Because multimodal treatment is associated with an improved prognosis, identifying patients who benefit from
multimodal treatment is crucial to the treatment of patients with metastatic colorectal cancer. Conversion
chemotherapy and ablative, interventional, and surgical
techniques can improve resectability and require a close
multidisciplinary cooperation. Sequencing of the treatment steps, the selection of patients who benefit from
resection or local treatment, and improved therapies to
treat remaining micrometastases are emergent open
questions in the treatment of patients with limited metastatic colorectal cancer.
References
1. Adam R, Delvart V, Pascal G, et al. Rescue surgery for unresectable
colorectal liver metastases downstaged by chemotherapy. Ann Surg.
2004;240:644-657.
2. Folprecht G, Gruenberger T, Bechstein WO, et al. Tumour response and
secondary resectability of colorectal liver metastases following neoadjuvant chemotherapy with cetuximab: the CELIM randomised phase
2 trial. Lancet Oncol. 2010;11:38-47.
3. Spelt L, Andersson B, Nilsson J, et al. Prognostic models for outcome
following liver resection for colorectal cancer metastases: a systematic
review. Eur J Surg Oncol. 2012;38:16-24.
e190
4. Fong Y, Fortner J, Sun RL, et al. Clinical score for predicting recurrence
after hepatic resection for metastatic colorectal cancer: analysis of
1001 consecutive cases. Ann Surg. 1999;230:309-318, discussion
318-321.
5. Masi G, Loupakis F, Pollina L, et al. Long-term outcome of initially unresectable metastatic colorectal cancer patients treated with
5-fluorouracil/leucovorin, oxaliplatin, and irinotecan (FOLFOXIRI) followed by radical surgery of metastases. Ann Surg. 2009;249:420-425.
6. Palma D a, Salama JK, Lo SS, et al. The oligometastatic state-separating
truth from wishful thinking. Nat Rev Clin Oncol. 2014;11: 549-557.
7. Ruers T, Punt C, Van Coevorden F, et al; EORTC Gastro-Intestinal Tract

Cancer Group, Arbeitsgruppe Lebermetastasen undtumoren in der
Chirurgischen Arbeitsgemeinschaft Onkologie, and the National Cancer
Research Institute Colorectal Clinical Study Group. Radiofrequency
ablation combined with systemic treatment versus systemic treatment
alone in patients with non-resectable colorectal liver metastases:
a randomized EORTC Intergroup phase II study (EORTC 40004). Ann
Oncol. 2012;23:2619-2626.
8. Ruers T, Punt CJA, van Coevorden F, et al. Radiofrequency ablation (RFA)
combined with chemotherapy for unresectable colorectal liver metastases (CRC LM): long-term survival results of a randomized phase II
study of the EORTC-NCRI CCSG-ALM Intergroup 40004 (CLOCC). J Clin
Oncol. 2015;33;(suppl; abstr 3501).
9. Folprecht G, Gruenberger T, Bechstein W, et al. Survival of patients with
initially unresectable colorectal liver metastases treated with FOLFOX/
cetuximab or FOLFIRI/cetuximab in a multidisciplinary concept (CELIM
study). Ann Oncol. 2014;25:1018-1025.
10. Rees M, Tekkis PP, Welsh FK, et al. Evaluation of long-term survival after
hepatic resection for metastatic colorectal cancer: a multifactorial
model of 929 patients. Ann Surg. 2008;247:125-135.
11. Kohne CH, van Cutsem E, Wils J, et al; European Organisation for Research and Treatment of Cancer Gastrointestinal Group. Phase III study
of weekly high-dose infusional fluorouracil plus folinic acid with or
without irinotecan in patients with metastatic colorectal cancer: European Organisation for Research and Treatment of Cancer Gastrointestinal Group Study 40986. J Clin Oncol. 2005;23:4856-4865.
12. de Gramont A, Bosset JF, Milan C, et al. Randomized trial comparing
monthly low-dose leucovorin and fluorouracil bolus with bimonthly
high-dose leucovorin and fluorouracil bolus plus continuous infusion for
advanced colorectal cancer: a French intergroup study. J Clin Oncol.
1997;15:808-815.
13. Takahashi S, Inoue K, Konishi M, et al. Prognostic factors for poor
survival after repeat hepatectomy in patients with colorectal liver
metastases. Surgery. 2003;133:627-634.
14. Shaw IM, Rees M, Welsh FKS, et al. Repeat hepatic resection for recurrent colorectal liver metastases is associated with favourable longterm survival. Br J Surg. 2006;93:457-464.
15. Lam VWT, Pang T, Laurence JM, et al. A systematic review of repeat
hepatectomy for recurrent colorectal liver metastases. J Gastrointest
Surg. 2013;17:1312-1321.
16. Makuuchi M, Thai BL, Takayasu K, et al. Preoperative portal embolization to increase safety of major hepatectomy for hilar bile duct
carcinoma: a preliminary report. Surgery. 1990;107:521-527.
17. Brouquet A, Abdalla EK, Kopetz S, et al. High survival rate after twostage resection of advanced colorectal liver metastases: responsebased selection and complete resection define outcome. J Clin
Oncol. 2011;29:1083-1090.
18. Lam VWT, Laurence JM, Johnston E, et al. A systematic review of twostage hepatectomy in patients with initially unresectable colorectal liver
metastases. HPB (Oxford). 2013;15:483-491.
19. Schnitzbauer AA, Lang SA, Goessmann H, et al. Right portal vein ligation
combined with in situ splitting induces rapid left lateral liver lobe
hypertrophy enabling 2-staged extended right hepatic resection in
small-for-size settings. Ann Surg. 2012;255:405-414.
20. Schadde E, Raptis DA, Schnitzbauer AA, et al. Prediction of mortality
after ALPPS stage-1: an analysis of 320 patients from the international
ALPPS registry. Ann Surg. 2015;262:780-785, discussion 785-786.
21. Evrard S, Poston G, Kissmeyer-Nielsen P, et al. Combined ablation and
resection (CARe) as an effective parenchymal sparing treatment for
extensive colorectal liver metastases. PLoS One. 2014;9:e114404.
22. Adams RA, Meade AM, Seymour MT, et al; MRC COIN Trial Investigators.
Intermittent versus continuous oxaliplatin and fluoropyrimidine
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
combination chemotherapy for first-line treatment of advanced

colorectal cancer: results of the randomised phase 3 MRC COIN trial.
Lancet Oncol. 2011;12:642-653.
Hegewisch-Becker S, Graeven U, Lerchenmuller CA, et al. Maintenance
strategies after first-line oxaliplatin plus fluoropyrimidine plus bevacizumab for patients with metastatic colorectal cancer (AIO 0207):
a randomised, non-inferiority, open-label, phase 3 trial. Lancet Oncol.
2015;16:1355-1369.
Simkens LHJ, van Tinteren H, May A, et al. Maintenance treatment with
capecitabine and bevacizumab in metastatic colorectal cancer
(CAIRO3): a phase 3 randomised controlled trial of the Dutch Colorectal
Cancer Group. Lancet. 2015;385:1843-1852.
Folprecht G, Grothey A, Alberts S, et al. Neoadjuvant treatment of
unresectable colorectal liver metastases: correlation between tumour
response and resection rates. Ann Oncol. 2005;16:1311-1319.
Jones RP, Hamann S, Malik HZ, et al. Defined criteria for resectability
improves rates of secondary resection after systemic therapy for liver
limited metastatic colorectal cancer. Eur J Cancer. 2014;50:1590-1601.
Loupakis F, Cremolini C, Masi G, et al. Initial therapy with FOLFOXIRI and
bevacizumab for metastatic colorectal cancer. N Engl J Med. 2014;371:
1609-1618.
Falcone A, Ricci S, Brunetti I, et al; Gruppo Oncologico Nord Ovest. Phase
III trial of infusional fluorouracil, leucovorin, oxaliplatin, and irinotecan
(FOLFOXIRI) compared with infusional fluorouracil, leucovorin, and
irinotecan (FOLFIRI) as first-line treatment for metastatic colorectal
cancer: the Gruppo Oncologico Nord Ovest. J Clin Oncol. 2007;25:
1670-1676.
Van Cutsem E, Kohne C-H, Hitre E, et al. Cetuximab and chemotherapy
as initial treatment for metastatic colorectal cancer. N Engl J Med. 2009;
360:1408-1417.
Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer.
N Engl J Med. 2004;350:2335-2342.
Kabbinavar FF, Hambleton J, Mass RD, et al. Combined analysis of
efficacy: the addition of bevacizumab to fluorouracil/leucovorin improves survival for patients with metastatic colorectal cancer. J Clin
Oncol. 2005;23:3706-3712.
Saltz LB, Clarke S, Daz-Rubio E, et al. Bevacizumab in combination with
oxaliplatin-based chemotherapy as first-line therapy in metastatic
colorectal cancer: a randomized phase III study. J Clin Oncol. 2008;26:
2013-2019.
Venook AP, Niedzwiecki D, Lenz H-J, et al. CALGB/SWOG 80405: phase III
trial of irinotecan/5-FU/leucovorin (FOLFIRI) or oxaliplatin/5-FU/
leucovorin (mFOLFOX6) with bevacizumab (BV) or cetuximab (CET)
for patients (pts) with KRAS wild-type (wt) untreated metastatic adenocarcinoma of the colon or rectum (MCRC). J Clin Oncol. 2014;32:5s
(suppl; abstr LBA3).
Heinemann V, von Weikersthal LF, Decker T, et al. FOLFIRI plus
cetuximab versus FOLFIRI plus bevacizumab as first-line treatment for
patients with metastatic colorectal cancer (FIRE-3): a randomised,
open-label, phase 3 trial. Lancet Oncol. 2014;15:1065-1075.
Schwartzberg LS, Rivera F, Karthaus M, et al. PEAK: a randomized,
multicenter phase II study of panitumumab plus modified fluorouracil,
leucovorin, and oxaliplatin (mFOLFOX6) or bevacizumab plus mFOLFOX6 in patients with previously untreated, unresectable, wild-type
KRAS exon 2 metastatic colorectal cancer. J Clin Oncol. 2014;32:
2240-2247.
Heinemann V, Stintzing S, Modest DP, et al. Early tumour shrinkage
(ETS) and depth of response (DpR) in the treatment of patients with
metastatic colorectal cancer (mCRC). Eur J Cancer. 2015;51:1927-1936.
Khattak MA, Martin H, Davidson A, et al. Role of first-line anti-epidermal
growth factor receptor therapy compared with anti-vascular
e191
GUNNAR FOLPRECT
38.
39.
40.
41.
42.
43.
44.
endothelial growth factor therapy in advanced colorectal cancer:

a meta-analysis of randomized clinical trials. Clin Colorectal
Cancer. 2015;14:81-90.
Venook A, Niedzwiecki D, Lenz H, et al. CALGB/SWOG 80405: analysis of
patients undergoing surgery as part of treatment strategy. Ann Oncol.
2014;25:(suppl 4; abstr LBA10).
Mitry E, Fields ALA, Bleiberg H, et al. Adjuvant chemotherapy after potentially curative resection of metastases from colorectal cancer: a pooled
analysis of two randomized trials. J Clin Oncol. 2008;26:4906-4911.
Nordlinger B, Sorbye H, Glimelius B, et al; EORTC Gastro-Intestinal Tract
Cancer Group; Cancer Research UK; Arbeitsgruppe Lebermetastasen
undtumoren in der Chirurgischen Arbeitsgemeinschaft Onkologie;
Australasian Gastro-Intestinal Trials Group; Federation Francophone de
Cancerologie Digestive. Perioperative FOLFOX4 chemotherapy and
surgery versus surgery alone for resectable liver metastases from colorectal cancer (EORTC 40983): long-term results of a randomised,
controlled, phase 3 trial. Lancet Oncol. 2013;14:1208-1215.
Sorbye H, Mauer M, Gruenberger T, et al; EORTC Gastro-Intestinal Tract
und-tumoren in der Chirurgischen Arbeitsgemeinschaft Onkologie;
Australasian Gastro-Intestinal Trials Group; Federation Francophone de
Cancerologie Digestive. Predictive factors for the benefit of perioperative FOLFOX for resectable liver metastasis in colorectal cancer
patients (EORTC Intergroup Trial 40983). Ann Surg. 2012;255:534-539.
Primrose J, Falk S, Finch-Jones M, et al. Systemic chemotherapy with or
without cetuximab in patients with resectable colorectal liver metastasis:
the New EPOC randomised controlled trial. Lancet Oncol. 2014;15:601-611.
Nordlinger B, Poston GJ, Goldberg RM. Should the results of the new
EPOC trial change practice in the management of patients with resectable metastatic colorectal cancer confined to the liver? J Clin Oncol.
2014;33:241-243.
Taieb J, Tabernero J, Mini E, et al; PETACC-8 Study Investigators.
Oxaliplatin, fluorouracil, and leucovorin with or without cetuximab in
e192
45.
46.
47.
48.
49.
50.
51.
patients with resected stage III colon cancer (PETACC-8): an open-label,

randomised phase 3 trial. Lancet Oncol. 2014;15:862-873.
Allegra CJ, Yothers G, OConnell MJ, et al. Phase III trial assessing
bevacizumab in stages II and III carcinoma of the colon: results of NSABP
protocol C-08. J Clin Oncol. 2011;29:11-16.
Vauthey JN, Pawlik TM, Ribero D, et al. Chemotherapy regimen predicts
steatohepatitis and an increase in 90-day mortality after surgery for
hepatic colorectal metastases. J Clin Oncol. 2006;24:2065-2072.
Nordlinger B, Sorbye H, Glimelius B, et al; EORTC Gastro-Intestinal Tract
und-tumoren in der Chirurgischen Arbeitsgemeinschaft Onkologie;
Australasian Gastro-Intestinal Trials Group; Fede ration Francophone de
Cancerologie Digestive. Perioperative chemotherapy with FOLFOX4 and
surgery versus surgery alone for resectable liver metastases from colorectal cancer (EORTC Intergroup trial 40983): a randomised controlled
trial. Lancet. 2008;371:1007-1016.
Karoui M, Penna C, Amin-Hashem M, et al. Influence of preoperative
chemotherapy on the risk of major hepatectomy for colorectal liver
metastases. Ann Surg. 2006;243:1-7.
Levi FA, Boige V, Hebbar M, et al; Association Internationale pour la
Recherche sur le Temps Biologique et la Chronotherapie. Conversion to
resection of liver metastases from colorectal cancer with hepatic artery
infusion of combined chemotherapy and systemic cetuximab in multicenter trial OPTILIV. Ann Oncol. 2015;27:267-274.
DAngelica MI, Correa-Gallego C, Paty PB, et al. Phase II trial of hepatic
artery infusional and systemic chemotherapy for patients with unresectable hepatic metastases from colorectal cancer: conversion to
resection and long-term outcomes. Ann Surg. 2015;261:353-360.
van Hazel GA, Heinemann V, Sharma NK, et al. SIRFLOX: randomized
phase III trial comparing first-line mFOLFOX6 (plus or minus bevacizumab) versus mFOLFOX6 (plus or minus bevacizumab) plus selective
internal radiation therapy in patients with metastatic colorectal cancer.
J Clin Oncol. Epub 2016 Feb 22.
Questions on Treatment of Locally

Advanced Rectal Cancer
CHAIR
Julio Garcia-Aguilar, MD, PhD
New York, NY
SPEAKERS
Deborah Schrag, MD, MPH
Boston, MA
Rob Glynne-Jones, MD, FRCP
Mount Vernon Cancer Centre
Middlesex, United Kingdom
GARCIA-AGUILAR, GLYNNE-JONES, AND SCHRAG
Multimodal Rectal Cancer Treatment: In Some Cases, Less

May Be More
Julio Garcia-Aguilar, MD, Rob Glynne-Jones, MBBS, FRCR, FRCP, and Deborah Schrag, MD, MPH
OVERVIEW
A series of clinical trials in the last several decades has resulted in the development of multimodality treatment of locally
advanced rectal cancer that includes neoadjuvant (preoperative) chemoradiotherapy, total mesorectal excision, and
postoperative adjuvant chemoradiotherapy. Owing to this regimen, patients with locally advanced rectal cancer have better
survival rates than patients with colon cancer, but at the cost of substantial morbidity and reduced quality of life. The
challenge is to identify treatment approaches that maintain or even improve oncologic outcomes while preserving quality of
life. We have identified different tumor characteristics that are associated with recurrence and probability of survival for
locally advanced rectal cancer. This risk stratification, based on baseline clinical staging and tumor response to chemoradiotherapy, has led us to question whether all patients with locally advanced rectal cancer require every component of the
multimodal regimen. In this article, we will review recent evidence that some patients with locally advanced rectal cancer
can be spared one or more treatment modalities without compromising long-term oncologic outcomes and while preserving
quality of life.
urgical excision of the rectum and its mesorectal envelope has been the mainstay of treatment for rectal cancer
for over a century.1 Despite advances in surgical technique
and perioperative care, total mesorectal excision remains
an operation associated with some mortality, substantial
morbidity, and sequelae that permanently impair quality
of life.2
CAN SURGERY BE AVOIDED FOR SELECTED

PATIENTS?
Some patients with locally advanced rectal cancer have a
pathologic complete response (pCR) to neoadjuvant chemoradiotherapy. Patients with pCR experience lower local
recurrence and improved survival rates compared with
patients who do not experience pCR, raising the question of
whether the former need surgery.3 Given that the mortality,
morbidity, and long-term sequelae from multimodal therapy
are related to excision of the rectum, avoiding total mesorectal excision for selective patients who obtain a sustained
response to chemoradiotherapy will improve their quality
of life.
Although the evidence suggesting that some rectal cancers
can be treated with radiation alone is almost a century old,
it is Angelita Habr-Gama from Sao Paulo who should be
credited with the suggestion that patients with rectal cancer

who experience a clinically complete response (cCR) to
neoadjuvant chemoradiotherapy could achieve long-term
local tumor control without surgery.4
These ideas were initially received with disbelief, but
reports from other institutions have confirmed that surgery
can be avoided for select patients with rectal cancer who
received treatment with chemoradiotherapy. However, the
evidence supporting this treatment approach is based on
small institutional series of heterogenous groups of patients
who were staged using different imaging modalities, treated
according to diverse radiation and chemotherapy regimens,
evaluated at different times after completion of the neoadjuvant therapy, selected for observation using different
criteria, and followed for relatively short periods of time. In
spite of these limitations, clinicians are starting to accept a
paradigm shift for this selected group of patients with rectal
cancer with a cCR after neoadjuvant chemoradiation. Acceptance has often been accelerated by patients motivated
to avoid the consequences of a low colorectal anastomosis
or a permanent colostomy. The treatment plan after neoadjuvant therapy that consists of close active surveillance,
rather than surgery, is called watch-and-wait, wait-andsee, or nonoperative management. We use the term
nonoperative management in this article.
From the Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY; Department of Medical Oncology, Mount Vernon Centre for Cancer Treatment, London, United
Kingdom; Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA.
Corresponding author: email: Julio Garcia-Aguilar, MD, Department of Surgery, Memorial Sloan Kettering Cancer Center, 1275 York Ave., New York, NY 10065;
email: garciaaj@mskcc.org.
92
RECTAL CANCER TREATMENT: LESS MAY BE MORE
Evidence Supporting Nonoperative Management

Most studies that have reported nonoperative management
outcomes in locally advanced rectal cancer have compared
recurrence and survival rates between patients with cCR
entered in a nonoperative management protocol and patients who had similar neoadjuvant treatment followed by
total mesorectal excision and subsequent pCR.
In Dr. Habr-Gamas protocol, clinical tumor response is
assessed 8 weeks after chemoradiotherapy.5 Patients with
persistent tumor undergo total mesorectal excision; those
with cCR undergo monthly evaluations including digital rectal
examination, proctoscopy, carcinoembryonic antigen (CEA)
blood testing, and biopsy of suspicious lesions. Patients found
to have tumor relapse are directed to surgery, whereas patients with a sustained cCR after 1 year continue surveillance
every 3 months for an additional year and every 6 months
thereafter. Of the patients treated according to this protocol,
27% experienced cCR after 1 year and were spared from total
mesorectal excision. Local relapse during follow-up beyond
1 year developed in 10% of patients treated with the
nonoperative management protocol, and all proceeded
to curative total mesorectal excision. The oncologic results in
this nonoperative management group were equivalent to
those of patients who had a pCR after total mesorectal
excision.
KEY POINTS
Treatment counseling should convey the risks and

benefits by using quantitative estimates expressed in
absolutenot relativeterms to empower informed
patient decision making.
Multimodality therapy provides excellent local tumor
control and long-term survival for patients with locally
advanced rectal cancer.
Phased-array MRI can define features associated with a
high risk of metastases and a high likelihood of local
recurrence.
Evidence supporting the use or non-use of adjuvant
chemotherapy in patients with rectal cancer is not
available, and so omission of this treatment can be
considered.
Select patients with rectal cancer who experience
complete clinical response after neoadjuvant therapy
can avoid total mesorectal excision, but the evidence is
still preliminary.
A selective approach to radiotherapy provision is
feasible with the provision that we use high-quality MRI
and optimal total mesorectal excision surgery.
Fluorouracil is best administered using an infusional
rather than a bolus schedule. Capecitabine is a
reasonable alternative to infusional 5-FU.
When FOLFOX is administered, it is essential to avoid
long-term toxicity by dose-reducing or discontinuing
oxaliplatin before peripheral neuropathy becomes
severe or persistent.
A group in the Netherlands reported their experience with

nonoperative management for 21 patients with cCR among
a total of 192 patients treated with chemoradiotherapy
between 2004 and 2010. After a mean follow-up of
25 months, local relapse developed in one patient who then
underwent curative salvage surgery; the other 20 patients
are alive without disease. Outcomes of patients treated with
the nonoperative management protocol who experienced
CR were similar to the outcomes of patients with pCR after
total mesorectal excision.6
Our group at the Memorial Sloan Kettering Cancer Center
reported the results of 32 patients with cCR treated with the
nonoperative management protocol who started neoadjuvant therapy between 2006 and 2010 and were followed for a median of 23 months. We compared these
results with 57 patients treated during the same period who
underwent total mesorectal excision and had a pCR. The
total mesorectal excision group had slightly more proximal
tumors and received adjuvant treatment more often, and
also had more advanced tumors compared with the patients
treated with the nonoperative management protcol. Six of
32 patients (21%) experienced disease relapse and underwent curative salvage surgery; distant metastasis developed in three of these patients. The 3-year disease-free
survival (DFS) and overall survival (OS) rates were not different
between groups.7 In an update to this series, we compared the
results of 73 patients with cCR treated with the nonoperative
management protocol and 72 patients treated with total
mesorectal excision who had pCR between 2006 and 2014.
Although relapse occurred among 19 of the 73 patients (26%)
treated with nonoperative management, 18 patients received
salvage therapy, and DFS and OS were equivalent between the
groups. In total, 56 of 73 patients (77%) who were treated
using the nonoperative management protocol preserved the
rectum after a median of 4 years of follow-up.8
Araujo et al reported a retrospective comparison of 42 patients with rectal cancer entered into a nonoperative management protocol with 69 patients treated with neoadjuvant
therapy and total mesorectal excision with subsequent pCR.
Nonoperative management was not an elective alternative and
was only considered for patients who refused surgery. Relapse
occurred in a total of 12 (28%) patients in the nonoperative
management group; local recurrence developed in five patients, distant metastasis developed in three patients, and both
local recurrence and distant metastasis occurred in four patients at a median of 2 years of follow-up. Four of the five
patients with isolated local recurrence had successful salvage
surgery. There was no difference in OS rates, but DFS was
reduced in the nonoperative management group compared
with the surgery group (61% vs. 83%). This study had several
unfortunate limitations including pretreatment tumor staging
performed by digital examination and a nonstandardized
definition of clinical response, and the groups were not
matched for important clinicopathologic variables such as
tumor distance from the anal verge.9
Li et al reported results from a cohort of 122 patients with
cCR among a total of 900 patients with stage II and III rectal
93
cancer treated with chemoradiotherapy. Of the 122 patients, 92 had rectal resection and 30 entered a nonoperative management protocol. Both groups of patients
were well-matched for clinical characteristics and received
similar treatment. Recurrence and DFS rates were not significantly different between groups.10
A group from the United Kingdom recently reported a
multi-institutional experience with a nonoperative management approach versus surgical resection for patients
with rectal cancer treated with chemoradiotherapy. In
contrast to the previous series, this study compared
the outcomes of 129 patients with cCR with 228 patients
who had surgical resection after chemoradiotherapy, independent of the pathologic stage. Neoadjuvant therapy
was similar between groups. After a median follow-up of
33 months from start of chemoradiotherapy, 44 (34%)
patients with cCR had local regrowth, corresponding to
an actuarial 3-year local regrowth rate of 38%. Similar to
previous findings, most local regrowths were on the bowel
wall, and most underwent successful salvage treatment. The
authors developed one-to-one paired cohorts (109 patients
in each group) using propensity-score matching for the key
confounders. The 3-year nonregrowth DFS rate (defined as
time until death, local recurrence, or distant metastasis, not
including local regrowths) was 88% for the nonoperative
management group and 78% for the surgical group (log rank;
p = .22). The colostomy-free survival rates were 74% and
47%, respectively. The authors concluded that nonoperative
management is safe in a multi-institutional setting, supporting the standard adoption of this protocol. However, the
results of this study should be interpreted with caution, as
tumors in patients treated with nonoperative management
had an earlier pretreatment tumor stage, were less likely to
have nodal involvement, rarely had unfavorable histologic
features, and were more likely to have normal CEA levels. In
addition, comparing patients with and without cCR, independent of the pathologic stage, introduces substantial
bias, as tumor response is associated with improved outcome compared with nonresponders.11
In summary, despite the relatively low-quality evidence,
these studies all suggest that most patients with cCR after
neoadjuvant chemotherapy can achieve prolonged local
tumor control without surgery. Tumor regrowth occurs in
some patients, but most can be effectively treated with
surgery. As the follow-up period in most of these series
is relatively short, long-term survival rates are still
unknown.
Challenges to Nonoperative Management of Rectal

Cancer
Although the above-mentioned studies all suggest that most
patients with cCR after neoadjuvant chemotherapy can
achieve prolonged local tumor control without surgery, a
number of questions must be answered before nonoperative
management can be considered a standard option for patients with locally advanced rectal cancer.
94
The proportion of patients who respond to chemoradiotherapy seems small, and the optimal time to assess
clinical response is unknown. Tumor response depends on
radiation dose, but doses beyond 54 Gy are rarely used in
these patients. Adding other drugs that are effective in colon
cancer as radiosensitizers beyond fluoropyrimidines has
been found to be ineffective or prohibitively toxic.12-15
Tumor response to chemoradiotherapy is closely associated
with time, and, for patients undergoing total mesorectal
excision after chemoradiotherapy, the proportion of tumors
with pCR increases with the time interval between chemoradiotherapy and surgery.16 As prolonging the interval to
surgery and postoperative systemic chemotherapy may be
unsafe for patients at risk for distant metastasis, attempts
have been made to deliver systemic chemotherapy immediately before or after chemoradiation.17,18 Delivering systemic chemotherapy before surgery, rather than after, has
been shown to increase tumor response without delaying
the treatment of potential micrometastatic disease. In these
patients, the assessment of clinical response and the recommendation of nonoperative management or surgery are
performed at the completion of both chemoradiotherapy
and systemic chemotherapy. This approach has resulted in
pCR rates as high as 38% for patients with clinical stage II and
III disease, and it has the added advantage of increasing
compliance with adjuvant systemic chemotherapy and
shortening ileostomy time for patients with locally advanced
rectal cancer.17,18 The experimental arm of the RAPIDO trial
uses short-term radiation and consolidation chemotherapy;
the trials results are forthcoming.19
The lack of a reliable and uniform method of distinguishing
post-treatment scar from residual tumor in the bowel wall or
regional lymph nodes is the main obstacle to nonoperative
management for patients treated with neoadjuvant therapy.
Most authors agree that digital rectal examination, endoscopy, and imaging studies should be used. A flat white scar
with or without telangiectasia and a normal digital examination are good predictors of pCR, whereas the presence of
superficial ulceration or a palpable modularity upon digital
rectal examination indicate an incomplete response.20,21
Although clinical assessment tends to underestimate tumor response, there is always a possibility that tumors
are concealed in or behind apparently normal scar tissue
in the rectal wall. 22 Endorectal ultrasound, CT, and
18
F-fludeoxyglucose PET provide an estimate of tumor
regression but are not sensitive enough to identify pCR.23
Conventional MRI morphologic sequences (e.g., T2- and
T1-weighted images) cannot differentiate residual tumor
from surrounding fibrosis, but diffusion-weighted MRI
sequences may improve the diagnostic performance of
morphologic MRI sequences for differentiating pCR from
residual tumor. 24
The definition of response undoubtedly influences clinical
outcomes: a strict definition reduces the proportion eligible
but increases the chance of success with nonoperative
management, whereas less-strict criteria increase the
number of eligible patients but also risk of local tumor
regrowth and distant metastasis. Although there are currently no validated criteria defining clinical and radiologic
tumor response, a new set of criteria categorizing response
in a three-tier system is currently being tested in a prospective clinical trial.25
A number of patients with apparent cCR develop tumor
regrowth during follow-up. As most regrowth occurs in the
bowel wall, repeated endoscopic examinations are essential.
Any suspicious changes in the scar should be biopsied. MRI
should also be performed regularly to detect nodal
regrowth. Changes in the size, contour, heterogeneity, or
restriction of diffusion should raise the possibility of relapse.
Repeated examinations and continuous monitoring are often necessary to confirm recurrence.
Ultimately, finding reliable predictors of response to
neoadjuvant therapy would help identify patients who
would most likely to benefit from nonoperative management as well as reduce toxicity for patients whose disease
will likely have a poor response. Tumor size and stage seem
to predict response, with smaller early-stage tumors being
more likely to develop pCR. The search for molecular predictors of tumor response has not resulted in breakthrough
findings so far. We have previously shown that rectal tumors
with a KRAS mutation are less likely to respond to neoadjuvant therapy.26 However, these findings must be validated in large independent cohorts.
In summary, the nonoperative management approach is
an attractive alternative for patients with rectal cancer who
experience cCR after neoadjuvant therapy. However, evidence supporting this approach is still relatively scarce and
too many questions about its safety remain to recommend
this approach outside of a clinical trial. Although a welldesigned phase III randomized trial comparing surgery
versus nonoperative management for patients with cCR
after neoadjuvant therapy may never be palatable to prospective patients, a number of prospective registries and
phase II trials are currently open to accrual.
recurrence observed among patients with rectal cancer

treated with total mesorectal excision alone challenged the
benefit of neoadjuvant therapy for patients with low-risk
stage II rectal tumors30-32 and called for a more selective
approach, particularly for upper rectal cancer.33-35 In addition, for patients with a high risk of metastatic disease,
the integration of more active chemotherapy into the
preoperative setting has become more attractive either in
addition to or as an alternative for chemoradiotherapy. The
clinical and imaging features and histopathologic analysis
of tumor determine the inherent risks of both local recurrence and distant metastasis.36,37 Variations in practice
and the different uptake of these modalities in rectal cancer
are common, as judgments are made on the balance of risk
and toxicity.
Surgery and radiotherapy often have permanent sequelae
resulting in an adverse effect on quality of life. Preoperative
chemoradiotherapy enhances local control and can be curative on its own for some patients, but it worsens the
constellation of changes in bowel function experienced by
many patients after sphincter-saving rectal cancer surgery, known as low anterior resection syndrome. Many
patients complain of chronic pain.38 Despite potentially
confounding factors such as older age or the effects of
prior surgical procedures in the pelvis, these symptoms
are worsened by the addition of short-course preoperative chemoradiotherapy (SCPRT) or conventional
chemoradiotherapy.39 In the Dutch total mesorectal excision trial, 62% of patients who received SCPRT before
total mesorectal excision reported fecal incontinence,
compared with 38% in the total mesorectal excision alone
group.40 Hence, a paradox: as we are advocating less chemoradiotherapy because surgery is getting better, we are
simultaneously advocating chemoradiotherapy to avoid
radical surgery.
Indications for Chemoradiotherapy
CAN WE AVOID NEOADJUVANT THERAPY OR

RADIOTHERAPY IN SELECTED PATIENTS?
Historically, surgeons reported a high pelvic recurrence rate
after radical surgery alone, resulting in a range of symptoms, including profuse mucinous discharge, bleeding, and
intractable pelvic pain.27 A landmark German study subsequently established preoperative chemoradiotherapy
as the standard of care.28 Although the rates of local recurrence reported before total mesorectal excision/
chemoradiotherapy came into widespread use would
not be acceptable today, these historical local recurrence
rates continue to dominate decision making.29 Many advocate the use of chemoradiotherapy for all patients staged
as cT3N0 or cT3N1 regardless of tumor site (low, middle, or
upper rectum), location (anterior or posterior), depth of
invasion, or proximity to other vital structures, despite the
ability of modern MRI to predict the risk of both local
recurrence and distant metastasis. The low rate of local
There are four main indications for delivering preoperative radiotherapy or chemoradiotherapy. These comprise
unresectable or borderline resectable cancers; resectable
cancers with a high risk of local recurrence; early rectal
cancers in older or frail patients as an alternative to
surgery; and treatment with palliative intent. The radiation oncologist should keep in mind that postoperative
chemoradiotherapy is associated with substantially more
acute and long-term morbidity compared with preoperative utilization.28 Hence, postoperative chemoradiotherapy
is not usually part of the planned management and usually
reflects poor preoperative decisions, unexpected surgical
or histopathologic findings, or both.
Tumors are considered unresectable or borderline resectable per initial MRI either because the primary tumor or
involved lymph nodes abut or breach the mesorectal fascia
or there is tumor outside the mesorectal fascia with local
extension to pelvic side-wall and sacrum or nodal involvement in the lateral pelvic lymph nodes. Lateral pelvic lymph
95
node involvement may reach as high as 15% when tumors

are low and bulky with multiple visible and presumed involved mesorectal nodes. These lateral pelvic lymph nodes,
although considered regional by the American Joint Committee on Cancer classification, are not routinely resected
during a standard total mesorectal excision operation. In
these circumstances, a degree of response to chemoradiotherapy is considered essential prior to surgery.41
Borderline resectable tumors are usually in the low or
midrectum. They demonstrate various adverse features
such as extension beyond the muscularis propria by 10 mm
to 15 mm, multiple nodal metastases in the mesorectum
(N2), or the presence of gross extramural vascular invasion.
These features are associated with a higher risk of local
recurrence and a high risk of metastatic disease. There are
sometimes less-evident clinical features that may alert the
surgeon to the potential difficulty of the operation and the
risk of not achieving an R0 resection. In particular, an anteriorly placed tumor at or below the level of the prostate,
in a large male with a narrow pelvis, is predictive of a
technically challenging dissection. The risk of a positive
circumferential resection margin in an anteriorly located
distal rectal cancer is as high as 30% even when treated with
an abdominoperineal excision. In these circumstances,
chemoradiotherapy to downsize the tumor will reduce the
risk of a circumferential resection margin or a defect in the
mesorectum.
Although the literature suggests that radiotherapy may
not fully compensate for a positive circumferential resection
margin, it may reduce the risk of local recurrence for patients
with a suboptimal total mesorectal excision surgery.42
Surgeons in the United Kingdom who are coming out of
training may have little experience and certainly face
a learning curve. As a radiation oncologist, without the
availability of records with multiple photographic images of
the surgical specimens to determine the quality of the
mesorectal excision in previous cases, I would likely recommend preoperative chemoradiotherapy. Recent data43
suggest that many patients with rectal cancer in the United
States are treated by surgeons who perform fewer than six
procedures per year. The circumferential resection margin
(when reported) in such cases has been documented as
positive in 17.2% of patients. Although SCPRT or chemoradiotherapy do not completely compensate for inadequate
surgery, there is sufficient historical data27,28,44,45 to argue
that surgeons entering the learning curve or practicing rectal
cancer surgery in a low-volume environment should offer
their patients neoadjuvant chemoradiotherapy or SCPRT,
even if the features mentioned in the preceding paragraph
are not necessarily present.
Preoperative radiation or chemoradiotherapy followed by
local excision is also an alternative to total mesorectal excision for patients with early stage rectal cancers treated
with curative intent.46,47 Chemoradiotherapy and local excision can also be used as a compromise treatment of patients with more advanced rectal cancers who refuse an
abdominoperineal excision of the rectum. Finally, radiation
96
or chemoradiotherapy with a brachytherapy boost may be

preferable to surgery for patients at high risk due to age or
multiple comorbidities.
MRI Is the Key to Selection

Modern phased-array coil MRI using various sequences and
multiple planes provides high-resolution tissue imaging and
excellent anatomic depiction of the rectum and other pelvic
structures relative to tumor location. MRI can predict involvement of the mesorectal fascia with a sensitivity of 77%
(95% CI, 57%90%) and specificity of 94% (95% CI, 88%
97%).48 The MERCURY trial, a prospective observational
study that assessed the accuracy of MRI in predicting a
curative resection in rectal cancer, reported 92% specificity
in predicting a negative circumferential resection margin.49
On multivariate analysis, MRI involvement of the circumferential resection margin was the only preoperative staging
parameter that remained predictive for local recurrence and
survival.50
Therefore, MRI is essential to stratify patients into risk
categories for local recurrence according to the proximity of
the tumor to the circumferential resection margin. These risk
categories are used to separate and select low-risk patients
who are unlikely to benefit from radiotherapy. MRI can also
define a group of patients with a high risk of recurrence
and metastases, typically suggested by extramural spread,
extramural vascular invasion, peritoneal reflection involvement, and tumor proximity to the levator muscle and
sphincter. High risk of metastatic disease means that the
use of chemotherapy at systemically effective doses would
seem essential to improve survival rates for patients with
locally advanced rectal cancer. This raises the question of
whether chemoradiotherapy is required if neoadjuvant
chemotherapy is effective in downstaging the tumor.
When Can Radiation Be Omitted?

The risk of local recurrence for patients with locally advanced
rectal cancer is dependent on tumor stage, the distance of
the tumor from the anal verge, and the proximity of the
tumor to the mesorectal fascia.50
Tumors located in the upper rectum, distant from the
mesorectal fascia, have a low risk of local recurrence when
treated with total mesorectal excision. The added benefit of
radiation therapy in these patients has been questioned, as
radiation is associated with substantial toxicity including
bowel obstruction, hip fractures, sexual and urinary dysfunction, and proctitis.51-53 A growing body of evidence
suggests that radiation therapy could be safely avoided for
patients with intermediate-risk rectal cancer, such as rectal
cancers located between 5 cm to 12 cm from the anal verge
that do not threaten the mesorectal fascia on MRI.54,55
In a pilot phase II trial conducted at Memorial Sloan
Kettering Cancer Center, 32 patients with resectable clinically staged II to III rectal cancer were treated with preoperative FOLFOX (oxaliplatin, 5-fluorouracil [5-FU], and
folinic acid) plus antivascular endothelial growth factor and
selective chemoradiotherapy based on tumor response.56

The 30 patients who completed preoperative chemotherapy
experienced tumor regression and underwent total mesorectal excision without preoperative chemoradiotherapy.
No local recurrences were noted at 4 years, and the DFS was
84%.56 These data were used as proof-of-concept for the
design of the phase II/III PROSPECT trial, which is now accruing worldwide. The overarching goal of this randomized
trial is to determine whether pelvic radiation therapy can be
used selectively for patients with locally advanced rectal
cancer.55 The trial utilizes selective rather than reflexive
chemoradiotherapy to individualize treatment based on
each patients response to neoadjuvant FOLFOX. The hope is
that a trial tailoring therapy more precisely, based on clinical
subgroups and tumor response to treatment, will help
eliminate the over- or undertreatment noted in previous
trials.54
In summary, chemoradiotherapy before total mesorectal
excision is currently the treatment of choice for patients with
high-risk locally advanced rectal cancer, including those with
involvement of the pelvic sidewall or mesorectal fascia. In
more straightforward cases, there is a question of whether
chemoradiotherapy alone or surgery alone should be considered given the long-term effects of both treatments. The
risks presented by either course should be estimated and
discussed with the patient. The argument for any of these
approaches rests on the provision of using good quality MRI,
good quality radiotherapy, and the surgeons record of good
performance of total mesorectal excision surgery within the
mesorectal plane.
ADJUVANT SYSTEMIC CHEMOTHERAPY FOR

LOCALLY ADVANCED (CLINICAL STAGE II AND III)
RECTAL CANCER
The controversy regarding adjuvant systemic chemotherapy
for rectal cancer is best understood in its historical context. In 1990, an American consensus conference57 recommended postoperative chemoradiotherapy and systemic
5-FU on the basis of two randomized trials that demonstrated better OS and reduced local and distant recurrence rates compared with surgery alone or to
surgery with radiation. 58,59 These trials did not distinguish between postoperative chemoradiotherapy and
postoperative chemotherapy, and therefore the benefits
realized were bundled as a multicomponent package. Since
then, adjuvant systemic treatment has been an entrenched
ingredient of curative-intent treatment for locally advanced rectal cancer and, indeed, has been adopted as a
measure of care quality. Unfortunately, the evidence underpinning this treatment recommendation is suboptimal,
particularly for patients with stage II locally advanced rectal
cancer.
The prevailing approach to curative-intent treatment of
rectal cancer for reasonably fit patients in the United States
is chemoradiotherapy followed by total mesorectal excision
and eight cycles of FOLFOX. The shift to neoadjuvant
chemoradiotherapy dates to 2004, when the German rectal

cancer trial demonstrated better local control with preoperative chemoradiotherapy with OS identical for pre- and
postoperative treatment.28,60 5-FUbased postoperative
chemotherapy was included in both arms of the German
trial. Also in 2004, the MOSAIC study in stage II and III colon
cancer compared 12 cycles of infusional 5-FU to 12 cycles
of FOLFOX and demonstrated superior survival with
FOLFOX.61,62 The benefits were not manifest in the stage II
subgroup. Although patients with rectal cancer were not
included, many clinicians have interpreted the results of
this study as pertinent to rectal cancer under the assumption that if FOLFOX is preferable for stage III colon
cancer, then logically it should be preferred for stage II and
III rectal cancer as well.
The persistent controversy regarding adjuvant treatment
of rectal cancer can be traced to varying interpretations of
these foundational studies. One school of thought considers
the clear benefit from the original trials conducted in the late
1980s as compelling evidence for postoperative 5-FU as
part of the package. Many also accept the legitimacy of
extrapolating from colon to rectal cancer and therefore
including oxaliplatin with 5-FU. Adherents of the opposing
viewpoint correctly point out that there is a dearth of
high-powered studies examining postoperative systemic
chemotherapy among patients treated with chemoradiotherapy and that there is compelling evidence that
patients with stage II tumors and/or cPR have very favorable
prognoses, particularly in the setting of high-quality preoperative imaging and optimal surgical technique, even in
the absence of systemic treatment. Posthoc analyses of
complete responders to chemoradiotherapy treated with
and without chemotherapy have failed to show the compelling advantage of the former approach.60,63,64
Does Systemic Chemotherapy Improve Rectal Cancer

Outcomes After Preoperative Chemoradiotherapy
and Surgical Resection?
Trials that have specifically addressed the benefit of systemic
chemotherapy compared with observation alone following
neoadjuvant chemoradiotherapy are few and have faced
considerable challenges. The EORTC 2991 study, which now
has 10 years of follow-up, used a two-by-two factorial design
to evaluate the inclusion of systemic chemotherapy in both
preoperative treatment and postoperative therapy.65 With
500 patients treated with adjuvant therapy and 500 without,
the power to detect small differences was necessarily limited. Although both DFS and OS rates favored the adjuvant
treatment arms, the differences were small: 3.3% for DFS
and 3.4% for OS at 10 years (Table 1). The primary impediment to interpretation of this study is that it used a
much older systemic treatment regimen, Mayo Clinicstyle
bolus 5-FU, which is now recognized to have less efficacy and
greater toxicity than infusional regimens.
The Dutch Colorectal Cancer Group planned the PROCTOR/
SCRIPT trial with 840 patients to detect a difference between
97
TABLE 1. Outcomes of Patients With Rectal Cancer After Receiving Neoadjuvant Therapy and Subsequent
Treatment With Nonoperative Management or Total Mesorectal Excision
Reference
cCRs
pCRs
(NOM) (OM)
Interval to
Salvage After
Regrowth in Regrowth in
Regrowth in
NOM (%)
NOM (months) NOM (%)*
Overall Survival
NOM vs. OM
(p value)
Disease-Free
Survival NOM vs.
OM (p value)
Permanent Stoma Evidence

NOM vs. OM
Level**
71
Habr-Gama
et al 20045
22
2 (3)
60
2 (100)
5-year; 100% vs.

88% (p = .01)
5-year; 92% vs.

83% (p = .09)
0% vs. 41%
3B
Maas et al
20116
21
20
1 (5)
22
1 (100)
2-year; 100% vs.

93% (p = .23)
2-year; 89% vs.

91% (p = .77)
0% vs. 45%
3B
Smith et al
20127
32
57
6 (19)
11
6 (100)
2-year; 96% vs.

2-year; 88% vs.
100% (p = .56)
98% (p = .27)
NS
3B
Araujo et al
20159
42
69
5 (12)
48
4 (80)
5-year; 72% vs.

90% (p = .32)
5-year; 61% vs.

83% (p = .04)
7% vs. 19%
92*
2 (7)
22
2 (100)
5-year; 100% vs.

96% (p = .26)
5-year; 90% vs.

94% (p = .51)
0% vs. 43%
Li et al 201510 30
*Patients with cCR subjected to surgery.

**Level according to Oxford Centre for Evidence-Based Medicine levels of evidence and grades of recommendations for therapeutic interventions.
Abbreviations: NOM, nonoperative management; TME, total mesorectal excision; cCR, clinically complete response; pCR, pathologic complete response; OM, operative management;
NS, not significant.
60% to 70% in OS.66 Unlike the EORTC 2291 trial, patients

were randomly assigned to receive 5-FU postoperatively
and after completing preoperative chemoradiotherapy.
PROCTOR/SCRIPT accrued poorly. It was amended to
substitute eight cycles of capecitabine for the original bolus
5-FU regimens, but it still did not accrue well and closed
after a total of 437 patients had been assigned. With 5 years
of follow-up, there is a 7% increase in 5-year DFS in the
adjuvant therapy arm. This difference did not reach significance. Moreover, OS is nearly identical between the arms.
An Italian trial reported by Cionini et al randomly assigned
635 patients with stage II/III rectal cancer to receive six
cycles of bolus 5-FU and leucovorin or observation, and it
found no difference in OS (Table 2).67
Overall, trials that have directly addressed the question of
the benefit of postoperative adjuvant systemic therapy for
patients with rectal cancer treated with neoadjuvant chemoradiotherapy have not demonstrated the superiority of
treatment to observation. However, it is also fair to say that
these studies have not demonstrated the absence of benefit.
These trials had trouble accruing, were underpowered, and
used antiquated bolus 5-FU/leucovorin regimens. The only
study that included oxaliplatin68 failed to complete accrual
and was closed prematurely. As a result, clinicians who treat

rectal cancer do not have an answer to this fundamental
question. However, the low accrual rates in both the United
Kingdom and the Netherlands, attributed to lack of equipoise,
make it unlikely that trials with an observation arm will ever
be conducted. Instead, focus has shifted to intensive efforts to
characterize the phenotypes and genotypic subsets of rectal
cancer that are associated with poor prognosis and/or greater
responsiveness to systemic therapy and to tailor treatment
thresholds to risk. Indeed, most recent rectal cancer clinical
trials examine strategies for tailoring treatment to risk profile
to strike a balance between either under- or overtreatment.
Does the Addition of Oxaliplatin to 5-FU Improve

Outcomes Compared With 5-FU Alone?
More recent trials have assumed that systemic therapy is
beneficial and have focused on addressing the benefit of
adding oxaliplatin to 5-FU in postoperative adjuvant regimens for rectal cancer. In the wake of MOSAIC, oncologists
have accepted the transitive principle and presumed that if
adding oxaliplatin is better than adjuvant infusional 5-FU/
leucovorin alone in colon cancer, the same must be true in
rectal cancer. Typically, the 12 cycles used in MOSAIC are
TABLE 2. Randomized Clinical Trials Comparing Postoperative Systemic Chemotherapy With Observation of
Patients With Rectal Cancer Treated with Preoperative Pelvic Radiotherapy
Disease-Free Survival
Study (Maturity)
No. Patients
EORTC2291 (10-year) 1,101

PROCTOR/SCRIPT
(5-year)
Overall Survival
Adjuvant Treatment
Rx+ vs. Rx-
Difference Rx+ vs. Rx-
4 cycles of bolus 5-FU/LV every

3 weeks
47 vs. 43.7
3.3%
437; closed early 8 cycles of capecitabine or bolus 62.7 vs. 55.4 7.3%
5-FU/LV
Difference Reference
51.8 vs. 48.4 3.4%
Bossett et al44,65
80.4
Breugom et al66
79.2%
CHRONICLE (3-year) 113; closed early 6 cycles of CaPOx
77.5 vs. 71.3 6.2%
87.8 vs. 88.8 21%
Glynne-Jones et al68
Cionini
N/A
68 vs. 64
Cionini et al67
635
6 cycles of bolus 5-FU/LV every

4 weeks
N/A
4%
Abbreviations: 5-FU, 5-fluorouracil; LV, leucovorin; Rx+, with treatment; RX-, without treatment; CaPOx, capecitabine and oxaliplatin; N/A, not applicable.
98
reduced to eight cycles. Somewhat arbitrarily, credit for

four cycles is given for chemoradiotherapy.
Notwithstanding the paucity of direct evidence that any
postoperative chemotherapy is superior to observation in
locally advanced rectal cancer, several randomized trials
have directly compared oxaliplatin with nonoxaliplatincontaining regimens for postoperative adjuvant treatment
of rectal cancer. Specifically, the ECOG E320169 study sought
to compare eight cycles of postoperative 5-FU and leucovorin with two to eight cycles of FOLFOX after standard
neoadjuvant chemoradiotherapy and surgery. The study
accrued fewer than 100 patients, with surveys indicating a
perceived lack of equipoise.
Fortunately, three non-U.S.based studies were more
successful. The PETACC-6 study randomly assigned 1,069
patients to receive either six cycles of adjuvant capecitabine
and oxaliplatin (CAPOX) or to capecitabine alone. Although
median follow-up was only 31 months, 3-year DFS rates were
nearly identical in both arms.70 The most recent study from
the German Rectal Cancer Group randomly assigned 1,265
patients to receive either four cycles of Mayo Clinicstyle
bolus 5-FU or eight cycles of postoperative FOLFOX. With
50 months of follow-up, 3-year DFS was 71.2% for 5-FU and
75.9% for FOLFOX, a difference of 4.7%.15 Mature data and a
final report have not yet been published. In contrast, the
recent randomized ADORE trial from South Korea revealed
an advantage for FOLFOX compared with Mayo Clinicstyle
bolus 5-FU/leucovorin. This study randomly assigned 321
patients and reported a 71.6% 3-year DFS rate in the FOLFOX
group compared with 62.9% in the 5-FU/leucovorin group.71
This difference was significant with a hazard ratio (HR) of
0.66 (95% CI, 0.430.99; p = .047). In the FOLFOX arm, the
3-year OS was 95% versus 85.7% in the control arm, a difference of 9.3% with a HR of 0.46 (95% CI, 0.220.97). This
study has been critiqued for its small sample size and the
different 5-FU backbones in the two arms. As a result, it is
unclear whether the better outcomes in the intervention
arm were attributable to the superiority of infusional versus
bolus 5-FU or to the addition of oxaliplatin.
Treatment Decision Making in the Absence of Perfect

Information
For patients treated with preoperative chemoradiotherapy,
decisions about postoperative systemic therapy must be
tailored to the risk of metastatic disease based on pathology.
They must also take into account the patients life expectancy independent of the cancer, as well as the patients
preferences. In some contexts, decision making is easy. For
example, extrapolating from the MOSAIC trial, patients with
node-positive rectal cancer are often recommended eight
cycles of FOLFOX. For patients with no evidence of nodal
involvement and excellent response, chemoradiotherapy
may be omitted. In this context, it is important to attenuate
or remove oxaliplatin altogether before peripheral neuropathy becomes severe (grade 3). Among patients who are
frail or do not tolerate oxaliplatin, some oncologists still
prefer to administer infusional 5-FU without oxaliplatin

rather than no adjuvant systemic therapy. Conversely, for
patients with substantial comorbidities that suggest life
expectancy of less than 5 years, omitting treatment seems
most appropriate.
The challenge arises in the treatment of patients with cT3
node-negative tumors with complete or favorable pathologic responses to neoadjuvant chemoradiotherapy including cT3N0 and yT0N0, T1N0 or T2N0. In this context,
there is reluctance to administer postoperative FOLFOX
given its toxicity and the imperfect evidence supporting its
use. The focus of recent and current randomized trials is to
evaluate alternative preoperative management strategies
and completion of systemic chemotherapy prior to surgery,
so it is possible that this conundrum will eventually have less
salience.
Nevertheless, how can todays clinicians arrive at informed
decisions in the face of imperfect information? There are
two strategies that may be helpful. The first is careful
consideration of the most appropriate treatment prior to
initiation of any treatment. Although not infallible, MRI
staging helps to identify relatively small rectal cancers with
low likelihood of nodal involvement. For rectal tumors that
are cT2N0 or cT3N0, it makes sense to consider up-front
surgical resection. Many of these patients will have nodenegative disease and can reasonably avoid systemic therapy.
Patients with node-positive disease require both postoperative chemoradiotherapy and systemic chemotherapy.
Although less well-tolerated than preoperative treatment, it
is important to recall that the German trial showed similar
OS irrespective of treatment timing.28,60
The second approach to adjuvant treatment recommendations in clinical stage II rectal cancer or in the setting of
pCR to neoadjuvant chemoradiation is to scrutinize the
phenotypic and genotypic tumor characteristics for high-risk
features. Recently, Dalerba et al identified the transcription
factor CDX2 as a negative prognostic factor in stage II and III
colon cancer.72 They found that 4.1% of colon cancers lack
CDX2 expression. In patients with stage II disease, 5-year DFS
was approximately 50% for the subset lacking CDX2 expression and over 80% for the majority with CDX2 expression. Moreover, the 5-year DFS was 91% for the 23 patients
with CDX2-positive stage II disease treated with adjuvant
chemotherapy and 56% for the 25 patients with CDX2positive disease who did not receive adjuvant treatment.
Though information on the prevalence of CDX2 in patients
with rectal cancer is not yet available, administration of
adjuvant systemic treatment to patients with rectal cancer
lacking CDX2 expression is justified even in the setting of
node-negative disease or pCR, given the magnitude of this
finding. This study highlights the value of including tissue
correlatives alongside clinical trials and using bioinformatics
to identify heterogeneity of risk and treatment responsiveness. Further discoveries like these practicechanging results are anticipated and will help us to tailor
adjuvant treatment recommendations to individual risk
profiles.
99
Practice Guidelines
A considerable body of evidence demonstrates favorable
outcomes among patients with average-risk stage II rectal
cancer treated with chemoradiotherapy who have complete or
near-complete pathologic response without adjuvant systemic
treatment. European and U.S. practice guidelines recommend
treatment decisions tailored to individual patient preferences
but interpret the evidence base somewhat differently. In the
United States, the 2015 National Comprehensive Cancer
Network guideline recommends postoperative adjuvant systemic therapy for all patients treated with neoadjuvant chemoradiotherapy irrespective of response.73 The inclusion of
oxaliplatin is suggested. In Europe, the European Society of
Medical Oncology guideline advises postoperative chemotherapy in stage III and high-risk stage II cancers.74 However, in
some countries, treatment is not consistently recommended in
the setting of complete pathologic response and a recent
European Society of Medical Oncology consensus statement
acknowledges persistent uncertainty.75
In summary, there is still much work to be done in elucidating the circumstances in which adjuvant systemic
chemotherapy can be safely omitted from treatment for
locally advanced rectal cancer. Although there is some evidence to suggest the safety of omitting this treatment for
some patients with stage II locally advanced rectal cancer,
this treatment is indispensable in many situations, notably
for patients with post-chemoradiotherapy invasive or nodeinvolved tumors and tumors lacking CDX2 expression.
CONCLUSION
All patients embarking on curative-intent therapy for rectal
cancer should be informed of the risks, benefits, and uncertainties of surgery, chemoradiotherapy, and adjuvant
systemic therapy, the three pillars of treatment. Modern
MRI technology is essential to informed decision making, but
patients comorbidities, natural life expectancy, and preferences must be considered. The conversations about
treatment recommendations must convey risks and benefits
in simple language that takes into account patients
numeracy. Clinical trial participation should always be

prioritized.
Total mesorectal excision, although effective in the
majority of locally advanced rectal cancer cases and still
the mainstay of curative-intent treatment of rectal cancer, harbors the risks of substantial mortality, morbidity,
and long-term quality-of-life issues. There is a growing
body of evidence that supports nonoperative management for patients whose disease demonstrates a cCR to
neoadjuvant chemoradiotherapy. However, this approach
must be paired with frequent and thorough follow-up,
especially in the first year after completion of chemoradiotherapy, and patients must be counseled about the
risks both of the nonoperative management approach and
the possibility that surgery may be necessary if recurrence
is detected.
Chemoradiotherapy should be offered to patients with
unresectable or borderline unresectable tumors and resectable tumors with high-risk features. Because of its
proven ability to eradicate some tumors and its less onerous
safety profile, chemoradiotherapy alone may be the most
appropriate choice for selected patients, particularly those
unfit or unwilling to undergo surgery.
For some patients with no evidence of clinical nodal involvement, no high-risk features, and excellent response to
chemoradiotherapy, the omission of postoperative chemotherapy may be reasonable provided that there has been
explicit discussion about the imperfect evidence supporting
this decision. However, when patients with clinical stage II
rectal cancer or patients with a good response to chemoradiotherapy decide to receive postoperative systemic
therapy, it is critical to dose-reduce and/or discontinue
oxaliplatin altogether in the setting of incipient peripheral
neuropathy to prevent long-term adverse effects; in some
patients, oxaliplatin may be avoided entirely. Patients must
understand that when neoadjuvant chemoradiotherapy is
administered, postoperative systemic treatment is part of
the package, because it is not always possible to know
whether the tumor did or did not involve lymph nodes at
the outset.
References
1. Heald RJ, Smedh RK, Kald A, et al. Abdominoperineal excision of the
rectuman endangered operation. Norman Nigro Lectureship. Dis
Colon Rectum. 1997;40:747-751.
2. Hoerske C, Weber K, Goehl J, et al. Long-term outcomes and quality of
life after rectal carcinoma surgery. Br J Surg. 2010;97:1295-1303.
3. Maas M, Nelemans PJ, Valentini V, et al. Long-term outcome in patients with a
pathological complete response after chemoradiation for rectal cancer:
a pooled analysis of individual patient data. Lancet Oncol. 2010;11:835-844.
4. Habr-Gama A, de Souza PM, Ribeiro U Jr, et al. Low rectal cancer: impact
of radiation and chemotherapy on surgical treatment. Dis Colon Rectum. 1998;41:1087-1096.
5. Habr-Gama A, Perez RO, Nadalin W, et al. Operative versus nonoperative
treatment for stage 0 distal rectal cancer following chemoradiation
100
6.
7.
8.
9.
therapy: long-term results. Ann Surg. 2004;240:711-717, discussion 717718.

Maas M, Beets-Tan RG, Lambregts DM, et al. Wait-and-see policy for
clinical complete responders after chemoradiation for rectal cancer.
J Clin Oncol. 2011;29:4633-4640.
Smith JD RJ, Ruby JA, Goodman KA, et al. Nonoperative management of
rectal cancer with complete clinical response after neoadjuvant therapy. Ann Surg. 2012;256:965-972.
Smith JJ, Chow OS, Eaton A, et al. Organ preservation in patients with
rectal cancer with clinical complete response after neoadjuvant therapy. J Clin Oncol. 2015;33 (suppl 3; abstr 509).
Araujo RO, Valad~ao M, Borges D, et al. Nonoperative management of
rectal cancer after chemoradiation opposed to resection after complete
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
clinical response. A comparative study. Eur J Surg Oncol. 2015;41:

1456-1463.
Li J, Liu H, Yin J, et al. Wait-and-see or radical surgery for rectal cancer
patients with a clinical complete response after neoadjuvant chemoradiotherapy: a cohort study. Oncotarget. 2015;6:42354-42361.
Renehan AG, Malcomson L, Emsley R, et al. Watch-and-wait approach
versus surgical resection after chemoradiotherapy for patients with
rectal cancer (the OnCoRe project): a propensity-score matched cohort
analysis. Lancet Oncol. 2015;17:174-183.
OConnell MJ CL, Colangelo LH, Beart RW, et al. Capecitabine and
oxaliplatin in the preoperative multimodality treatment of rectal
cancer: surgical end points from National Surgical Adjuvant Breast and
Bowel Project trial R-04. J Clin Oncol. 2014;32:1927-1934.
Aschele C, Cionini L, Lonardi S, et al. Primary tumor response to preoperative chemoradiation with or without oxaliplatin in locally advanced rectal cancer: pathologic results of the STAR-01 randomized
phase III trial. J Clin Oncol. 2011;29:2773-2780.
Gerard JP, Azria D, Gourgou-Bourgade S, et al. Comparison of two
neoadjuvant chemoradiotherapy regimens for locally advanced rectal
cancer: results of the phase III trial ACCORD 12/0405-Prodige 2. J Clin
Oncol. 2010;28:1638-1644.
Rodel C, Liersch T, Becker H, et al; German Rectal Cancer Study Group.
Preoperative chemoradiotherapy and postoperative chemotherapy
with fluorouracil and oxaliplatin versus fluorouracil alone in locally
advanced rectal cancer: initial results of the German CAO/ARO/AIO-04
Probst CP, Becerra AZ, Aquina CT, et al; Consortium for Optimizing the
Surgical Treatment of Rectal Cancer (OSTRiCh). Extended intervals after
neoadjuvant therapy in locally advanced rectal cancer: the key to
improved tumor response and potential organ preservation. J Am Coll
Surg. 2015;221:430-440.
Garcia-Aguilar J, Renfro LA, Chow OS, et al. Organ preservation for
clinical T2N0 distal rectal cancer using neoadjuvant chemoradiotherapy
and local excision (ACOSOG Z6041): results of an open-label, single-arm,
multi-institutional, phase 2 trial. Lancet Oncol. 2015;16:1537-1546.
Cercek A, Goodman KA, Hajj C, et al. Neoadjuvant chemotherapy first,
followed by chemoradiation and then surgery, in the management of
locally advanced rectal cancer. J Natl Compr Canc Netw. 2014;12:
513-519.
Nilsson PJ, van Etten B, Hospers GA, et al. Short-course radiotherapy
followed by neo-adjuvant chemotherapy in locally advanced rectal
cancerthe RAPIDO trial. BMC Cancer. 2013;13:279.
Habr-Gama A, Perez RO, Wynn G, et al. Complete clinical response after
neoadjuvant chemoradiation therapy for distal rectal cancer: characterization of clinical and endoscopic findings for standardization. Dis
Colon Rectum. 2010;53:1692-1698.
Smith FM, Wiland H, Mace A, et al. Clinical criteria underestimate
complete pathological response in rectal cancer treated with neoadjuvant chemoradiotherapy. Dis Colon Rectum. 2014;57:311-315.
Duldulao MP, Lee W, Streja L, et al. Distribution of residual cancer cells
in the bowel wall after neoadjuvant chemoradiation in patients with
rectal cancer. Dis Colon Rectum. 2013;56:142-149.
Samdani T, Garcia-Aguilar J. Imaging in rectal cancer: magnetic resonance imaging versus endorectal ultrasonography. Surg Oncol Clin N
Am. 2014;23:59-77.
Lambregts DM, Lahaye MJ, Heijnen LA, et al. MRI and diffusionweighted MRI to diagnose a local tumour regrowth during long-term
follow-up of rectal cancer patients treated with organ preservation
after chemoradiotherapy. Eur Radiol. Epub 2015 Oct 30.
Smith JJ, Chow OS, Gollub MJ, et al; Rectal Cancer Consortium. Organ
preservation in rectal adenocarcinoma: a phase II randomized controlled trial evaluating 3-year disease-free survival in patients with
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
locally advanced rectal cancer treated with chemoradiation plus induction or consolidation chemotherapy, and total mesorectal excision
or nonoperative management. BMC Cancer. 2015;15:767.
Garcia-Aguilar J, Chen Z, Smith DD, et al. Identification of a biomarker
profile associated with resistance to neoadjuvant chemoradiation
therapy in rectal cancer. Ann Surg. 2011;254:486-492, discussion 492493.
Swedish Rectal Cancer Trial. Improved survival with preoperative radiotherapy in resectable rectal cancer. N Engl J Med. 1997;336:980-987.
Sauer R, Becker H, Hohenberger W, et al; German Rectal Cancer Study
Group. Preoperative versus postoperative chemoradiotherapy for
rectal cancer. N Engl J Med. 2004;351:1731-1740.
Quirke P, Steele R, Monson J, et al; MRC CR07/NCIC-CTG CO16 Trial
Investigators; NCRI Colorectal Cancer Study Group. Effect of the plane
of surgery achieved on local recurrence in patients with operable rectal
cancer: a prospective study using data from the MRC CR07 and NCICCTG CO16 randomised clinical trial. Lancet. 2009;373:821-828.
Law WL, Ho JW, Chan R, et al. Outcome of anterior resection for stage II
rectal cancer without radiation: the role of adjuvant chemotherapy. Dis
Colon Rectum. 2005;48:218-226.
Frasson M, Garcia-Granero E, Roda D, et al. Preoperative chemoradiation may not always be needed for patients with T3 and T2N+
rectal cancer. Cancer. 2011;117:3118-3125.
Mathis KL, Larson DW, Dozois EJ, et al. Outcomes following surgery
without radiotherapy for rectal cancer. Br J Surg. 2012;99:137-143.
Popek S, Tsikitis VL, Hazard L, et al. Preoperative radiation therapy for
upper rectal cancer T3,T4/Nx: selectivity essential. Clin Colorectal
Cancer. 2012;11:88-92.
Kreis ME, Junginger T, Rodel C, et al. The optimult study concept selective neoadjuvant chemoradiation therapy based on preoperative
MRI. Zentralbl Chir. 2010;135:302-306.
Strassburg J, Ruppert R, Ptok H, et al. MRI-based indications for neoadjuvant radiochemotherapy in rectal carcinoma: interim results of a
prospective multicenter observational study. Ann Surg Oncol. 2011;18:
2790-2799.
Syk E, Glimelius B, Nilsson PJ. Factors influencing local failure in rectal
cancer: analysis of 2315 patients from a population-based series. Dis
Colon Rectum. 2010;53:744-752.
Kusters M, Marijnen CA, van de Velde CJ, et al. Patterns of local recurrence in rectal cancer; a study of the Dutch total mesorectal excision
trial. Eur J Surg Oncol. 2010;36:470-476.
Feddern ML, Jensen TS, Laurberg S. Chronic pain in the pelvic area or
lower extremities after rectal cancer treatment and its impact on quality
of life: a population-based cross-sectional study. Pain. 2015;156:
1765-1771.
Bregendahl S, Emmertsen KJ, Lous J, et al. Bowel dysfunction after low
anterior resection with and without neoadjuvant therapy for rectal
cancer: a population-based cross-sectional study. Colorectal Dis. 2013;
15:1130-1139.
Peeters KC, Tollenaar RA, Marijnen CA, et al; Dutch Colorectal Cancer
Group. Risk factors for anastomotic failure after total mesorectal excision of rectal cancer. Br J Surg. 2005;92:211-216.
Oh HK, Kang SB, Lee SM, et al. Neoadjuvant chemoradiotherapy affects
the indications for lateral pelvic node dissection in mid/low rectal
cancer with clinically suspected lateral node involvement: a multicenter
retrospective cohort study. Ann Surg Oncol. 2014;21:2280-2287.
Gosens MJEM, Klaassen RA, Tan-Go I, et al. Circumferential margin
involvement is the crucial prognostic factor after multimodality
treatment in patients with locally advanced rectal carcinoma. Clin
Cancer Res. 2007;13:6617-6623.
Rickles AS, Dietz DW, Chang GJ, et al; Consortium for Optimizing the
Treatment of Rectal Cancer (OSTRiCh). High rate of positive
101
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
102
circumferential resection margins following rectal cancer surgery: a call

to action. Ann Surg. 2015;262:891-898.
Kapiteijn E, Marijnen CA, Nagtegaal ID, et al; Dutch Colorectal Cancer
Group. Preoperative radiotherapy combined with total mesorectal
excision for resectable rectal cancer. N Engl J Med. 2001;345:638-646.
Bosset JF, Collette L, Calais G, et al; EORTC Radiotherapy Group Trial
22921. Chemotherapy with preoperative radiotherapy in rectal cancer.
N Engl J Med. 2006;355:1114-1123.
Garcia-Aguilar J, Renfro LA, Chow OS, et al. Organ preservation for
clinical T2N0 distal rectal cancer using neoadjuvant chemoradiotherapy
and local excision (ACOSOG Z6041): results of an open-label, single-arm,
multi-institutional, phase 2 trial. Lancet Oncol. 2015;16:1537-1546.
Verseveld M, de Graaf EJ, Verhoef C, et al. Chemoradiation therapy for
rectal cancer in the distal rectum followed by organ-sparing transanal
endoscopic microsurgery (CARTS study). Br J Surg. 2015;102:853-860.
Al-Sukhni E, Milot L, Fruitman M, et al. Diagnostic accuracy of MRI for
assessment of T category, lymph node metastases, and circumferential
resection margin involvement in patients with rectal cancer: a systematic review and meta-analysis. Ann Surg Oncol. 2012;19:2212-2223.
MERCURY Study Group. Diagnostic accuracy of preoperative magnetic
resonance imaging in predicting curative resection of rectal cancer:
prospective observational study. BMJ. 2006;333:779.
Taylor FG, Quirke P, Heald RJ, et al; MERCURY study group. Preoperative
high-resolution magnetic resonance imaging can identify good prognosis
stage I, II, and III rectal cancer best managed by surgery alone: a prospective, multicenter, European study. Ann Surg. 2011;253:711-719.
Peeters KC, van de Velde CJ, Leer JW, et al. Late side effects of short-course
preoperative radiotherapy combined with total mesorectal excision for
rectal cancer: increased bowel dysfunction in irradiated patientsa Dutch
colorectal cancer group study. J Clin Oncol. 2005;23:6199-6206.
Birgisson H, Pahlman L, Gunnarsson U, et al; Swedish Rectal Cancer Trial
Group. Adverse effects of preoperative radiation therapy for rectal
cancer: long-term follow-up of the Swedish Rectal Cancer Trial. J Clin
Oncol. 2005;23:8697-8705.
Joye I, Haustermans K. Early and late toxicity of radiotherapy for rectal
cancer. Recent Results Cancer Res. 2014;203:189-201.
Gunderson LL, Sargent DJ, Tepper JE, et al. Impact of T and N stage and
treatment on survival and relapse in adjuvant rectal cancer: a pooled
analysis. J Clin Oncol. 2004;22:1785-1796.
Schrag D. Evolving role of neoadjuvant therapy in rectal cancer. Curr
Treat Options Oncol. 2013;14:350-364.
Schrag D, Weiser M, Goodman KA, et al. Neoadjuvant chemotherapy
without routine use of radiation therapy for patients with locally advanced rectal cancer: a pilot trial. J Clin Oncol. 2014;32:513-518.
NIH consensus conference. Adjuvant therapy for patients with colon
and rectal cancer. JAMA. 1990;264:1444-1450.
Fisher B, Wolmark N, Rockette H, et al. Postoperative adjuvant chemotherapy or radiation therapy for rectal cancer: results from NSABP
protocol R-01. J Natl Cancer Inst. 1988;80:21-29.
Gastrointestinal Tumor Study Group. Prolongation of the disease-free
interval in surgically treated rectal carcinoma. N Engl J Med. 1985;312:
1465-1472.
Sauer R, Liersch T, Merkel S, et al. Preoperative versus postoperative
chemoradiotherapy for locally advanced rectal cancer: results of the
German CAO/ARO/AIO-94 randomized phase III trial after a median
follow-up of 11 years. J Clin Oncol. 2012;30:1926-1933.
Andre T, Boni C, Mounedji-Boudiaf L, et al; Multicenter International
Study of Oxaliplatin/5-Fluorouracil/Leucovorin in the Adjuvant Treatment
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
of Colon Cancer (MOSAIC) Investigators. Oxaliplatin, fluorouracil, and

leucovorin as adjuvant treatment for colon cancer. N Engl J Med. 2004;
350:2343-2351.
Andre T, Boni C, Navarro M, et al. Improved overall survival with
oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment in stage II
or III colon cancer in the MOSAIC trial. J Clin Oncol. 2009;27:3109-3116.
Garca-Albeniz X, Gallego R, Hofheinz RD, et al. Adjuvant therapy sparing
in rectal cancer achieving complete response after chemoradiation.
World J Gastroenterol. 2014;20:15820-15829.
Bujko K, Glynne-Jones R, Bujko M. Does adjuvant fluoropyrimidinebased chemotherapy provide a benefit for patients with resected rectal
cancer who have already received neoadjuvant radiochemotherapy?
A systematic review of randomised trials. Ann Oncol. 2010;21:
1743-1750.
Bosset JF CG, Calais G, Mineur L, et al; EORTC Radiation Oncology Group.
Fluorouracil-based adjuvant chemotherapy after preoperative chemoradiotherapy in rectal cancer: long-term results of the EORTC 22921
randomised study. Lancet Oncol. 2014;15:184-190.
Breugom AJ, van Gijn W, Muller EW, et al; Cooperative Investigators of
Dutch Colorectal Cancer Group and Nordic Gastrointestinal Tumour
Adjuvant Therapy Group. Adjuvant chemotherapy for rectal cancer
patients treated with preoperative (chemo)radiotherapy and total
mesorectal excision: a Dutch Colorectal Cancer Group (DCCG) randomized phase III trial. Ann Oncol. 2015;26:696-701.
Cionini L, Sainato A, De Paoli A, et al. Final results of randomized trial on
adjuvant chemotherapy after preoperative chemoradiation in rectal
cancer. Radiother Oncol. 2010;96:S113-S114.
Glynne-Jones R, Counsell N, Quirke P, et al. Chronicle: results of a
randomised phase III trial in locally advanced rectal cancer after
neoadjuvant chemoradiation randomising postoperative adjuvant
capecitabine plus oxaliplatin (XELOX) versus control. Ann Oncol. 2014;
25:1356-1362.
Benson AB, Catalan P, Meropol NJ, et al. ECOG E3201: Intergroup
randomized phase III study of postoperative irinotecan, 5-fluorouracil
(FU), leucovorin (LV) (FOLFIRI) vs oxaliplatin, FU/LV (FOLFOX) vs FU/LV
for patients (pts) with stage II/III rectal cancer receiving either pre or
postoperative radiation (RT)/FU. J Clin Oncol. 2006;24 (suppl; abstr
3526).
Schmoll HJ, Haustermans K, Price TJ, et al. Preoperative chemoradiotherapy and postoperative chemotherapy with capecitabine and
oxaliplatin versus capecitabine alone in locally advanced rectal cancer:
Disease-free survival results at interim analysis. J Clin Oncol. 2014;32:5s
Hong YS, Nam BH, Kim KP, et al. Oxaliplatin, fluorouracil, and leucovorin
versus fluorouracil and leucovorin as adjuvant chemotherapy for locally
advanced rectal cancer after preoperative chemoradiotherapy
(ADORE): an open-label, multicentre, phase 2, randomised controlled
Dalerba P, Sahoo D, Paik S, et al. CDX2 as a prognostic biomarker in stage
II and stage III colon cancer. N Engl J Med. 2016;374:211-222.
Benson AB III, Venook AP, Bekaii-Saab T, et al. Rectal cancer, version
2.2015. J Natl Compr Canc Netw. 2015;13:719-728, quiz 728.
Glimelius B, Tiret E, Cervantes A, et al; ESMO Guidelines Working Group.
Rectal cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2013;24:vi81-vi88.
Schmoll HJ, Van Cutsem E, Stein A, et al. ESMO Consensus Guidelines for
management of patients with colon and rectal cancer. a personalized
approach to clinical decision making. Ann Oncol. 2012;23:2479-2516.
Biliary Tract Cancer: The New and

the Old
CHAIR
Mary F. Mulcahy, MD
Robert H. Lurie Comprehensive Cancer Center of Northwestern University
Chicago, IL
SPEAKERS
John A. Bridgewater, MD, PhD
University College London Cancer Institute
Karyn A. Goodman, MD
University of Colorado School of Medicine
Aurora, CO
BRIDGEWATER ET AL
Biliary Tract Cancer: Epidemiology, Radiotherapy, and

Molecular Profiling
John A. Bridgewater, MD, PhD, Karyn A. Goodman, MD, Aparna Kalyan, MD, and Mary F. Mulcahy, MD
OVERVIEW
Biliary tract cancer, or cholangiocarcinoma, arises from the biliary epithelium of the small ducts in the periphery of the liver
(intrahepatic) and the main ducts of the hilum (extrahepatic), extending into the gallbladder. The incidence and epidemiology
of biliary tract cancer are fluid and complex. It is shown that intrahepatic cholangiocarcinoma is on the rise in the Western
world, and gallbladder cancer is on the decline. Radiation therapy has emerged as an important component of adjuvant therapy
for resected disease and definitive therapy for locally advanced disease. The emerging sophisticated techniques of imaging
tumors and conformal dose delivery are expanding the indications for radiotherapy in the management of bile duct tumors. As
we understand more about the molecular pathways driving biliary tract cancers, targeted therapies are at the forefront of new
therapeutic combinations. Understanding the gene expression profile and mutational burden in biliary tract cancer allows us to
better discern the pathogenesis and identify promising new developmental therapeutic targets.
iliary tract cancer, or cholangiocarcinoma, arises from

the biliary epithelium of the small ducts in the periphery
of the liver (intrahepatic) and the main ducts of the hilum
(extrahepatic). Extrahepatic biliary tract cancers include
gallbladder cancer, ampullary cancer, and cancer of the
pancreatic biliary ducts. Although extrahepatic cancers arise
from similar epithelia, their etiology can be very different
because of their anatomy.
The incidence of cholangiocarcinoma is modest in the western
world, between 0.35 to 2 per 100,000 annually; however, in
China and Thailand, the incidence can be up to 40 times the rate
observed in the United Kingdom and, thus, poses significant
public health questions (Fig. 1).1,2 The incidence of gallbladder
cancer tends to be closely associated with its primary etiology,
cholelithiasis. As such, the incidence is uniform for most of the
Western world, however, disease clusters are found in northern
India, Japan, and the Andes region (Fig. 2).3
The incidence of intrahepatic cholangiocarcinoma in the
Western world is rising.1 Data from the United Kingdom,
United States, and other countries show a consistent and
steady rise in the incidence of intrahepatic cholangiocarcinoma
from 0.1 to 0.6 per 100,000 over the last 30 years (Figs. 3
and 4).4-10 Several reasons have been cited for this change.
The International Classification of Diseases (ICD) for cholangiocarcinoma has been confusing.11 There are multiple
codes for cholangiocarcinoma depending on histology and
topography. This coding has recently changed three times (ICD0-1 to ICD-0-2 in 1993 and ICD-0-3 in 2001) and has been
adopted at different times in different countries. In ICD-0-2,
hilar cholangiocarcinoma could be classified pathologically using a histology code, rather than an anatomic code, but was
cross-referenced to the anatomic code for intrahepatic cholangiocarcinoma rather than extrahepatic cholangiocarcinoma.
Furthermore, hilar cholangiocarcinoma could also be correctly
reported as extrahepatic cholangiocarcinoma by using other
histology codes.12 Additionally, there have been local changes in
data collection, such as the loss of independent verification of
cause of death in the United Kingdom in 1993. The terminology
may also be difficult in certain locations such as in England and
Wales, where only 1% of cholangiocarcinomas were classified
as intrahepatic.5
This can be considered as an abrupt step change in the
recorded incidence, such as the step change in incidence in
2001 for intrahepatic cholangiocarcinoma.12 Despite these
noted step changes, the overall pattern is a rise in incidence demonstrated across multiple international data sets.
Changes in technology, such as improved stenting, would
similarly present as step changes, although these are not
discernable in the data. Additionally, the incidence appears
consistent across different tumor sizes and mortality. The
overall picture presents an increase in incidence that appears to override short-term changes.
From the UCL Cancer Institute, London, United Kingdom; Department of Radiation Oncology, University of Colorado School of Medicine, Aurora, CO; Northwestern University, Chicago,
IL; Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago, IL.
Corresponding author: Mary F. Mulcahy, MD, Robert H. Lurie Comprehensive Cancer Center of Northwestern University, 676 North St., Clair Suite 850, Chicago, IL 60611;
email: mmulcahy@nm.org.
e194
BILIARY TRACT CANCER
FIGURE 1. Incidence of Cholangiocarcinoma
One refuted hypothesis to explain the increase in intrahepatic cholangiocarcinoma is the increase in diagnoses
of intrahepatic cholangiocarcinomas that were previously
thought to be metastases from a carcinoma of unknown
origin. Again, if true, this is unlikely to have a substantial
impact on incidence, as the numbers are modest. A number
of criteria have been proposed to help differentiate between
KEY POINTS
The incidence of intrahepatic cholangiocarcinoma in the

Western world is rising, and the incidence of gallbladder
cancer is on the decline.
The tendency for cholangiocarcinomas to recur locally
provides a rationale for additional local therapy after
definitive surgery.
A recent meta-analysis of 10 retrospective studies
evaluating adjuvant radiotherapy after curative
resection for extrahepatic cholangiocarcinomas
demonstrated a significant benefit in overall survival for
patients receiving adjuvant therapy.
The landscape of molecular mutations in biliary tract
cancers has demonstrated multiple new targetable
mutations.
The majority of biliary tract cancer mediators seem to
affect the epigenetics and transcription.
intrahepatic cholangiocarcinoma and carcinoma of unknown

origin.13
The incidence of cholangiocarcinoma in Thailand demonstrates the impact of chronic infection with liver fluke.
Liver cancers rival HIV/AIDS, road traffic accidents, and
stroke as a primary cause of death in Thai males.2 Endemic
liver fluke infection (Opisthorchis viverrini) is related to
eating raw or poorly cooked fish over 20 years or longer. The
picture in China and Korea is similar, although the description of data is less detailed, and the culprit organism is
Clonorchis sinensis.14 Fluke-related cancer has a distinctive
microRNA and mutational profile compared with cholangiocarcinoma of the Western world. The relationship to
malignancy appears to be related to the generation of radical
nitrogen species demonstrated in a murine model.
In the Western world, cholangiocarcinoma is associated
with chronic inflammation of the biliary tree and hepatic
parenchyma. The lifetime incidence of cholangiocarcinoma
in hepatitis C is modest at 3.5% at 10 years. The data on
primary sclerosing cholangitis (PSC) suggest a lifetime risk of
7% and rising, corresponding with the increasing incidence
of PSC. As for PSC, this increase is primarily among men age 40
or younger and appears to spare the small ducts.
Gallbladder cancer has a different pathophysiology and is
related primarily, but not entirely, to cholecystitis. The incidence varies from England and Wales (0.1%) to the Czech
Republic (3.7%6.5%) and La Paz in Bolivia (7.5%13.2% per
100,000). The risk factors and demographics for gallbladder
cancer and gallstone disease closely parallel each other. As
e195
BRIDGEWATER ET AL
FIGURE 2. Incidence of Gallbladder Cancer
such, there is a predominance of women and known risk factors

for gallstone disease (age, obesity, multiple pregnancies, family
history of gallstones, and low levels of physical activity). Chronic
infection is also implicated with the relative risk consequent on
Salmonella (typhi and paratyphi) and Helicobacter (bilis and
pylori) isolation, between 2.6 and 7.5. Interestingly, only 1% of
patients with gallstones develop gallbladder cancer, suggesting
that any screening program should be targeted to higher-risk
groups.
The incidence of gallbladder cancer is decreasing in the
Western world and is related to the increase in routine
cholecystectomy. This is not the case in Chile, where public
health programs have diverted funding for cholecystectomy
to maternity services, resulting in a slow but substantial increase in the rate of gallbladder cancer. Nevertheless, prophylactic cholecystectomy programs have been proposed.
Primary sclerosing cholangitis and natural killer cell polymorphisms have been associated with gallbladder cancer.
Choledochal cysts have a 1% to 15% lifetime risk of developing into gallbladder cancer and cholangiocarcinoma. The
risk of finding malignancy is higher when the diagnosis is made
in adults, but recurrence is rare if it is resected.
ADJUVANT AND DEFINITIVE RADIOTHERAPY

FOR RESECTABLE AND LOCALLY ADVANCED
Role of Adjuvant Therapy for Biliary Tract Tumors
Complete resection remains the gold standard of treatment and offers the best chance of cure for patients with
biliary tract cancers. The type and extent of resection are
e196
determined by the location of the tumor and the extent of

invasion into adjacent structures. Biliary tract tumors are
generally divided into intrahepatic cholangiocarcinomas,
perihilar and extrahepatic cholangiocarcinomas, and gallbladder tumors. For intrahepatic cholangiocarcinomas, which
are less common, a liver resection based on the anatomic
involvement of the liver is performed, whereas for extrahepatic cholangiocarcinoma, a pancreaticoduodenectomy may
be required depending on the level of disease within the
biliary tree. Gallbladder cancers are often incidentally found
after a nononcologic surgery, and a repeat oncologic resection
to obtain adequate margins at the liver parenchyma may be
necessary.
Even with a complete resection, the 5-year overall survival rates range from 21% to 63% for intrahepatic cholangiocarcinoma, 30% to 40% for perihilar lesions, and
20% to 54% for distal cholangiocarcinomas managed by
pancreaticoduodenectomy.15-17 The recurrence patterns
differ by primary site in the biliary tract. Investigators at the
Memorial Sloan Kettering Cancer Center reviewed the
patterns of recurrence among 156 patients with extrahepatic cholangiocarcinoma and gallbladder tumors who underwent definitive resection. Two-thirds of the patients in
each group developed disease recurrence, with an isolated
locoregional disease as the first site of failure in 15% of
patients with gallbladder cancer compared with 59% of
patients with extrahepatic cholangiocarcinoma (p , .001).
By contrast, 85% of gallbladder cancers had distant spread as
the first site of failure compared with 41% of extrahepatic
cholangiocarcinomas (p , .001).18
FIGURE 3. Age-Standardized Mortality Rates per

100,000 in England and Wales5
FIGURE 4. Age-Adjusted (1970 U.S. Standard)

Incidence Rates of Primary Intrahepatic
Cholangiocarcinoma in the United States, 197319971
Reprinted with permission. 2001 American Association for the Study of Liver
Diseases. All rights reserved. Patel T: Hepatology. 2001;33:1353-1357.
Reprinted by permission from Macmillan Publishers Ltd: Gut. Taylor-Robinson SD

et al. Increase in mortality rates from intrahepatic cholangiocarcinoma in England and
Wales 1968-1998. Gut. 2001;48:816-820. 2001.
The tendency for cholangiocarcinomas to recur locally

provides a rationale for additional local therapy after definitive surgery. Although several phase II and retrospective
series suggest a benefit, phase III data are lacking.19-22

Moreover, the indications for postoperative radiation
therapy are not yet clear. The most important prognosticator for survival is a complete (R0) resection23,24; nonetheless, the available data suggest isolated locoregional
recurrence rates of up to 60% after curative resection.25-27
A recent meta-analysis of 10 retrospective studies evaluating adjuvant radiotherapy after curative resection for
extrahepatic cholangiocarcinomas demonstrated a significant benefit in overall survival for patients receiving adjuvant therapy.28 The pooled hazard ratio (HR) for
overall survival for the addition of adjuvant radiation
therapy versus surgery alone was 0.62 (95% CI, 0.480.78),
despite the fact that the radiotherapy cohorts more frequently had positive margins (69% vs. 31%; p , 0.001).
The authors concluded that adjuvant radiotherapy appeared beneficial, however, this meta-analysis ultimately only included eight studies on extrahepatic
cholangiocarcinomas. Conversely, a Surveillance, Epidemiology, and End Results (SEER) analysis of adjuvant radiotherapy failed to show a survival benefit, but the authors
caution that key data, including margin status and the use
of combined chemotherapy that might impact on overall
outcomes for these patients, were not available through the
SEER database.29
The recently reported Intergroup phase II study, led
by the Southwest Oncology Group (SWOG), evaluated
the use of adjuvant chemotherapy (gemcitabine and
capecitabine) followed by chemoradiation (concurrent
radiotherapy and capecitabine) for resected extrahepatic
e197
BRIDGEWATER ET AL
cholangiocarcinoma or gallbladder tumors. There were a

total of 79 evaluable patients, 21 (32%) of whom underwent an R1 resection and 49 (62%) of whom had extrahepatic cholangiocarcinoma. Two-year overall survival
was 65% for all patients (67% and 60% in R0 and R1, respectively). Median overall survival was 35 months, and
only 14 patients developed a local recurrence.30 Grade 3
and 4 adverse events were observed for 53% and 11% of
patients. These promising data demonstrate the feasibility
of adjuvant chemotherapy followed by chemoradiation for
resected extrahepatic cholangiocarcinoma and gallbladder
tumors.
Adjuvant radiotherapy may also benefit patients with
intrahepatic cholangiocarcinoma due to the high risk of
positive resection margins.15 An analysis of intrahepatic
cholangiocarcinoma patients from the SEER database demonstrated an improvement in median survival for adjuvant radiotherapy versus surgery alone (11 vs. 6 months,
respectively).31
Definitive Radiotherapy for Biliary Tract Tumors

In cases of unresectable nonmetastatic cholangiocarcinoma,
the use of radiotherapy may improve pain control and biliary decompression. A recent retrospective review of 37 patients with unresectable extrahepatic cholangiocarcinomas
treated with radiotherapy at Duke University reported 2-year
actuarial overall survival and local control rates of 22% and
71%, respectively. Two patients were alive without evidence
of recurrence at the time of analysis, thus demonstrating longterm survival in a small subset of patients.32 For intrahepatic
cholangiocarcinoma, a dosimetric analysis of patients treated
at The University of Texas MD Anderson Cancer Center
suggests that a biologically effective dose of greater than
80.5 Gy is associated with improved local control and longterm survival.33
Several small, single-institution, prospective studies have
evaluated the use of stereotactic body radiotherapy for
hilar and intrahepatic cholangiocarcinomas, suggesting
that the higher dose per fraction is feasible and results in
encouraging local control rates for both peripheral and
central biliary tumors.34-40 Stereotactic body radiotherapy
allows for highly focal dose distributions and sparing of the
surrounding normal tissue; however, given the close
proximity of hilar masses to highly radiosensitive organs
such as the small bowel and stomach, stereotactic body
radiotherapy for hilar cholangiocarcinomas should be used
with caution and preferably on a clinical trial. A recent
multi-institutional phase II study of hypofractionated
proton therapy (median dose 58 Gy) for intrahepatic
cholangiocarcinoma demonstrated a local control rate of
94% and 2-year overall survival of 46%.41 Short-term
morbidity associated with radiotherapy includes nausea,
vomiting, malaise, fatigue, or weight loss and intermittent
fevers due to cholangitis. Long-term complications following radiotherapy are usually related to duodenal bleeding
or stricture.
e198
Brachytherapy
Brachytherapy has been used at numerous institutions.
Overall, there appears to be a small trend toward improved
survival with its use as a boost.42,43 Dose distributions can be
quite satisfactory with the use of Ir-192 catheters. Ideally, an
external beam dose designed to eradicate microscopic
disease (i.e., 45 to 50 Gy) is given first, followed by seed
placement designed to give an additional high dose to the
site of gross or highly suspect disease. Long-term complications secondary to the high doses and use of stents can
include stricture, bowel obstruction, and bleeding.
Bridge to Transplant
The role of liver transplantation for unresectable hilar
cholangiocarcinoma has been established by the excellent
results from the Mayo Clinic using 4,500 cGy neoadjuvant
chemoradiation and concurrent fluorouracil and a brachytherapy boost followed by liver transplantation.44 The 5-year
survival rate was 82%, which compared favorably to another
group of patients who underwent resection with a 5-year
survival rate of only 21%. It must be emphasized that
transplantation is reserved for highly selected patients, and
selection criteria are quite stringent.
Advances in Radiotherapy Techniques

The advancement of radiotherapy over the last 2 decades
has been a remarkable achievement that has improved
delivery, outcomes, and quality of care for patients with
primary biliary tumors. Historically, radiotherapy for abdominal tumors has been limited by the tolerance of the
surrounding normal organs to even relatively low doses of
radiation. With the emergence of more conformal radiotherapy techniques and sophisticated treatment-planning
software, more focal treatment fields are now possible.
Classic approaches to portals and fractionation have been
abandoned in favor of more focal techniques of irradiating
liver tumors, such as intensity-modulated radiotherapy,
image-guided radiotherapy, and the use of motion management. When planning treatment using intensity-modulated
radiotherapy, the radiation dose is prescribed to the target
volume and strict dose constraints are placed on normal
tissues. A computer-optimized algorithm generates a plan to
deliver radiation of varying intensity to the target volume via
multiple beams or even via multiple arcs to meet the requirements for the target volume coverage and normal
tissue constraints. The overall result is a highly conformal
dose distribution that is customized to the shape of the
tumor. The delivery of conformal intensity-modulated radiotherapy plans is predicated on the ability to accurately
target the tumor during treatment, which has led to the
development of image-guided radiotherapy to reduce the
uncertainties of tumor positioning. Diagnostic-quality imaging can now be performed on the linear accelerator to
allow real-time assessment of tumor positioning while
the patient is on the table prior to treatment delivery.
Abdominal tumors can move with respiration, and, thus,
adaptations to account for motion or reduce motion have

been developed to allow accurate delivery of the focal radiotherapy fields while patients are breathing. These motion
management techniques include using an abdominal compression belt to minimize the excursion of the diaphragm
during respiration to reduce motion or to use respiratory
gating where the linear accelerator turn on and off with the
respiratory cycle. In summary, the emerging sophisticated
techniques of imaging tumors and conformal dose delivery
are expanding the indications for radiotherapy in the
management of bile duct tumors.
MOLECULAR TARGETS IN BILIARY TRACT

CANCERS
A majority of biliary tract cancers (. 90%) are welldifferentiated, mucin-producing adenocarcinomas, whereas
a very small proportion are squamous or small cell in
origin.45,46
Patterns of Molecular Drivers in Biliary Tract Cancer

As we understand more about the molecular pathways involved in biliary tract cancers, targeted therapies are at the
forefront of new therapeutic combinations. The landscape
of molecular mutations in biliary tract cancers has demonstrated multiple new targetable mutations. To date, Ross
et al has performed the most comprehensive review of the
molecular profile of biliary tract cancers.47 In this study, DNA
from 412 intrahepatic cholangiocarcinomas, 57 extrahepatic cholangiocarcinomas, and 85 gallbladder cancers were
extracted and next-generation sequencing (NGS) was performed on these specimens. Genomic profiling encompassed 182 cancer-related genes plus 37 introns from 14
genes frequently rearranged in cancer. These results demonstrate that biliary tract cancers all share genomic
aberrations in cell cycle regulators (specifically CDKN2B)
and chromatin remodeling (ARID1A). Intrahepatic cholangiocarcinomas feature FGFR fusions, IDH1/2 substitutions, BRAF substitutions, and MET amplifications with a
low KRAS mutational frequency.48 ERBB2 amplification and
PIK3CA/mTOR pathway aberrations were more common in

extrahepatic cholangiocarcinomas and gallbladder tumors.
Mutations in TP53 have been long established in biliary tract cancers, with incidence ranging around 44% to
47%. Incidence is seen more frequently in extrahepatic
cholangiocarcinoma compared with intrahepatic cholangiocarcinoma, with rates as high as 17.5% and 8.6%,
respectively.48 Multivariate analysis identified that TP53 and
KRAS are independent predictors of survival.48 Similar frequency and mutational burden were seen in a study by Churi
et al in which NGS was performed using a panel of 46 cancerrelated genes.49 In addition to corroborating the findings by
Ross et al, Churi et al found that genetic aberrations in
chromatin-modulating genes, BAP1 and PBRM1, were associated with worse overall survival and higher frequency of
bony metastases for patients with extrahepatic cholangiocarcinoma. KRAS mutations have been shown to
occur at a higher rate in extrahepatic cholangiocarcinoma
(42%47%) compared with intrahepatic cholangiocarcinoma
(15.7%22%) and gallbladder cancer (11%19.2%), confirmed by whole-exome sequencing studies and genomic
profiling studies (Table 1).47-50
Chromatin-remodeling genes BAP1 (encoding a nuclear deubiquitinase), ARID1A (encoding a subunit of the
SWI/SNF chromatin-remodeling complexes), and PBRM1
(encoding a subunit of the ATP-dependent SWI/SNF chromatinremodeling complexes) have been observed in biliary tract
cancer. The incidence of ARID1A ranges from 11% to 20% in
intrahepatic cholangiocarcinoma specimens, although its
incidence is 12% in extrahepatic cholangiocarcinoma and
11% to 13% in gallbladder cancer.47-49,51 Similarly, BAP1 is
seen in 9% to 25% of intrahepatic cholangiocarcinomas
and 4% to 13% of gallbladder cancers. The incidence of
PBRM1 was 11% to 17% in intrahepatic cholangiocarcinoma,
7.7% in gallbladder cancer, and 3.5% to 5% in extrahepatic
cholangiocarcinoma. The frequency of these chromatinremodeling genes indicates that not only is chromatin
remodeling a key component in cholangiocarcinoma, but
these mutations may lend themselves to be sensitive to
therapeutic agents targeting the chromatin-remodeling
genes.47-49,51
TABLE 1. Incidence of Molecular Mutations in Biliary Tract Cancer as Determined by Genomic Sequencing
Mutation
Intrahepatic Cholangiocarcinoma
Extrahepatic Cholangiocarcinoma
Gallbladder Cancer
16%
ERBB2 Amplification
3%
11%
BRAF Substitution
5%
3%
KRAS
PI3KCA Substitution
15%22%
5%
FGFR1-3 Fusion
11%12.5%
CDKN2A/B Loss
18%
IDH1/2 Substitution
15%23%
ARID1A Alteration
11%20%
42%47%
7%
1%
11%19.2%
14%
3%
17%
19%
3%4%
12%
0
11%13%
MET
4%
BAP1
9%25%
4%13%
e199
BRIDGEWATER ET AL
Molecular Targets
EGFR. The EGFR pathway is commonly expressed in biliary
tract cancer. According to one study, it is expressed in 100%
of intrahepatic cholangiocarcinomas, 52.6% of extrahepatic
cholangiocarcinomas, and 38.5% of gallbladder cancers.52
Erlotinib, a selective and reversible inhibitor of EGFR, has been
studied in the management of advanced biliary tract cancer in
several phase II studies. In a small phase II study evaluating
erlotinib for 42 patients, overall survival was 7.5 months and
median progression-free survival was 2.6 months.52 However,
definitive conclusions between HER1/EGFR status and response rates could not be assessed due to the small sample
size. A large phase III study evaluating the efficacy of gemcitabine and oxaliplatin with the addition of erlotinib was
conducted among patients with newly diagnosed metastatic
biliary tract cancer.53 A total of 268 patients were randomly
selected to receive either chemotherapy alone or in combination with erlotinib. The median progression-free survival
was 5.8 months in the combination arm compared with
4.2 months in the chemotherapy-alone arm (p = .087). Overall
survival was the same at 9.5 months in both arms (p = .611).
Erlotinib with chemotherapy failed to demonstrate superiority in this study.
Monoclonal antibodies targeting EGFR used in combination
with chemotherapy have shown some success. Gruenberger
et al performed a phase II study using cetuximab in combination with gemcitabine and oxaliplatin (GEMOX).54 The
median progression-free survival was 8.3 months and overall
survival was 12.7 months. Interestingly, the response rate
was a remarkable 63%, with 30% of patients undergoing
curative surgery after treatment.
HER2. HER2 (ERBB2) mediates its signaling via the MAPK
and PI3K pathways. Expression of HER2 is seen in approximately 10% of gallbladder cancers and 26% of extrahepatic
cholangiocarcinomas.52 The California Consortium group
performed a phase II study of lapatinib, a dual HER2 and
EGFR inhibitor, for patients with advanced biliary tract
cancer.55 Among the 17 patients enrolled, no response was
seen in any of the patients. Similarly disappointing results
have been reported in other phase I studies of lapatinib.
VEGF. The VEGF pathway mediates its tumorigenic potential
not just by angiogenesis or vascular permeability, but also by
cell signaling for tumor initiation.50,56 VEGF expression has
been correlated with increasing grade, metastatic potential,
and overall prognosis.57
Bevacizumab, a recombinant humanized VEGF antibody,
has been studied in phase II trials for its efficacy in treatment
of biliary tract cancers. The combination of gemcitabine,
oxaliplatin, and bevacizumab (GEMOX-B) was evaluated
among 35 patients with advanced biliary tract cancer.58
FDG-PET scans demonstrated a significant decrease in
the standardized uptake value (SUV) of the tumors after
two cycles of treatment (p , .0001). The median
progression-free survival was 7.0 months. More recently, a
small first-line study evaluating the combination of
e200
gemcitabine, capecitabine, and bevacizumab had a median

progression-free survival of 8.1 months and overall survival
of 11.3 months. These results are similar to those observed with traditional chemotherapy. Other antiangiogenic
agents, including sorafenib and sunitinib, have been evaluated in advanced biliary tract cancers, with response rates
less than 10% when given as either as monotherapy or as
combination therapy.59-61
PIK3/mTOR. The PI3K/mTOR signaling cascade is crucial for
cell growth and survival. It is a key component of many
cancers. A study documented the rate of activating mutations in PI3K to be approximately 12.5%, exclusively in
gallbladder cancer. BKM120, an oral PIK3CA inhibitor, was
recently under clinical investigation, but the study was
closed due to poor accrual (NCT01501604). The mTOR
pathway has been targeted in several malignancies but has
shown disappointing results in small phase II studies. In an
Italian phase II study of everolimus, the median progressionfree survival was 3.2 months and overall survival was
7.7 months.62 Everolimus was used for patients with firstline metastatic disease, with a median progression-free
survival of 6.0 months.63 It is unclear if larger studies will
have more promising results.
FGFR. The fibroblast growth factor (FGF) pathway and
fibroblast growth factor receptor (FGFR) genes are involved in multiple biologic processes, ranging from cell
transformation, angiogenesis, and tissue repair to embryonic development. This pathway has been targeted
more recently in several malignancies including cholangiocarcinoma. Genomic profiling has demonstrated
that FGFR gene alterations are exclusively seen in intrahepatic cholangiocarcinoma, and its natural history is more
indolent.49 FGFR mutations and fusions have been observed
in 13.6% of intrahepatic cholangiocarcinomas and in 45%
of intrahepatic cholangiocarcinomas. The documented fusions to date include: FRGR2-TACC3, FGFR2-BICC1, FGFR2AHCYL1, FGFR2-PPHLN1, and FGFR2-MGEA5.49,64,65 The
fusion protein results in morphologic changes at the
cellular level causing abnormal cellular proliferation.
There are a number of clinical trials currently underway using FGFR antibodies (NCT02508467, NCT02150967,
NCT01752920) and specific FGFR2 antibodies (NCT02368951,
NCT02265341, NCT02450136).
IDH1. Somatic mutations in IDH1 and IDH2 have been well
documented in several different studies and particularly in
intrahepatic cholangiocarcinoma. These mutations cause a
single amino acid change at a conserved arginine residue
within the isocitrate binding site of IDH1 (R132) or IDH2
(R172, R140), resulting in decreased enzymatic activity for
oxidative decarboxylation of isocitrate to a-ketoglutarate.66
It has been shown that inhibition of IDH gain of function mutations results in reversal of epigenetic methylation and cancer cell differentiation.67-69 Mutation rates for
IDH1/2 have been observed in 18% to 24% of intrahepatic
cholangiocarcinomas. Further, the rates of IDH1 (15%23%)

are far higher than IDH2 (3%4%) across all reported
intrahepatic cholangiocarcinoma sequencing.47-49,51 Interestingly, 2-hydroxyglutarate, the by-product of enzymatic
activity of IDH1 and IDH2, can be detected in the serum and,
hence, can potentially be used as a biomarker.67,69,70 Studies
evaluating IDH1 inhibitors in cholangiocarcinoma are ongoing (NCT02481154, NCT02073994, NCT02496741)
MEK pathway. Activated RAS starts a phosphorylation
cascade, which involves RAF kinase, MEK1/2, and ERK1/2,
and, ultimately, these affect cellular function. Inhibition of
the MEK/ERK signaling pathway lends itself as a therapeutic
target for biliary tract cancer and other solid malignancies. A
phase II study of selumetinib in advanced biliary tract cancer
had a median progression-free survival of 3.7 months and
overall survival of 9.8 months, with a response rate of 12%.71
Interestingly, 68% of patients had stable disease; at least
12% had stable disease for longer than 1 year. In addition,
the absence of ERK staining was associated with no response. These results clearly point toward the MEK pathway
as a potentially promising targetable agent. Clinical studies
using other MEK inhibitors are underway.
CONCLUSION
With dismal overall prognosis for most cases of biliary tract
cancer, new novel therapeutic targets are clearly needed.
Biliary tract cancer is often associated with hypo- and
hypermethylation of promoters. Although several targets
(such as selumetinib and sunitinib) have focused on the
EGFR/HER2/VEGF pathways, the specificity of drugs targeting these agents is likely to be low because the majority of
biliary tract cancer mediators seem to affect epigenetics and
transcription. Understanding the gene expression profile
and mutational burden in biliary tract cancer allows us to
better discern the pathogenesis and identify promising new
developmental therapeutic targets.
References
1. Patel T. Increasing incidence and mortality of primary intrahepatic cholangiocarcinoma in the United States. Hepatology. 2001;33:1353-1357.
2. Sripa B, Kaewkes S, Sithithaworn P, et al. Liver fluke induces cholangiocarcinoma. PLoS Med. 2007;4:e201.
3. Zatonski WA, Lowenfels AB, Boyle P, et al. Epidemiologic aspects of
gallbladder cancer: a case-control study of the SEARCH Program of the
International Agency for Research on Cancer. J Natl Cancer Inst. 1997;
89:1132-1138.
4. West J, Wood H, Logan RF, et al. Trends in the incidence of primary liver
and biliary tract cancers in England and Wales 1971-2001. Br J Cancer.
2006;94:1751-1758.
5. Taylor-Robinson SD, Toledano MB, Arora S, et al. Increase in mortality
rates from intrahepatic cholangiocarcinoma in England and Wales
1968-1998. Gut. 2001;48:816-820.
6. von Hahn T, Ciesek S, Wegener G, et al. Epidemiological trends in incidence and mortality of hepatobiliary cancers in Germany. Scand J
Gastroenterol. 2011;46:1092-1098.
7. Alvaro D, Crocetti E, Ferretti S, et al; AISF Cholangiocarcinoma committee. Descriptive epidemiology of cholangiocarcinoma in Italy. Dig
Liver Dis. 2010;42:490-495.
8. Witjes CD, Karim-Kos HE, Visser O, et al. Intrahepatic cholangiocarcinoma in a low endemic area: rising incidence and improved
survival. HPB (Oxford). 2012;14:777-781.
9. Lee T-Y, Lin J-T, Kuo KN, et al. A nationwide population-based study
shows increasing incidence of cholangiocarcinoma. Hepatol Int. 2013;7:
226-232.
10. Utada M, Ohno Y, Tamaki T, et al. Long-term trends in incidence and
mortality of intrahepatic and extrahepatic bile duct cancer in Japan.
J Epidemiol. 2014;24:193-199.
11. Welzel TM, McGlynn KA, Hsing AW, et al. Impact of classification of hilar
cholangiocarcinomas (Klatskin tumors) on the incidence of intra- and
extrahepatic cholangiocarcinoma in the United States. J Natl Cancer
Inst. 2006;98:873-875.
12. Shaib YH, Davila JA, McGlynn K, et al. Rising incidence of intrahepatic
cholangiocarcinoma in the United States: a true increase? J Hepatol.
2004;40:472-477.
13. Al Ansari N, Kim BS, Srirattanapong S, et al. Mass-forming cholangiocarcinoma and adenocarcinoma of unknown primary: can
they be distinguished on liver MRI? Abdom Imaging. 2014;39:
1228-1240.
14. Sithithaworn P, Yongvanit P, Duenngai K, et al. Roles of liver fluke
infection as risk factor for cholangiocarcinoma. J Hepatobiliary Pancreat
Sci. 2014;21:301-308.
15. Carpizo DR, DAngelica M. Management and extent of resection for
intrahepatic cholangiocarcinoma. Surg Oncol Clin N Am. 2009;18:
289-305, viii-ix.
16. Jarnagin WR, Shoup M. Surgical management of cholangiocarcinoma.
Semin Liver Dis. 2004;24:189-199.
17. Fong Y, Blumgart LH, Lin E, et al. Outcome of treatment for distal bile
duct cancer. Br J Surg. 1996;83:1712-1715.
18. Jarnagin WR, Ruo L, Little SA, et al. Patterns of initial disease recurrence
after resection of gallbladder carcinoma and hilar cholangiocarcinoma:
implications for adjuvant therapeutic strategies. Cancer. 2003;98:
1689-1700.
19. Ben-David MA, Griffith KA, Abu-Isa E, et al. External-beam radiotherapy
for localized extrahepatic cholangiocarcinoma. Int J Radiat Oncol Biol
Phys. 2006;66:772-779.
20. Cheng Q, Luo X, Zhang B, et al. Predictive factors for prognosis of hilar
cholangiocarcinoma: postresection radiotherapy improves survival. Eur
J Surg Oncol. 2007;33:202-207.
21. Kim TH, Han SS, Park SJ, et al. Role of adjuvant chemoradiotherapy for
resected extrahepatic biliary tract cancer. Int J Radiat Oncol Biol Phys.
2011;81:e853-e859.
22. Todoroki T, Ohara K, Kawamoto T, et al. Benefits of adjuvant radiotherapy after radical resection of locally advanced main hepatic duct
carcinoma. Int J Radiat Oncol Biol Phys. 2000;46:581-587.
23. Schoenthaler R, Phillips TL, Castro J, et al. Carcinoma of the extrahepatic
bile ducts. The University of California at San Francisco experience. Ann
Surg. 1994;219:267-274.
24. Buskirk SJ, Gunderson LL, Schild SE, et al. Analysis of failure after curative irradiation of extrahepatic bile duct carcinoma. Ann Surg. 1992;
215:125-131.
e201
BRIDGEWATER ET AL
25. Cameron JL, Pitt HA, Zinner MJ, et al. Management of proximal cholangiocarcinomas by surgical resection and radiotherapy. Am J Surg.
1990;159:91-97.
26. Gonzalez Gonzalez D, Gerard JP, Maners AW, et al. Results of radiation
therapy in carcinoma of the proximal bile duct (Klatskin tumor). Semin
Liver Dis. 1990;10:131-141.
27. Schoenthaler R, Castro JR, Halberg FE, et al. Definitive postoperative
irradiation of bile duct carcinoma with charged particles and/or photons. Int J Radiat Oncol Biol Phys. 1993;27:75-82.
28. Bonet Beltran M, Allal AS, Gich I, et al. Is adjuvant radiotherapy needed
after curative resection of extrahepatic biliary tract cancers? A systematic review with a meta-analysis of observational studies. Cancer
Treat Rev. 2012;38:111-119.
29. Vern-Gross TZ, Shivnani AT, Chen K, et al. Survival outcomes in resected
extrahepatic cholangiocarcinoma: effect of adjuvant radiotherapy in a
surveillance, epidemiology, and end results analysis. Int J Radiat Oncol
Biol Phys. 2011;81:189-198.
30. Ben-Josef E, Guthrie KA, El-Khoueiry AB, et al. SWOG S0809: a phase II
Intergroup trial of adjuvant capecitabine and gemcitabine followed by
radiotherapy and concurrent capecitabine in extrahepatic cholangiocarcinoma and gallbladder carcinoma. J Clin Oncol. 2015;33:
2617-2622.
31. Shinohara ET, Mitra N, Guo M, et al. Radiation therapy is associated with
improved survival in the adjuvant and definitive treatment of intrahepatic cholangiocarcinoma. Int J Radiat Oncol Biol Phys. 2008;72:
1495-1501.
32. Ghafoori AP, Nelson JW, Willett CG, et al. Radiotherapy in the treatment
of patients with unresectable extrahepatic cholangiocarcinoma. Int J
33. Tao R, Krishnan S, Bhosale PR, et al. Ablative radiotherapy doses lead
to a substantial prolongation of survival in patients with inoperable
intrahepatic cholangiocarcinoma: a retrospective dose response
34. Jung DH, Kim MS, Cho CK, et al. Outcomes of stereotactic body radiotherapy for unresectable primary or recurrent cholangiocarcinoma.
Radiat Oncol J. 2014;32:163-169.
35. Mahadevan A, Dagoglu N, Mancias J, et al. Stereotactic body radiotherapy (SBRT) for intrahepatic and hilar cholangiocarcinoma. J Cancer.
2015;6:1099-1104.
36. Barney BM, Olivier KR, Miller RC, et al. Clinical outcomes and toxicity
using stereotactic body radiotherapy (SBRT) for advanced cholangiocarcinoma. Radiat Oncol. 2012;7:67.
37. Ibarra RA, Rojas D, Snyder L, et al. Multicenter results of stereotactic
body radiotherapy (SBRT) for non-resectable primary liver tumors. Acta
Oncol. 2012;51:575-583.
38. Polistina FA, Guglielmi R, Baiocchi C, et al. Chemoradiation treatment with gemcitabine plus stereotactic body radiotherapy for
unresectable, non-metastatic, locally advanced hilar cholangiocarcinoma.
Results of a five year experience. Radiother Oncol. 2011;99:120-123.
39. Goodman KA, Wiegner EA, Maturen KE, et al. Dose-escalation study of
single-fraction stereotactic body radiotherapy for liver malignancies. Int
J Radiat Oncol Biol Phys. 2010;78:486-493.
40. Tse RV, Hawkins M, Lockwood G, et al. Phase I study of individualized stereotactic body radiotherapy for hepatocellular carcinoma and intrahepatic cholangiocarcinoma. J Clin Oncol. 2008;26:
657-664.
41. Hong TS, Wo JY, Yeap BY, et al. Multi-institutional phase II study of
high-dose hypofractionated proton beam therapy in patients with
localized, unresectable hepatocellular carcinoma and intrahepatic
cholangiocarcinoma. J Clin Oncol. Epub 2015 Dec 14.
42. Jarnagin WR. Cholangiocarcinoma of the extrahepatic bile ducts. Semin
Surg Oncol. 2000;19:156-176.
e202
43. Wolkov HB, Graves GM, Won M, et al. Intraoperative radiation therapy
of extrahepatic biliary carcinoma: a report of RTOG-8506. Am J Clin
Oncol. 1992;15:323-327.
44. Rea DJ, Heimbach JK, Rosen CB, et al. Liver transplantation with
neoadjuvant chemoradiation is more effective than resection for hilar
cholangiocarcinoma. Ann Surg. 2005;242:451-461.
45. Marino D, Leone F, Cavalloni G, et al. Biliary tract carcinomas: from
chemotherapy to targeted therapy. Crit Rev Oncol Hematol. 2013;85:
136-148.
46. Zhu AX, Hezel AF. Development of molecularly targeted therapies in
biliary tract cancers: reassessing the challenges and opportunities.
Hepatology. 2011;53:695-704.
47. Ross JS, Wang K, Catenacci DVT, et al. Comprehensive genomic profiling
of biliary tract cancers to reveal tumor-specific differences and genomic
alterations. J Clin Oncol. 2015;33(suppl; abstr 231).
48. Simbolo M, Fassan M, Ruzzenente A, et al. Multigene mutational
profiling of cholangiocarcinomas identifies actionable molecular subgroups. Oncotarget. 2014;5:2839-2852.
49. Churi CR, Shroff R, Wang Y, et al. Mutation profiling in cholangiocarcinoma: prognostic and therapeutic implications. PLoS One.
2014;9:e115383.
50. Merla A, Liu KG, Rajdev L. Targeted therapy in biliary tract cancers. Curr
Treat Options Oncol. 2015;16:48.
51. Jiao Y, Pawlik TM, Anders RA, et al. Exome sequencing identifies frequent inactivating mutations in BAP1, ARID1A and PBRM1 in intrahepatic cholangiocarcinomas. Nat Genet. 2013;45:1470-1473.
52. Pignochino Y, Sarotto I, Peraldo-Neia C, et al. Targeting EGFR/HER2
pathways enhances the antiproliferative effect of gemcitabine in biliary
tract and gallbladder carcinomas. BMC Cancer. 2010;10:631.
53. Lee J, Park SH, Chang HM, et al. Gemcitabine and oxaliplatin with or
without erlotinib in advanced biliary-tract cancer: a multicentre, openlabel, randomised, phase 3 study. Lancet Oncol. 2012;13:181-188.
54. Gruenberger B, Schueller J, Heubrandtner U, et al. Cetuximab, gemcitabine, and oxaliplatin in patients with unresectable advanced or
metastatic biliary tract cancer: a phase 2 study. Lancet Oncol. 2010;11:
1142-1148.
55. Ramanathan RK, Belani CP, Singh DA, et al. A phase II study of lapatinib
in patients with advanced biliary tree and hepatocellular cancer. Cancer
Chemother Pharmacol. 2009;64:777-783.
56. Goel HL, Mercurio AM. VEGF targets the tumour cell. Nat Rev Cancer.
2013;13:871-882.
57. Quan ZW, Wu K, Wang J, et al. Association of p53, p16, and vascular
endothelial growth factor protein expressions with the prognosis and
metastasis of gallbladder cancer. J Am Coll Surg. 2001;193:380-383.
58. Zhu AX, Meyerhardt JA, Blaszkowsky LS, et al. Efficacy and safety of
gemcitabine, oxaliplatin, and bevacizumab in advanced biliary-tract
cancers and correlation of changes in 18-fluorodeoxyglucose PET
with clinical outcome: a phase 2 study. Lancet Oncol. 2010;11:
48-54.
59. Bengala C, Bertolini F, Malavasi N, et al. Sorafenib in patients with
advanced biliary tract carcinoma: a phase II trial. Br J Cancer. 2010;102:
68-72.
60. Moehler M, Kanzler S, Woerns S, et al. A randomized, double-blind,
multicenter phase II AIO trial with gemcitabine plus sorafenib versus
gemcitabine plus placebo in patients with chemotherapy-naive advanced or metastatic biliary tract cancer: first safety and efficacy data. J
Clin Oncol. 2011;29 (suppl; abstr 4077).
61. Yi JH, Thongprasert S, Lee J, et al. A phase II study of sunitinib as a
second-line treatment in advanced biliary tract carcinoma: a multicentre, multinational study. Eur J Cancer. 2012;48:196-201.
62. Buzzoni R, Pusceddu S, Bajetta E, et al. Activity and safety of RAD001
(everolimus) in patients affected by biliary tract cancer progressing
63.
64.
65.
66.
after prior chemotherapy: a phase II ITMO study. Ann Oncol. 2014;25:

1597-1603.
Yeung YH, Chionh FJM, Price TJ, et al. Phase II study of everolimus
monotherapy as first-line treatment in advanced biliary tract cancer:
RADichol. J Clin Oncol. 2014;32:5s (suppl; abstr 4101).
Sia D, Losic B, Moeini A, et al. Massive parallel sequencing uncovers
actionable FGFR2-PPHLN1 fusion and ARAF mutations in intrahepatic
cholangiocarcinoma. Nat Commun. 2015;6:6087.
Arai Y, Totoki Y, Hosoda F, et al. Fibroblast growth factor receptor 2
tyrosine kinase fusions define a unique molecular subtype of cholangiocarcinoma. Hepatology. 2014;59:1427-1434.
Borger DR, Tanabe KK, Fan KC, et al. Frequent mutation of isocitrate
dehydrogenase (IDH)1 and IDH2 in cholangiocarcinoma identified
67.
68.
69.
70.
71.
through broad-based tumor genotyping. Oncologist. 2012;17:

72-79.
Razumilava N, Gores GJ. Cholangiocarcinoma. Lancet. 2014;383:2168-2179.
Rohle D, Popovici-Muller J, Palaskas N, et al. An inhibitor of mutant IDH1
delays growth and promotes differentiation of glioma cells. Science.
2013;340:626-630.
Keum YS, Choi BY. Isocitrate dehydrogenase mutations: new opportunities for translational research. BMB Rep. 2015;48:266-270.
Reitman ZJ, Parsons DW, Yan H. IDH1 and IDH2: not your typical oncogenes. Cancer Cell. 2010;17:215-216.
Bekaii-Saab T, Phelps MA, Li X, et al. Multi-institutional phase II study of
selumetinib in patients with metastatic biliary cancers. J Clin Oncol.
2011;29:2357-2363.
e203
Designing Clinical Trials to

Achieve Breakthrough Results in
Upper Gastrointestinal
Malignancies
CHAIR
Lynn M. Matrisian, PhD, MBA
Pancreatic Cancer Action Network
Manhattan Beach, CA
SPEAKERS
Jordan D. Berlin, MD
Nashville, TN
Markus Frederic Renschler, MD
Celgene Corporation
Summit, NJ
LESSONS FROM PANCREATIC CANCER CLINICAL TRIALS
The Past, Present, and Future of Pancreatic Cancer Clinical

Trials
Lynn M. Matrisian, PhD, MBA, and Jordan D. Berlin, MD
OVERVIEW
Upper gastrointestinal malignancies comprise half of the deadliest cancers as defined by those with a 5-year survival rate less
than 50%. Using pancreatic adenocarcinoma (PAC) as an example, we retrospectively evaluated the success of phase III
clinical trials, examined the current landscape of clinical trials, and identified emerging areas that foretell the future for this
disease. Pancreatic and liver cancers are on the rise and will be the second and third leading causes of cancer deaths in 2030.
A total of 35 different agents or combinations have been tested in randomized phase III clinical trials for patients with
advanced PAC over the past 25 years, but only 11% have been incorporated into clinical practice. There has been a 37%
increase in the number of PAC trials open in the United States between 2011 to 2012 and 2014 to 2015. Enrollment has also
increased slightly, from 3.85% of the newly diagnosed cases in 2011 to 4.15% in 2014. However, the demand for patients far
exceeds the number of patients available for these trials. On the horizon is the realization that stratification of patients with
PAC using biomarkers that predict a high probability of a response could reallocate patients to faster, smaller trials with a
greater chance of a survival benefit. The current landscape of PAC clinical trials and the launch of the Pancreatic Cancer
Action Networks Know Your Tumor initiative indicate this shift is starting to occur, with particular emphasis on targeted
therapies, immunotherapies, and agents that disrupt the stroma.
istorically, the big four cancers in the United States have

been considered lung, colorectal, breast, and prostate
because of their high incidence and mortality rates. However, trends are changing dramatically, and cancers that
have not been top of mind are becoming increasing
threats to public health in the United States. Eight major
cancers are considered the deadliest, characterized by
5-year survival rates below 50%.1,2 These eight cancers will
account for half of the cancer deaths in the United States this
year, and four of the eight cancers are diseases of the upper
gastrointestinal tract (i.e., pancreas, liver, esophagus, and
stomach).3 This article focuses on PAC, representing 95% of
pancreatic cancer diagnoses and defines the problem from
the perspective of a patient with PAC through the lens of an
advocacy organization and a medical oncologist working in
this field for many years. We outline some of the historical
challenges to successful clinical trial design and execution,
describe efforts currently underway, and highlight future
directions that may bring urgently needed progress to this
deadly disease. These insights provide lessons learned
relevant to deadly upper gastrointestinal malignancies in
general.
Although overall incidence and death rates for cancer in
the United States are declining, pancreatic cancer is on the
rise. It is the only major cancer with a 5-year relative survival

in the single digits. Despite pancreatic cancer being the 11th
most commonly diagnosed cancer in men and 9th in women,
deaths from it surpassed breast cancer and moved to the
third leading cause of cancer-related death in the United
States this year. An estimated 53,070 Americans will be
diagnosed with pancreatic cancer in 2016 and 41,780 are
expected to lose their lives to the disease.3
The threat of pancreatic cancer is only expected to increase in the coming years. A 2014 study predicted that
pancreatic cancer deaths would surpass breast cancer and
further projects that pancreatic cancer deaths will exceed
those caused by colorectal cancer around 2020, positioning
pancreatic cancer as the second leading cause of cancerrelated deaths in the United States before 2030 (Fig. 1).4
Moreover, deaths from liver cancer have surpassed prostate
cancer this year, placing it among the top-five cancer killers
in the United States. Projections show that liver cancer
deaths will continue to rise quickly, leading to it becoming
the third leading cause of cancer-related death in the United
States by 2030.4
The clinical advances required to slow the alarming trajectory of pancreatic cancer deaths will be made as a result
of clinical research. National Comprehensive Cancer Network
From the Pancreatic Cancer Action Network, Manhattan Beach, CA; Vanderbilt-Ingram Cancer Center, Vanderbilt University, Nashville, TN.
Corresponding author: Lynn M. Matrisian, PhD, MBA, The Pancreatic Cancer Action Network, 1500 Rosecrans Ave., Suite 200, Manhattan Beach, CA 90266; email: lmatrisian@pancan.org.
e205
MATRISIAN AND BERLIN
FIGURE 1. Projected Number of Cancer Deaths, 2010 to 2030
guidelines indicate that a clinical trial is the preferred course

of treatment at all stages of PAC.5 The Pancreatic Cancer
Action Network encourages all patients who call its Patient
Central call center to consider clinical trials when making
treatment decisions. But enrollment rates remain low, and
clinical trials do not always match patients needs, leading to
slow progress and continued poor outcomes for patients. The
KEY POINTS
e206
Deaths from pancreatic cancer are rising, making the

disease the third leading cause of cancer deaths in the
United States in 2016.
Randomized phase III trials over the past 25 years have
resulted in a new standard of care for PAC 11% of the
time.
The number of clinical trials for PAC open in the United
States has increased by 37% over the past 5 years.
Although there has been a slight increase in the number
of patients who enroll in PAC trials from 3.85% in 2011 to
4.15% in 2014, the demand for patients to enroll in trials
far exceed the number of patients available, and it will
take, on average, more than 6 years to accrue those
trials open in 2014.
The landscape of current PAC clinical trials is changing,
with an increase in the number of neoadjuvant trials,
particularly focusing on radiation therapy and an
increase in the number of trials for previously treated
patients, such as for targeted therapies.
The Pancreatic Cancer Action Networks Know Your
Tumor initiative has offered molecular profiling
opportunities to more than 500 individuals from 38
states; 40% of them demonstrate an actionable
alteration that suggests a therapeutic option that
would not otherwise be identified for these
patients.
following sections evaluate the past, present, and future of

pancreatic cancer clinical trials.
THE PAST
Progress in advanced PAC can be measured by treatment
options considered clinically meaningful enough to be
available as standard of care to patients: gemcitabine (approved by the U.S. Food and Drug Administration [FDA]
in 1996), FOLFIRINOX (positive phase III data published in
2011),6 gemcitabine with nab-paclitaxel (FDA approved in
2013) for first-line treatment, and the recent FDA approval
of irinotecan liposome injection (2015) for patients previously treated with gemcitabine. To achieve this degree of
success, a total of 35 different agents or combinations were
tested in 39 phase III clinical trials in advanced-stage PAC
between 1997 and 2015 (Fig. 2, unpublished data, LR, PhD
2016). Twenty-seven of these were tested in patients with
treatment-naive disease, and eight of the 35 agents or
combinations were tested in patients with previously
treated disease. Although the FDA approved the combination of gemcitabine and erlotinib in 2005 based on a hazard
ratio (HR) of 0.82 for overall survival, the modest improvement in median survival of 6.2 months compared with
5.9 months with gemcitabine alone7 is generally viewed as
being insufficient to warrant the additional toxicity and
expense of erlotinib. Thus, the overall success rate of PAC
phase III trials is a dismal 11%, despite a prior phase II trial in
the majority of cases.
In the purest sense of trial design, the phase II trial is
written with the ultimate goal that (1) if the primary endpoint is positive, it will lead to a phase III trial, and (2) if
negative, the drug or regimen will not go further in that
setting. However, in 85% of the cases where we could
identify a prior phase II trial, a phase III trial was pursued
irrespective of the phase II not meeting its primary endpoint
(Rahib et al, unpublished data, 2016). The decision to
FIGURE 2. Agents/Combinations in Phase III Clinical Trials for Pancreatic Adenocarcinoma, 1992 to 2015
Abbreviation: FDA, U.S. Food and Drug Administration.

Adapted from Rahib et al (unpublished data, 2016).
proceed was based on encouraging secondary endpoints,

subset analyses results, or a second phase II trial conducted
elsewhere that showed promise. Factors that are not
mentioned but contribute to corporate go/no-go decisions
involve economic and logistic considerations by the trial
sponsors that are influenced by whether the trial is the first
indication for a new drug and by the size of the company.8 If
the past is prelude to the future, there is considerable
concern that the trend of negative phase III trials will
continue and thwart the chance for any major advancement
in PAC survival.
THE PRESENT
The Pancreatic Cancer Action Network maintains a database of pancreatic cancer clinical trials in the United States
that is constantly updated to be able to give those who
contact the Patient Central call center accurate, up-to-date
information on clinical trials that match the patients stage
of disease and that are open within a distance they are
willing to travel. We analyzed the portfolio of pancreatic
cancer clinical trials in 2011 and 2012 by type, phase,
disease stage, and treatment approaches.9 We extended
this study to include the years 2013 to 2015 and report the
results in this article. We are encouraged to find that the
number of PAC-specific clinical trials open in the United
States has steadily increased between 2011 and 2015
(Fig. 3A). This trend is observed for all phases of clinical
trials, with the most dramatic growth observed in phase I
trials (Fig. 3B).
During 2011 to 2012, the majority of PAC clinical trials
were designed for patients with treatment-naive, metastatic
disease (51% of total PAC trials for these 2 years; Fig. 4).
Since then, there has been a dramatic shift toward a greater
number of trials to accommodate patients who have already
been treated. In fact, the majority of the trials during 2014 to
2015 focused on patients with refractory and previously
treated disease (38%), whereas the percentage of first-line
metastatic trials during 2014 to 2015 (22%) decreased by
more than two fold compared with 2011 to 2012. This shift
from metastatic/treatment-naive trials to trials for patients
with previously treated disease is viewed very positively
from a patient perspective. In 2011, we observed that twothirds of the patients that call the Pancreatic Cancer Action
Networks Patient Central were ineligible for 90% of the

clinical trials that were open because they had already
started treatment.9 The increase in clinical trials that include
patients with previously treated disease opens up more
options and encourages a multistep treatment plan that
provides several opportunities for interventions that may
improve survival. In this respect, the study of patients with
more refractory disease in later line is encouraging, but we
acknowledge that due to the severity of this illness risks
missing potentially active agents if had they been studied in
an earlier line of therapy.
For patients with early-stage resectable disease, 13% of
trials were active for neoadjuvant treatment and 10% for
adjuvant treatment from 2011 to 2012. From 2014 to
2015, there was an approximately two fold increase in
trials for neoadjuvant treatment (24% of active trials)
and a slight decrease to 7% of the trials requiring patients
for adjuvant studies (Fig. 4). The increase in neoadjuvant
trials may be attributed to the recognition that even
patients with resectable PAC have dismal outcomes because of the rapid development of metastatic disease
a clinical observation that was supported with preclinical
evidence in 2012.10 In addition, the eligibility criteria for
neoadjuvant trials have shown an expansion to patients
with borderline resectable disease with the hope that
neoadjuvant treatment shrinks the tumor and permits a
negative margin resection.
The landscape of PAC clinical trials was also examined by
treatment type (Fig. 5A). Trials that focused on therapies
targeted to specific molecular pathways increased markedly
between 2011 and 2015, comprising 29% of the PAC
landscape from 2011 to 2012 and 40% from 2014 to 2015.
The majority of these targeted trials are for patients with
refractory disease and those who received previous treatments (Fig. 5B) and include targets for PARP, JAK, WEE1,
mTOR, and several others.
In addition, trials focused on radiation therapy increased
from 13% of total PAC trials during 2011 to 2012 to 18%
during 2014 to 2015 (Fig. 5A). The majority of these trials
were in the neoadjuvant setting, which comprised 58% of
the total radiation trials from 2014 to 2015 (Fig. 5B). Many of
these neoadjuvant trials (. 50%) used intensity-modulated
radiotherapy or stereotactic body radiotherapy in combination with chemotherapy.
e207
FIGURE 3. Pancreatic Adenocarcinoma Trials Open in the United States from 2011 to 2015
(A) Trials open (2011-2015) (B) Trials open by phase.
Trials focused on an immunotherapeutic approach were

relatively steady, with an average of 14 trials per year
over the past 5 years (Fig. 5A). There has been a change,
however, in the stage of disease of treatment. From 2011 to
2012, the majority of trials were for patients with metastatic disease (52%), whereas from 2014 to 2015 the
majority of immunotherapy trials were for patients with
refractory disease and those who received previous
treatments (62%; Fig. 5B). Approximately 60% of these
trials involved a vaccine treatment, 20% included T-cell
modified therapy, and 7% included checkpoint blockade
therapy.
e208
As a result of regularly updating clinical trial information

stored in the organizations comprehensive database, the
Pancreatic Cancer Action Network has established strong
relationships with trial sponsors. These relationships
allowed the accumulation of data to estimate the number of
patients with pancreatic cancer in the United States who
were accrued to clinical trials in 2011,9 and this analysis was
repeated for 2014 clinical trials. Considering only those
patients with a PAC diagnosis who enrolled in trials (in
contrast to the previous analysis, which considered all
pancreatic cancer diagnoses), 1,612 patients with PAC enrolled in clinical trials in 2011 (93% of trials reporting), and
FIGURE 4. Pancreatic Adenocarcinoma Clinical Trials Open in the United States from 2011 to 2015 by Disease
Stage
1,835 patients with PAC enrolled in clinical trials in 2014

(94% of trials reporting). As a percentage of the estimated
number of new cases of PAC in that year (95% of the estimated total pancreatic cancer incidence),3 we calculate
that 3.85% of patients with PAC went on a clinical trial in
2011, and 4.16% went on a trial in 2014.
Although the slight increase is encouraging, the time it will
take to enroll patients with PAC in clinical trials based on this
accrual rate is unacceptably slow. For example, the number of
newly diagnosed patients with resectable disease who could
be eligible for clinical trials for neoadjuvant or adjuvant
treatment approaches in 2014 is estimated to be 4,178
(Table 1). However, the 47 trials that were open in these
categories required 3,437 patients, indicating that 82% of the
eligible patients in the United States would need to go on one
of these clinical trials to complete enrollment during that year.
The actual percentage of enrollment for these trials for 2014
was 10%, indicating that it would take 9.7 years to complete
enrollment of the neoadjuvant/adjuvant trials open in 2014.
On the other hand, it is encouraging that the time necessary to
complete enrollment of trials for patients with locally advanced and previously treated metastatic disease has decreased, from 9.0 years to 3.8 years and 7.1 years to 6.0 years,
respectively (Table 1). Nevertheless, for a disease with a 71%
1-year mortality, it is clear we must identify a new paradigm
for clinical advancement to accelerate progress.
THE FUTURE
There has been a palpable shift away from the nihilistic
attitude toward PAC that bodes well for the future for these
patients. In October 2015, the FDA approved irinotecan liposomal injection for patients with metastatic PAC who had
received prior chemotherapy. Irinotecan liposome injection

was shown to improve overall survival by nearly 2 months
when administered in combination with fluorouracil and
leucovorin. This approval was a historic achievement, as it
was the first drug combination approved with an indication
specific to this patient population.11 As described above, in
the past few years, the amount of clinical trials focused on
patients with previously treated disease has increased. As
oncologists and patients are often eager to begin chemotherapy treatments as soon as possible upon diagnosis,
providing expanded options for patients with treatmentrefractory PAC fills a critical gap and has the potential to
positively affect patient outcomes.
Clinical research provides the evidence for treatments that
prolong survival. The randomized phase III trial is the gold
standard for FDA approval of new treatments and acceptance as standard of care. However, progress in pancreatic
cancer using this paradigm has been agonizingly slow. A
transformative shift in approach is emerging as a result of
the incorporation of molecular profiling as a basis of identifying subsets of patients with cancer with a high probability
of responding to a specific treatment. For example, the care
of patients with breast cancer is routinely stratified by their
estrogen receptor and HER2 status, and the genotyping of
nonsmall cell lung cancer for ALK translocations as an indicator of specific chemotherapeutic treatments is becoming routine.
The ability to identify those patients whose disease has a
high probability of responding strongly to a specific treatment has enormous implications for clinical trial design. For
example, a treatment with a true HR of 0.4 requires only
70 patients if all the patients enrolled in the trial have the
drug target.12 However, if the trial occurs in an unselected
e209
FIGURE 5. Pancreatic Adenocarcinoma Clinical Trials Open in the United States from 2011 to 2015 by Treatment
Type
A
5
60
4
1
50
22
Number of Active Trials
Phase 0
Phase I
Phase II
Phase III
Phase IV
21
2
24
40
2 19
2
30
2
2
1
3
18
20
36
1
1
10
15
6
0
0
1
9
0
0
1
1
0
4
0
0
1
2
0
4
0
3
0
0
2
3
0
0
2
0
2
21
12
12
16 19
3
0
0
4
2
3
5
0
19
17
8
0
17
19
2
1
2
0
1
3
1
0
0
1
4
0
2
0
5
0
1
3
1
10 4
0
3
0
2011
2012
2013
2014
2015
2011
2012
2013
2014
2015
2011
2012
2013
2014
2015
2011
2012
2013
2014
2015
2011
2012
2013
2014
2015
2011
2012
2013
2014
2015
2011
2012
2013
2014
2015
2011
2012
2013
2014
2015
10
24
11
39
31
18
Immunotherapy
Targeted
Stroma
Metabolism
Radiaon
Delivery
Opmizaon
Miscellaneous
70
60
Neoadjuvant
Adjuvant
Locally Advanced
Number of Active Trials
50
31 35
7
40
Metastac; treatment nave
27
Metastac; previously treated
30
3
0
4
4
7
2
1
2 6 5 3
25
20
21
8 8 6
7
0
1
1 1 3
0
1
3 3
2
1 1 2 1
19 24 21
6 3 2
5
0
3
3
4
1
9
19 2
0
4
4
1
3
8 5
0
1
15 16
2
0 5
10
1 1
5 5 2
0
2
1 1 8 9
3
7 6 2 0
2 0 3 3 3
2
7
0 2 7
0
2
3 2
5
1
0
2 2 0
2 2 2
2
4 4
1 1
3 0
0
0
0
2
2
1
1
1
1
1 1 2
1
1
1
1
1
0
0
0
0
0 0 0
0
2
5
8 4
6
7 6 1
6
1
1 2 5
4 6
1
5
11 4
12 6
9
3
0
7
7
2
1 2
2 2
0
0 1
9
2 6
2
0 0
1
1
0
2011
2012
2013
2014
2015
2011
2012
2013
2014
2015
2011
2012
2013
2014
2015
2011
2012
2013
2014
2015
2011
2012
2013
2014
2015
2011
2012
2013
2014
2015
2011
2012
2013
2014
2015
2011
2012
2013
2014
2015
10
Immunotherapy
Targeted
Stroma
Metabolism
Radiation
Delivery
Optimization Miscellaneous
(A) Trials open by treatment type and phase (B) Trials open by treatment type and disease stage
patient population, the number of patients needed is

proportional to the prevalence of the target. Further, the
observed HR decreases with proportion of patients. The
same treatment above requiring 70 selected patients, if
evaluated in an unselected population containing 20% of
patients with the target, requires 1,750 patients, and the
HR observed is only 0.84. A trial requires 29,620 patients if
the marker is present in only 5% of the unselected population. Past trials, in large numbers of unselected patients
e210
with pancreatic cancer, have resulted in relatively small

observed HR and improvement in survival. Given that
pancreatic cancer is heterogeneous, it seems possible that
a missing component is identification of patients who are
most likely to respond to specific treatments. Continuing
with the paradigm of treating all patients with the same
chemotherapeutic agents will continue to reap low rewards
and squander the precious resource of patients willing and
eligible to enroll in clinical trials.
TABLE 1. Anticipated Total and Actual Accrual in the United States in 2011 and 2014
Stage at Diagnosis
Distribution
Year (%)
Est. No.
Patients*
No. of
Trials
Potential
Enrollment
Enrollment Capacity
(%)**
Enrolled
(%)
Years to
Completion
Localized ([Neo]
Adjuvant)
2011
3,963
29
2,874
73
15
6.9
2014
4,178
47
3,437
82
10
9.7
Regional (Locally
Advanced)
2011 28
12,328
16
756
13
9.0
2014 28
12,998
12
800
27
3.8
Distant (Metastatic)
2011 53
23,336
62
6,394
27
18
6.2
2014 53
24,603
38
3,535
14
15
6.6
2011 27
11,668
12
524
14
7.1
2014 27
12,301
59
4,364
35
17
6.0
Metastatic; Previously
Treated
The distribution by disease stage was determined using the Surveillance, Epidemiology, and End Results (SEER) 18 2005-2011 Registry, All Races, both Sexes by SEER Summary Stage
2000 for localized, locally advanced, and metastatic. The number of estimated patients diagnosed with pancreatic cancer for each year was multiplied by 95% to account for pancreatic
adenocarcinomas only. The distribution of previously treated metastatic disease was determined by multiplying the percentage of patients with metastatic disease by 50%, as it is
reported that about 50% of patients with pancreatic cancer are eligible for second-line therapy.36-38
*The number of new patient cases of pancreatic cancer was estimated to be 44,030 for 2011 and 46,420 for 2014.
**Enrollment capacity is total potential enrollment for open trials divided by the estimated number of available patients for 2011 and 2014.
Percentage enrolled is the number of patients enrolled divided by the potential enrollment for trials within that subgroup for which accrual numbers were available for 2011 and
2014.
Years to completion is the number of potential enrollment divided by the accrual numbers (for each subgroup) for which accrual numbers were available for 2011 and
2014.
Is molecular stratification feasible in pancreatic cancer?

Recent genomics analyses have revealed high numbers of
genetic changes that contribute to the initiation and progression of PAC. Although we know that oncogenic KRAS
drives 95% of PAC cases, more recent studies have shown
greater genetic diversity than previously expected. Efforts
remain underway to devise methods to directly attack KRAS,
which would certainly be game-changing for the treatment
of PAC.13 In the meantime, identifying genetic changes and
signatures that could predict patient response to targeted or
more traditional therapies could have immediate and major
effect on patient care.
Results from the Australian Pancreatic Cancer Genome
Initiative within the International Cancer Genome Consortium defined four subtypes of PAC: stable, locally rearranged, scattered, and unstable. Of particular interest is the
unstable phenotype, which was observed in 14% of the
100 patient samples analyzed. Unstable genomes were found
to contain more than 200 structural variation events, which
often suggests damage to the DNA repair pathways.14 This
group and others have shown that patients with PAC who
have DNA damage repair alterations may be particularly
sensitive to platinum-containing chemotherapeutics and/or
PARP inhibitors. 15-17 A recent paper further refines the
molecular subtypes of pancreatic cancer based on genomic
analysis.18
An additional targeted therapeutic approach involves
assessing patients levels of hyaluronan (HA), which is a
glycosaminoglycan present in the microenvironment surrounding PAC tumors. Hyaluronan contributes to elevated
interstitial pressure, and its inhibition with an agent called
PEGPH20 leads to the expansion of tumor-associated blood
vessels, allowing delivery of other drugs to the tumor.18,19
Interim analyses of a phase II randomized study of patients
with previously untreated metastatic PAC treated with
PEGPH20, nab-paclitaxel, and gemcitabine (PAG) or nabpaclitaxel plus gemcitabine (AG) were presented at the
2015 American Society of Clinical Oncology Annual Meeting. Among patients with high HA levels treated with PAG,
median progression-free survival was 9.2 months, which
was more than double that observed with AG alone
(4.3 months).20 This ongoing trial represents an example of
customizing treatment to a patients tumor characteristics
and provides promising preliminary evidence in favor of this
approach.
Finally, multiple studies are underway to investigate immunotherapy for the treatment of PAC. Pancreatic adenocarcinoma is known to be a highly immunosuppressive
disease, rendering immunotherapeutic approaches that
have shown success in other cancer types ineffective as
single agents in PAC. To combat this immunosuppression
and recruit T cells to the PAC microenvironment, immune
checkpoint inhibitors such as antiCTLA-4, antiPD-1, and
antiPD-L1 antibodies are being tested. Preliminary results
suggest that these tactics can be successful in combating
the immunosuppression but do not lead to cancer cell
death. Therefore, focus has shifted to combination of checkpoint inhibitors with a therapeutic vaccine strategy or
other agents such as chemotherapy or radiation (e.g.,
NCT02303990).21-25
Correlative studies are revealing stratification approaches
that indicate patients most likely to respond to immunotherapeutic approaches, including antiCTLA-4 and anti
PD-1 therapy. A study published in 2015 showed that the
overall mutational load, neoantigen load, and expression of
cytolytic markers in the immune microenvironment of patients with metastatic melanoma were significantly associated with clinical benefit to ipilimumab.26 Another study,
also published in 2015 that included patients with nonsmall
cell lung cancer, melanoma, and renal cell cancer, showed
e211
that nine out of 25 patients with PD-L1positive tumors had

an objective response to antiPD-1 therapy, while all 17
patients with PD-L1negative tumors had no response.27,28
Standardized testing of PD-L1 on tumor cells for the prediction and evaluation of treatment efficacy is needed. In
addition, mismatch repairdeficient tumors may be sensitive to immune checkpoint blockade. Currently, multiple
trials are underway testing the sensitivity of antiPD-1
therapy in this subset of patients in several gastrointestinal cancers, including a phase II trial in patients with
previously treated advanced colorectal cancer,29 a phase III
first-line randomized trial comparing antiPD-1 therapy to
standard chemotherapy in patients with metastatic colorectal cancer,30 and a phase II trial for patients with
previously treated colorectal and other gastrointestinal
cancers. This phase II trial reported promising response to
antiPD-1 therapy in noncolorectal gastrointestinal cancers
with mismatch repair deficiency, including ampullary, pancreas, biliary small bowel, and gastric cancers.31
As we gather more genomic, proteomic, and immune cell
information about PAC and accumulate evidence from this
and other cancer types about molecularly targeted treatment approaches, efforts such as the Pancreatic Cancer
Action Networks Know Your Tumor precision medicine
initiative will play an important role in providing patients
access to personalized treatment approaches so that hypotheses on the predictive value of these markers can be
tested. Patients with pancreatic cancer across the United
States are able to access Know Your Tumor for multi-omic
molecular profiling information on their tumor by contacting the Pancreatic Cancer Action Networks Patient Central.
A highly trained Patient Central associate determines the
eligibility and interest of the patient in enrolling in the
program and transfers suitable patients to a partnering
organization, Personalized Cancer Therapy, Inc. (doing
business as [d.b.a] Perthera). Perthera obtains patient
consent, talks to the treating physician, facilitates obtaining
suitable biopsy tissue according to standard operating
procedures, and sends the sample to appropriate diagnostic
laboratories. Multi-omic testing is performed, including
genomic, protein, and phosphoprotein analysis. The results
are returned to Perthera for data analysis and integration,
and an expert medical review is conducted that combines
this information with current literature and knowledge of
pancreatic cancer clinical trial results and the patients
history. The medical review team includes a medical oncologist with considerable experience and expertise in
treating pancreatic cancer. Treatment options are ranked
and include clinical trials with up-to-date information on
availability from the Pancreatic Cancer Action Network
database, solid tumor phase I trials, off-label treatments,
and standard of care options. The report is sent to the
treating physician, the patient, and the Pancreatic Cancer
Action Network, and outcomes data are collected from both
the patient and the treating physician. All patients using
Know Your Tumor are encouraged to participate in the
Pancreatic Cancer Action Networks patient-reported
e212
outcomes registry to obtain more detailed information on

their experience and the outcome of treatment. Since its
inception in June 2014 to December 2015, Know Your Tumor
has enrolled 496 patients with pancreatic cancer from 38
different states.32
There were 175 molecular profiles from Know Your Tumor
participants evaluated as of December 31, 2015. Of the 175
patients with completed reports, 58% were treated at an
academic institution or high-volume center, while 42% were
treated at a community center (Fig. 6). The distribution of
pathologic genomic alterations as determined by analysis
with the Foundation One panel of 343 genes is shown in
Fig. 7.32 Forty percent of patients had at least one actionable
alteration as defined based on a high response rate in patients with an identified molecular abnormality in any cancer
type, or based on a mechanism/pathway-defined implication of response to treatment.33 For example, 18% of patients tumors had alterations in genes involved in DNA
damage repair, including BRCA2, PALB2, ATM, and others
indicated in Fig. 7. These alterations suggest treatment
with a DNA-damaging agent such as platinum and/or a PARP
inhibitor. Based on patient follow-up as of December 31,
2015, 34 patients received their next line of treatment
consistent with the Perthera report, eight of whom enrolled
in a clinical trial and three of whom used an off-label
treatment. In addition, six patients received therapy that
was not indicated by the report. Although the numbers are
small, preliminary analysis suggests an improved outcome
for patients who followed options based on the Perthera
report compared with those who received treatment not
listed on the report (median overall survival 45 vs.
32 weeks).33
Clinical trials for pancreatic cancer are incorporating
biomarkers for patient selection. In January 2016, an analysis
of the Pancreatic Cancer Action Networks clinical trial database revealed 21 of 174 (12%) clinical trials required a
molecular indicator for enrollment. It is anticipated that this
percentage will increase over time as more knowledge is
gained and a targeted approach is applied to the deadliest
cancers.
OTHER UPPER GASTROINTESTINAL CANCERS

Similar to pancreatic cancer, other upper gastrointestinal
cancers have also had limited successes. Biliary tract cancer
and hepatocellular cancers each have only one standard
regimen. After initial success with sorafenib as first-line
therapy for hepatocellular carcinoma, multiple other vascular endothelial growth factor receptor (VEGFR) inhibitors
were unsuccessfully tested in phase III trials.34 Although it is
not surprising that other VEGFR inhibitors were tested, it is
remarkable that the sheer number of trials were conducted
and no other targets were evaluated in a substantive manner.
Similarly, based on almost no preclinical evidencesimply
overexpression of the proteinEGFR inhibitors were tested
in multiple randomized studies in gastroesophageal cancer,
all of which were negative.
FIGURE 6. Know Your Tumor Recruitment Across the United States and Community and Academic/High-Volume
Sites
The question for the future of all upper gastrointestinal

cancersand all cancers in generalis whether we can start
conducting smarter trials. In 2009, the State of the Science
meeting results for pancreatic cancer were published.35 The
meeting emphasized having preclinical data guide the development of the clinical trials, rather than becoming an
afterthought used as rationalization for a planned study. As

noted above, designing clinical trials in the era of personalized medicine may allow for smaller trials with greater
effect. Others have recommended n of 1 studies, testing
different agents or combinations for an individual patient,
but how this can be done without simply being an anecdote
FIGURE 7. Pathogenic Alterations Observed by Next-Generation Sequencing in Know Your Tumor Participants
e213
has yet to be defined. Smarter trials can be smaller, but it

unlikely that one patient can define the utility of a drug.
CONCLUSION
The lessons we have learned from pancreatic cancer reflect
the potential that exists in upper gastrointestinal malignancies in general. Although we celebrate the few successes,
it is essential that we learn from the failures. Clinical trials
must be smarter in their design to conserve the precious
resource of patients willing to enroll in clinical trials. Applying biomarkers that identify patients with a high probability of response is one way to leverage remarkable
advances in cancer research with the practical outcome of
reducing the number of patients needed to identify a
clinically meaningful result. Initiatives such as the Pancreatic Cancer Action Networks Know Your Tumor provide a
mechanism to disseminate the practice of interrogating the
molecular characteristics of the tumor from patients across

the United States and learn from them. This will set the stage
for vast improvements in the current approach and rapid
advances toward improving the survival of those with the
deadliest cancers.
ACKNOWLEDGMENT
Pancreatic Cancer Action Network staff and associates that
contributed to this work include Lynn Matrisian, PhD, MBA
(concepts, writing, oral presentation); Lola Rahib, PhD (data
analysis, writing, figure preparation); Porsha James, MPH,
and Kyla Holmes (data collection for clinical trial landscape
and accrual); Allison Rosenzweig, PhD (writing); William
Hoos, MBA (concepts); and our partners from Personalized
Cancer Therapy, Inc. (Perthera) and the Know Your Tumor
initiative. We thank all trial sponsors that provided accrual
information.
References
1. Deadliest Cancers Coalition. http://www.deadliestcancers.org/. Accessed
February 1, 2016.
2. Pancreatic Cancer Action Network. The Recalcitrant Cancer Research
Act: An Important Step Toward Improving Pancreatic Cancer Survival.
https://www.pancan.org/wp-content/uploads/2015/07/fact-sheetRecalcitrant-Cancer-Research-Act.pdf. Accessed February 2, 2016.
2016;66:7-30.
4. Rahib L, Smith BD, Aizenberg R, et al. Projecting cancer incidence and
deaths to 2030: the unexpected burden of thyroid, liver, and pancreas
cancers in the United States. Cancer Res. 2014;74:2913-2921.
5. National Comprehensive Cancer Network. Pancreatic Adenocarcinoma,
version 2.2015. http://www.nccn.org/. Accessed February 2, 2016.
6. Conroy T, Desseigne F, Ychou M, et al. FOLFIRINOX versus gemcitabine
for metastatic pancreatic cancer. N Engl J Med. 2011;364:1817-1825.
7. Moore MJ, Goldstein D, Hamm J, et al. Erlotinib plus gemcitabine
compared with gemcitabine alone in patients with advanced pancreatic
cancer: a phase III trial of the National Cancer Institute of Canada Clinical
Trials Group. J Clin Oncol. 2007;25:1960-1966.
8. Feuerstein A, Ratain MJ. Oncology micro-cap stocks: caveat emptor!
9. Hoos WA, James PM, Rahib L, et al. Pancreatic cancer clinical trials and
accrual in the United States. J Clin Oncol. 2013;31:3432-3438.
10. Rhim AD, Mirek ET, Aiello NM, et al. EMT and dissemination precede
pancreatic tumor formation. Cell. 2012;148:349-361.
11. Wang-Gillam A, Li CP, Bodoky G, et al. Nanoliposomal irinotecan with
fluorouracil and folinic acid in metastatic pancreatic cancer after
previous gemcitabine-based therapy (NAPOLI-1): a global, randomised,
open-label, phase 3 trial. Lancet. 2016;387:545-557.
12. Sleijfer S, Bogaerts J, Siu LL. Designing transformative clinical trials in
the cancer genome era. J Clin Oncol. 2013;31:1834-1841.
13. National Cancer Institute. The RAS Initiative. http://www.cancer.gov/
research/key-initiatives/ras. Accessed February 2, 2016.
14. Waddell N, Pajic M, Patch A-M, et al. Whole genomes redefine the
mutational landscape of pancreatic cancer. Nature. 2015;518:495-501.
e214
16. Lohse I, Borgida A, Cao P, et al. BRCA1 and BRCA2 mutations sensitize to
chemotherapy in patient-derived pancreatic cancer xenografts. Br J
Cancer. 2015;113:425-432.
17. Kindler HL, Locker GY, Mann H, et al. POLO: a randomized phase III trial
of olaparib tablets in patients with metastatic pancreatic cancer (mPC)
and a germline BRCA1/2mutation (gBRCAm) who have not progressed
following first-line chemotherapy. J Clin Oncol. 2015;33:15s (suppl;
abstr TPS4149).
18. Bailey P, Chang DK, Nones K, et al. Genomic analyses identify molecular
subtypes of pancreatic cancer. Nature. 2016;531:47-52.
19. Jacobetz MA, Chan DS, Neesse A, et al. Hyaluronan impairs vascular
function and drug delivery in a mouse model of pancreatic cancer. Gut.
2013;62:112-120.
20. Hingorani SR, Harris PW, Hendifar AE, et al. High response rate and PFS
with PEGPH20 added to nab-paclitaxel/gemcitabine in stage IV previously untreated pancreatic cancer patients with high-HA tumors:
Interim results of a randomized phase II study. J Clin Oncol. 2015;33
21. Foley K, Kim V, Jaffee E, Zheng L. Current progress in immunotherapy for
pancreatic cancer. Cancer Lett. Epub 2015 Dec 23.
22. Le DT, Lutz E, Uram JN, et al. Evaluation of ipilimumab in combination
with allogeneic pancreatic tumor cells transfected with a GM-CSF
gene in previously treated pancreatic cancer. J Immunother. 2013;36:
382-389.
23. Le DT, Wang-Gillam A, Picozzi V, et al. Safety and survival with GVAX
pancreas prime and Listeria Monocytogenes-expressing mesothelin
(CRS-207) boost vaccines for metastatic pancreatic cancer. J Clin Oncol.
2015;33:1325-1333.
24. Le DT, Crocenzi TS, Uram JN, et al. Randomized phase II study of the
safety, efficacy, and immune response of GVAX pancreas (with cyclophosphamide) and CRS-207 with or without nivolumab in patients with
previously treated metastatic pancreatic adenocarcinoma (STELLAR). J
Clin Oncol. 2016;34:4s (suppl; abstr TPS486).
25. Twyman-Saint Victor C, Rech AJ, Maity A, et al. Radiation and dual
checkpoint blockade activate non-redundant immune mechanisms in
cancer. Nature. 2015;520:373-377.
26. Van Allen EM, Miao D, Schilling B, et al. Genomic correlates of response to
CTLA-4 blockade in metastatic melanoma. Science. 2015;350:207-211.
2443-2454.
28. Zheng P, Zhou Z. Human cancer immunotherapy with PD-1/PD-L1
blockade. Biomark Cancer. 2015;7:15-18.
29. Le DT, Yoshino T, Jager D, et al. KEYNOTE-164: phase II study of
pembrolizumab (MK-3475) for patients with previously treated,
microsatellite instability-high advanced colorectal carcinoma. J Clin
Oncol. 2016;34;4s (suppl; abstr TPS787).
30. Diaz LAJ, Le DT, Yoshino T, et al. KEYNOTE-177: first-line, open-label,
randomized, phase III study of pembrolizumab (MK-3475) versus
investigator-choice chemotherapy for mismatch repair deficient or
microsatellite instability-high metastatic colorectal carcinoma. J Clin
Oncol. 2016;34;4s (suppl; abstr TPS789).
31. Le DT, Uram JN, Wang H, et al. PD-1 blockade in mismatch repair
deficient non-colorectal gastrointestinal cancers. J Clin Oncol. 2016;34:
4s (suppl; abstr 195).
32. Engebretson A, Brody JR, Rahib L, et al. The Know Your Tumor (KYT)
initiative: a national program of multi-omic molecular profiling (MoP)
for patients (Pts) with pancreatic cancer (PDA). J Clin Oncol. 2016;34;4s
(suppl; abstr 279).
33. Pishvaian MJ, Brody JR, Matrisian LM, et al. Multi-omic profiling (MoP)
for patients (pts) with pancreatic cancer (PDA): initial results of the
Know Your Tumor (KYT) initiative. J Clin Oncol. 2016;34:4s (suppl; abstr
282).
34. Llovet JM, Villanueva A, Lachenmayer A, et al. Advances in targeted
therapies for hepatocellular carcinoma in the genomic era. Nat Rev Clin
Oncol. 2015;12:436.
35. Philip PA, Mooney M, Jaffe D, et al. Consensus report of the national
cancer institute clinical trials planning meeting on pancreas cancer
treatment. J Clin Oncol. 2009;27:5660-5669.
36. Rahma OE, Duffy A, Liewehr DJ, et al. Second-line treatment in advanced pancreatic cancer: a comprehensive analysis of published
clinical trials. Ann Oncol. 2013;24:1972-1979.
37. Teague A, Lim KH, Wang-Gillam A. Advanced pancreatic adenocarcinoma: a review of current treatment strategies and developing therapies. Ther Adv Med Oncol. 2015;7:68-84.
38. Schrag D, Archer L, Wang X, et al. A patterns-of-care study of postprogression treatment (Rx) among patients (pts) with advanced pancreas cancer (APC) after gemcitabine therapy on Cancer and Leukemia
Group B (CALGB) study #80303. J Clin Oncol. 2007;25:18s (suppl; abstr
4524).
e215
HER2, VEGFR, and Beyond:

Genomic Profiling of Upper
Gastrointestinal Tract Cancers
and the Future of Personalized
Treatment
CHAIR
Jaffer A. Ajani, MD
Houston, TX
SPEAKERS
Adam J. Bass, MD
Boston, MA
Jeeyun Lee, MD, PhD
Samsung Medical Center
Seoul, South Korea
LEE, BASS, AND AJANI
Gastric Adenocarcinoma: An Update on Genomics, Immune

System Modulations, and Targeted Therapy
Jeeyun Lee, MD, PhD, Adam J. Bass, MD, and Jaffer A. Ajani, MD
OVERVIEW
Gastric adenocarcinoma (GAC) is a global health burden on all societies, and it was the third-leading cause of cancer-related
mortality in 2012, causing 723,000 deaths worldwide. The prognosis of patients with metastatic GAC remains poor, with a
median overall survival of less than 1 year in patients treated with currently available therapies. A limited number of
therapeutic agents is currently available. Recent additions to the armamentarium include trastuzumab and ramucirumab,
which have shown some survival advantage when added to cytotoxic(s). Genomic analyses have defined various genotypes
of GACs. The novel genomic knowledge can lead to discovery of novel targets and novel therapeutic agents. In this update,
we focus on the current genomic data, targeted therapies including immune system modulators, and expand on HER2/neu
testing and the use of agents against this target. Several other facets of GAC and its therapy are not to be included in this
review but have been discussed elsewhere.
eeper analysis of the genomic and molecular features of

GAC can provide multiple benefits to the study and
development of therapy for GAC that is clearly a heterogeneous entity. Gastric adenocarcinoma has been divided
into two primary histologic variants: the intestinal and
diffuse types. Beyond histology, there is discussion regarding
the differences in the GAC in Eastern and Western patients,
different features of GAC in the distal versus proximal
stomach, and differences relating to distinct molecular
etiologies and molecular subtypes of GACs. In this context,
modern genomic and molecular analyses can help delineate
subtypes of GAC and identify salient molecular features of
these distinct classes, information that is increasingly essential to guide therapy, especially emerging targeted and
biologic agents.1-9
Recent years have witnessed a revolution in the capacity to
genomically and molecularly describe cancer. As these
technologies have become less costly, there have been
several large-scale efforts to molecularly detail this cancer.
Here, we will discuss two recent substantial efforts in this
realm, studies emerging from the Asian Cancer Research
Group10 (ACRG) and from The Cancer Genome Atlas11
(TCGA). TCGA, a consortium from the National Institutes of
Health, collected primary untreated gastric samples from
institutions across the world. These samples were all subjected to a diverse array of molecular analyses including
whole-exome sequencing, microarray profiling of genomic

amplifications and deletions, messenger RNA (mRNA) and
micro-RNA expression analysis, and genome-wide DNA
methylation analysis with some samples also analyzed with
whole-genome sequencing or proteomic analysis. By contrast, the ACRG GAC study accrued from a single referral
hospital in South Korea, providing a more homogenously
treated cohort with richer clinical annotation relative to the
TCGA study. The ACRG samples were subjected to more
limited analysis including mRNA expression, somatic copy
number, and focused gene sequencing.
Each of these projects developed a four-group classification structure for GAC, with these classification approaches having some overlapping but also distinct features.
The ACRG categories were determined on the basis of gene
expression profiling with the following features of GAC: (1)
microsatellite instability, (2) epithelial-to-mesenchymal transition (EMT) phenotype, (3) a p53 signature (CDKN1A and
MDM2 expressing), or (4) no p53 signature. Of these groups,
the cohort with microsatellite instability had improved survival, whereas the EMT group, which was composed mostly of
diffuse-type tumors, showed inferior survival. Furthermore,
after surgical resection, those with the EMT signatures also
showed greater rates of recurrence and an association with
subsequent peritoneal metastases. Although this EMT group
was largely diffuse-type tumors, the tumors in this group
From the Division of Hematology-Oncology, Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Kangnamgu, Seoul, South Korea;
Division of Molecular and Cellular Oncology, Dana-Farber Cancer Institute, Boston, MA; Department of Gastrointestinal Medical Oncology, The University of Texas MD Anderson
Cancer Center, Houston, TX.
Corresponding author: Jaffer A. Ajani, MD, Department of Gastrointestinal Medical Oncology, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd., Houston,
TX 77030; email: jajani@mdanderson.org.
104
GASTRIC ADENOCARCINOMA
lacked other features that have been seen commonly in

diffuse GAC, such as mutation of CDH1, the gene responsible
for hereditary diffuse GAC, and recently identified DGC mutations of the RHOA GTPase.12,13 As predicted, the cohort of
samples with deficient p53 activity possessed greater aneuploidy and accompanying amplifications of oncogenes including ERBB2 (HER2).
The TCGA classification derivation differed from the ACRG
in that it was based upon an integration of six distinct molecular platforms as opposed to the reliance upon mRNA
expression. Based upon observations from the molecular
data, tumors were divided into four groups: (1) GACs positive
for Epstein-Barr virus (EBV), (2) microsatellite instability, (3)
chromosomally unstable, and (4) genome stable. The latter
genome stable group represented the remnants after the
groups with EBV and microsatellite instability were excluded
and the remaining GACs were divided based upon possession
of aneuploidy (chromosomally unstable) or its absence (genome stable). A first notable difference compared with the
ACRG was that EBV-positive tumors emerged as a distinct
entity. This difference likely follows the inclusion of DNA
methylation analysis, which revealed a strong and highly
distinguishing methylation signature in this cohort. Beyond
methylation, however, EBV-positive GACs possess other
clinically relevant findings, including highly recurrent PIK3CA
mutations and genomic amplification of the genes encoding
PD-1 pathway activators PD-L1 and PD-L2, suggesting the
possible roles for PI3K pathway inhibitors, emerging immune
checkpoint agents, and, perhaps, DNA hypomethylating
agents. Like microsatellite instability GACs, EBV-positive GACs
have also been described to have a favorable phenotype.14
KEY POINTS
Patients with advanced GAC continue to have poor

quality of life, short survival, and limited therapeutic
choices. Many patients do not have access to clinical
trials when standard regimens have failed.
The empiric approaches that assume one-size-fits-all
have been ineffective methods to improve the outcome
of patients with GAC. Alternatives and customized
approaches must be considered.
The Cancer Genome Atlas study of GACs has uncovered
four genotypes with unique features. This discovery
needs to be fully exploited to find novel targets and
effective drugs.
Trastuzumab and ramucirumab (in combination with
cytotoxic[s]) have produced only minimum/modest
benefit for patients with GAC, emphasizing the need for
better therapies.
Since early results from checkpoint inhibitors show
promising results in some patients with GAC, rationally
designed multiprong approaches that combine immune
modulation, targeted agent(s), and conventional
cytotoxics may prove more advantageous than all
previous approaches.
Microsatellite instability was found as a distinct group by

the TCGA as well as ACRG effort. Microsatellite instabilitypositive GACs possess marked hypermutation as a
result of the underlying defect in DNA mismatch repair.
These tumors also possess a relatively bland genomic copy
number pattern lacking frequent genomic amplifications of
genes such as ERBB2. However, a number of likely pathogenic mutations exist within this cohort, including mutations in targetable genes such as ERBB2, ERBB3, and
PIK3CA, as well as oncogenes such as KRAS that lack effective inhibitors. A notable contrast with MSI-positive
colorectal cancer was the lack of activating BRAF mutations
within these GACs. One notable question regarding
microsatellite instability tumors, whose high mutation
burdens generate large number of neoantigens, will be the
potential efficacy of immune checkpoint inhibitors as have
shown efficacy in microsatellite instabilitypositive colorectal cancers.15
Like the EMT group from the ACRG, the TCGAs genome
stable group primarily consist of those with diffuse histology.
The GACs in this so-called genome stable group get their name
from their lack of the hypermethylation prevalent in EBVpositive, the hypermutation of microsatellite instability, or the
marked aneuploidy in chromosomally unstable GACs. Although both the genome stable and EMT tumors were largely
diffuse type, their molecular features did differ as evidenced
by the common finding of both CDH1 and RHOA mutations in
the TCGA genome stable set. Despite these differences, results
from these two classification schemes provide insights into
diffuse GAC. First, GACs with diffuse histology span molecular
subtypes including those with microsatellite instability, EBV,
and marked aneuploidy. However, diffuse types are generally
less enriched in classic oncogenes and therapeutic targets such
as KRAS or ERBB2 that are more prevalent in other groups of
GACs, speaking to the clear remaining need to identify new
therapeutic approaches and vulnerabilities for these highly
aggressive tumors.
The largest group in the TCGA analysis, the chromosomally
unstable tumors, matches most closely the p53-deficient
tumors from the ACRG schema. As their names suggest,
the salient features of these GACs are the high degree of
structural genomic aberrations, leading to a high frequency of
genomic amplifications of bona fide oncogenes such as receptor tyrosine kinases ERBB2, EGFR, MET, or FGFR2; cell cycle
mediators such as CCND1, CCNE1, and CDK6; and transcription
factor oncogenes such as GATA4, GATA6, and MYC. These
genomic data speak for the potential for inhibitors for targets
other than HER2 to ultimately have roles in GACs. Indeed, the
emergence of approved inhibitors targeting specific cell cycle
kinases, such as the CDK4/6-directed agent palbociclib, should
be evaluated in specific sets of patients guided by genomic
data.
Comparing the TCGA chromosomally unstable group and
the ACRG p53-negative group, a notable difference is that
the TCGA chromosomally unstable class comprises a much
more substantial fraction of the total tumors than does the
p53-negative group. This difference may reflect not only the
105
TABLE 1. Major Clinical Trials in Gastric Adenocarcinoma With Targeted Agents

Patient
Selection
Method
Results
(Primary
Endpoint)
Reference
Target
Trial
Type of
Study/Line
HER2
ToGa
Phase III/first
HER2 IHC
5-FU/capecitabine +
cisplatin 6 trastuzumab
Positive (OS)
Bang et al 20105
HER2
LOGIC
Phase III/first
HER2
amplification
Lapatinib + XELOX
Negative (OS)
Hecht et al16
HER2
TYTAN
Phase III/second
HER2
amplification
Paclitaxel + lapatinib vs. paclitaxel
Negative (OS)
Bang et al (2013)17
EGFR
EXPAND
Phase III/first
All comer
Cetuximab/XP vs. placebo/XP
Negative (OS)
Lordick et al18
EGFR
REAL-III
Phase III/first
All comer
Panitumumab/EOC vs. EOC
Negative (OS)
Waddell et al19
EGFR
Nimotuzumab
Phase II/second
All comer
Nimotuzumab/irinotecan vs.
irinotecan
Negative
Kim et al20
VEGF
AVAGAST
Phase III/first
All comer
XP/bevacizumab vs. XP
Negative (OS)
Van Cutsem et al21
MET
RILOMET-1
Phase III/first
MET IHC
Rilotumumab/ECX vs. ECX
Negative (OS)
Iveson et al22
MET
METGastric
Phase III
MET IHC
Onartuzumab/FOLFOX vs. FOLFOX
Negative (OS)
Shah et al23
FGFR2
SHINE
R-Phase II/second
FGFR2
amplification
AZD4547/paclitaxel vs. paclitaxel
Negative (PFS)
Bang et al (2015)24
mTOR
GRANITE
Phase III/second
or third
All comer
Everolimus vs. placebo
Negative (OS)
Ohtsu et al25
AKT
MK2206
Phase II/second
All comer
MK-2206
Response rate, 1%
Ramanathan et al26
ATM
Olaparib
R-Phase II/second
ATM IHC
Paclitaxel/olaparib vs.
paclitaxel/placebo
Negative (PFS)
Bang et al (2015)27
VEGF
MEGA
R-Phase II/first
All comer
FOLFOX/aflibercept vs. FOLFOX
Negative (6-mo PFS)
Enzinger et al28
HER2
GATSBY
Phase II/III/ second
HER2 IHC
TDM1 vs. paclitaxel or docetaxel
Negative (OS)
Kang et al29
VEGFR-2
RAINBOW
Phase III/second
All comer
Paclitaxel/ramucirumab vs.
paclitaxel/placebo
Positive (OS)
Wilke et al30
VEGFR-2
REGARD
Phase III/third
All comer
Ramucirumab vs. placebo
Positive (OS)
Fuchs et al31
Regimen
XELOX
Abbreviations: OS, overall survival; XP, capecitabine (Xeloda) and cisplatin; EOC, epirubicin, oxaliplatin, and capecitabine; IHC, immunohistochemistry; ECX, epirubicin, cisplatin, and
capecitabine; PFS, progression-free survival.
different classification schema but also the differences in the

populations studied. Chromosomally unstable tumors were
more prevalent in the proximal stomach, whereas other
groups were more prevalent in more distal regions. It is
well recognized that GAC in Western populations has, in
recent years, become increasingly a disease of the proximal
stomach, mirroring the rise of esophageal adenocarcinoma.
As this subset may be a more aneuploidy variant of GACs, a
cohort such as the TCGA group with more Western patients
would be expected to harbor higher degrees of aneuploidy.
Indeed, in the ACRG cohort, the percentage of diffuse-type
GACs was higher and proximal GACs lower than in the TCGA
study. Regardless of this difference, the prevalence of distinct oncogene amplification events in these tumors does
speak for the potential of these different alterations to serve
as biomarkers to guide therapy.
Overall, the most important lesson of these differences is
that it is beyond the time to try to devise singular approaches
for GAC. There are substantial molecular and etiologic differences across these GACs. Furthermore, even within a
single molecular subtype, there remains highly relevant
heterogeneity. For example, even among the highly aneuploidy GACs, optimal treatment of an ERBB2-amplified and
MET-amplified GACs will clearly differ. As we move forward
106
to developing new therapies, we will have to take into account this clear molecular heterogeneity that relates to
different etiologies of genomic instability, different microbial etiologies, and GACs possessing distinct driving somatic
alterations. These differences will impact our approaches
both to standard conventional therapy, emerging targeted
therapies, and immunotherapy.
TARGETED THERAPIES AND IMMUNE SYSTEM

MODULATION TO TREAT GASTRIC
ADENOCARCINOMA
Table 1 summarizes representative important clinical trial
results in GAC.
Negative trials with targeted agents have substantially
outnumbered the positive trials (the ToGA, REGARD, and
RAINBOW trials) in GAC in the past decade. Among other
factors, major factors accounting for this negative outcome
may be (1) many trials did not select the patient population
based on specific target (lacking any biomarker) and did not
account for distinct molecular subtypes and how they may
impact response to therapy, (2) inaccurate biomarker for
selection patient selection (i.e., HER2, fluorescence in situ
hybridization [FISH] vs. HER2 immunohistochemistry [IHC];
FGFR2 FISH vs. FGFR2 IHC; MET FISH vs. MET IHC). There may
be many other factors influencing the failure of various drugs

when investigated in the clinic.
In general, for the targeted agents, clinical trials in GAC for
unselected patient population have been largely disappointing. One clear example is the EXPAND trial, where EGFR
inhibitors were given to all patients despite the prevalence of
EGFR gene amplification occurs in only approximately 5% of
GACs and EGFR protein overexpression in approximately 15%
to 20% of patients with GAC. Another EGFR-directed study
in unselected patients, the nimotuzumab/irinotecan secondline trial, also failed to demonstrate survival benefit. However,
in subgroup analysis of the EXPAND trial, a subset of patients
with high EGFR overexpression by IHC showed survival benefit
from cetuximab compared with placebo.32 In line with this,
in the randomized phase II trial of nimotuzumab/irinotecan
versus irinotecan, a subset of patients with EGFR 2+/3+ by
IHC demonstrated superior progression-free survival (PFS)
and overall survival (OS).33 Based on this finding, a phase III
trial of nimotuzumab/irinotecan versus irinotecan with patients selected based upon EGFR2/3-positive IHC is currently
ongoing.
Recently, another disappointing result came from the
phase III RILOMET-1 trial that compared rilotumumab in
combination with epirubicin/cisplatin/capecitabine (ECX)
versus ECX alone as first-line chemotherapy administered
in patients with MET-positive GAC. There was worse survival in the arm containing the anti-HGFtargeted agent
rilotumumab.34 Moreover, there were no subgroups shown
to benefit from rilotumumab including those patients with
higher percentages of cells with greater than or equal to +1
MET expression by IHC.34 It is not yet reported whether
genomic amplification of the gene encoding MET influenced
survival or response to this agent. However, a phase I trial of
the MET kinase inhibitor AMG337 with 13 patients with
MET-amplified GAC demonstrated a response rate of 62%.35
Intriguingly, 40% to 50% of patients with MET-amplified GAC
enrolled on AMG337 trial harbored coamplification of HER2
and/or EGFR concurrently, suggestive of patients who might
benefit from combination targeted therapy.36 To further add
to the complexity, there was a substantial heterogeneity and
discordance in MET amplifications between the primary and
metastatic lesions in a single studied patient with GAC.
Hence, molecular profiling of a single tumor biopsied may
not be inadequate to guide targeted therapy in GAC.
Another lesson can be learned from the SHINE trial that
compared AZD4547, an FGFR2 inhibitor, administered in
patients with FGFR2-amplified GAC in the second-line
setting. Despite strong preclinical evidence for AZD4547
in FGFR2-amplified GAC preclinical models, the SHINE trial
failed to meet the primary endpoint, PFS, when AZD4547
was added to paclitaxel.24 The trial screened 960 patients
and randomly assigned 71 patients with 9% prevalence for
FGFR2 amplification. Although that trial was negative, in another phase II trial with AZD4547 in FGFR1-, 2-, and 3-amplified
cancer, three of nine (33%) patients with FGFR2-amplified
GAC experienced a partial response. Notably, all responding
patients had FGFR2 copy number gain demonstrated in
plasma DNA and high FGFR2 copy number gain,37 suggesting

that optimal use of biomarkers may mark those for whom
this target may be effective. Hence, although the SHINE trial
did not meet its primary endpoint, it is not yet conclusive to
preclude FGFR2 inhibitor from clinical development, but
more optimized biomarker selection criteria are needed for
the future FGFR2 inhibitor trials.
There is emerging evidence that pembrolizumab/nivolumab
blocks binding of PD-L1 to its receptor, PD-1, resulting in T-cell
activation/proliferation. KEYNOTE-012, a phase IB study of
pembrolizumab in patients with metastatic GAC and PD-L1
tumor cell positivity by IHC demonstrated promising results.38
MEDI4736, another antiPD-L1 antibody, also has demonstrated promising activity in GAC in an initial analysis.
Is It Time for Umbrella Trials in GAC?

Recently, even further complicating the clinical development of targeted agents in GAC, TCGA and ACRG10,11
published comprehensive genomic data, and both groups
have reported on molecular subtypes of GAC (detailed
earlier). Given the tumor heterogeneity among patients
with GAC and within each patient, an umbrella trial has
been recently introduced.39-41 An umbrella (or basket) trial
investigates a single tumor type selected according to the
biomarkers relevant to one or more of the candidate drugs
and patients and directed toward different arms of the study
according to the molecular characteristics of the tumor.39
The Personalized Antibodies from Gastro-Esophageal Adenocarcinoma (PANGEA) trial selects patients with GAC
according to biomarker status, such as MET amplification,
FGFR2 amplification, and KRAS amplification, to different
arms in first-line setting. Another trial, the Targeted Agent
Evaluation in Gastric Cancer Basket Korea (VIKTORY), assigns
patients with metastatic GAC according to RAS amplification,
RAS mutation, TP53 mutation, PIK3CA mutation, MEK high
versus low signature, MET amplification, and MET protein
overexpression to corresponding treatment arms in secondline treatment. Patients without any of the designated
markers receive either mTOR inhibitor or Akt inhibitor with
paclitaxel.
HER2/NEU AS TARGET IN GASTRIC

ADENOCARCINOMA AND AN UPDATE ON HER2/
NEU TESTING
A small fraction of GAC in patients overexpresses HER2/neu
protein or have increase copy number of the ERBB2 gene.
The HER2/neu receptor is one of the four receptors in the
EGFR family. The intracellular domain has tyrosine kinase
activity. The HER2/neu receptor can be targeted by an
antibody (such as trastuzumab or pertuzumab, which can
prevent homodimerization or heterodimerization with other
family members) against the extracellular domain or receptor tyrosine kinase small-molecule inhibitors (such as
lapatinib). There have been several prospective phase III
trials that have investigated inhibition of HER2/neu signaling
in patients with HER2/neu-positive GAC.5,16,42 However,
only the ToGA trial showed statistically significant benefit
107
(p = .0046).5 Therefore, the use of trastuzumab in combination with chemotherapy in patients with HER2/neu-positive metastatic GAC is now considered a standard. We
discuss various aspects of targeting HER2/neu pathway and
HER2 testing.
The Receptor
The HER2/neu receptor is unique in that it lacks the ligand
binding cleft for extracellular growth factors that other
ERBB3 family members have. Its dimerization loop is extended like others, and the intracellular domain has ATPbinding cleft.43 The ERBB2 gene, encoding HER2/neu, is
often amplified in many cancers including a fraction of GACs
(including in some squamous cell carcinoma of the esophagus). The drugs that can inhibit HER2/neu receptors are
well described in a recent review.43 However, primary and
secondary resistance to HER2/neu inhibition, as with other
single targets, is a likely inevitable and universal phenomenon.44 Common mechanisms of HER2/neu directed therapy resistance have not been extensively studied in GAC,
but, in breast cancer, it has been seen because of PTEN loss,
truncation of the receptor, activating mutations in PIK3CA,
or upregulation of other oncogenes (IGF1 or ERBB3).43,44
HER2/neu Test Results in Various Settings

In routine use, the presence of strong (3+) IHC staining has
been shown to have strong concordance with the presence
of amplification of ERBB2, leading this protein test alone to
be sufficient to guide the use of trastuzumab.5,45 However,
one issue to consider is the considerable heterogeneity in
staining for HER2/neu and the potential adequacy of a single
small biopsy to accurately assess HER2 status (as opposed
to a surgical specimen46-48). How many unique specimens
should be tested also remains unresolved.49-51 Similarly,
technical issues regarding the optimal copy numbers cutoffs
to use to guide use of anti HER2/neu agents are also a subject
of debate.52 In addition, whether there is concordance in
HER2/neu results if metastatic cancer or primary cancer is
tested, remains generally unresolved.53-55 Regarding prognostic significance, HER2/neu positivity in a GAC does not
confer poor prognosis, but there are reports to refute this as
well.56-63
Testing for HER2/neu

Currently, we recommend the guidelines outlined by the
National Comprehensive Cancer Network (NCCN)64,65 until
formal guidelines to be published jointly by the American
Society of Clinical Oncology, College of American Pathologists, and American Society of Clinical Pathologists become
available in early 2016. Patients with gastroesophageal
adenocarcinoma who have unresectable, advanced cancer
should be tested for HER2/neu, starting with IHC. If it is 3+
(positive), 1+ (negative), or 0 (negative), no further testing is
required. However, if it is 2+ (equivocal), then FISH for
amplification of ERBB2 is recommended. Currently, any
available tumor tissue should be tested.
108
In a biopsy specimen of GAC (or gastroesophageal adenocarcinoma), if 5% of cells adequately stain for HER2/neu,
this is currently considered adequate to designate a positive
HER2/neu status; however, the presence of positive staining
in such a small fraction of the sampled area raises obvious
questions of sampling and whether such few cells really
represent the entire tumor (including metastatic tumor).
However, from surgical specimen, at least 10% of cells must
stain for HER2/neu designation. This is a bit better but still
raises the same questions.
Clinical Trials
The ToGA trial screened 3,803 patients with untreated,
advanced gastroesophageal adenocarcinoma and randomly
assigned 594 patients with HER2/neu positive tumors (as
defined for the purposes of this trial) to receive either
chemotherapy (fluoropyrimidine/cisplatin) and trastuzumab or chemotherapy. The primary endpoint of improved
overall survival was achieved (13.8 months vs. 11.1 months;
p = .0046). Other endpoints (PFS and increase in response
rate) were also met. Trastuzumab was found safe to deliver
with standard chemotherapy. This trial led to worldwide
regulatory approval of trastuzumab and clinical adoption.
However, the benefit from trastuzumab diminished considerably when the results were reanalyzed after a longer
follow-up by the U.S. Food and Drug Administration (the
hazard ratio increased from 0.73 to 0.80, and the overall
survival difference narrowed to a meager 1.4 months). This
would suggest considerable heterogeneity among patients
with HER2/neu-positive GACs and need for better agents.
This modest impact also suggests the need to refine our
identification of possible gradations of HER2/neu-positive
GACs. Furthermore, these results speak strongly to the need
to develop additional treatments for patients with HER2/
neu-positive GAC, both for up-front therapies and when the
resistance emerges.
In contrast to the ToGA trial of the therapeutic antibody
trastuzumab, results have been disappointing for HER2directed kinase inhibitors. Two trials studied the reversible EGFR/ERBB2 inhibitor lapatinib in the second-line
setting42 and in the first-line setting,16 both yielding
negative results. Ado-trastuzumab emtansine (T-DM1) is
an antibody-drug conjugate consisting of antibody
(trastuzumab) and cytotoxic (DM1). The EMILIA trial for
patients with HER2/neu-positive breast cancer was positive, resulting in a survival benefit of 5.8 months; however, a T-DM1 trial for HER2/neu-positive GAC in the
second-line setting was a disappointment.66 Currently,
trials are ongoing with higher doses of trastuzumab to
improve exposure and another that adds the second
antibody pertuzumab to trastuzumab in comparison with
trastuzumab (the JACOB trial). Additional mechanistic
study into the molecular effects of specific classes of antiHER2/neu agents and the mechanistic etiologies of both
de novo and acquired resistance are required to
establish a foundation for rational development of new
approaches for to target HER2-positive GACs.
FUTURE DIRECTIONS
Gastric adenocarcinoma remains a deadly disease with
significant molecular and histologic heterogeneity. Trials of
empirical chemotherapy, largely with a one-size-fits-all
approach have led to modest survival benefits and limited
options. Newer insights into the underlying etiology of
GACs does bring possible promise of new, improved
therapies. However, substantial work needs to be completed to identify optimal targets, optimal biomarkers, and
the development of effective combination therapies to

exploit these targets. Furthermore, we will have to evaluate the potential for tumor heterogeneity and tumor
evolution/resistance to be a barrier to these therapies.
Moreover, emerging immunotherapies possess great
promise. Ultimately, linking immune, cytotoxics, and other
targeted therapies will likely be needed as part of multipronged approaches to provide maximum benefits to our
patients.
References
1. Torre LA, Bray F, Siegel RL, et al. Global cancer statistics, 2012. CA Cancer
J Clin. 2015;65:87-108.
2. Al-Batran SE, Hartmann JT, Probst S, et al; Arbeitsgemeinschaft
Internistische Onkologie. Phase III trial in metastatic gastroesophageal
adenocarcinoma with fluorouracil, leucovorin plus either oxaliplatin or
cisplatin: a study of the Arbeitsgemeinschaft Internistische Onkologie.
J Clin Oncol. 2008;26:1435-1442.
3. Cunningham D, Starling N, Rao S, et al; Upper Gastrointestinal Clinical
Studies Group of the National Cancer Research Institute of the United
Kingdom. Capecitabine and oxaliplatin for advanced esophagogastric
4. Kang YK, Kang WK, Shin DB, et al. Capecitabine/cisplatin versus 5-fluorouracil/
cisplatin as first-line therapy in patients with advanced gastric cancer:
a randomised phase III noninferiority trial. Ann Oncol. 2009;20:666-673.
5. Bang YJ, Van Cutsem E, Feyereislova A, et al; ToGA Trial Investigators.
Trastuzumab in combination with chemotherapy versus chemotherapy
alone for treatment of HER2-positive advanced gastric or gastrooesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlled trial. Lancet. 2010;376:687-697.
6. Wilke H, Muro K, Van Cutsem E, et al; RAINBOW Study Group.
Ramucirumab plus paclitaxel versus placebo plus paclitaxel in patients
with previously treated advanced gastric or gastro-oesophageal junction adenocarcinoma (RAINBOW): a double-blind, randomised phase 3
7. Shah MA. Update on metastatic gastric and esophageal cancers. J Clin
Oncol. 2015;33:1760-1769.
8. Cohen DJ, Leichman L. Controversies in the treatment of local and
locally advanced gastric and esophageal cancers. J Clin Oncol. 2015;33:
1754-1759.
9. Tan P, Yeoh KG. Genetics and molecular pathogenesis of gastric adenocarcinoma. Gastroenterology. 2015;149:1153-1162 e3.
10. Cristescu R, Lee J, Nebozhyn M, et al. Molecular analysis of gastric
cancer identifies subtypes associated with distinct clinical outcomes.
Nat Med. 2015;21:449-456.
11. Cancer Genome Atlas Research Network. Comprehensive molecular
characterization of gastric adenocarcinoma. Nature. 2014;513:202-209.
12. Kakiuchi M, Nishizawa T, Ueda H, et al. Recurrent gain-of-function
mutations of RHOA in diffuse-type gastric carcinoma. Nat Genet. 2014;
46:583-587.
13. Wang K, Yuen ST, Xu J, et al. Whole-genome sequencing and comprehensive molecular profiling identify new driver mutations in gastric
cancer. Nat Genet. 2014;46:573-582.
14. van Beek J, zur Hausen A, Klein Kranenbarg E, et al. EBV-positive gastric
adenocarcinomas: a distinct clinicopathologic entity with a low frequency of lymph node involvement. J Clin Oncol. 2004;22:664-670.
16. Hecht JR, Bang YJ, Qin SK, et al. Lapatinib in combination with capecitabine plus oxaliplatin in human epidermal growth factor receptor
2-positive advanced or metastatic gastric, esophageal, or gastroesophageal adenocarcinoma: TRIO-013/LOGiC-A randomized phase III
trial. J Clin Oncol. 2016;34:443-451.
17. Bang Y-J RX, Taroh S, et al. A randomized, open-label, phase III study of
lapatinib in combination with weekly paclitaxel versus weekly paclitaxel
alone in the second-line treatment of HER2 amplified advanced gastric
cancer in Asian population: TyTAN study. J Clin Oncol. 2013;30 (suppl 34;
abstr 11).
18. Lordick F, Kang YK, Chung HC, et al; Arbeitsgemeinschaft Internistische
Onkologie and EXPAND Investigators. Capecitabine and cisplatin with or
without cetuximab for patients with previously untreated advanced
gastric cancer (EXPAND): a randomised, open-label phase 3 trial. Lancet
Oncol. 2013;14:490-499.
19. Waddell T, Chau I, Cunningham D, et al. Epirubicin, oxaliplatin, and
capecitabine with or without panitumumab for patients with previously
untreated advanced oesophagogastric cancer (REAL3): a randomised,
open-label phase 3 trial. Lancet Oncol. 2013;14:481-489.
20. Kim YH, Sasaki Y, Lee KH, et al. Randomized phase II study of nimotuzumab, an anti-EGFR antibody, plus irinotecan in patients with 5fluorouracil-based regimen-refractory advanced or recurrent gastric
cancer in Korea and Japan: preliminary results. J Clin Oncol. 2011;29:4s
(suppl 4; abstr 87).
21. Van Cutsem E, de Haas S, Kang YK, et al. Bevacizumab in combination
with chemotherapy as first-line therapy in advanced gastric cancer:
a biomarker evaluation from the AVAGAST randomized phase III trial.
J Clin Oncol. 2012;30:2119-2127.
22. Iveson T, Donehower RC, Davidenko I, et al. Rilotumumab in
combination with epirubicin, cisplatin, and capecitabine as first-line
treatment for gastric or oesophagogastric junction adenocarcinoma: an open-label, dose de-escalation phase 1b study and a
double-blind, randomised phase 2 study. Lancet Oncol. 2014;15:
1007-1018.
23. Shah MA, Bang Y-J, Lordick F, et al. METGastric: A phase III study of
onartuzumab plus mFOLFOX6 in patients with metastatic HER2negative (HER22) and MET-positive (MET+) adenocarcinoma of the
stomach or gastroesophageal junction (GEC). J Clin Oncol. 2015;33
24. Bang Y-J, Cutsem EV, Mansoor W, et al. A randomized, open-label phase
II study of AZD4547 (AZD) versus paclitaxel (P) in previously treated
patients with advanced gastric cancer (AGC) with fibroblast growth
factor receptor 2 (FGFR2) polysomy or gene amplification (amp): SHINE
study. J Clin Oncol. 2015;33 (suppl; abstr 4014).
25. Ohtsu A, Ajani JA, Bai YX, et al. Everolimus for previously treated advanced gastric cancer: results of the randomized, double-blind, phase III
GRANITE-1 study. J Clin Oncol. 2013;31:3935-3943.
109
26. Ramanathan RK, McDonough SL, Kennecke HF, et al. Phase 2 study of
MK-2206, an allosteric inhibitor of AKT, as second-line therapy for
advanced gastric and gastroesophageal junction cancer: A SWOG cooperative group trial (S1005). Cancer. Epub 2015 Mar 30.
27. Bang YJ, Im SA, Lee KW, et al. Randomized, double-blind phase II trial
with prospective classification by ATM protein level to evaluate the
efficacy and tolerability of olaparib plus paclitaxel in patients with
recurrent or metastatic gastric cancer. J Clin Oncol. 2015;33:3858-3865.
28. Enzinger PC, McCleary NJ, Zheng H, et al. Multicenter double-blind
randomized phase II: FOLFOX + ziv-aflibercept/placebo for patients (pts)
with chemo-naive metastatic esophagogastric adenocarcinoma
(MEGA). J Clin Oncol. 2016;34 (suppl 4S; abstr 4).
29. Kang Y-K, Shah MA, Ohtsu A, et al. A randomized, open-label, multicenter, adaptive phase 2/3 study of trastuzumab emtansine (T-DM1)
versus a taxane (TAX) in patients (pts) with previously treated HER2positive locally advanced or metastatic gastric/gastroesophageal
junction adenocarcinoma (LA/MGC/GEJC). J Clin Oncol. 2016;34
(suppl 4S; abstr 5).
30. Wilke H, Van Cutsem E, Oh SC, et al. RAINBOW: A global, phase III,
randomized, double-blind study of ramucirumab plus paclitaxel versus
placebo plus paclitaxel in the treatment of metastatic gastroesophageal
junction (GEJ) and gastric adenocarcinoma following disease progression on first-line platinum- and fluoropyrimidine-containing combination therapy rainbow IMCL CP12-0922 (I4T-IE-JVBE). J Clin Oncol.
2014;32 (suppl 3; abstr LBA7).
31. Fuchs CS, Tomasek J, Yong CJ, et al; REGARD Trial Investigators.
Ramucirumab monotherapy for previously treated advanced gastric or
gastro-oesophageal junction adenocarcinoma (REGARD): an international, randomised, multicentre, placebo-controlled, phase 3 trial.
Lancet. 2014;383:31-39.
32. Lordick F, Kang Y-K, Salman P, et al. Clinical outcome according to tumor
HER2 status and EGFR expression in advanced gastric cancer from the
EXPAND study. J Clin Oncol. 2013;31 (suppl; abstr 4021).
33. Satoh T, Lee KH, Rha SY, et al. Randomized phase II trial of nimotuzumab
plus irinotecan versus irinotecan alone as second-line therapy for
patients with advanced gastric cancer. Gastric Cancer. 2015;18:
824-832.
34. Cunningham D, Tebbutt NC, Davidenko I, et al. Phase III, randomized,
double-blind, multicenter, placebo (P)-controlled trial of rilotumumab
(R) plus epirubicin, cisplatin and capecitabine (ECX) as first-line therapy
in patients (pts) with advanced MET-positive (pos) gastric or gastroesophageal junction (G/GEJ) cancer: RILOMET-1 study. J Clin Oncol.
2015;33 (suppl; abstr 4000).
35. Kwak EL, LoRusso P, Hamid O, et al. Clinical activity of AMG 337, an oral
MET kinase inhibitor, in adult patients (pts) with MET-amplified gastroesophageal junction (GEJ), gastric (G), or esophageal (E) cancer. J Clin
Oncol. 2015;33 (suppl 3; abstr 1).
36. Kwak EL, Ahronian LG, Siravegna G, et al. Molecular heterogeneity and
receptor coamplification drive resistance to targeted therapy in METamplified esophagogastric cancer. Cancer Discov. 2015;5:1271-1281.
37. Smyth EC, Turner NC, Pearson A, et al. Phase II study of AZD4547 in FGFR
amplified tumours: gastroesophageal cancer (GC) cohort pharmacodynamic and biomarker results. J Clin Oncol. 2016;34 (suppl 4S; abstr
154).
38. Muro K, Bang Y-J, Shankaran V, et al. Relationship between PD-L1
expression and clinical outcomes in patients (Pts) with advanced gastric
cancer treated with the anti-PD-1 monoclonal antibody pembrolizumab
(Pembro; MK-3475) in KEYNOTE-012. J Clin Oncol. 2015;33 (suppl 3,
abstr 3).
39. Biankin AV, Piantadosi S, Hollingsworth SJ. Patient-centric trials for
therapeutic development in precision oncology. Nature. 2015;526:
361-370.
110
40. Catenacci DV. Next-generation clinical trials: Novel strategies to address

the challenge of tumor molecular heterogeneity. Mol Oncol. 2015;9:
967-996.
41. Lee J, Kim KM, Kang WK, et al. Innovative personalized medicine in
gastric cancer: time to move forward. Clin Genet. 2014;86:37-43.
42. Satoh T, Xu RH, Chung HC, et al. Lapatinib plus paclitaxel versus paclitaxel alone in the second-line treatment of HER2-amplified advanced
gastric cancer in Asian populations: TyTANa randomized, phase III
study. J Clin Oncol. 2014;32:2039-2049.
43. Yarden Y, Pines G. The ERBB network: at last, cancer therapy meets
systems biology. Nat Rev Cancer. 2012;12:553-563.
44. Holohan C, Van Schaeybroeck S, Longley DB, et al. Cancer
drug resistance: an evolving paradigm. Nat Rev Cancer. 2013;13:
714-726.
45. Chen M, Li Y, Ming Z, et al. Comparison of HER2 status by fluorescence
in situ hybridisation and immunohistochemistry in gastric cancer.
Contemp Oncol (Pozn). 2014;18:95-99.
46. Yoshida H, Yamamoto N, Taniguchi H, et al. Comparison of HER2 status
between surgically resected specimens and matched biopsy specimens
of gastric intestinal-type adenocarcinoma. Virchows Arch. 2014;465:
145-154.
47. Wang T, Hsieh ET, Henry P, et al. Matched biopsy and resection
specimens of gastric and gastroesophageal adenocarcinoma show high
concordance in HER2 status. Hum Pathol. 2014;45:970-975.
48. van Hagen P, Biermann K, Boers JE, et al. Human epidermal growth
factor receptor 2 overexpression and amplification in endoscopic biopsies and resection specimens in esophageal and junctional adenocarcinoma. Dis Esophagus. 2015;28:380-385.
49. Grillo F, Fassan M, Ceccaroli C, et al. The reliability of endoscopic biopsies in assessing HER2 status in gastric and gastroesophageal junction
cancer: a study comparing biopsies with surgical samples. Transl Oncol.
2013;6:10-16.
50. Ge X, Wang H, Zeng H, et al. Clinical significance of assessing Her2/neu
expression in gastric cancer with dual tumor tissue paraffin blocks. Hum
Pathol. 2015;46:850-857.
51. Qiu Z, Sun W, Zhou C, et al. HER2 expression variability between primary
gastric cancers and corresponding lymph node metastases. Hepatogastroenterology. 2015;62:231-233.
52. Kumarasinghe MP, de Boer WB, Khor TS, et al. HER2 status in gastric/
gastro-oesophageal junctional cancers: should determination of gene
amplification by SISH use HER2 copy number or HER2: CEP17 ratio?
Pathology. 2014;46:184-187.
53. Kochi M, Fujii M, Masuda S, et al. Differing deregulation of HER2 in
primary gastric cancer and synchronous related metastatic lymph
nodes. Diagn Pathol. 2013;8:191.
54. Cho EY, Park K, Do I, et al. Heterogeneity of ERBB2 in gastric carcinomas:
a study of tissue microarray and matched primary and metastatic
carcinomas. Mod Pathol. 2013;26:677-684.
55. Bozzetti C, Negri FV, Lagrasta CA, et al. Comparison of HER2 status in
primary and paired metastatic sites of gastric carcinoma. Br J Cancer.
2011;104:1372-1376.
56. Chen XZ, Zhang WH, Yao WQ, et al. Immunohistochemical HER2 expression not associated with clinicopathological characteristics of stage IIII gastric cancer patients. Hepatogastroenterology. 2014;61:1817-1821.
57. Geng Y, Chen X, Qiu J, et al. Human epidermal growth factor receptor-2
expression in primary and metastatic gastric cancer. Int J Clin Oncol.
2014;19:303-311.
58. Geppert CI, Rummele P, Sarbia M, et al. Multi-colour FISH in oesophageal adenocarcinoma-predictors of prognosis independent of stage
and grade. Br J Cancer. 2014;110:2985-2995.
59. Aizawa M, Nagatsuma AK, Kitada K, et al. Evaluation of HER2-based

biology in 1,006 cases of gastric cancer in a Japanese population. Gastric
Cancer. 2014;17:34-42.
60. Gomez-Martin C, Plaza JC, Pazo-Cid R, et al. Level of HER2 gene amplification predicts response and overall survival in HER2-positive advanced gastric cancer treated with trastuzumab. J Clin Oncol. 2013;31:
4445-4452.
61. Okines AF, Thompson LC, Cunningham D, et al. Effect of HER2 on
prognosis and benefit from peri-operative chemotherapy in early
oesophago-gastric adenocarcinoma in the MAGIC trial. Ann Oncol.
2013;24:1253-1261.
62. Yoon HH, Shi Q, Sukov WR, et al. Adverse prognostic impact of
intratumor heterogeneous HER2 gene amplification in patients with
esophageal adenocarcinoma. J Clin Oncol. 2012;30:3932-3938.
63. Janjigian YY, Werner D, Pauligk C, et al. Prognosis of metastatic gastric and
gastroesophageal junction cancer by HER2 status: a European and USA
International collaborative analysis. Ann Oncol. 2012;23:2656-2662.
64. Ajani JA, Bentrem DJ, Besh S, et al; National Comprehensive Cancer
Network. Gastric cancer, version 2.2013: featured updates to the NCCN
Guidelines. J Natl Compr Canc Netw. 2013;11:531-546.
65. Ajani JA, DAmico TA, Almhanna K, et al; National Comprehensive
Cancer Network. Esophageal and esophagogastric junction cancers,
version 1.2015. J Natl Compr Canc Netw. 2015;13:194-227.
66. Kang YK, Shah MA, Ohtsu A, et al. A randomized, open-label, multicenter, adaptive phase 2/3 study of trastuzumab emtasine (T-DM1)
versus a taxane (TAX) in patients with previously treated Her2 positive
locally advanced or metastatic gastric/gastroesophageal junction adenocarcinoma. J Clin Oncol. 2016;34 (suppl 4S; abstr 5).
111
Localized Pancreatic Cancer:

Multidisciplinary Management
With Incorporation of ASCO
Guidelines
CHAIR
E. Gabriela Chiorean, MD
Seattle, WA
SPEAKERS
Diane M. Simeone, MD
Ann Arbor, MI
Joseph M. Herman, MD, MSc
Baltimore, MD
MULTIDISCIPLINARY TREATMENT IN PANCREATIC CANCER
Localized Pancreatic Cancer: Multidisciplinary Management

Andrew L. Coveler, MD, Joseph M. Herman, MD, MSc, Diane M. Simeone, MD, and E. Gabriela Chiorean, MD
OVERVIEW
Pancreatic cancer is an aggressive cancer that continues to have single-digit 5-year mortality rates despite advancements in
the field. Surgery remains the only curative treatment; however, most patients present with late-stage disease deemed
unresectable, either due to extensive local vascular involvement or the presence of distant metastasis. Resection guidelines
that include a borderline resectable group, as well as advancements in neoadjuvant chemotherapy and radiation that
improve resectability of locally advanced disease, may improve outcomes for patients with more invasive disease. Multiagent chemotherapy regimens fluorouracil, leucovorin, irinotecan, and oxaliplatin (FOLFIRINOX) and nab-paclitaxel with
gemcitabine improved response rates and survival in metastatic pancreatic cancer and are now being used in earlier
stages for patients with localized potentially resectable and unresectable disease, with goals of downstaging tumors to allow
margin-negative resection and reducing systemic recurrence. Chemoradiotherapy, although still controversial for both
resectable and unresectable pancreatic cancer, is being used in the context of contemporary chemotherapy backbone
regimens, and novel radiation techniques such as stereotactic body frame radiation therapy (SBRT) are studied on the premise
of maintaining or improving efficacy and reducing treatment duration. Patient selection for optimal treatment designation is
currently provided by multidisciplinary tumor boards, but biomarker discovery, in blood, tumors, or through novel imaging, is
an area of intense research. Results to date suggest that some patients with unresectable disease at the outset have survival
rates as good as those with initially resectable disease if able to undergo surgical resection. Long-term follow-up and improved
clinical trials options are needed to determine optimal treatment modalities for patients with localized pancreatic cancer.
pproximately 277,000 new cases of pancreatic cancer

are diagnosed each year in the world,1 among which
approximately 49,000 occur in both the United States2 and
Europe.3 Localized disease accounts for approximately 50%
of newly diagnosed patients, with the largest fraction, 35%,
being locally advanced, unresectable pancreatic cancer
(LAUPC) and 10% to 15% resectable or borderline resectable
pancreatic cancer (BRPC). The goals of care for patients with
localized pancreatic cancer are increasing the likelihood of
margin-negative resection, preventing metastatic spread,
converting unresectable into potentially resectable disease,
and overall improving patients quality of life. In this update
on multidisciplinary treatment of localized pancreatic cancer, we highlight standard and novel approaches, overview
ongoing controversies, and emphasize current clinical trials.
INTEGRATING MULTIMODALITY THERAPY IN

THE TREATMENT OF LOCALIZED AND LOCALLY
ADVANCED PANCREATIC CANCER
Resectable Pancreatic Cancer
Localized resectable pancreatic cancer represents less than
15% of all new diagnoses, and the National Comprehensive
Cancer Network (NCCN) guidelines recommend surgery followed by adjuvant chemotherapy with gemcitabine or fluorouracil (5-FU) as the current standard of care.4 Despite
perioperative mortality rates of less than 2% in experienced
centers, patients undergoing surgical resection followed by
adjuvant chemotherapy (gemcitabine or 5-FU) or chemotherapy plus chemoradiotherapy experience survival rates of
24 months and 5-year overall survival (OS) of 20% on average,
unchanged in the past few decades, mostly due to high rates
of systemic relapse.58 The EORTC 40013 randomized phase
II trial compared adjuvant gemcitabine (four cycles) with
gemcitabine (two cycles) plus gemcitabine-based radiotherapy and showed similar OS rates (24 months), but local
relapse was improved in the chemoradiotherapy arm (15%
vs. 31%).9 Because 30% of patients with pancreatic cancer
experience relapse only locally after surgery,5 the potential
benefit of adjuvant chemoradiotherapy needs to be addressed
in a larger randomized trial. With many historical trials highly
criticized due to suboptimal chemoradiotherapy treatment
delivery (e.g., GITSG, EORTC, and ESPAC-1),6,10,11 the phase III
study RTOG-0848 (NCT01013649) in the United States is
currently evaluating the role of adjuvant chemoradiotherapy
From the Division of Hematology-Oncology, University of Washington, Fred Hutchinson Cancer Research Center, Seattle, WA; Department of Radiation Oncology and Molecular
Radiation Sciences, Johns Hopkins University, Baltimore, MD; University of Michigan, Ann Arbor, MI.
Corresponding author: E. Gabriela Chiorean, MD, Fred Hutchinson Cancer Research Center, 825 Eastlake Ave. E, G4-833, Seattle, WA 98109-6248; email: gchiorea@uw.edu.
e217
COVELER ET AL
when added after 6 months of gemcitabine compared with

gemcitabine alone (6 months). Until results from RTOG-0848
more clearly define the benefit of adjuvant chemoradiotherapy
in this patient population, in clinical practice, adjuvant
chemoradiotherapy is most commonly reserved for patients with margin-positive (R1 or R2) resection and/or
lymph nodepositive disease.12,13 Intensifying systemic
chemotherapy after pancreatic cancer resection is being
addressed in the adjuvant APACT study (NCT01964430),
using nab-paclitaxel with gemcitabine versus gemcitabine
alone, and the ACCORD 24 study (NCT01526135) with
modified FOLFIRINOX (mFOLFIRINOX) versus gemcitabine.
None of these latter studies incorporates chemoradiotherapy
in the adjuvant regimen.
Data in genetically engineered mouse models of pancreatic cancer suggest metastatic spread precedes local tumor
formation.14 Combined with evidence that there is a high
rate of disease recurrence following resection of stage I/II
disease suggests that even in the absence of radiologic
evidence of metastasis, more effective systemic chemotherapy should confer an increased likelihood of cure after
resection by successfully treating micrometastases as early
as possible. On this premise, neoadjuvant approaches have
been tested in resectable pancreatic cancer with the
KEY POINTS
e218
Neoadjuvant multimodality therapy should be

considered for patients with borderline resectable and
locally advanced unresectable pancreatic cancer.
Induction chemotherapy for at least 2 to 3 months
selects patients without rapid disease metastasis who
may benefit most from the addition of radiotherapy.
Ongoing clinical trials will help define the role of
adjuvant radiotherapy for resectable patients
(RTOG0848), as well as for locally advanced
unresectable patients (RTOG1201).
Select patients with initially localized but unresectable
disease may undergo potentially curable surgical
resection after neoadjuvant therapy and have outcomes
comparable to patients with initially resectable disease.
Radiologic RECIST response does not adequately predict
resectability for patients with borderline resectable or
locally advanced unresectable pancreatic cancer.
Pathologic complete or near-complete response (often
defined as < 5% viable cells) to neoadjuvant therapy may
predict increased survival.
It is currently unknown whether borderline resectable
or locally advanced unresectable patients undergoing
induction (neoadjuvant) multi-agent chemotherapy will
derive additional benefit with the addition of
radiotherapy.
Novel combinations of systemic treatment and
radiotherapy modalities such as SBRT and IMRT may
provide superior resectability and local control.
Blood and tissue biomarkers may assist in determining
optimal neoadjuvant strategies.
assumption of several advantages: prompt use of systemic

therapy to treat occult micrometastases, allowing a period of
time to observe the tumor biology during chemotherapy
exposure and avoiding unnecessary resection for rapidly
metastasizing tumors, improving the likelihood of marginnegative resection, and improved chemotherapy delivery
compared with postresection, when complications may
delay or worsen chemotherapy tolerability. In a few institutional prospective studies of neoadjuvant chemotherapy (gemcitabine/cisplatin and gemcitabine/oxaliplatin)
with or without chemoradiotherapy in radiographic resectable disease, approximately 70% of patients underwent
surgical resection, and OS reached 27 to 34 months for
resected patients, possibly due to patient selection.1517
However, most patients in these studies still experienced
relapsed with metastatic disease. It is therefore not clear
that neoadjuvant therapy with standard chemotherapy
regimens confers higher OS compared with adjuvant approaches, and with no randomized studies to date, the NCCN
guidelines do not yet endorse neoadjuvant therapy for
patients with clearly resectable disease on baseline staging
CT scans, outside of clinical trials. For patients who receive
neoadjuvant therapy followed by resection, further adjuvant
therapy is recommended. This approach is being adopted by
the recently opened SWOG S1505 randomized phase II
study (NCT02243007), which compares mFOLFIRINOX with
nab-paclitaxel with gemcitabine for 3 months before and
3 months after surgery. A recent report suggests that a
subset of patients treated with neoadjuvant therapy who
undergo surgical resection and are found to have low lymph
node ratio (number of positive lymph nodes/number total
lymph nodes resected , 0.15) may have better survival
outcomes when also treated with adjuvant chemotherapy
(OS: 72 vs. 33 months; p = .008).18 In the absence of definitive
guidelines, clinical criteria for considering neoadjuvant
therapy in resectable pancreatic cancer are: large primary
tumors, high carbohydrate antigen 19-9 (CA19-9) levels
(. 1,000 U/mL), peripancreatic lymph node involvement by
cross-sectional imaging or endoscopic ultrasound, or equivocal radiologic features.19 Randomized phase II/III studies are
ongoing in resectable pancreatic cancer that may help clarify
the role of neoadjuvant versus adjuvant therapies: NEOPAC
(NCT01314027) compares adjuvant gemcitabine versus
neoadjuvant gemcitabine/oxaliplatin followed by adjuvant
gemcitabine, NEPAFOX (NCT02172976) compares adjuvant
gemcitabine versus neoadjuvant/adjuvant FOLFIRINOX,
and NEONAX (NCT02047513) compares adjuvant versus
neoadjuvant/adjuvant nab-paclitaxel with gemcitabine.
Defining a more effective neoadjuvant or adjuvant regimen would provide great benefit in improving survival
among resectable patients, in whom the therapy is geared
toward targeting micrometastatic disease.
Borderline Resectable Pancreatic Cancer

Patients with borderline-resectable pancreatic cancer (BRPC)
have localized disease, but less likely to be resected with
negative margins due to proximity or direct involvement of

venous and/or arterial vascular structures (# 180).20 In this
setting, although no particular chemotherapy with or without
a chemoradiotherapy regimen can be considered superior, the
NCCN guidelines do not recommend up-front surgery in this
category of patients when there is a high likelihood of positive
margins and instead recommend neoadjuvant therapy,
preferably at a high-volume center. Consensus opinions for
BRPC have recommended neoadjuvant multimodality treatment mostly based on historical retrospective studies, which
showed increased probability of response (10%30%), resectability (30%60%), and R0 resection (80%90%) with this
approach. 21-26 It is well described that R0 resections
are associated with improved OS rates (1823 months)
when compared with R1 (1415 months) or R2 resections
(1011 months).27-29 Recent retrospective studies with contemporary therapies FOLFIRINOX or nab-paclitaxel with
gemcitabine with or without chemoradiotherapy, often including both patients BRPC or LAUPC, note encouraging results
with both regimens: resectability rates of 60% to 80% for
patients with BRPC and 80% to 90% R0 resections among

those resected.30-40 Patients undergoing surgical resection
had median progression-free survival (PFS) rates of 16 to
23 months and median OS rates of 24 to 30 months, similar to
historical results for patients with up-front resectable disease
(Table 1); in contrast, for patients whose disease remained
unresectable after neoadjuvant treatment, PFS and OS rates
were 9 to 10 and 12 to 15 months, respectively.30,32 Nevertheless, many of the retrospective studies included both patients with localized resectable and BRPC/LAUPC, making it
difficult to identify distinct results for individual subsets of
patients.
The prospective Alliance A021101 study (NCT01821612)
used neoadjuvant mFOLFIRINOX (four cycles) followed by
capecitabine-based chemoradiotherapy for patients with
BRPC.41 Resectable patients underwent surgery followed by
two cycles of adjuvant gemcitabine. Fifteen of the 22 treated
patients (68%) underwent surgical resection, among which
93% were R0 resections. The pathologic complete response
(pCR) and near pCR (, 5% residual tumor cells) were 9% and
TABLE 1. Multimodality Clinical Trials for BRPC and LAUPC With Emphasis on FOLFIRINOX and Nab-Paclitaxel/
Gemcitabine
Study
26
Katz et al
Hosein et al30
No. of BRPC/LAUPC
Treatment
Resected (%)
OS (Overall) Months
OS (Resected) Months
129/0
Gem-platinum + CRT
66
22
33
FOLFIRINOX 6 CRT
75 (BRPC)
Not reported
Not reported
18.4 (BRPC)
25.4
14/4
43 (LAUPC)
Kim et al25
39/6
Gem-oxaliplatin 6 CRT
63
9.4 (LAUPC)
Faris et al51
0/22
31
12/13
Boone et al
FOLFIRINOX + CRT
23
3 year, 7%
Not reached
FOLFIRINOX 6 SBRT
64 (BRPC)
Not reported
Not reported
20 (LAUPC)
Christians36
18/0
FOLFIRINOX 6 CRT
67
Not reported
22 months, 58%
Rose et al24
64/0
Gem-docetaxel 6 CRT
48
23.6
22 months, 81%
Paniccia et al
18/0
FOLFIRINOX 6 CRT
94
25
Not reported
Katz et al41
23/0
FOLFIRINOX + CRT
68
18 months, 50%
Not reported
Blazer et al38
18/25
FOLFIRINOX
61 (BRPC)
21.2
Not reached
Khushman et al50
11/40
FOLFIRINOX 6 CRT
22
35.4
3-year 67%
Ferrone et al39
14/33
FOLFIRINOX 6 CRT
85
34
Not reached
Marthey et al
0/77
FOLFIRINOX 6 CRT
36
22 months, 77%
24.9 months, not reached
Nitsche et al53
0/14
FOLFIRINOX
29
1 year, 64%
31 months, not reached
Nanda et al40
15/29
FOLFIRINOX + CRT
41
1 year, 66%
18.6 months, not reached
FOLFIRINOX 6 CRT
30
26, if no PD on chemo
Not reached
FOLFIRINOX 6 CRT
84
27.4
Not reached
24
53
37
44 (LAUPC)
52
Sadot et al55
32
Idrees et al
0/101
58/28
nab-P/Gem 6 CRT
Portales et al33
16/19
FOLFIRINOX 6 CRT
63 (BRPC)
Kim et al34
26/0
FOLFIRINOX
77
PFS 22.5 months
Not reported
Peterson et al35
14/6
nab-P/Gem
25
Not reported
Not reported
nab-P/Gem 6 CRT
19
Not reported
27.5
37 (LAUPC)
54
Dean et al
0/42
Abbreviations: BRPC, borderline resectable pancreatic cancer; LAUPC, locally advanced unresectable pancreatic cancer; OS, overall survival; CRT, chemoradiotherapy; SBRT,
stereotactic body radiotherapy; PD, progressive disease; chemo, chemotherapy; PFS, progression-free survival; nab-P, nab-paclitaxel; Gem, gemcitabine.
e219
COVELER ET AL
47%, respectively, and the RECIST response after preoperative therapy was 28% (two CR/four PR). Overall survival at 18 months was 50%. Based on the encouraging
resectability results, the prospective randomized phase II
Alliance A021501 study in BRPC using mFOLFIRINOX with or
without hypofractionated radiotherapy, either SBRT or
hypofractionated image-guided radiotherapy (HIGRT), is to
begin shortly. Some reports indicate that near-complete
pathologic complete response (near-pCR, with , 5% residual viable cells) or pCR to neoadjuvant therapy is associated with long-term survival for resected patients (OS:
53 months or longer)32,42,43; however, analysis of larger
cohorts of patients will be needed to verify this observation.
Although it has been assumed that multi-agent chemotherapy will confer high response rates, radiologic RECIST
response after 2 to 3 months of neoadjuvant chemotherapy
for localized disease has not consistently shown meaningful
downsizing.26,41 At this time, lack of downstaging after neoadjuvant chemotherapy with or without chemoradiotherapy
for patients with BRPC should not deter exploration and
attempt at resection in the absence of local or metastatic
progression or other clinical deterioration factors in appropriate surgical candidates.
Locally Advanced Unresectable Pancreatic Cancer

It has been observed in an autopsy series that up to one-third
of patients with pancreatic cancer die due to complications
from local progression and not metastatic disease, suggesting
that more aggressive local therapy may be appropriate in a
subset of patients.44 Locally advanced unresectable pancreatic cancer has traditionally been treated with both chemotherapy and chemoradiotherapy, in various sequencing
approaches, with OS rates of 9 to 13 months and with some
studies (ECOG 4201: gemcitabine-based chemoradiotherapy
followed by gemcitabine versus gemcitabine alone),45 but not
others (FFCD-SFRO: 5-FU/cisplatin-based chemoradiotherapy
followed by gemcitabine vs. gemcitabine alone),46 showing
OS benefit with the addition of up-front chemoradiotherapy
(ECOG 4201: OS 11.1 vs. 9.2 months, p = .017 in favor of the
chemoradiotherapy arm, and FFCD-SFRO: OS of 13 vs.
8.6 months, p = .03 in favor of the chemotherapy arm). In a
meta-analysis of GERCOR studies, Huguet et al47 noted that
induction chemotherapy for 3 to 4 months may select patients
without rapid metastatic progression who may be the most
likely to benefit from additional chemoradiotherapy compared
with continuing chemotherapy alone. Based on these data, the
randomized phase III LAP-07 trial used 4 months of induction
chemotherapy with gemcitabine with or without erlotinib
and, if no progression, randomly assigned patients to continuing chemotherapy (for an additional 2 months) versus
capecitabine-based chemoradiotherapy. The OS rates for
patients able to undergo the randomization after induction
chemotherapy were similar: 16.7 versus 15.3 months (p = .83)
in the chemotherapy arm compared with the chemotherapy
plus chemoradiotherapy arm.48 Nevertheless, local progression
rates were improved with CRT, 34% vs. 65% (p , .0001).49
e220
Similar to the efforts of intensifying multimodality therapy

in localized and BRPC, both FOLFIRINOX and nab-paclitaxel
with gemcitabine, with the addition of chemoradiotherapy
for select patients, have been extensively adopted for patients with LAUPC, to increase the chance of downstaging
and the potential for curative resection. Several institutional
and multi-institutional studies with FOLFIRINOX or nabpaclitaxel with gemcitabine reported radiologic response
rates of 20% to 40%, resectability rates of 20% to 44%, with
R0 resection rates of 40% to 80% among resected patients
(Table 1).30-33,35,38-40,50-55 Notably, patients whose disease
becomes resectable after induction/neoadjuvant chemotherapy with or without chemoradiotherapy have an encouraging median OS of approximately 30 months and
2-year OS of 60% to 70%.38,39,49,50,54,55 A Massachusetts
General Hospital retrospective report compared patients
with initially resectable pancreatic cancer who underwent
surgery (87 patients) with patients who had initially unresectable disease with either BRPC or LAUPC and who were
treated with neoadjuvant FOLFIRINOX with or without
chemoradiotherapy followed by surgery with or without
intraoperative radiation therapy if deemed resectable
(40 patients), and the authors noted that R0 resections were
86% versus 92%, and OS rates were longer in the initially
unresectable group (2-year OS: 5% vs. 70%; p = .008).39
As is the case with retrospective data, most of the institutional studies were heterogenous and included both
patients with BRPC and LAUPC. Prior to the use of novel
multi-agent chemotherapy regimens, a meta-analysis of
111 studies of 4,394 patients with resectable or BRPC/LAUPC
treated with neoadjuvant chemotherapy (with or without
chemoradiotherapy) noted overall responses of 34%, resection rates of 74% in initially resectable disease, and 33%
in BRPC/LAUPC, among which R0 resection rates were approximately 80%.22 Interestingly, OS after resection was
similar in both the initially resectable and unresectable
groups (23.3 months [range 1254 months] versus
20.5 months [range 962 months]). Nevertheless, patients
with disease deemed unresectable after neoadjuvant
therapy had poor outcomes, with OS of 8.4 months (range
614 months) among those initially resectable and
10.2 months (range 621 months) for those with initial
BRPC/LAUPC. A more recent meta-analysis of FOLFIRINOXbased neoadjuvant therapy (with or without chemoradiotherapy)
in BRPC and LAUPC among 253 patients across 13 studies noted
an overall resection rate of 43% (of which 85% were R0 resections) in all patients and 26% among those with LAUPC, but
long-term outcomes are not available.56
Compared with historical chemotherapy regimens that
rendered approximately 20% to 30% of the patients with
locally advanced and borderline pancreatic cancers resectable, but with high recurrence rates and OS rates less than
24 months,22 preliminary evidence from new retrospective
and small prospective studies using the highly active regimens FOLFIRINOX and nab-paclitaxel with gemcitabine
seem to confer 10% to 20% higher resection rates and more
durable PFS and OS outcomes.
In the context of the novel multi-agent chemotherapy

backbones, the role of adding chemoradiotherapy has not
yet been defined. DPC4/SMAD4 has been identified as a
putative biomarker predictive of local progression when
intact or metastatic dissemination when mutated.44 The
RTOG1201 phase II randomized study (NCT01921751) will
prospectively test whether the DPC4 gene status identifies
patients with LAUPC more likely to benefit from chemoradiotherapy. In this study, 346 patients will be randomly
assigned to nab-paclitaxel with gemcitabine alone versus
nab-paclitaxel with gemcitabine plus standard-dose 50 Gy
of capecitabine-based radiotherapy (three-dimensional
conformal or intensity modulated radiotherapy [IMRT])
versus nab-paclitaxel with gemcitabine plus high-dose
60 Gy of capecitabine-based IMRT. In Europe, CONKO-007
(NCT01827553) will assess induction FOLFIRINOX or gemcitabine followed by gemcitabine-based radiotherapy versus continuing chemotherapy, and the SCALOP-2 study
(NCT02024009) will use nab-paclitaxel with gemcitabine
(three cycles) followed by capecitabine-based radiotherapy
with or without the addition of nelfinavir, an HIV protease
inhibitor with radiosensitizing activity, versus nab-paclitaxel
with gemcitabine alone (six cycles). LAPACT (NCT02301143) is
an international randomized phase II study that will evaluate
induction nab-paclitaxel with gemcitabine for six cycles
and, if no progression, allow investigator choice of surgical
resection, chemoradiotherapy, or continued nab-paclitaxel
with gemcitabine.
Choosing the best radiosensitizing agent has been the subject
of many investigations. Gemcitabine has been found to be as
effective but better tolerated than 5-FU.57,58 More recently, the
randomized phase II SCALOP study using induction gemcitabine/
capecitabine followed by capecitabine-based or gemcitabinebased radiotherapy found capecitabine to be more effective
than gemcitabine in prolonging OS (15.2 vs. 13.4 months;
p = .012) and improving tolerability.59
At this time, the NCCN guidelines recommend multi-agent
chemotherapy (at least 23 months and up to 68 months)
with the possibility of adding chemoradiotherapy for select
patients with LAPC who do not progress systemically after a
course of induction chemotherapy. The use of up-front
chemoradiotherapy should be reserved for patients presenting with poorly controlled pain or local invasion with
bleeding (NCCN guidelines).
Novel Therapies in Pancreatic Cancer

Several studies have incorporated immunotherapeutics in
early-stage pancreatic cancer to prevent metastasis and improve survival. IMPRESS (NCT01072981) was a phase III randomized study among 722 patients with resected pancreatic
cancer treated with gemcitabine and fluoropyrimidine-based
radiotherapy with and without algenpantucel-L, a hyperacute
vaccine of two allogenic pancreatic cancer cell lines transduced
with the a-galactosyl transferase gene. Although results are
eagerly awaited (median OS: at least 28.5 in both arms
combined), this study will be the first to show the potential
benefit of incorporating an immunotherapy to standard

therapies to reduce pancreatic cancer recurrence.60 The GVAX
pancreas vaccine, composed of granulocyte macrophage
colony-stimulating factorsecreting pancreatic cancer cell
lines, is currently being studied in combination with SBRT and
FOLFIRINOX for patients with resected pancreatic cancer
(NCT01595321). Given potential synergism, neoadjuvant and
adjuvant GVAX will be studied with and without immune
checkpoint inhibitors and chemoradiotherapy for patients
with resectable disease (NCT02451982), and in LAUPC, GVAX
with an immune checkpoint inhibitor will be added to SBRT
(NCT02648282). Therapies targeting the tumor stroma, including the immune-suppressive compartment, are being
actively investigated for resectable and unresectable localized
pancreatic cancer. For example, tumor-associated macrophages (TAM)targeted therapies such as the CD40 agonist
RO7009789 alone and with nab-paclitaxel with gemcitabine
are being studied as neoadjuvant/adjuvant therapy in resectable pancreatic cancer (NCT02588443). PF-04136309, a
novel chemokine (C-C motif) receptor 2 inhibitor capable of
depleting inflammatory immunosuppressive TAM in combination with FOLFIRINOX, showed encouraging preliminary
results, with 52% radiologic response rates among patients
with BRPC/LAUPC,61 and further studies are anticipated.
Stroma-depleting pegylated recombinant human hyaluronidase, which targets tumor hyaluronan to potentially improve
chemotherapy delivery, is being tested in a neoadjuvant trial
with nab-paclitaxel with gemcitabine for patients with BRPC
(NCT02487277). Several other agents, including PARP inhibitors, are being studied with the goal of improved radiosensitization by preventing DNA repair (NCT01908478) or
among patients with BRCA mutations who may be more
therapeutically responsive to this class of inhibitors.62-65 Two
recent studies of large numbers of pancreatic cancers demonstrating novel molecular subtypes, including an immune
subtype, may help us direct more personalized, effective
regimens to patients most likely to benefit.66,67
LOCALIZED PANCREATIC CANCER: HOW DOES

THE RADIATION ONCOLOGIST CONTRIBUTE?
What Is SBRT?
Radiation therapy has historically been delivered to patients with pancreatic cancer to improve local control,
delay disease progression, and ameliorate locally obstructive
symptoms such as pain. SBRT is a novel technique that allows
precise delivery of radiation to the tumor while minimizing
dose to surrounding normal structures. Recent advances in
image guidance, respiratory motion management, and radiation technology over the past few decades allow for
millimeter accuracy that is necessary for this emerging
modality. Due to the high doses of radiation delivered in
such a short amount of time, SBRT requires close supervision, and advanced measures have been acquired to minimize treatment-related toxicity. Image guidance throughout
the process of SBRT planning and delivery is achieved with
gold fiducial markers (or fiducials) that are placed in or
near the pancreatic tumor under endoscopic guidance.
e221
COVELER ET AL
A four-dimensional CT simulation scan can be used to determine which patients require management of respiratory
motion to safely achieve an ablative dose to the tumor.
Personalization of treatment plans with dose modification
based on the location of the tumor to critical structures such
as the stomach or duodenum may be used to reduce radiotherapy exposure to those tissues. Radiotherapy quality
assurance is also used to maintain consistent setup each day
such that SBRT can be safely and accurately delivered.
What Are the Potential Advantages of SBRT

Compared With Standard Chemoradiotherapy?
SBRT has become a standard treatment modality in several
malignancies such as lung, liver, brain, spine, and prostate
cancers, but it has not yet been established as such in the
management of pancreatic cancer. Nevertheless, SBRT has
been increasingly adopted in institutions worldwide for
patients with pancreatic cancer due to several advantages:
(1) SBRT can be delivered as a hypofractionated regimen
over 3 to 5 days (with dose ranging from 2540 Gy) in
comparison with 25 to 30 days with conventional chemoradiotherapy (dose range, 5054 Gy), (2) SBRT allows for
optimal local control while limiting the delay of alternative
therapies such as systemic full-dose chemotherapy or surgical resection, (3) SBRT results in minimal acute side effects
and has been shown to improve pain while preserving
quality of life,68 and (4) the radiobiology of SBRT along with
the ability to escalate the dose to at least 50 Gy at the tumor
vessel interface may increase the likelihood of a marginnegative resection. Therefore, patients with pancreatic
cancer, especially those with localized disease, have an
increased likelihood of achieving aggressive trimodality
therapy with a hope for prolonged survival. Because SBRT is
well tolerated and only involves a week of treatment, patients are able to focus their energy on achieving personal
treatment goals, whether this includes extending survival by
resuming chemotherapy, undergoing surgical resection, or
maximizing quality of life by returning to daily activities.
SBRT in Localized Pancreatic Cancer

Early studies reporting high rates of late gastrointestinal
toxicity sparked controversy over pancreatic SBRT69-71;
however, the abovementioned methods to prioritize safety
have emerged, and SBRT is now gaining traction as a favorable and highly effective local technique. Recent studies
report acceptable rates of late toxicity in concordance with
minimal acute toxicity.69,72-74 Thus far, there have been
more than 800 published cases of pancreatic cancer treated
with SBRT, and the toxicity, survival, and patterns of failure
data are promising.
Role of Neoadjuvant SBRT in Borderline Resectable

Pancreatic Cancer
A retrospective study included 73 patients with localized
pancreatic cancer (57 boderline resectable and 16 LAUPC) who
received induction chemotherapy followed by five-fraction
e222
SBRT (2530 Gy) at Moffitt Cancer Center was published

in 2013.75 Most patients received induction gemcitabine/
docetaxel/capecitabine (66%) or gemcitabine alone (25%),
whereas only 5% received induction FOLFIRINOX. Among the
patients with BRPC, 56% underwent surgery, with 97%
achieving a margin-negative (R0) resection and 9% pCR.
Median OS among the patients who underwent marginnegative resection was significantly higher than for unresected patients (19.3 vs. 12.3 months; p = .03). Furthermore,
this approach was well tolerated with no acute grade 3 or
worse toxicity reported and only 5.3% late grade 3 or worse
toxicity. More recently, this single-institution series was
updated to include a total of 159 patients (110 with BRPC
and 49 with LAPC).76 Among the patients with BRPC, 51%
underwent surgery with 97% achieving a R0 resection and
7% pCR. Median OS in the patients with resected BRPC was
34 months. Acute grade 3 toxicity or worse was 2%, and late
grade 3 toxicity or worse was 5%. The Johns Hopkins University School of Medicine reported on their study among
88 patients (14 with BRPC and 74 with LAPC) who received
induction chemotherapy followed by SBRT (2533 Gy). The
majority of patients (76%) received gemcitabine-based
chemotherapy as opposed to 5-FU-based (specifically,
FOLFIRINOX) chemotherapy. Of the 19 patients (22%) who
underwent surgery following SBRT, 84% had a R0 resection,
and 16% of patients had pCR.77 Median OS of resected
patients was 20.2 versus 12.3 months in unresected patients
(p = .07). Of note, only 11% of resected patients had BRPC.
Role of SBRT for Locally Advanced, Unresectable

Pancreatic Cancer
Results of SBRT among patients with LAUPC are encouraging. Early phase I/II studies using single-fraction SBRT
(25 Gy in one fraction) demonstrated excellent freedom
from local progression (FFLP) at 1 year (. 90%) and minimal
acute toxicity, but high rates of late grade 2 to 4 gastrointestinal toxicity were reported.70,71,78-80 A single-arm phase
II multi-institutional study evaluated whether gemcitabine
with fractionated SBRT (in five fractions of 6.6 Gy, for a total
33.0 Gy) could achieve reduced late grade 2 to 4 gastrointestinal toxicity compared with a historical cohort of patients
treated with gemcitabine and a single 25 Gy-fraction of
SBRT.68 Forty-nine patients with LAUPC received up to three
doses of gemcitabine (1,000 mg/m2) followed by a 1-week
break and SBRT (33.0 Gy in five fractions). Following SBRT,
patients continued gemcitabine until progression or toxicity.
Rates of acute and late toxicities (primary endpoint) grade
2 or worse gastritis, fistula, enteritis, or ulcers were 2% and
11%, respectively, and QLQ-C30 global quality-of-life scores
remained stable. Patients reported a notable improvement
in pancreatic pain (p , .001) 4 weeks after SBRT on the QLQPAN26 questionnaire. Median OS was 13.9 months (95% CI,
10.216.7), and FFLP at 1 year was 78%. Four patients (8%)
with LAUPC at diagnosis underwent R0 resections and had
node-negative status; a fifth patient who was deemed resectable declined surgery.
With the consideration that patients originally deemed

unresectable may become resectable, a shift toward neoadjuvant therapy in LAUPC has recently emerged. Of the
49 patients with LAUPC who were treated with induction
chemotherapy and SBRT at Moffitt Cancer Center, five (10%)
were resected with a 100% R0 resection rate.76 Patients with
LAUPC survived a median of 13.2 months. Johns Hopkins
reported 15 of 74 (20%) patients with LAUPC who went to
surgery with an 80% R0 resection rate and 13% pCR.77
Median OS was 18.4 months.
How Can We Improve the Impact of SBRT in Localized

Pancreatic Cancer?
In an initial pilot cooperative group study in BRPC (Alliance
A021101), FOLFIRINOX-based multimodality therapy was
shown to be well tolerated, and resectability and preliminary
OS were encouraging.41 The follow-up study (Alliance
A021501) will evaluate survival among patients with BRPC
randomly assigned to neoadjuvant mFOLFIRINOX with or
without hypofractionated radiotherapy, either SBRT or
HIGRT.
As we await results from cooperative group studies, it is
essential to pursue further understanding of the tumor
biology and immune response associated with pancreatic
cancer. It has been shown that radiotherapy not only induces
cell death through DNA damage, but also promotes antitumor reactivity in certain solid tumors.81-83 SBRT specifically, with its highly ablative doses over a short course, has
been shown to be associated with radiation-induced
immunomodulation.84-86 Preclinical data suggest that radiation increases antigen presentation by altering the immunephenotype of tumor infiltrating inflammatory cells, increases
T-cell infiltration, and induces proinflammatory cytokines that
promote tumor cell death.87,88
How Can We Optimally Measure Response to

Therapy, and Which Patients Should Be Taken to
Surgery?
The difficulty of selecting patients with localized pancreatic
cancer who are candidates for surgery following neoadjuvant therapy lies in the inadequacy of radiographic
imaging to depict treatment response and vessel involvement. Two series have published on patients with BRPC who
have been successfully resected despite the lack of tumor
downstaging or tumor response observed on imaging.26,89

The University of Texas MD Anderson Cancer Center reported on 122 patients with BRPC who were restaged after
neoadjuvant therapy, with 85 patients (66%) undergoing
surgery despite only one patient (1%) having been downstaged to have resectable disease.26 Patients received
gemcitabine-based induction chemotherapy followed by
conventional chemoradiotherapy. Similarly, Johns Hopkins
revealed that 29 of 50 patients (58%) with BRPC (25 patients)
and LAPC (four patients) underwent successful surgical resection despite no considerable change in vessel involvement after neoadjuvant therapy.89 Interestingly, three (15%)
patients were noted to have progressive disease on imaging
but were successfully resected. The difficulty in evaluating
tumor response after neoadjuvant therapy is presumably
due to the amount of treatment-related fibrosis that
develops following the delivery of therapy, particularly
chemoradiotherapy.
Given the morbidity and mortality associated with a pancreatectomy, careful patient selection is warranted; however,
as shown above, it is important to consider patients who
appear to otherwise be unresectable after therapy. If the
patient is willing, surgery can be attempted, and is recommended, in the following circumstances following a course of
neoadjuvant therapy of at least 3 months, ideally: (1) absence
of distant progression, (2) stable or improved CA 19-9, and
(3) good performance status (ECOG # 1). Insight into
predictive and prognostic biomarkers and imaging modalities such as PET may allow us to further guide management decisions and practice personalized care.
CONCLUSION
Although the results of ongoing clinical trials are likely to
reshape clinical practice, improvement in systemic and
chemoradiotherapy modalities and the development and
routine use of more predictive blood, tumor, and imaging
biomarkers for the selection of patients most likely to
benefit are areas of high need and intense research. New
understanding of the genetic complexity and intertumoral
heterogeneity among patients and the development of
biomarker-driven clinical trials using novel combinations of
promising therapeutic agents are likely to begin to change
the landscape to improved outcomes in patients with this
dreadful disease.
References
1. Ferlay J, Soerjomataram I, Dikshit R, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN
2012. Int J Cancer. 2015;136:E359-E386.
2015;65:5-29.
3. Bosetti C, Bertuccio P, Malvezzi M, et al. Cancer mortality in Europe,
2005-2009, and an overview of trends since 1980. Ann Oncol. 2013;24:
2657-2671.
4. Tempero MA, Malafa MP, Behrman SW, et al. Pancreatic adenocarcinoma, version 2.2014: featured updates to the NCCN guidelines. J Natl
Compr Canc Netw. 2014;12:1083-1093.
5. Oettle H, Neuhaus P, Hochhaus A, et al. Adjuvant chemotherapy with
gemcitabine and long-term outcomes among patients with resected pancreatic cancer: the CONKO-001 randomized trial. JAMA. 2013;310:1473-1481.
6. Neoptolemos JP, Stocken DD, Friess H, et al; European Study Group
for Pancreatic Cancer. A randomized trial of chemoradiotherapy and
e223
COVELER ET AL
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
chemotherapy after resection of pancreatic cancer. N Engl J Med. 2004;

350:1200-1210.
Regine WF, Winter KA, Abrams R, et al. Fluorouracil-based chemoradiation with either gemcitabine or fluorouracil chemotherapy after
resection of pancreatic adenocarcinoma: 5-year analysis of the U.S.
Intergroup/RTOG 9704 phase III trial. Ann Surg Oncol. 2011;18:
1319-1326.
Liao WC, Chien KL, Lin YL, et al. Adjuvant treatments for resected
pancreatic adenocarcinoma: a systematic review and network metaanalysis. Lancet Oncol. 2013;14:1095-1103.
Van Laethem JL, Hammel P, Mornex F, et al. Adjuvant gemcitabine alone
versus gemcitabine-based chemoradiotherapy after curative resection
for pancreatic cancer: a randomized EORTC-40013-22012/FFCD-9203/
GERCOR phase II study. J Clin Oncol. 2010;28:4450-4456.
Gastrointestinal Tumor Study Group. Further evidence of effective
adjuvant combined radiation and chemotherapy following curative
resection of pancreatic cancer. Cancer. 1987;59:2006-2010.
Klinkenbijl JH, Jeekel J, Sahmoud T, et al. Adjuvant radiotherapy and
5-fluorouracil after curative resection of cancer of the pancreas and
periampullary region: phase III trial of the EORTC gastrointestinal tract
cancer cooperative group. Ann Surg. 1999;230:776-782, discussion 782-784.
Herman JM, Swartz MJ, Hsu CC, et al. Analysis of fluorouracil-based
adjuvant chemotherapy and radiation after pancreaticoduodenectomy
for ductal adenocarcinoma of the pancreas: results of a large, prospectively collected database at the Johns Hopkins Hospital. J Clin Oncol.
2008;26:3503-3510.
You DD, Lee HG, Heo JS, et al. Prognostic factors and adjuvant chemoradiation therapy after pancreaticoduodenectomy for pancreatic
adenocarcinoma. J Gastrointest Surg. 2009;13:1699-1706.
Rhim AD, Mirek ET, Aiello NM, et al. EMT and dissemination precede
pancreatic tumor formation. Cell. 2012;148:349-361.
Evans DB, Varadhachary GR, Crane CH, et al. Preoperative gemcitabinebased chemoradiation for patients with resectable adenocarcinoma of
the pancreatic head. J Clin Oncol. 2008;26:3496-3502.
Varadhachary GR, Wolff RA, Crane CH, et al. Preoperative gemcitabine
and cisplatin followed by gemcitabine-based chemoradiation for resectable adenocarcinoma of the pancreatic head. J Clin Oncol. 2008;26:
3487-3495.
OReilly EM, Perelshteyn A, Jarnagin WR, et al. A single-arm, nonrandomized phase II trial of neoadjuvant gemcitabine and oxaliplatin in
patients with resectable pancreas adenocarcinoma. Ann Surg. 2014;
260:142-148.
Roland CL, Katz MH, Tzeng CW, et al. The addition of postoperative
chemotherapy is associated with improved survival in patients with
pancreatic cancer treated with preoperative therapy. Ann Surg Oncol.
2015;22:1221-1228 (suppl 3).
Heestand GM, Murphy JD, Lowy AM. Approach to patients with pancreatic cancer without detectable metastases. J Clin Oncol. 2015;33:
1770-1778.
Katz MH, Marsh R, Herman JM, et al. Borderline resectable pancreatic
cancer: need for standardization and methods for optimal clinical trial
design. Ann Surg Oncol. 2013;20:2787-2795.
Abrams RA, Lowy AM, OReilly EM, et al. Combined modality treatment
of resectable and borderline resectable pancreas cancer: expert consensus statement. Ann Surg Oncol. 2009;16:1751-1756.
Gillen S, Schuster T, Meyer Zum Buschenfelde C, et al. Preoperative/
neoadjuvant therapy in pancreatic cancer: a systematic review and
meta-analysis of response and resection percentages. PLoS Med. 2010;
7:e1000267.
Katz MH, Pisters PW, Evans DB, et al. Borderline resectable pancreatic
cancer: the importance of this emerging stage of disease. J Am Coll Surg.
2008;206:833-846, discussion 846-848.
e224
24. Rose JB, Rocha FG, Alseidi A, et al. Extended neoadjuvant chemotherapy
for borderline resectable pancreatic cancer demonstrates promising
postoperative outcomes and survival. Ann Surg Oncol. 2014;21:
1530-1537.
25. Kim EJ, Ben-Josef E, Herman JM, et al. A multi-institutional phase 2 study
of neoadjuvant gemcitabine and oxaliplatin with radiation therapy in
patients with pancreatic cancer. Cancer. 2013;119:2692-2700.
26. Katz MH, Fleming JB, Bhosale P, et al. Response of borderline pancreatic
cancer to neoadjuvant therapy is not reflected by radiographic criteria.
Cancer. 2012;118:5749-5756.
27. Fatima J, Schnelldorfer T, Barton J, et al. Pancreatoduodenectomy for
ductal adenocarcinoma: implications of positive margin on survival.
Arch Surg. 2010;145:167-172.
28. Winter JM, Cameron JL, Campbell KA, et al. 1423 pancreaticoduodenectomies for pancreatic cancer: A single-institution experience.
J Gastrointest Surg. 2006;10:1199-1210, discussion 1210-1211.
29. Konstantinidis IT, Warshaw AL, Allen JN, et al. Pancreatic ductal adenocarcinoma: is there a survival difference for R1 resections versus
locally advanced unresectable tumors? What is a true R0 resection?
Ann Surg. 2013;257:731-736.
30. Hosein PJ, Macintyre J, Kawamura C, et al. A retrospective study of
neoadjuvant FOLFIRINOX in unresectable or borderline-resectable locally advanced pancreatic adenocarcinoma. BMC Cancer. 2012;12:199.
31. Boone BA, Steve J, Krasinskas AM, et al. Outcomes with FOLFIRINOX for
borderline resectable and locally unresectable pancreatic cancer. J Surg
Oncol. 2013;108:236-241.
32. Idrees K, Parikh AA, McLendon Postlewait L, et al. Treatment of borderline resectable and locally advanced pancreatic cancer in the era of
FOLFIRINOX and gemcitabine plus nab-paclitaxel: a multi-institutional
study. J Clin Oncol. 2016;34 (suppl 4S; abstr 451).
33. Portales F, Gagniard B, Thezenas S, et al. Feasibility and impact on
resectability of FOLFIRINOX in locally-advanced and borderline pancreatic cancer. J Clin Oncol. 2016;34 (suppl 4S; abstr 318).
34. Kim SS, Nakakura EK, Wang ZJ, et al. Is neoadjuvant chemoradiation
important in borderline resectable pancreatic cancer (BRPC)? Clinical
and surgical outcomes associated with preoperative FOLFIRINOX alone
in BRPC. J Clin Oncol. 2016;34 (suppl 4S; abstr 351).
35. Peterson S, Loaiza-Bonilla A, Ben-Josef E, et al. Neoadjuvant nabpaclitaxel and gemcitabine (AG) in borderline resectable (BR) or
unresectable (UR) locally advanced pancreatic adenocarcinoma
(LAPC) in patients ineligible for FOLFIRINOX. J Clin Oncol. 2016;34
(suppl 4S; abstr 328).
36. Christians KK, Tsai S, Mahmoud A, et al. Neoadjuvant FOLFIRINOX for
borderline resectable pancreas cancer: a new treatment paradigm?
Oncologist. 2014;19:266-274.
37. Paniccia A, Edil BH, Schulick RD, et al. Neoadjuvant FOLFIRINOX application in borderline resectable pancreatic adenocarcinoma: a retrospective cohort study. Medicine (Baltimore). 2014;93:e198.
38. Blazer M, Wu C, Goldberg RM, et al. Neoadjuvant modified FOLFIRINOX
for locally advanced unresectable and borderline resectable adenocarcinoma of the pancreas. Ann Surg Oncol. 2015;22:1153-1159.
39. Ferrone CR, Marchegiani G, Hong TS, et al. Radiological and surgical
implications of neoadjuvant treatment with FOLFIRINOX for locally
advanced and borderline resectable pancreatic cancer. Ann Surg. 2015;
261:12-17.
40. Nanda RH, El-Rayes B, Maithel SK, et al. Neoadjuvant modified
FOLFIRINOX and chemoradiation therapy for locally advanced pancreatic
cancer improves resectability. J Surg Oncol. 2015;111:1028-1034.
41. Katz MHG, Shi Q, Ahmad SA, et al. Preoperative modified FOLFIRINOX
followed by chemoradiation for borderline resectable pancreatic
cancer: initial results from Alliance trial A021101. J Clin Oncol. 2015;33
42. Chatterjee D, Katz MH, Rashid A, et al. Histologic grading of the extent of
residual carcinoma following neoadjuvant chemoradiation in pancreatic ductal adenocarcinoma: a predictor for patient outcome. Cancer.
2012;118:3182-3190.
43. Zhao Q, Rashid A, Gong Y, et al. Pathologic complete response to
neoadjuvant therapy in patients with pancreatic ductal adenocarcinoma is associated with a better prognosis. Ann Diagn Pathol. 2012;16:
29-37.
44. Iacobuzio-Donahue CA, Fu B, Yachida S, et al. DPC4 gene status of the
primary carcinoma correlates with patterns of failure in patients with
pancreatic cancer. J Clin Oncol. 2009;27:1806-1813.
45. Loehrer PJ Sr, Feng Y, Cardenes H, et al. Gemcitabine alone versus
gemcitabine plus radiotherapy in patients with locally advanced pancreatic cancer: an Eastern Cooperative Oncology Group trial. J Clin
Oncol. 2011;29:4105-4112.
46. Chauffert B, Mornex F, Bonnetain F, et al. Phase III trial comparing
intensive induction chemoradiotherapy (60 Gy, infusional 5-FU and
intermittent cisplatin) followed by maintenance gemcitabine with
gemcitabine alone for locally advanced unresectable pancreatic cancer.
Definitive results of the 2000-01 FFCD/SFRO study. Ann Oncol. 2008;19:
1592-1599.
47. Huguet F, Andre T, Hammel P, et al. Impact of chemoradiotherapy after
disease control with chemotherapy in locally advanced pancreatic
adenocarcinoma in GERCOR phase II and III studies. J Clin Oncol. 2007;
25:326-331.
48. Hammel P, Huguet F, van Laethem JL, et al. Comparison of chemoradiotherapy (CRT) and chemotherapy (CT) in patients with locally
advanced pancreatic cancer (LAPC) controlled after 4 months of
gemcitabine with or without erlotinib: final results of the international
phase III LAP 07 study. J Clin Oncol. 2013;33:15s (suppl; abstr LBA4003).
49. Huguet F, Hammel P, Vernerey D, et al. Impact of chemoradiation on
local control and time without treatment in patients with locally advanced pancreatic cancer included in the international phase III LAP 07
study. J Clin Oncol. 2014;32 (suppl 5S; abstr 4001).
50. Khushman M, Dempsey N, Maldonado JC, et al. Full dose neoadjuvant
FOLFIRINOX is associated with prolonged survival in patients with locally advanced pancreatic adenocarcinoma. Pancreatology. 2015;15:
667-673.
51. Faris JE, Blaszkowsky LS, McDermott S, et al. FOLFIRINOX in locally
advanced pancreatic cancer: the Massachusetts General Hospital
Cancer Center experience. Oncologist. 2013;18:543-548.
52. Marthey L, Sa-Cunha A, Blanc JF, et al. FOLFIRINOX for locally advanced
pancreatic adenocarcinoma: results of an AGEO multicenter prospective observational cohort. Ann Surg Oncol. 2015;22:295-301.
53. Nitsche U, Wenzel P, Siveke JT, et al. Resectability after first-line
FOLFIRINOX in initially unresectable locally advanced pancreatic cancer: a single-center experience. Ann Surg Oncol. 2015;22:1212-1220
(suppl 3).
54. Dean A, McGrath A, Youd J, Spry N. Nab-paclitaxel plus gemcitabine
followed by radiotherapy with concurrent 5-FU in locally advanced
unresectable pancreatic cancer. J Clin Oncol. 2016;34 (suppl 4S; abstr
430).
55. Sadot E, Doussot A, OReilly EM, et al. FOLFIRINOX induction therapy for
stage 3 pancreatic adenocarcinoma. Ann Surg Oncol. 2015;22:
3512-3521.
56. Petrelli F, Coinu A, Borgonovo K, et al; Gruppo Italiano per lo Studio dei
Carcinomi dellApparato Digerente (GISCAD). FOLFIRINOX-based neoadjuvant therapy in borderline resectable or unresectable pancreatic
cancer: a meta-analytical review of published studies. Pancreas. 2015;
44:515-521.
57. Crane CH, Abbruzzese JL, Evans DB, et al. Is the therapeutic index better
with gemcitabine-based chemoradiation than with 5-fluorouracil-based
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
chemoradiation in locally advanced pancreatic cancer? Int J Radiat

Oncol Biol Phys. 2002;52:1293-1302.
Huang J, Robertson JM, Margolis J, et al. Long-term results of full-dose
gemcitabine with radiation therapy compared to 5-fluorouracil with
radiation therapy for locally advanced pancreas cancer. Radiother
Oncol. 2011;99:114-119.
Mukherjee S, Hurt CN, Bridgewater J, et al. Gemcitabine-based or
capecitabine-based chemoradiotherapy for locally advanced pancreatic
cancer (SCALOP): a multicentre, randomised, phase 2 trial. Lancet
Oncol. 2013;14:317-326.
Coveler AL, Rossi GR, Vahanian NN, et al. Algenpantucel-L immunotherapy in pancreatic adenocarcinoma. Immunotherapy. 2016;8:
117-125.
Wang-Gillam A, Nyweing TM, Sanford DE, et al. Phase IB study of
FOLFIRINOX plus PF-04136309 in patients with borderline resectable
and locally advanced pancreatic adenocarcinoma. J Clin Oncol. 2015;33
(suppl 3; abstr 338).
Salo-Mullen EE, OReilly EM, Kelsen DP, et al. Identification of germline
genetic mutations in patients with pancreatic cancer. Cancer. 2015;121:
4382-4388.
Lowery MA, Kelsen DP, Smith SC, et al. Phase II trial of veliparib in
patients with previously treated BRCA or PALB2 mutated pancreas
adenocarcinoma. J Clin Oncol. 2015;33 (suppl 3; abstr 358).
Pishvaian MJ, Wang H, Zhuang T, et al. A phase I/II study of ABT-888 in
combination with 5-fluorouracil and oxaliplatin in patients with metastatic pancreatic cancer. J Clin Oncol. 2013;31 (suppl 4; abstr 147).
OReilly EM, Lowery MA, Segal MF, et al. Phase Ib trial of cisplatin,
gemcitabine, and veliparib in patients with known or potential BRCA or
PALB2-mutated pancreas adenocarcinoma. J Clin Oncol. 2014;32 (suppl
5S; abstr 4023).
Waddell N, Pajic M, Patch AM, et al; Australian Pancreatic Cancer
Genome Initiative. Whole genomes redefine the mutational landscape
of pancreatic cancer. Nature. 2015;518:495-501.
Bailey P, Chang DK, Nones K, et al; Australian Pancreatic Cancer Genome
Initiative. Genomic analyses identify molecular subtypes of pancreatic
cancer. Nature. 2016;531:47-52.
Herman JM, Chang DT, Goodman KA, et al. Phase 2 multi-institutional
trial evaluating gemcitabine and stereotactic body radiotherapy for
patients with locally advanced unresectable pancreatic adenocarcinoma. Cancer. 2015;121:1128-1137.
Hoyer M, Roed H, Sengelov L, et al. Phase-II study on stereotactic
radiotherapy of locally advanced pancreatic carcinoma. Radiother
Oncol. 2005;76:48-53.
Schellenberg D, Goodman KA, Lee F, et al. Gemcitabine chemotherapy
and single-fraction stereotactic body radiotherapy for locally advanced
pancreatic cancer. Int J Radiat Oncol Biol Phys. 2008;72:678-686.
Schellenberg D, Kim J, Christman-Skieller C, et al. Single-fraction stereotactic body radiation therapy and sequential gemcitabine for the
treatment of locally advanced pancreatic cancer. Int J Radiat Oncol Biol
Phys. 2011;81:181-188.
Tozzi A, Comito T, Alongi F, et al. SBRT in unresectable advanced
pancreatic cancer: preliminary results of a mono-institutional experience. Radiat Oncol. 2013;8:148.
Mahadevan A, Jain S, Goldstein M, et al. Stereotactic body radiotherapy
and gemcitabine for locally advanced pancreatic cancer. Int J Radiat
Oncol Biol Phys. 2010;78:735-742.
Timmerman RD, Kavanagh BD, Cho LC, et al. Stereotactic body radiation
therapy in multiple organ sites. J Clin Oncol. 2007;25:947-952.
Chuong MD, Springett GM, Freilich JM, et al. Stereotactic body radiation
therapy for locally advanced and borderline resectable pancreatic
cancer is effective and well tolerated. Int J Radiat Oncol Biol Phys. 2013;
86:516-522.
e225
COVELER ET AL
76. Mellon EA, Hoffe SE, Springett GM, et al. Long-term outcomes of
induction chemotherapy and neoadjuvant stereotactic body radiotherapy for borderline resectable and locally advanced pancreatic
adenocarcinoma. Acta Oncol. 2015;54:979-985.
77. Moningi S, Dholakia AS, Raman SP, et al. The role of stereotactic body
radiation therapy for pancreatic cancer: A single-institution experience.
Ann Surg Oncol. 2015;22:2352-2358.
78. Koong AC, Le QT, Ho A, et al. Phase I study of stereotactic radiosurgery in
patients with locally advanced pancreatic cancer. Int J Radiat Oncol Biol
Phys. 2004;58:1017-1021.
79. Koong AC, Christofferson E, Le QT, et al. Phase II study to assess the
efficacy of conventionally fractionated radiotherapy followed by a
stereotactic radiosurgery boost in patients with locally advanced
pancreatic cancer. Int J Radiat Oncol Biol Phys. 2005;63:320-323.
80. Chang DT, Schellenberg D, Shen J, et al. Stereotactic radiotherapy for
unresectable adenocarcinoma of the pancreas. Cancer. 2009;115:665-672.
81. Reits EA, Hodge JW, Herberts CA, et al. Radiation modulates the peptide
repertoire, enhances MHC class I expression, and induces successful
antitumor immunotherapy. J Exp Med. 2006;203:1259-1271.
82. Gameiro SR, Jammeh ML, Wattenberg MM, et al. Radiation-induced
immunogenic modulation of tumor enhances antigen processing and
calreticulin exposure, resulting in enhanced T-cell killing. Oncotarget.
2014;5:403-416.
e226
83. Gupta A, Probst HC, Vuong V, et al. Radiotherapy promotes tumorspecific effector CD8+ T cells via dendritic cell activation. J Immunol.
2012;189:558-566.
84. Zeng J, Harris TJ, Lim M, Drake CG, Tran PT. Immune modulation and
stereotactic radiation: Improving local and abscopal responses. Biomed
Res Int. 2013;2013:658126.
85. Neefjes J, Jongsma ML, Paul P, et al. Towards a systems understanding
of MHC class I and MHC class II antigen presentation. Nat Rev Immunol.
2011;11:823-836.
86. Finkelstein SE, Timmerman R, McBride WH, et al. The confluence of
stereotactic ablative radiotherapy and tumor immunology. Clin Dev
Immunol. 2011;2011:439752.
87. Sharabi AB, Nirschl CJ, Kochel CM, et al. Stereotactic radiation therapy
augments antigen-specific PD-1-mediated antitumor immune responses via cross-presentation of tumor antigen. Cancer Immunol
Res. 2015;3:345-355.
88. Matsumura S, Demaria S. Up-regulation of the pro-inflammatory
chemokine CXCL16 is a common response of tumor cells to ionizing
radiation. Radiat Res. 2010;173:418-425.
89. Dholakia AS, Hacker-Prietz A, Wild AT, et al. Resection of borderline
resectable pancreatic cancer after neoadjuvant chemoradiation does
not depend on improved radiographic appearance of tumor-vessel
relationships. J Radiat Oncol. 2013;2:413-425.
Elderly Patients With Bladder

Cancer: Perspectives From a
Surgeon, Medical, and Radiation
Oncologist
CHAIR
Supriya Gupta Mohile, MD, MS
University of Rochester Medical Center
Rochester, NY
SPEAKERS
Eila C. Skinner, MD
Stanford, CA
Nicholas D. James, BSc, MBBS, PhD
University of Warwick and Queen Elizabeth Hospital
Coventry, United Kingdom
EILA C. SKINNER
Treatment of Muscle-Invasive Bladder Cancer in Older

Patients
Eila C. Skinner, MD
OVERVIEW
Treatment of muscle-invasive bladder cancer in older patients is challenging. Definitive therapy of localized disease requires
either surgery or radiation therapy, ideally combined with systemic chemotherapy. However, current population data
suggest that less than half of patients older than age 70 are offered such treatments. We will review tools available to assess
the fitness of older patients for surgery, alternatives, and tips for perioperative patient treatment.
ladder cancer is a disease of older patients, with increasing incidence with age, and rarely seen before age
40. The majority of patients present with noninvasive tumors that have a high recurrence rate but rarely develop
metastases. These are managed with transurethral resection
and intravesical therapy, both of which are generally welltolerated in older patients. However, 20% to 30% of patients
will present with an invasive, high-grade tumor that has a
high potential for metastasis. In general, muscle-invasive
tumors that are not treated will cause increasingly severe
local symptoms and, in most cases, death from metastatic
disease within 2 years. In recent years, urologists and radiation oncologists have been called on to treat an increasing
number of patients older than age 75 with this type of
aggressive disease. Definitive therapy requires either radical
cystectomy or radiation therapy, with or without systemic
chemotherapy. However, current reports indicate that only
65% of patients age 71 to 80 and 35% of patients age 81 to 90
receive these treatments.1 A multidisciplinary approach is
required to identify older patients who should be offered
such treatments.
PATIENT EVALUATION
Initial evaluation of the patient with high-risk bladder cancer
requires careful evaluation of both the tumor and the patient. Tumor evaluation is identical regardless of age, and
it includes review of prior treatments, complete tumor
staging, review of pathology slides, and possible repeat
resection. Repeat resection with examination under anesthesia has been shown to improve accuracy of local staging.2
A full radiographic staging workup of a muscle-invasive
cancer should include a CT chest/abdomen/pelvis with
contrast and bone scan, or PET/CT scan. Although there are
mixed reports regarding the advantage of PET/CT over

standard CT with contrast, PET may improve detection of
occult metastases, which might be particularly useful in
older patients in whom the risk of surgery may not be
warranted if cure is unlikely.3
Patient evaluation in older patients requires some assessment of fitness for surgery and a relative estimate of
complications related to the operation. There has been
considerable progress in this regard over the past few years
with the development of tools for assessment of older
patients. Ideally, the patient may be evaluated by a trained
geriatrician to perform a comprehensive geriatric assessment.4 However, such specialists may not be readily available and the full assessment is not practical for the average
urologic oncology clinic.
A number of health domains have been shown to contribute to overall frailty in older patients. In addition to
chronological age, these include cognitive function, dependence in activities of daily living, nutritional status,
physical fitness, medical comorbidities, social support,
mood, and history of falls. Several authors have combined
factors that were predictive of morbidity and mortality from
surgery into simplified tests that could be applied in a
urology clinic with minimal extra training.5-14 Some of these
assessments focus on medical comorbidities, functional
status, cognition, and laboratory tests such as serum albumin. These assessments also have been used in retrospective studies and applied to large databases such as the
National Surgery Quality Improvement Program.10,11,14
Others use actual tests of physical strength and mobility
(such as the get up and go test and grip-strength measure),
cognitive testing, and patient-reported signs such as weight
loss. These measures require a trained clinician to directly
From Stanford University School of Medicine, Stanford, CA.

Corresponding author: Eila C. Skinner, MD, Stanford University School of Medicine, 300 Pasteur Dr., Suite S-287, Stanford, CA 94305; email: skinnere@stanford.edu.
e228
TREATMENT OF MUSCLE-INVASIVE BLADDER CANCER IN OLDER PATIENTS
evaluate the patient prospectively.9,13 Regardless of the

particular measures used, most of these assessments have
correlated with surgical outcomes and have improved
predictive power compared with the surgeons intuition
after seeing the patient seated on the examination table.
Table 1 shows an example of one such tool that is relatively
straightforward to administer.9
Patients who score poorly on these tests have been shown
to have higher complications and higher mortality from
major abdominal surgery and, even in some cases, in lessinvasive procedures such as transurethral resection of the
prostate or sling.11 The results from these evaluations must
be interpreted in light of the severity of patient symptoms
and the likelihood that a nonoperative approach will be
successful. In addition, the degree to which these factors are
modifiable in a reasonably short time is unclear. However,
the knowledge about risks of surgery can help patients and
families decide on treatment options.
CHOICE OF TREATMENT
Three options should be discussed with every older patient
who presents with invasive localized bladder cancer: (1)
observation with symptom management alone; (2) radiation
therapy, with or without concomitant chemotherapy; and
(3) radical cystectomy, with or without chemotherapy
Partial cystectomy is appealing in this age group because it
has so much less morbidity, but it is only effective in rare
cases of an isolated tumor near the dome of the bladder and
is often associated with local recurrence. Similarly, transurethral resection of the bladder tumor (TURBT) alone has
been associated with at least 50% local recurrence and
progression, even in highly selected patients.15 Only 5% to
KEY POINTS
Less than 50% of patients older than age 70 are currently

receiving definitive therapy for localized muscleinvasive bladder cancer.
Mortality from radical cystectomy is approximately 10%
in older patients by 90 days.
A number of combined measures of frailty have been
shown to be predictive of morbidity and mortality of
major surgeries in older patients, including radical
cystectomy.
Combination treatment with chemotherapy plus
radiation therapy has been shown to be effective in
older patients, with similar overall survival to most
cystectomy series. However, toxicity may be significant,
and 10% to 20% of patients will require delayed
cystectomy because of invasive local recurrence or
complications from radiation.
A multidisciplinary approach with surgeons, medical
oncologists, and radiation oncologists can optimize
outcomes in this group of patients.
10% of patients with reasonable life expectancy are candidates for either of these approaches.
Observation
The primary advantage of observation is avoiding the potential risks and side effects associated with treatment, with
the implicit hope/expectation that the patient will die of
another illness before the bladder cancer causes morbidity.
Local symptoms may include bleeding, frequency and urgency,
incontinence, bladder pain, outlet obstruction, and uppertract obstruction. Symptom management may include repeat
TURBT or fulguration for bleeding, anticholinergics, catheter
drainage, and stents or percutaneous nephrostomy tube
placement. Unfortunately, once local symptoms become
problematic they often are refractory to these treatments
and patients may require multiple trips to the emergency
department, hospital admissions, transfusions, and painful
procedures. It is not uncommon for the 88-year-old patient
who is not thought to be a surgical candidate to end up
requiring a palliative cystectomy a year later when he or she
has metastatic disease. Thus, a patient who is believed to
have a 2-year life expectancy and is likely to survive the
surgery is often best treated with definitive therapy with
cystectomy or radiation.
Radiation Therapy
The advantage of radiation therapy is the avoidance of the
morbidity of surgery and the requirement for a urostomy.
Radiation therapy is most effective when combined with
sensitizing chemotherapy.16 The two most common regimens are either weekly cisplatin or 5-fluorouracil and mitomycin C for patients who are ineligible for cisplatin.16 The
two regimens have not been compared directly, but they
appear to have similar outcomes. There are also no randomized studies directly comparing cystectomy and chemoradiation, and retrospective comparisons are fraught
with selection bias. Nevertheless, overall survival rates appear similar to cystectomy series. Age alone does not appear
to affect the success of chemoradiation.17
Initial response to induction chemoradiation is approximately 70% in most patients.16-19 Local recurrence is observed
in 30% to 40% with approximately 10% to 30% of patients
requiring a subsequent cystectomy for persistent or recurrent
invasive disease. Noninvasive cancers also are common
during follow-up, emphasizing the need for continued cystoscopic surveillance. Endoscopic evaluation of the bladder
after radiation therapy can be difficult because of mucosal
ulceration and hyperemia that may require biopsy to differentiate from recurrent cancer, especially carcinoma in situ.
Short-term bladder and bowel side effects are common
during radiation treatments, with grade 3 to 4 genitourinary toxicity ranging from 2% to 20%, and gastrointestinal toxicity approximately 10%. Approximately 10%
to 20% of patients also will have grade 3 to 4 hematologic
complications from the chemotherapy. Long-term grade 1
to 2 genitourinary side effects are observed in 10% to 25%
e229
EILA C. SKINNER
TABLE 1. The Edmonton Frail Scale9

Domain
Item
0 Points
1 Point
2 Points
Cognition
Imagine that this pre-drawn circle is a clock.

I would like you to place the numbers in
correct position then place the hands to
indicate a time of 10 after 11.
No errors
Minor spacing errors
Other errors
General Health Status
In the past year how many times have you

been admitted to a hospital?
12
.2
In general how would you describe your

health?
Excellent
Fair
Poor
Very good
Good
Functional Independence
With how many of the following activities

do you require help (meal preparation,
transportation, telephone, housekeeping, laundry, managing money,
taking medication)?
01
24
58
Social Support
When you need help, can you count on

someone who is willing and able to
meet your needs?
Always
Sometimes
Never
Medications
Do you use five or more different

prescription medications on a regular
basis?
No
Yes
At times do you forget to take your

prescription medications?
No
Yes
Nutrition
Have you recently lost weight such that

your clothing has become looser?
No
Yes
Mood
Do you often feel sad or depressed?
No
Yes
Continence
Do you have a problem with losing control

of urine when you dont want to?
No
Yes
Functional Performance
I would like you to sit in this chair with

your back and arms resting. When I say
Go please get up and walk at a safe
and comfortable pace to the mark on
the floor (approximately 3 m) and then
return to the chair and sit down.
010 seconds
1120 seconds
. 20 seconds,
unable or
requires
assistance
Total Score (of 17)
of patients, with grade 3 to 4 symptoms in 3% to 8%.

Approximately 2% to 3% require cystectomy for severe
local symptoms.20
Radiation therapy is most effective in patients with lowervolume disease. Clinical cT2 (versus cT3), visibly complete
transurethral resection, and absence of hydronephrosis and
carcinoma in situ have been shown to be predictors of
better initial response.17,20 A number of radiation series in
the literature have excluded patients with these high-risk
characteristics.
Shipley and colleagues popularized endoscopic re-evaluation
of patients undergoing chemoradiation after approximately
4 weeks of treatment (40 Gy).17 At that point, salvage cystectomy has minimally increased morbidity from the radiation
effects. This may be a reasonable approach in an older patient
who is still a good surgical candidate. This approach has not been
directly compared with a more standard practice of completing
all therapy and offering cystectomy for persistent or recurrent
disease. If a patient truly is not a surgical candidate, then there is
no rationale for an intermediate assessment and the follow-up
e230
schedule also may be less intense. Salvage cystectomy after

chemoradiation is feasible but has increased complications
compared with nonirradiated patients, and neobladder reconstruction is only possible in selected patients.21
Radical Cystectomy
The advantages of cystectomy include management of local
symptoms and the potential for cure. The latter depends on
the pathologic stage of the tumor. For tumors that are organconfined on final pathology, the overall cure rate is above 70%
at 5 years. On the other hand, tumors that are already
metastatic to pelvic lymph nodes are only cured about 30%
of the time with surgery, even with the addition of adjuvant
chemotherapy.22 Neoadjuvant chemotherapy has been
shown to result in a 5% absolute improvement in overall
survival in patients undergoing cystectomy, and it is particularly helpful in patients with hydronephrosis, lymphovascular
invasion, a palpable mass under anesthesia, or invasion of the
prostate or uterus (cT3 or T4 disease). However, extravesical
extension is encountered on final pathology in 50% of patients

even in the absence of any preoperative risk factors.23
Fitness of the patient to receive a cisplatin-based chemotherapy regimen should be assessed by a medical oncologist
because those regimens have the highest effectiveness.
However, up to 60% of patients older than age 70 may have an
estimated glomerular filtration rate of less than 60, which may
make them ineligible for cisplatin therapy.24 Other challenges
of delivery of effective systemic therapy include cardiac disease, hearing loss, neuropathy, and cognitive dysfunction.25
Expert geriatric assessment and treatment by oncologists experienced with treating older patients are both helpful. Patients of all ages tolerate chemotherapy better before surgery
than afterward. In the patient who is not able to receive a
standard cisplatin combination regimen, the decision between
surgery alone versus neoadjuvant chemotherapy with an alternate regimen, such as split-dose cisplatin, gemcitabine plus
carboplatin, or a clinical trial, must be made on an individual
basis with consultation among the patient, oncologist, and
surgeon.
The morbidity of cystectomy and diversion is substantial,
with approximately 60% of patients suffering some type of
complication and about 15% to 20% suffering high-grade
complications.26-28 Overall mortality rate is about 1% to 2%,
but, in patients older than age 75, it is approximately 10% by
90 days following surgery.7,27-30 The most common major
complications include cardiovascular events, sepsis, and bowel
complications. The use of minimally invasive techniques generally has not improved outcomes except for decreased transfusion requirements.31,32 Full recovery can take several months,
especially for older patients. A major concern is the potential loss
of capacity for independent living, both as a result of the basic
surgical recovery and difficulty in managing the urinary diversion.
Perioperative management. Once a patient and his or her
family have decided to pursue surgery, optimizing physiologic function before surgery is critical to minimizing risk.
These might include the following:
1. Nutritional assessment with improvement of malnutrition if possible with dietary supplementation
2. Cardiovascular and/or pulmonary evaluation to rule
out reversible ischemia or chronic obstructive pulmonary disease
3. Optimization of other medical conditions such as diabetes and renal dysfunction (e.g., with drainage of an
obstructed kidney)
4. Smoking cessation at least 8 weeks before surgery
(ideally). There is no clear benefit and there may be an
increased risk of quitting for less than 8 weeks.33
Specific conditions that significantly increase risk of cystectomy
in older patients include cirrhosis, malnutrition, morbid obesity,
and chronic immunosuppression. None of these conditions is
easily reversed in the short period of time required for cancer
treatment. However, patients and their families must be aware
of the increased risk and, together with their physicians, should
take this into consideration when making decisions about
treatment.
Because the average age of patients undergoing cystectomy

today is approximately age 70, most centers performing a
large number of cystectomies have experience with treatment of older patients. Careful fluid management is critical in
this age group, especially in patients with cardiac disease.
Postoperative delirium is also more common in older patients
and avoidance of narcotics may be helpful.34 A low threshold
for diagnosis of infections is necessary because fever and
leukocytosis may be suppressed in this age group. Early
ambulation and aggressive physical therapy is helpful in
avoiding progressive weakness from inactivity, and often the
physical therapy should be continued at home.
Perioperative management strategies recently have focused on a multimodality approach called the ERAS (Enhanced Recovery After Surgery) system. This was first
described in the colorectal literature and it has been adapted
to patients undergoing cystectomy.35 The ERAS system includes aspects of preoperative, intraoperative, and postoperative management. The addition of an antagonist to
the m morphine receptor in the gut (alvimopam) has been a
critical element of this system, and it has been shown in a
randomized cystectomy trial to decrease hospital stay by at
least 1 day and markedly decrease ileus and the need for
nasogastric suction.36 Daneshmand and colleagues reported
on 110 patients treated with a modified ERAS protocol and
observed a median hospital stay of 4 days without an increase in readmission.37 We have observed a similar improvement in average postoperative hospital stay at
Stanford from a median of 8 to 5 days with adoption of these
techniques.
Discharge management is also critical to the patients full
recovery from such major surgery. Patients of all ages who
have undergone cystectomy require assistance with basic
activities of daily living for at least 2 to 3 weeks after surgery.
Many older patients live with an older or frail spouse who is
unable to provide such support, or live alone without family
or other resources for such assistance. They may need to be
discharged to a convalescent facility. Discharge planning
should begin before surgery with a frank discussion of the
help required and resources available to the patient.
OUTCOMES
Major complications and death following radical cystectomy
are more common in patients older than age 70 than in
younger patients. A number of recent reports identify 90-day
mortality after cystectomy to be in the range of 10% to as high
as 32% for patients over age 85.7,27,28 Nevertheless, most older
patients recover after cystectomy, and the primary risk of
death in the first 2 to 3 years after surgery is metastatic disease
rather than competing causes of mortality. Continent diversion
has been performed successfully in selected older patients
who appear to have outcomes similar to younger patients.
Regaining continence may take longer in older patients who
undergo neobladder reconstruction. Although overall survival,
of course, is worse in older patients, there is controversy over
whether cancer-specific survival is worse in this age group.38,39
e231
EILA C. SKINNER
CONCLUSION
Older patients with muscle-invasive bladder cancer
represent a particularly challenging group of patients to
treat. Patient evaluation and risk assessment are critical in
helping patients and families decide about treatment options and the advantage or disadvantage of aggressive
treatment. Careful geriatric assessment and a multidisciplinary approach can optimize patient outcomes.
References
1. Gray PJ, Fedewa SA, Shipley WU, et al. Use of potentially curative
therapies for muscle-invasive bladder cancer in the United States:
results from the National Cancer Data Base. Eur Urol. 2013;63:823-829.
2. Herr HW. Role of repeat resection in non-muscle-invasive bladder
cancer. J Natl Compr Canc Netw. 2015;13:1041-1046.
3. Rais-Bahrami S, Pietryga JA, Nix JW. Contemporary role of advanced
imaging for bladder cancer staging. Urol Oncol. 2016;34:124-133.
4. Hurria A, Cirrincione CT, Muss HB, et al. Implementing a geriatric assessment in cooperative group clinical cancer trials: CALGB 360401.
J Clin Oncol. 2011;29:1290-1296.
5. Kim SW, Han HS, Jung HW, et al. Multidimensional frailty score for the
prediction of postoperative mortality risk. JAMA Surg. 2014;149:
633-640.
6. Revenig LM, Canter DJ, Kim S, et al. Report of a simplified frailty score
predictive of short-term postoperative morbidity and mortality. J Am
Coll Surg. 2015;220:904-911.e1.
7. Morgan TM, Keegan KA, Barocas DA, et al. Predicting the probability of
90-day survival of elderly patients with bladder cancer treated with
radical cystectomy. J Urol. 2011;186:829-834.
8. Robinson TN, Wu DS, Pointer L, et al. Simple frailty score predicts
complications across surgical specialties. Am J Surg. 2013;206:544-50.
9. Rolfson DB, Majumdar SR, Tsuyuki RT, et al. Validity and reliability of the
Edmonton Frail Scale. Age Ageing. 2006;35:526-529.
10. Lascano D, Pak JS, Kates M, et al. Validation of a frailty index in patients
undergoing curative surgery for urologic malignancy and comparison
with other risk stratification tools. Urol Oncol. 2015;33:426.e1-426.
e.12.
11. Suskind AM, Walter LC, Jin C, et al. Impact of frailty on complications in
patients undergoing common urological procedures: a study from the
American College of Surgeons National Surgical Quality Improvement
database. BJU Int. Epub. 2016 Jan 17.
12. Grubmueller B, Seitz C, Shariat SF. The treatment of muscle-invasive
bladder cancer in geriatric patients. Curr Opin Urol. 2016;26:160-164.
13. Fried LP, Tangen CM, Walston J, et al; Cardiovascular Health Study
Collaborative Research Group. Frailty in older adults: evidence for a
phenotype. J Gerontol A Biol Sci Med Sci. 2001;56:M146-M157.
14. Amrock LG, Neuman MD, Lin HM, et al. Can routine preoperative data
predict adverse outcomes in the elderly? Development and validation
of a simple risk model incorporating a chart-derived frailty score. J Am
Coll Surg. 2014;219:684-694.
15. Cha EK, Donahue TF, Bochner BH. Radical transurethral resection alone,
robotic or partial cystectomy, or extended lymphadenectomy: can we
select patients with muscle invasion for less or more surgery? Urol Clin
North Am. 2015;42:189-199, viii.
16. James ND, Hussain SA, Hall E, et al; BC2001 Investigators. Radiotherapy
with or without chemotherapy in muscle-invasive bladder cancer.
N Engl J Med. 2012;366:1477-1488.
17. Efstathiou JA, Spiegel DY, Shipley WU, et al. Long-term outcomes of
selective bladder preservation by combined-modality therapy for invasive bladder cancer: the MGH experience. Eur Urol. 2012;61:705-711.
18. Mak RH, Hunt D, Shipley WU, et al. Long-term outcomes in patients with
muscle-invasive bladder cancer after selective bladder-preserving
e232
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
combined-modality therapy: a pooled analysis of Radiation Therapy

Oncology Group protocols 8802, 8903, 9506, 9706, 9906, and 0233.
J Clin Oncol. 2014;32:3801-3809.
Rodel C, Grabenbauer GG, Kuhn R, et al. Combined-modality treatment
and selective organ preservation in invasive bladder cancer: long-term
results. J Clin Oncol. 2002;20:3061-3071.
Ploussard G, Daneshmand S, Efstathiou JA, et al. Critical analysis of
bladder sparing with trimodal therapy in muscle-invasive bladder
cancer: a systematic review. Eur Urol. 2014;66:120-137.
Eisenberg MS, Dorin RP, Bartsch G, et al. Early complications of
cystectomy after high dose pelvic radiation. J Urol. 2010;184:
2264-2269.
Stein JP, Lieskovsky G, Cote R, et al. Radical cystectomy in the treatment
of invasive bladder cancer: long-term results in 1,054 patients. J Clin
Oncol. 2001;19:666-675.
Culp SH, Dickstein RJ, Grossman HB, et al. Refining patient selection for
neoadjuvant chemotherapy before radical cystectomy. J Urol. 2014;
191:40-47.
Dash A, Galsky MD, Vickers AJ, et al. Impact of renal impairment on
eligibility for adjuvant cisplatin-based chemotherapy in patients
with urothelial carcinoma of the bladder. Cancer. 2006;107:
506-513.
Galsky MD. How I treat bladder cancer in elderly patients. J Geriatr
Oncol. 2015;6:1-7.
Shabsigh A, Korets R, Vora KC, et al. Defining early morbidity of radical
cystectomy for patients with bladder cancer using a standardized
reporting methodology. Eur Urol. 2009;55:164-174.
Comploj E, West J, Mian M, et al. Comparison of complications from
radical cystectomy between old-old versus oldest-old patients. Urol Int.
2015;94:25-30.
Berger I, Martini T, Wehrberger C, et al. Perioperative complications and
90-day mortality of radical cystectomy in the elderly (75+): a retrospective, multicentre study. Urol Int. 2014;93:296-302.
Schiffmann J, Gandaglia G, Larcher A, et al. Contemporary 90-day
mortality rates after radical cystectomy in the elderly. Eur J Surg Oncol.
2014;40:1738-1745.
Zakaria AS, Santos F, Tanguay S, et al. Radical cystectomy in patients
over 80 years old in Quebec: a population-based study of outcomes.
J Surg Oncol. 2015;111:917-922.
Yuh BE, Nazmy M, Ruel NH, et al. Standardized analysis of frequency and
severity of complications after robot-assisted radical cystectomy. Eur
Urol. 2012;62:806-813.
Bochner BH, Dalbagni G, Sjoberg DD, et al. Comparing open radical
cystectomy and robot-assisted laparoscopic radical cystectomy: a randomized clinical trial. Eur Urol. 2015;67:1042-1050.
Warner MA, Offord KP, Warner ME, et al. Role of preoperative cessation
of smoking and other factors in postoperative pulmonary complications: a blinded prospective study of coronary artery bypass patients.
Mayo Clin Proc. 1989;64:609-616.
Large MC, Reichard C, Williams JT, et al. Incidence, risk factors, and
complications of postoperative delirium in elderly patients undergoing
radical cystectomy. Urology. 2013;81:123-129.
35. Cerantola Y, Valerio M, Persson B, et al. Guidelines for perioperative

care after radical cystectomy for bladder cancer: Enhanced Recovery
After Surgery (ERAS()) society recommendations. Clin Nutr. 2013;32:
879-887.
36. Lee CT, Chang SS, Kamat AM, et al. Alvimopan accelerates gastrointestinal recovery after radical cystectomy: a multicenter randomized
placebo-controlled trial. Eur Urol. 2014;66:265-272.
37. Daneshmand S, Ahmadi H, Schuckman AK, et al. Enhanced recovery protocol after radical cystectomy for bladder cancer. J Urol. 2014;192:50-56.
38. Horovitz D, Turker P, Bostrom PJ, et al. Does patient age affect survival
after radical cystectomy? BJU Int. 2012;110(11 Pt B):E486-E493.
39. Leveridge MJ, Siemens DR, Mackillop WJ, et al. Radical cystectomy and
adjuvant chemotherapy for bladder cancer in the elderly: a populationbased study. Urology. 2015;85:791-798.
e233
Renal Cell Carcinoma: Systemic

Treatment, Evolving TKIs, and
Immuno-Oncology
CHAIR
Thomas E. Hutson, DO, PharmD, FACP
Texas Oncology
Dallas, TX
SPEAKERS
Brian I. Rini, MD, FACP
Cleveland Clinic Taussig Cancer Institute
Cleveland, OH
Robert A. Figlin, MD, FACP
Cedars-Sinai Medical Center
Los Angeles, CA
THE EVOLUTION OF SYSTEMIC THERAPY IN mRCC
The Evolution of Systemic Therapy in Metastatic Renal Cell

Carcinoma
Thomas E. Hutson, DO, PharmD, FACP, Gregory R. Thoreson, MD, Robert A. Figlin, MD, FACP, and
Brian I. Rini, MD, FACP
OVERVIEW
The treatment landscape for renal cell carcinoma (RCC) is a dynamic process that has seen considerable change in recent
years. We have seen a rebirth of original breakthroughs with immune checkpoint inhibitors showing promise in patients with
treatment-refractory disease. The optimal sequencing of treatments and incorporation of novel therapeutics are actively
being investigated and have yet to be determined. The clinical challenges of this evolving treatment paradigm can be
attributed to cost considerations, toxicity, and defining endpoints in the management of advanced RCC. As novel therapeutics emerge, finding the optimal treatment regimen for patients will have an increasing focus on patient-centered
outcomes and improvement in quality of life in addition to improving survival.
anagement of RCC continues to evolve, with many of

the recent advancements occurring in the treatment
of patients with metastatic disease. Approximately one in
three patients present with either regionally advanced or
metastatic disease, excluding candidacy for extirpative
surgery with curative intent and ultimately requiring systemic therapy as their only option.1 The earliest forms of
systemic therapy emerged in the 1980s and were based on
adoptive immunotherapy. Observations at the National Institutes
of Health showed objective responses with the administration of
lymphokine-activated killer cells and recombinant interleukin-2
(IL-2) across four histologic cell types, including RCC.2 By the early
1990s, IL-2, interferon, or a combination of the two were widely
adopted, and in the early 2000s, this became the standard of
care.3,4 To date, the only durable complete responses observed in
metastatic RCC (mRCC) are with this modality of treatment.5 The
latter part of the decade ushered in the era of targeted therapies,
with pivotal clinical trials showing improved survival when
compared with interferon.6-8 This line of therapy is predicated
on inhibiting VEGF or mTOR. Recently, the treatment paradigm
has shifted back toward immunotherapy, with an immune
checkpoint inhibitor showing superiority over an mTOR inhibitor in a select patient population that demonstrated disease progression while on antiangiogenic therapy.9
This influx of new anticancer therapies in mRCC has
provided new opportunities to explore the therapeutic
potential of combination therapies. To date, combination
therapy in mRCC has not provided clinical benefit. Several
trials over the past decade have explored combinations,

largely of VEGF inhibitors plus VEGF inhibitors or VEGF inhibitors plus mTOR inhibitors.10-14 Collectively, these trials
failed because of either excessive toxicity preventing administration of full doses of each agent/toxicity in general, or
lack of a proven efficacy advantage over monotherapy.
On this landscape, recent data regarding the combination
of a VEGF receptor inhibitor (lenvatinib) and an mTOR inhibitor (everolimus) is notable.15 A phase II study randomly
assigned 153 patients with mRCC who had received one
prior VEGF-targeted therapy to lenvatinib plus everolimus
compared with lenvatinib monotherapy compared with everolimus monotherapy with a primary endpoint of progressionfree survival (PFS). Progression-free survival as initially reported
per investigator assessment, was 14.6 months for the combination arm, 7.4 months for the lenvatinib monotherapy arm,
and 5.5 months for the everolimus arm. Both lenvatinibcontaining arms were significantly longer in terms of PFS
than the everolimus arm (lenvatinib plus everolimus vs.
everolimus, hazard ratio [HR] 0.40; p = .0005; lenvatinib
vs. everolimus, HR 0.61; p = .048) but not significantly
different from each other (lenvatinib plus everolimus vs.
lenvatinib, HR 0.66; p = .12). Objective response rates also
favored the combination arm (43% vs. 27% vs. 6%). There was
considerable toxicity in the combination arm necessitating
that 71% of patients reduce their lenvatinib dose and 24%
discontinue for toxicity. Importantly, updated data using independent radiologic review of CT scans has been published.16
From the Texas A&M Health Science Center, Bryan TX; Baylor-Sammons Cancer Center, Dallas, TX; US Oncology, Texas Oncology, Charles A. Sammons Cancer Center, Baylor University
Medical Center, Dallas, TX; Samuel Oschin Comprehensive Cancer Institute, Cedars-Sinai Medical Center, Los Angeles, CA; Cleveland Clinic Taussig Cancer Institute, Cleveland, OH.
Corresponding author: Thomas E. Hutson, DO, PharmD, FACP, Texas Oncology, 3410 Worth St., Suite 400, Dallas, TX 75254; email: thomas.hutson@usoncology.com.
113
HUTSON ET AL
These data demonstrate a reduction in the combination arm

PFS (to 12.8 months) and the response rate (to 35%). Thus, the
benefit of this specific combination over monotherapy remains
to be established in larger trials. The biology of why this
combination could have benefit where other similar combinations have failed is unknown and also requires further study.
Cabozantinib, an oral small molecule inhibitor of tyrosine
kinases including MET, VEGF, and AXL, has shown recent
promise in patients with mRCC that has failed first-line
therapy with VEGFR-targeting tyrosine kinase inhibitors.
In a phase III clinical trial with 658 patients, cabozantinib
demonstrated improved PFS (7.4 vs. 3.8 months), overall
survival (HR 0.67), and objective response rates (21% vs. 5%)
when compared with everolimus.17 Cabozantinib, which is
currently approved by the U.S. Food and Drug Administration
(FDA) for the treatment of patients with progressive metastatic medullary thyroid cancer, has previously been granted
Breakthrough Therapy and Fast Track designations and is
likely to be approved for treatment of refractory disease.18
As noted, checkpoint inhibition has undergone rapid
clinical development in mRCC with the FDA approval of nivolumab for refractory disease. Several combination regimens of
immunotherapy or immunotherapy plus VEGF therapy have
been initially explored (Table 1). There is clinical rationale for
such combinations given the activity of each approach as
monotherapy, but also a biologic rationale, as VEGF may lead to
immune suppression and thus VEGF blockade may allow for
enhanced effect of immunotherapy.19,20 Nivolumab was
combined in a multicohort phase II trial with sunitinib and
separately with pazopanib. The pazopanib cohort was not
expanded because of liver function test (LFT) abnormalities
in the initial cohort of patients. The sunitinib cohort expanded and demonstrated antitumor activity as evidenced
by a robust objective response rate.
KEY POINTS
114
The treatment of advanced renal cell carcinoma is a

dynamic process that has undergone considerable
change in the past 40 years with a recent rebirth of
immuno-oncology drugs.
Optimal sequence and combinations of novel therapies
are currently being investigated and pose considerable
challenges to providers.
Toxicity profiles of targeted therapies and immunooncology agents are unique, and effective therapy
management must balance treatment dosing, duration,
and adverse event management.
A value-based framework for cancer care requires the
incorporation of effectiveness, safety, and efficiency
while maintaining incentives for the continued research
and development of novel oncologic therapeutics.
Defining endpoints for patients with metastatic renal
cell carcinoma continue to evolve, with an increasing
emphasis on patient-centered outcomes in addition to
survival data.
This combination, however, was not further pursued

clinically because of development of the nivolumab plus ipilimumab combination. This latter combination was evaluated
as part of the same clinical trial and also demonstrated remarkable clinical activity (Table 1). Toxicity was higher in the
arm with the higher ipilimumab dosing and was also
unacceptably high in a small cohort (six patients) who
received a higher dose of both drugs. This combination is
being further pursued in a front-line phase III trial compared with sunitinib. Accrual to this trial has been completed and results are pending.
Additional PD-1 or PD-L1 inhibitors have been initially
combined with anti-VEGF therapy. Pembrolizumab, a PD-1
inhibitor, was combined with pazopanib. Several patients
experienced LFT abnormalities necessitating interruption of
therapy, and other patients have been treated with a lower
initial dose of pazopanib. The safety of this particular combination awaits further experience. Pembrolizumab has also
been combined with axitinib. A small safety cohort of 11 patients demonstrated tolerability and early signs of efficacy. An
expansion cohort has been accrued to further clarify safety and
efficacy.
An alternative approach to checkpoint inhibition is an
antibody that blocks the ligand for the PD-1 receptor, PD-L1.
Atezolizumab is a monoclonal antibody that has been
studied in many solid tumors including RCC. An initial phase I
study of atezolizumab demonstrated the safety of monotherapy in RCC and other solid tumors, and a small expansion
cohort in RCC tested the combination of atezolizumab and
bevacizumab. This combination was well tolerated, with
early signs of efficacy in the form of tumor burden reduction.
A randomized phase II trial was then undertaken in front-line
mRCC randomly assigning patients to this combination,
atezolizumab monotherapy, or standard sunitinib with a
primary endpoint of PFS. This trial has completed accrual
with results pending. Importantly, this trial contains an
arm with antiPD-L1 monotherapy. Clinical and correlate
results from this arm may help identify if there is a subset
of patients with mRCC who can safely and effectively receive
single-agent checkpoint inhibition. A randomized phase III
trial has also begun with this combination in front-line mRCC,
randomly assigning patients to the combination or sunitinib
with a PFS/overall survival coprimary endpoint.
Challenges for managing RCC are not only considerations
of clinical algorithm development but also opportunities to
demonstrate value-based cancer care, especially in a disease
with multiple treatment approvals.21-23 The Institute of Medicine
has delineated six elements of value in cancer care: safety,
effectiveness, patient centeredness, timeliness, efficiency, and
equity. The American Society of Clinical Oncology has selected
three of these for its value-based framework for cancer care:
clinical benefit (effectiveness), toxicity (safety), and cost (efficiency). The Memorial Sloan Kettering Cancer Center has added
to the dialogue by including research and development costs and
population health burden, and the National Comprehensive
Cancer Network has included elements of affordability. Health
economics research has started to delineate how patients with
TABLE 1. Select Studies of Combination Therapy in Renal Cell Carcinoma

Trial
No.
Patients
Patient
Population
Study Design
ORR
Median PFS
Comments
Nivolumab 2 mg/kg every

3 weeks + sunitinib 50 mg 4/2
52%
49 weeks; 78%
progression-free
at 24 weeks
Nivolumab 2 mg/kg every

3 weeks + pazopanib 800 mg/day
45%
31 weeks; 55%
progression-free
at 24 weeks
Pazopanib cohort not

expanded because of
hepatic toxicity
Immunotherapy Plus Anti-VEGF Nivolumab Plus TKI

NCT01472081,
Phase I/II34
33
Front-line and
refractory mRCC
20
Pembrolizumab Plus TKI

NCT02014636,
Phase I/II35
20
Front-line mRCC
Pembrolizumab plus
pazopanib 600 mg every day
(10 patients) or 800 mg every
day (10 patients)
40%
NR
AST and/or ALT elevations

. 5x ULN observed in
13 patients
NCT02133742,
Phase IB36
11
Front-line mRCC
Pembrolizumab 2 mg/kg
every 3 weeks plus
axitinib 5 mg twice daily
55%
NR
41 additional patients
enrolled for further
safety/efficacy
investigation
Atezolizumab Plus Bevacizumab

NCT01633970,
Phase IB37
10
Front-line mRCC
1,200 mg IV every 3 weeks atezolizumab +

bevacizumab 15 mg/kg every
3 weeks
40%
SD $ 24 weeks
in 4 patients
9/10 patients remain

on treatment
NCT01984242
Randomized
Phase II
300
Front-line mRCC
1,200 mg IV every 3 weeks

atezolizumab +/2 bevacizumab
15 mg/kg every 3 weeks vs. sunitinib
NR
NR
Accrual complete
NCT02420821,
Phase III
550
Front-line mRCC

3 weeks vs. sunitinib
NR
NR
Accrual ongoing
NCT02420821,
Phase III
550
Front-line mRCC

3 weeks vs. sunitinib
NR
NR
Accrual ongoing
Immunotherapy Plus Immunotherapy

Nivolumab Plus Ipilimumab
NCT01472081,
Phase I/II38
47
NCT02231749,
Phase III
1,070
47
Front-line and
refractory mRCC
Front-line mRCC
N3 + I1*
38%
33.3 weeks
N1 + I3*
40%
47.1 weeks
Increased toxicity in
the I3
N3 + I1
NR
NR
Accrual complete
*Arm N3 + I1; nivolumab 3 mg/kg + ipilimumab 1 mg/kg; Arm N1 + I3; nivolumab 1 mg/kg + ipilimumab 3 mg/kg.
Abbreviations: AST, aspartate transaminase; ALT, alanine aminotransferase; NR, no response; PFS, progression-free survival; ORR, overall response rate; ULN, upper limit of normal;
TKI, tyrosine kinase inhibitor; mRCC, metastatic renal cell carcinoma.
cancer value hope and the implications for cost-effectiveness

assessments of high-cost cancer therapies.24 A patients willingness to pay for a hopeful therapy may be influenced by their
economic realities.25 In addition, consideration of qualityadjusted cost of care has been used to measure the growth
and new innovation costs associated with novel therapies.26,27
Unfortunately, none of these growing health economic approaches to cancer care have been applied uniformly to the
growing portfolio of cancer treatments for RCC. This remains a
unique challenge in a disease with nine new drugs approved by
the FDA over the past decade.
The toxicity profiles of targeted therapies and immunooncology agents are quite different, and their management
continues to evolve. Toxicities can generally be broken down
into three categories: side effects produced by the VEGF
inhibitors/tyrosine kinase inhibitors (i.e., sunitinib, pazopanib,
axitinib, sorafenib, and bevacizumab), mTOR inhibitors
(i.e., everolimus, temsirolimus), and more recently immunooncology drugs (i.e., nivolumab). Effective therapy management must be a balance between dosing, treatment duration,
and adverse event management.
The common adverse events associated with VEGF inhibitors include fatigue, hand/foot syndrome, hypertension,
hepatotoxicity, diarrhea, stomatitis, myelosuppression, and
proteinuria. For the mTOR inhibitors, fatigue, diarrhea, stomatitis, myelosuppression, pneumonitis, and infections are
the most common adverse events. Management of these
chronic toxicities is associated with improved dosing and
treatment duration and is best supported by aggressive adverse event management.
Recent evidence has suggested that, for example, with
sunitinib, adjusting the dosing schedule to a 2:1 schedule
from the original 4:2 schedule allows for full dosing
maintenance of concentrations of drug with a possible
115
HUTSON ET AL
amelioration of toxicity. Black box warnings for pazopanib

and sunitinib for potential liver toxicity must guide treatment management, as well as pneumonitis, a class effect
from mTOR inhibition characterized by noninfectious,
nonmalignant infiltrates associated with cough, dyspnea,
and radiographic findings of ground glass opacities and
focal consolidation. Several published summaries will guide
recommended adverse event management in mTOR-treated
RCC, especially for noninfectious pneumonitis, stomatitis, and
infections. The side effect profile from targeted agents is
usually quite predictable, and the optimal management will
increase drug exposure and improve overall outcome. Toxicity
is often treatable and can be managed with schedule changes
and treatment interruptions before dose reduction is
considered.
With the recently approved immuno-oncology agent, we are
now presented with a different spectrum of toxicitiesusually
of low frequencycalled immune-related adverse events,
which have timing that is less predictable than with adverse
events with targeted agents, can appear without warning, and
often require immune suppression with corticosteroids where
holding the drug is typically insufficient.
Several websites now have guides to immune-mediated
adverse reaction management, and physicians should familiarize themselves with these side effects. Interestingly,
the time to onset of selected treatment-related adverse
events for immuno-oncology agents such as nivolumab can
sometimes be weeks to months, and, as such, careful and
continual monitoring of the patient is required. Although
most side effects are low grade, grade 3 and 4 toxicities
can occur, especially immune-mediated pulmonary, hepatic, gastrointestinal, and renal adverse reactions. Immunemediated endocrinopathies can be life threatening if not
approached appropriately. Recent evidence suggests that
immune suppression may not reduce the efficacy of immunooncology approaches and, as such, might allow for the continued administration of these agents. These new therapies
have the potential to improve and prolong patients lives,
and optimizing toxicity management will mean keeping the

patient on drug while maintaining quality of life. The early
recognition and intervention for the successful management
of immune-related adverse events is critical with this new
class of drugs.
The traditional endpoints for approval of targeted therapies in RCC have been PFS. For patients treated with targeted agents, it is clear that improvements in PFS have been
associated with improvements in overall survival. It is interesting to note that with the newer immuno-oncology
agents, overall survival has been the primary endpoint. With
the approval of nivolumab, there is evidence that overall
survival is benefited, while there is no apparent difference in
PFS, and as such, we may be seeing an evolution in endpoint
evaluations for patients with RCC.28-30 It is clear from a
patient perspective that overall survival benefits are critical.
This presents a challenge to the practicing physician because
we are typically used to using imaging as a surrogate intermediate endpoint for the evaluation of benefit for patients. With circumstances such as pseudo-progression or
early growth followed by regression, we now must reconsider the use of imaging in deciding when and if to take a
patient off of systemic therapies.21 Further evolution in
endpoint evaluation in RCC will evolve, especially if biomarkers become available to help identify populations of
patients most likely to benefit from systemic therapy.
The treatment landscape for RCC is a dynamic process that
has seen considerable change in recent years. We have
seen a rebirth of original breakthroughs with immune
checkpoint inhibitors showing promise in treatment refractory disease, which may play a crucial role in combination therapies. The clinical challenges of this evolving
treatment paradigm can be attributed to cost considerations, toxicity, and defining endpoints in the management
of advanced RCC. As novel therapeutics emerge, identifying
the optimal sequence and combinations of therapies will be
based on individual patient characteristics with an emphasis on improving quality of life and survival.
References
1. Howlader NA, Krapcho M, Garshell J. SEER Cancer Statistics Review,
1975--2012. Bethesda: National Cancer Institute; 2014.
2. Rosenberg SA, Lotze MT, Muul LM, et al. Observations on the systemic
administration of autologous lymphokine-activated killer cells and
recombinant interleukin-2 to patients with metastatic cancer. N Engl J
Med. 1985;313:1485-1492.
3. Negrier S, Escudier B, Lasset C, et al. Recombinant human interleukin-2,
recombinant human interferon alfa-2a, or both in metastatic renal-cell
carcinoma. Groupe Franais dImmunotherapie. N Engl J Med. 1998;
338:1272-1278.
4. Yang JC, Sherry RM, Steinberg SM, et al. Randomized study of high-dose
and low-dose interleukin-2 in patients with metastatic renal cancer.
J Clin Oncol. 2003;21:3127-3132.
5. McDermott DF, Regan MM, Clark JI, et al. Randomized phase III trial of
high-dose interleukin-2 versus subcutaneous interleukin-2 and interferon
116
6.
7.
8.
9.
10.
in patients with metastatic renal cell carcinoma. J Clin Oncol. 2005;23:

133-141.
Motzer RJ, Hutson TE, Tomczak P, et al. Sunitinib versus interferon alfa
in metastatic renal-cell carcinoma. N Engl J Med. 2007;356:115-124.
Escudier B, Eisen T, Stadler WM, et al; TARGET Study Group. Sorafenib in
advanced clear-cell renal-cell carcinoma. N Engl J Med. 2007;356:125-134.
Hudes G, Carducci M, Tomczak P, et al; Global ARCC Trial. Temsirolimus,
interferon alfa, or both for advanced renal-cell carcinoma. N Engl J Med.
2007;356:2271-2281.
Motzer RJ, Escudier B, McDermott DF, et al; CheckMate 025 Investigators. Nivolumab versus everolimus in advanced renal-cell carcinoma. N Engl J Med. 2015;373:1803-1813.
Rini BI, Garcia JA, Cooney MM, et al. Toxicity of sunitinib plus bevacizumab in renal cell carcinoma. J Clin Oncol. 2010;28:e284-e285;
author reply e286-e287.
11. Rini BI, Bellmunt J, Clancy J, et al. Randomized phase III trial of temsirolimus and bevacizumab versus interferon alfa and bevacizumab in
metastatic renal cell carcinoma: INTORACT trial. J Clin Oncol. 2014;32:
752-759.
12. Ravaud A, Barrios CH, Alekseev B, et al. RECORD-2: phase II randomized
study of everolimus and bevacizumab versus interferon a-2a and
bevacizumab as first-line therapy in patients with metastatic renal cell
carcinoma. Ann Oncol. 2015;26:1378-1384.
13. Negrier S, Gravis G, Perol D, et al. Temsirolimus and bevacizumab, or
sunitinib, or interferon alfa and bevacizumab for patients with advanced renal cell carcinoma (TORAVA): a randomised phase 2 trial.
Lancet Oncol. 2011;12:673-680.
14. Flaherty KT, Manola JB, Pins M, et al. BEST: a randomized phase II study
of vascular endothelial growth factor, RAF kinase, and mammalian
target of rapamycin combination targeted therapy with bevacizumab,
sorafenib, and temsirolimus in advanced renal cell carcinomaa trial of
the ECOG-ACRIN Cancer Research Group (E2804). J Clin Oncol. 2015;33:
2384-2391.
15. Motzer RJ, Hutson TE, Glen H, et al. Lenvatinib, everolimus, and the
combination in patients with metastatic renal cell carcinoma: a randomised, phase 2, open-label, multicentre trial. Lancet Oncol. 2015;16:
1473-1482.
16. Motzer RJ, Hutson TE, Ren M, et al. Independent assessment of lenvatinib plus everolimus in patients with metastatic renal cell carcinoma.
Lancet Oncol. 2016;17:e4-e5.
17. Choueiri TK, Escudier B, Powles T, et al; METEOR Investigators.
Cabozantinib versus everolimus in advanced renal-cell carcinoma.
N Engl J Med. 2015;373:1814-1823.
18. Viola D, Cappagli V, Elisei R. Cabozantinib (XL184) for the treatment of
locally advanced or metastatic progressive medullary thyroid cancer.
Future Oncol. 2013;9:1083-1092.
19. Shrimali RK, Yu Z, Theoret MR, et al. Antiangiogenic agents can
increase lymphocyte infiltration into tumor and enhance the effectiveness of adoptive immunotherapy of cancer. Cancer Res. 2010;70:
6171-6180.
20. Kusmartsev S, Su Z, Heiser A, et al. Reversal of myeloid cell-mediated
immunosuppression in patients with metastatic renal cell carcinoma.
Clin Cancer Res. 2008;14:8270-8278.
21. Zon RT, Frame JN, Neuss MN, et al. American Society of Clinical Oncology Policy Statement on Clinical Pathways in Oncology. J Oncol Pract.
2016;12:261-266.
22. Saltz LB. The value of considering cost, and the cost of not considering
value. J Clin Oncol. 2016;34:659-660.
23. Young RC. Value-based cancer care. N Engl J Med. 2015;373:2593-2595.
24. Neumann PJ, Cohen JT. Measuring the value of prescription drugs.
N Engl J Med. 2015;373:2595-2597.
25. Lakdawalla DN, Romley JA, Sanchez Y, et al. How cancer patients value
hope and the implications for cost-effectiveness assessments of highcost cancer therapies. Health Aff (Millwood). 2012;31:676-682.
26. Lakdawalla D, Shafrin J, Lucarelli C, et al. Quality-adjusted cost of care:
a meaningful way to measure growth in innovation cost versus the value
of health gains. Health Aff (Millwood). 2015;34:555-561.
27. Stevens W, Philipson TJ, Khan ZM, et al. Cancer mortality reductions
were greatest among countries where cancer care spending rose the
most, 1995-2007. Health Aff (Millwood). 2015;34:562-570.
28. Hoos A, Wolchok JD, Humphrey RW, et al. CCR 20th Anniversary
Commentary: immune-related response criteriacapturing clinical activity in immuno-oncology. Clin Cancer Res. 2015;21:4989-4991.
29. Wolchok JD, Hoos A, ODay S, et al. Guidelines for the evaluation of
immune therapy activity in solid tumors: immune-related response
criteria. Clin Cancer Res. 2009;15:7412-7420.
30. Nishino M, Giobbie-Hurder A, Gargano M, et al. Developing a common
language for tumor response to immunotherapy: immune-related response criteria using unidimensional measurements. Clin Cancer Res.
2013;19:3936-3943.
31. Sonpavde G, Choueiri TK, Escudier B, et al. Sequencing of agents for
metastatic renal cell carcinoma: can we customize therapy? Eur Urol.
2012;61:307-316.
32. Motzer RJ, Hutson TE, Cella D, et al. Pazopanib versus sunitinib in
metastatic renal-cell carcinoma. N Engl J Med. 2013;369:722-731.
33. Escudier B, Szczylik C, Porta C, et al. Treatment selection in metastatic
renal cell carcinoma: expert consensus. Nat Rev Clin Oncol. 2012;9:
327-337.
34. Amin A, Plimack ER, Infante JR, et al. Nivolumab (anti--PD--1; BMS-936558, ONO--4538) in combination with sunitinib or pazopanib in
patients (pts) with metastatic renal cell carcinoma (mRCC). J Clin Oncol.
2014;32:5s (suppl; abstr 5010).
35. McDermott DF, Infante JR, Chowdhury S, et al. A phase I/II study to
assess the safety and efficacy of pazopanib (paz) and pembrolizumab
(pembro) in patients (pts) with advanced renal cell carcinoma (aRCC).
Eur J Cancer. 2015;51:S519-S520.
36. Atkins MB, Gupta S, Choueiri TK, et al. Phase lb dose--finding study of
axitinib plus pembrolizumab in treatment--naive patients with advanced renal cell carcinoma. J Immunother Cancer. 2015;3:P353.
37. McDermott, D, Sznol M, Sosman JA, et al. 809O -- Immune correlates
and long term follow up of a phase Ia study of MPDL3280A, an engineered PD--L1 antibody, in patients with metastatic renal cell carcinoma
(mRCC). Ann Oncol. 2014;25:iv280-iv304.
38. Hammers H, Plimack ER, Infante JR, et al. Expanded cohort results from
CheckMate 016: A phase I study of nivolumab in combination with
ipilimumab in metastatic renal cell carcinoma (mRCC). J Clin Oncol.
2015;33 (suppl; abstr 4516).
117
Contemporary Active Surveillance

for Prostate Cancer: Do We Need
Better Imaging and Molecular
Testing?
CHAIR
Edward M. Schaeffer, MD, PhD
Northwestern University
Chicago, IL
SPEAKERS
Jonathan I. Epstein, MD
Baltimore, MD
Peter L. Choyke, MD
National Cancer Institute at the National Institutes of Health
Bethesda, MD
Stacy Loeb, MD
New York University Langone Medical Center
New York, NY
ACTIVE SURVEILLANCE FOR PROSTATE CANCER
Active Surveillance of Prostate Cancer: Use, Outcomes,

Imaging, and Diagnostic Tools
Jeffrey J. Tosoian, MD, MPH, Stacy Loeb, MD, MSc, Jonathan I. Epstein, MD, Baris Turkbey, MD,
Peter L. Choyke, MD, and Edward M. Schaeffer, MD, PhD
OVERVIEW
Active surveillance (AS) has emerged as a standard management option for men with very low-risk and low-risk prostate
cancer, and contemporary data indicate that use of AS is increasing in the United States and abroad. In the favorable-risk
population, reports from multiple prospective cohorts indicate a less than 1% likelihood of metastatic disease and prostate
cancerspecific mortality over intermediate-term follow-up (median 56 years). Higher-risk men participating in AS appear
to be at increased risk of adverse outcomes, but these populations have not been adequately studied to this point. Although
monitoring on AS largely relies on serial prostate biopsy, a procedure associated with considerable morbidity, there is a need
for improved diagnostic tools for patient selection and monitoring. Revisions from the 2014 International Society of Urologic
Pathology consensus conference have yielded a more intuitive reporting system and detailed reporting of low-intermediate
grade tumors, which should facilitate the practice of AS. Meanwhile, emerging modalities such as multiparametric magnetic
resonance imaging and tissue-based molecular testing have shown prognostic value in some populations. At this time,
however, these instruments have not been sufficiently studied to consider their routine, standardized use in the AS setting.
Future studies should seek to identify those platforms most informative in the AS population and propose a strategy by
which promising diagnostic tools can be safely and efficiently incorporated into clinical practice.
ctive surveillance of prostate cancer with curative intent

was described in the mid-1990s, and early AS experiences were reported in 2002.1,2 Under the AS approach, men
with favorable-risk cancers are monitored, and curative
intervention is pursued upon evidence of higher-risk
disease. Over the last 2 decades, AS has emerged as a
standard management option for men with very low-risk
and low-risk prostate cancer.3,4 Observations from two
large, prospective AS cohorts reach nearly 20 years of
follow-up and indicate very low likelihood of metastatic
disease or prostate cancerspecific mortality in appropriately selected men.5,6 Despite its utility in reducing
overtreatment, AS is not without morbidity. 7 The contemporary practice of AS remains largely based on frequent clinical examination, serum prostate-specific antigen
(PSA) testing, and prostate biopsy,8 a procedure associated with patient discomfort and serious complications
including infection.9,10 Furthermore, these methods lack
sensitivity for detection of higher-risk disease, as evidenced by the substantial proportion of men meeting AS
criteria who demonstrate high-risk features at radical
prostatectomy.11,12
As such, there is a substantial need for more accurate methods

of patient selection and monitoring. An ideal diagnostic tool
would impart valuable diagnostic and prognostic information
with limited associated morbidity at a reasonable cost. Although
the ideal platform does not currently exist, advances in technology and improved understanding of the molecular basis of
prostate cancer have initiated progress toward that goal.13-16 For
example, the use and utility of multiparametric MRI (mpMRI)
of the prostate has increased substantially in recent years.17
Furthermore, clinically validated molecular tests have established a prognostic role in some clinical contexts.18 These
platforms, along with a recently updated system of pathologic
grading, present a unique opportunity to improve the practice of
AS.19,20 This article aims to review the contemporary practice of
AS, including trends in use, outcomes, pathologic grading, MRI,
and tissue-based molecular testing.
TRENDS IN USE OF ACTIVE SURVEILLANCE

Although the AS approach was previously underutilized,21,22
recent data from multiple countries have confirmed increasing use, corresponding with its inclusion in multiple
national guidelines and the availability of more data on
From the Brady Urological Institute, Departments of Urology and Pathology, Johns Hopkins University School of Medicine, Baltimore, MD; Department of Urology and Population Health,
New York University, New York, NY; Molecular Imaging Program, National Cancer Institute, Bethesda, MD; Department of Urology, Northwestern University, Chicago, IL.
Corresponding author: Edward M. Schaeffer, MD, PhD, Northwestern University, 303 East Chicago Ave., Tarry Building 16-713, Chicago, IL 60611; email: e-schaeffer@
northwestern.edu.
e235
TOSOIAN ET AL
long-term outcomes.4-6 In the United States, as of 2006, lowrisk prostate cancer was managed conservatively for only 10%
of men.23 Since that time, however, there has been a major
expansion in use. By 2011, the New Hampshire State Cancer
Registry reported that this disease was managed expectantly
for 42% of low-risk patients.24 In a large registry from
Michigan, 49% of low-risk prostate cancers diagnosed in 2012
to 2013 were managed on AS.25 Finally, new data from the
CaPSURE clinical practice registry reported an increase in
conservative management for up to 40% of low-risk cases
from 2010 to 2013.26 Similarly, studies from Canada have
reported a reduction in the proportion of low-risk patients
undergoing radical prostatectomy.27
Corroborative findings have been observed in several
European studies. In the National Prostate Cancer Register
of Sweden from 2007 to 2011, AS was selected by 59% of
very low-risk patients and 41% of low-risk patients.28
Meanwhile, data from Germany demonstrated a decline in
the proportion of men with Gleason score 6 disease at
radical prostatectomy from 2000 to 2014.29 Despite these
favorable trends suggesting a reduction in the overtreatment of low-risk prostate cancer, there continues to be
substantial variability in management patterns between and
within various clinical practice settings.30,31 Furthermore, in
some parts of the world, the use of AS continues to remain
more limited.32 Nevertheless, the indications for AS continue to expand in national guidelines, 33 and it can be
expected that this management paradigm will become
increasingly offered for favorable-risk disease in the future.
OUTCOMES OF ACTIVE SURVEILLANCE

The modern approach to AS was first described in 1995, and
numerous programs have now reported follow-up outcomes
KEY POINTS
e236
Active surveillance represents a standard management

option for men with very low-risk and low-risk prostate
cancer.
With follow-up extending to 20 years (median 56
years), multiple AS cohorts report metastasis and
prostate cancer mortality in less than 1% of favorablerisk men; these outcomes appear to be more frequent in
men who do not meet low-risk criteria, but higher-risk
populations have not been adequately studied.
Revisions to pathologic grading from the 2014
International Society of Urologic Pathology consensus
conference should facilitate a clearer understanding of
pathologic data and more accurate assessment of risk in
men bordering AS criteria.
Multiparametric MRI and tissue-based molecular
testing may improve extant monitoring protocols but
first require additional study and validation.
As the practice of AS continues to evolve, it is important
that modifications of protocols are based on evidence of
patient benefit.
at 5 years and beyond.8 In Toronto, Klotz et al5 initiated a

program of AS aimed at low-risk and select intermediate-risk
patients. Their monitoring protocol includes PSA measurements every 3 months for 2 years and then every 6 months,
with a confirmatory biopsy during the first year and then
every 3 to 4 years until age 80. Intervention was initially
recommended for histologic upgrading on rebiopsy and/or
PSA doubling time less than 3 years, but PSA changes now
trigger additional work-up rather than immediate intervention. The most recent update of this cohort was
published in 2015, with a median follow-up of 6.4 years
(ranging up to 19.8 years). Of the 993 men to enroll in the
program since 1995, 28 (2.8%) developed metastases and 15
(1.5%) died of prostate cancer. Meanwhile, treatment rates
at 10 and 15 years were 36% and 45%, respectively.
More recently, extended follow-up was reported from the
Johns Hopkins AS program, which was also initiated in 1995.6
This program aims to enroll very low-risk patients (with a PSA
density , 0.15, clinical stage T1c, and Gleason score # 6 in a
maximum of two cores with # 50% cancer involvement in
any core). The protocol includes PSA and digital rectal examination (DRE) every 6 months and a yearly prostate biopsy. At the most recent analysis, the cohort was composed
of 71% very low-risk men and 29% low-risk men. Triggers for
treatment included an increase in grade (Gleason score . 6)
or volume of cancer on biopsy. The 10- and 15-year cumulative incidence of treatment was 50% and 57%, respectively, with a median treatment-free survival of 8.5
years. Meanwhile, 15-year metastasis-free and cancerspecific survival rates were 99.4% and 99.9%, respectively.
Table 1 presents published data from these two large
prospective cohorts.
Other U.S. centers have similarly reported intermediateterm outcomes from AS. At the University of California, San
Francisco, prostate cancer has been managed among 810
men using an AS protocol based on quarterly PSA testing,
repeat biopsy within 12 months, and follow-up biopsy every
1 to 2 years depending on medical risk.34 Of these men, 69%
met strict eligibility criteria (PSA # 10 ng/ml, clinical stage
T1/T2, Gleason score # 6, # 33% positive biopsy cores,
and # 50% involvement of any core with cancer). The authors reported a 5-year treatment-free survival of 60%, and
there were no prostate cancer deaths during a median
follow-up of 60 months.
At Royal Marsden in the United Kingdom, the eligibility
criteria for AS include age 5080, clinical stage T1/T2 disease,
PSA less than 15 ng/ml, Gleason score of 6 or less (as well as
Gleason 3 + 4 in men older than age 65), and 50% positive
biopsy cores or less.35 PSA and DRE are performed every
3 months in year 1, every 4 months in year 2, and every
6 months thereafter. A biopsy is performed at 18 to
24 months from diagnosis and then every 2 years. Triggers
for treatment are a PSA velocity greater than 1 ng/ml per
year, a Gleason score above 3 + 4, and/or more than 50%
positive cores on repeat biopsy. Among 471 men in the
program, 323 (68.6%) remained on AS at a median follow-up
of 5.7 years, and there were two prostate cancer deaths.
TABLE 1. Basic Characteristics of Two Large Prospective Active Surveillance Cohorts

Patient Selection
Cohort
Toronto5
JHH6
Monitoring
Outcomes
GS 7, %
Frequency of
Biopsy, Years
993
13
34
6.4
36
1.9
2.8
1,298
5.0
50
0.1
0.4
No. of Patients
Median Follow-up,
Years
10-Year Treated, %
10-Year PCSM, %
Metastatic
Disease, %*
Abbreviations: JHH, Johns Hopkins Hospital; GS, Gleason score; PCSM, prostate cancerspecific mortality.
*Proportion of patients during total follow-up; time-adjusted values not available.
Finally, 6-year follow-up data were reported for 439 men

from the Goteborg randomized trial of prostate cancer
screening who elected AS.36 The cohort composition was
51% very low risk, 36.7% low risk, 21% intermediate risk, and
1.4% high risk. The monitoring protocol included PSA testing
every 3 to 6 months and rebiopsy at varying intervals,
depending on clinical characteristics. Nearly one-half of the
cohort remained free from treatment at 10 years, and a
single patient died of prostate cancer at 12.7 years from
diagnosis. The most common reason for discontinuing AS
was an increase in grade or cancer involvement on repeat
biopsy, followed by increases in PSA.
In summary, these results demonstrate the durability and
safety of AS with careful patient selection and follow-up. At 5
years, the majority of men enrolling in these programs remain free from treatment, allowing preservation of quality
of life. Furthermore, the development of metastatic disease
or prostate cancer death is rare within the first 10 to 15 years
of AS. Available data suggest that inclusion of higher-risk
men and less frequent monitoring biopsy may be associated
with increased risk of adverse outcomes. An understanding
of these principles is crucial to appropriate patient education
and counseling.
UPDATED INTERNATIONAL SOCIETY OF

UROLOGIC PATHOLOGY GRADING AND
IMPLICATIONS ON ACTIVE SURVEILLANCE
In 1966, Donald Gleason37 developed the classification of
prostatic carcinomas using the architectural pattern rather
than cytology for assigning the grade. The underlying
principles of the Gleason grading system and its contributions to prostate cancer clinical management retain relevance and influence over half a century from the time of
its development. However, a number of new clinical and
pathologic discoveries, changes in prostate cancer screening
and detection, and development of new methodologies
justified revisions of the original grading system at the 2005
and 2014 International Society of Urologic Pathology (ISUP)
consensus conferences.38,39
As it relates to AS, there are several major changes to the
Gleason grading system. The reporting of Gleason scores 25
has virtually disappeared from current clinical practice. In
Gleasons original data, Gleason scores 25 were seen in
27.9% of patients.40 Pathologists rendered a diagnosis of
Gleason score 24 in 22% to 24% of patients in the early
1990s compared with just 1.6% to 2.4% a decade later.41-43
Helpap and Egevad44 demonstrated that from 1996 to 2000
and then to 2005, reported Gleason scores 24 decreased

from 2.7% to 0% and reported Gleason score 5 decreased
from 12.2% to 0.3%.
Another major divergence from the original Gleason
system is in the assignment of grade to cribriform glands.
Within Gleasons original illustrations of his cribriform
pattern 3, he depicts large cribriform glands.45 Cases graded
prior to 2005 as Gleason pattern 3 included large cribriform
glands that today would uniformly be called Gleason pattern
4.46,47 Historically, a diagnosis of Gleason score 6 cancer was
not as predictive of good behavior, with a higher rate of
progression and some men even dying of prostate cancer.47,48 Currently, it is recommended that all cribriform
glands be considered as Gleason pattern 4. Ill-defined glands
with poorly formed glandular lumina were not discussed or
depicted by Gleason as either Gleason pattern 3 or 4. It was
the consensus of the 2005 conference that poorly formed
glands should not be considered Gleason pattern 3. Consequently, only individual well-formed glands are currently
graded as Gleason pattern 3.
The consequence of the above changes is that contemporaneously graded Gleason score 3 + 3 = 6 cancers are
incapable of metastatic behavior.47 Although occasional
publications (usually from large prostate cancer databases)
infrequently report otherwise, these reports are subjected
to incomplete submission of the prostate for histologic
examination, errors in data records, use of the older grading
system, and lack of recorded tertiary patterns, among other
limitations.49 Gleason pattern 3 in the setting of surrounding
Gleason pattern 4 may have metastatic potential as recently
shown in a case report supported by molecular evidence of
the clonal origin of the lethal cancer from a focus of Gleason
pattern 3 in the middle of higher-grade cancer.50
There has been a call to change the grading system for
prostate cancer, replacing Gleason score 3 + 3 = 6 disease
(and lower grades) with the term indolent lesion of epithelial origin to mitigate anxiety and curtail overtreatment
of potentially indolent cancers.51 There exist numerous
clinical and morphologic reasons why Gleason pattern 3
should be classified as cancer.52 Rather than changing
Gleason score 6 to a noncancerous diagnosis, there is a
need to change the confusing and inconsistent manner in
which the pathologic grade is reported. One of the major
deficiencies of the current Gleason grading system is that the
lowest grade assigned in the contemporary setting is 6,
despite a scale which ranges from 2 to 10; such practice
implies that a Gleason score 6 is in the middle of the grading
e237
TOSOIAN ET AL
scale in terms of aggressiveness. Similarly, patients with

Gleason score 3 + 4 = 7 may be overly concerned because a
score of 7 is closer to the maximum score of 10 than the
lowest score of 2.
To address these deficiencies with the original Gleason
grading system, along with other problems not relevant to
the practice of AS, a new grading system has been proposed.
This system is founded in the original Gleason system, yet it
is based on extensive subsequent research that has improved the original system in its definition and application. If
one were developing a new prostate cancer grading system
de novo, the goal would be a simple system with the fewest
number of distinct grades that can adequately represent
established differences in prognosis.
The updated grade groups were originally developed in
2013 by Dr. Jonathan Epstein based on data from 7,869
patients who underwent radical prostatectomy at the Johns
Hopkins Hospital.19 This contemporary grading system was
subsequently validated in a cohort of 20,845 patients from
five academic institutions.20 Based on these data, the rates of
5-year biochemical recurrence-free survival for grade groups
15 based on radical prostatectomy grade were 96%, 88%,
63%, 48%, and 26%, respectively. The grade groups were also
predictive of outcomes based on biopsy grade when biopsy
was followed by radical prostatectomy or radiation therapy.
These grade groups were recently validated in a populationbased setting using data from the National Prostate Cancer
Registry of Sweden.53 With the new grading system, patients
can be assured that they have a grade group 1 of 5, which is
the lowest grade, or a grade group 2 of 5, which is still a
relatively low grade. The new grades would, for the foreseeable future, be used in conjunction with the Gleason
system. For example, Gleason score 3 + 3 = 6 is equivalent to
grade group 1. These definitions are listed in Table 2.
Another recommendation of the 2014 ISUP Consensus
conference was for pathologists to record the percentage of
Gleason pattern 4 present in Gleason score 7 cancers. This
recommendation is directly related to the practice of AS.
Traditionally, low-volume Gleason score 3 + 3 = 6 cancers
have been widely considered for a surveillance approach.
Depending on patient age, comorbidity, and other clinicopathologic characteristics, however, there may be a proportion of Gleason score 3 + 4 = 7 cancers that could be
reasonably considered for AS if pattern 4 disease is minimal.
Currently, this information is not transparent in pathology
reports, in which the percentage pattern 4 in Gleason score
3 + 4 = 7 cancers can range from 1% to close to 50%. Based on
these and other improvements, the new grading system has
been accepted by the World Health Organization for the
2016 edition of Pathology and Genetics of Tumours of the
Urinary System and Male Genital Organs.39,54
MULTIPARAMETRIC MRI IN ACTIVE

SURVEILLANCE
Multiparametric MRI offers the ability to detect, measure, and
monitor prostate cancers in vivo, as well as direct biopsies to a
specific target or lesion.17 Currently, the main role of mpMRI in
e238
TABLE 2. Histologic Definition of the Updated Grading

System
Grade
Group
Gleason Score
Equivalent
Characteristic Features
3+3=6
Only individual discrete well-formed glands
3+4=7
Predominantly well-formed glands with

a lesser component of poorly formed/
fused/cribriform glands
4+3=7
Predominantly poorly formed/fused/cribriform glands with a lesser component of

well-formed glands*
Only poorly formed/fused/cribriform

glands, predominantly well-formed
glands and a lesser component lacking
glands, or predominantly lacking glands
and a lesser component of well-formed
glands
910
Lack of gland formation (or necrosis) with or

without poorly formed/fused/cribriform
glands*
*For patients with greater than 95% poorly formed/fused/cribriform glands or lack of
glands on a core or at RP, the component of less than 5% well-formed glands is not
factored into the grade.
Poorly formed/fused/cribriform glands can be a more minor component.
AS is to determine the appropriateness of patient candidates.55

For instance, patients with apparently small, low-grade tumors
on random biopsy may in fact harbor larger, clinically significant
cancers that would render them inappropriate for AS. If
concerning lesions are detected on mpMRI, they can be accurately sampled using various image-guided methods such as
transrectal ultrasonography (TRUS)/MRI fusion.
Multiparametric MRI has a high negative predictive value
for intermediate- and high-risk prostate cancers. Thus, a
negative MRI scan is good evidence that a clinically significant cancer is not present. Furthermore, when a lesion is
seen, mpMRI can provide insight regarding tumor behavior.56,57 For instance, mpMRI has been used to predict
Gleason scores mainly through apparent diffusion coefficient (ADC) values derived from diffusion-weighted MRI.
Although there are strong inverse correlations between ADC
and Gleason grade, mpMRI cannot be considered a replacement for biopsy at this time.58,59 Nevertheless, ADC
values can provide useful information in stratifying patient
risk prior to biopsy. For instance, a lesion with very low ADC
values that is read as a low-grade tumor on biopsy should be
considered for rebiopsy based on the high risk of such a
lesion harboring a higher-grade tumor.
Despite its utility in lesion detection and targeting, mpMRI
does not routinely appear in decision-making algorithms and
clinical nomograms for AS. To date, there are no long-term
prospective studies regarding the use of mpMRI in AS. Using a
220 patient cohort, Shukla-Dave et al60 developed a nomogram based on clinicopathological data and mpMRI findings;
this nomogram demonstrated a high area under the receiver
operating characteristic curve (AUC) of 0.854 for selecting
patients appropriate for AS. A follow-up study of 181 patients
revealed an AUC of only 0.738 for low-risk disease (defined
as # pT2, Gleason grade , 4, and tumor volume # 0.5 cc)
for those who underwent radical prostatectomy.61 The

National Cancer Institute nomogram, which uses only
mpMRI criteria, generated an AUC of 0.71 for predicting
candidates for AS in a cohort of 85 patients, 60 of whom had
very low-risk disease (# cT1c, PSA density , 0.15, biopsy
Gleason score # 6, # 2 positive biopsy cores, and # 50%
cancer involvement in any biopsy core).62 This nomogram was
developed with biopsy pathology as an endpoint, which is a
method that has not been validated. Thus, although mpMRI
appears to provide some incremental value in selecting
candidates for AS, it is far from perfect and remains limited by
an inability to consistently detect clinically significant lesions,
with reported false-negative rates of up to 16%.63
The utility of mpMRI for monitoring men on AS has not
been well established. Walton-Diaz et al64 reported a series
of 58 patients on AS with a median follow-up of 16.1 months
who underwent serial mpMRI with TRUS/MRI fusion-guided
biopsy. Upgrading to Gleason score 3 + 4 = 7 was documented for 17 men (29%), and the positive and negative
predictive values of mpMRI for Gleason score progression
were 53% (95% CI, 28%77%) and 80% (95% CI, 65%91%),
respectively. One recent retrospective study by Felker et al65
included 49 consecutive men with Gleason score 6 prostate
cancer who underwent mpMRI and targeted prostate biopsy
at baseline and again more than 6 months later. Over a mean
follow-up of 28.3 months (range, 1143 months), Gleason
score progression occurred for 19 patients (39%). The addition of serial mpMRI improved the detection of Gleason
score progression during follow-up.
Preliminary data indicate that mpMRI and targeted biopsy
provide incremental value in the selection of candidates for
AS. Performing mpMRI prior to entering AS can reduce the
number of inappropriate AS candidates by 29% (Figs. 1 and
2).62 Once a patient has initiated AS, mpMRI can be useful in
demonstrating lesion stability (Fig. 3) but currently lacks the
sensitivity to reliably detect higher-risk lesions during followup. As such, continued serial biopsies are recommended
until mature data indicate that a stable mpMRI may in fact
preclude subsequent biopsy. Ultimately, additional studies
with longer follow-up are needed to establish the role of
mpMRI in the context of monitoring. This practice will
continue to evolve as larger studies are published.
MOLECULAR TESTING IN ACTIVE SURVEILLANCE

Several molecular profiling tests have been presented for use in
the AS setting. Unfortunately, these tools have yet to be
prospectively assessed in AS populations or with repeat longitudinal measures. Based on existing data, however, some of
which are derived from conservatively managed cohorts, it is
both reasonable and biologically plausible to consider some
molecular tests in the AS setting.18 Here, we briefly review the
molecular tests proposed for use in decisions about AS (Table 3).
Oncotype DX: Genomic Prostate Score

The Oncotype DX (Genomic Health, Redwood City, CA)
genomic prostate score (GPS) measures expression of a
17-gene panel, including 12 genes from four pathways associated with carcinogenesis and five reference genes. Gene
expression levels are incorporated into an algorithm yielding
the GPS, which is measured from 0 to 100.66 The GPS is then
considered in the setting of traditional clinicopathologic risk
categories to determine a predicted likelihood of favorable
pathology. In an initial validation study, GPS from 395 men
with low- to low intermediaterisk disease were incorporated with preoperative models including the Cancer of
the Prostate Risk Assessment (CAPRA) score and National
Comprehensive Cancer Network (NCCN) risk categories.67 In
both models, a 20-unit increase in GPS was associated with
an approximate twofold increased odds of adverse pathology at prostatectomy (model with CAPRA: odds ratio
[OR], 2.1; 95% CI, 1.43.2; model with NCCN risk classification: OR,1.9; 95% CI, 1.32.8), defined as primary Gleason
pattern 4, any Gleason pattern 5, or nonorgan-confined
disease. In a subsequent study of 402 men with low to
intermediate-risk disease,68 a 20-point increase in GPS
similarly conveyed a significant increase in risk of biochemical recurrence after treatment (hazard ratio [HR] 2.93;
95% CI, 2.034.15; p , .001) over a median follow-up of 5.2
years; GPS was also an independent predictor of biochemical
recurrence (BCR) in multivariable models with baseline
clinical factors. Although the test is yet to be assessed in an
AS cohort, one clinical utility study demonstrated a 10%
increased use of AS when GPSs were incorporated into
clinical decision making.69 However, the long-term effect of
these changes in management are unknown.
Prolaris: Cell Cycle Progression

The Prolaris platform (Myriad Genetics, Salt Lake City, UT) is
based on tumor cell proliferation as measured by quantitative reverse-transcription polymerase chain reaction
of a 46-gene panel, consisting of 31 cell cycle genes and
15 housekeeping genes.70 The output is reported as a cell
cycle progression (CCP) score categorized clinically as less
than 0, 01, 12, 23, and greater than 3. CCP scores were
retrospectively assessed for 349 men conservatively
managed within the Transatlantic Prostate Group.71
Ten-year rates of prostate cancer death associated with
CCP score groups less than 0, 01, 12, 23, and greater
than 3 were 19.3%, 19.8%, 21.1%, 48.2%, and 74.9%, respectively. In multivariable analysis, the CCP score (HR for
one-unit change: 1.65; 95% CI, 1.312.09), Gleason score
greater than 7 (HR 1.90; 95% CI, 1.183.07), and PSA (HR
1.37; 95% CI, 1.051.79) significantly predicted prostate
cancer death. Perhaps most pertinent to the AS setting,
however, the CCP score was not independently predictive of death for men with Gleason score 6 cancers (HR
1.30; 95% CI, 0.612.77). Two subsequent analyses
have been performed among treated populations. In one
multi-institutional study of men treated with radical
prostatectomy,72 a single-unit increase in CCP score on
biopsy was associated with increased risk of BCR (HR 1.47;
95% CI, 1.231.76) and metastasis (HR 4.19; 95% CI,
2.088.45) on multivariable analysis. In men treated with
e239
TOSOIAN ET AL
FIGURE 1. Use of MRI Prior to Enrollment in Active Surveillance
A 71-year-old man with a serum PSA of 5.47 ng/ml was found to have Gleason score 3 + 3 cancer (5%) on random biopsy. Due to rising PSA levels, he underwent MRI scan
prior to enrollment in active surveillance. (A) An axial T2-weighted MRI scan shows a PI-RADS 5 lesion in the left mid-anterior transition zone (arrow). (BD) The lesion is
positive on the apparent diffusion coefficient map (B), diffusion-weighted MRI (b = 2,000 mm/s2) (C), and dynamic contrast-enhanced MRI (D; arrows). A TRUS/MRI fusion-guided
biopsy revealed a high volume Gleason 3 + 4 score (100% core involvement with cancer with 30% Gleason pattern 4 of the entire cancer). The patient became an active treatment
candidate after multiparametric MRI and a TRUS/MRI fusion-guided biopsy approach.
Abbreviations: PI-RADS, Prostate Imaging Reporting and Data System; PSA, prostate-specific antigen; TRUS, transrectal ultrasonography.
radiation therapy, a one-unit increase in the CCP score was

similarly associated with a twofold increased risk of BCR
(HR 2.11; 95% CI, 1.054.25).73 Based on surveys of ordering physicians, use of the CCP score led to a change in
management selection in one-third to two-thirds of cases;
however, as with the clinical utility studies of Oncotype DX,
the long-term effect of these changes in management
based on Prolaris results is unknown.74,75
ProMark
The ProMark test (Metamark Genetics Inc, Augusta, GA) is
based on a quantitative proteomics profile uniquely
designed to yield prognostic value regardless of sampling
error. To do this, two tissue microarrays (TMAs) were derived from a series of prostatectomy specimens, one using
the highest Gleason pattern observed in the specimen and a
second using the lowest.76,77 Both TMAs were tested using a

12-protein panel that demonstrated similar ability to discriminate tumor aggressiveness (Gleason score $ 7, pT3b,
N1, or M1) based on the AUC (using the highest Gleason
pattern: AUC, 0.70; 95% CI, 0.620.77; using the lowest
Gleason pattern: AUC, 0.72; 95% CI, 0.640.79).77 This initial
panel was reduced to an eight-protein assay and validated
for predicting unfavorable pathology (Gleason score $ 4 + 3
or nonorgan-confined disease) in a cohort of matched
biopsy and prostatectomy specimens.78 Remarkably, using a
threshold score of less than 0.33 the test was 90% sensitive
for predicting favorable pathology, and above a score of 0.8
the test was 95% specific for unfavorable pathology. On
multivariable analysis, a score increase of 0.25 was associated with 3.1-fold increased odds of unfavorable disease
(OR, 3.1; 95% CI, 2.24.4).
FIGURE 2. Use of Various Modalities to Determine Treatment Approach
A comparison of pathology-defined candidates (top row) for active treatment (yellow) and active surveillance (orange) versus definitions based on multiparametric MRI, the
Epstein criteria, the DAmico criteria, and CAPRA score. Each patient is represented by a column of the array. Note that multiparametric MRI results most closely mirrored pathology
results with fewest false-positive decisions for active treatment in good active surveillance candidates.
Abbreviations: CAPRA, Cancer of the Prostate Risk Assessment; MP, multiparametric.
Reprinted from Turkbey et al55 with permission from the publisher.
e240
FIGURE 3. Use of MRI in Monitoring a Prostate Lesion During Active Surveillance
A 59-year-old man presenting with a serum PSA of 4.68 ng/ml. (AC) An axial T2-weighted MRI scan (A), apparent diffusion coefficient map (B), and b1500 diffusion-weighted MRI
(C) show a lesion in the left mid-peripheral zone (arrows). This lesion was biopsied and found to harbor Gleason 3 + 3 prostate cancer (10%). The patient elected active surveillance.
(DF) A 1-year follow-up MRI scan (serum PSA = 5.10 ng/ml) shows that the lesion is stable in size and MRI signal features. The follow-up biopsy also revealed Gleason 3 + 3 prostate
cancer (15%). MRI may become a tool for monitoring patients on active surveillance.
Abbreviation: PSA, prostate-specific antigen.
Phosphatase and Tensin Homolog Testing

The phosphatase and tensin homolog (PTEN) gene is located
on chromosome 10q and functions as a tumor suppressor in
the phosphoinositide 3-kinase/AKT pathway.79 PTEN inactivation is an early event in carcinogenesis observed in up
to 40% of prostate cancers. Specifically, PTEN loss appears to
be associated with increasing likelihood of advanced grade
and stage, as well as early recurrence after treatment.80-82 As
such, PTEN loss detected by either fluorescence in situ
hybridization or immunohistochemical staining has been
proposed for prognostic use. One report from the Transatlantic Prostate Group investigated PTEN in men with
conservatively managed prostate cancer with favorable
disease characteristics.83 In the population of men with a
Gleason score below 7, PTEN was highly predictive of
prostate cancer death (HR 8.13; 95% CI, 2.823.2). Limiting

its utility, however, PTEN loss was only present in 3% of this
population. Other contemporary studies have described
PTEN loss in 11% to 12% of favorable-risk study populations.84,85 Findings from these studies have consistently
supported an association between PTEN loss on biopsy and
adverse clinicopathologic outcomes.84,85 Considering a
strong correlation with poor outcomes but its less than 15%
prevalence in potential AS candidates, there may be a role
for PTEN testing whereby PTEN loss is considered a high-risk
feature that may preclude management on AS.
Current Status of Molecular Testing

The true promise of molecular tests lies in their ability to
both improve upon and expand beyond biopsy. In addition
TABLE 3. Molecular Tests Proposed for Use in Active Surveillance

Test
Manufacturer
Platform
Biologic Process
Endpoints Assessed
Oncotype DX GPS
Genomic Health (Redwood City, CA)
qRT-PCR
RNA quantification (gene expression)
Likelihood of favorable
pathology at RP*
Prolaris CCP score
Myriad Genetics (Salt Lake City, UT)
qRT-PCR
RNA quantification (gene expression)
10-year risk of PCSM
ProMark
Metamark Genetics Inc. (Augusta, GA)
QMPI
Protein quantification
Likelihood of unfavorable
pathology
Abbreviations: GPS, genomic prostate score; CCP, cell cycle progression; qRT-PCR, quantitative reverse-transcription polymerase chain reaction; QMPI, quantitative multiplex
proteomics imaging; RP, radical prostatectomy; PCSM, prostate cancerspecific mortality.
*Freedom from: primary Gleason pattern 4, any Gleason pattern 5, and nonorgan-confined disease.
Gleason score of 4 + 3 or greater or nonorgan-confined disease ($ pT3, N1, or M1).
e241
TOSOIAN ET AL
to practical considerations, the use of biopsy is limited by

sampling error and a dependence on subjective grading.86
Conversely, Long et al87 have demonstrated consistent gene
expression across multiple biopsy cores of a single patient,
supporting previous work revealing limited variation in gene
expression levels of highly expressed genes across multiple
biopsy cores, including both malignant and benign stromal
tissue.88 As others have noted, such consistent gene expression, despite tissue and tumor heterogeneity, supports
the notion that one can be confident in the overall prostate
biology based on analysis of a limited sample.89
Certainly there are many reasons for optimism in exploring
molecular tests in the AS setting. Contrary to pathologic
grading, molecular tests provide an objective result that
remains consistent across institutions. Beyond the platforms
discussed above, additional tests continue to emerge and
may prove useful. For example, the Decipher genomic
classifier (GenomeDx Biosciences, Vancouver, BC, Canada),
which has been previously validated for predicting metastasis after surgery based on prostatectomy tissue,90-92 has
more recently demonstrated prognostic value in biopsy
specimens.93 At the same time, the majority of current
evidence has been gleaned from retrospective analyses of
cohorts that were not established with a surveillance approach in mind. Furthermore, these tests remain expensive
and their incremental value in the context of other diagnostics such as serum markers and MRI is unknown. For
example, the serum Prostate Health Index has demonstrated
predictive ability using baseline and longitudinal samples in
the AS setting and currently represents a less-invasive, more
affordable option.94,95 Acknowledging the entirety of the
data, we lack a definitive answer to the most fundamental
questions about molecular testing: does it work, and is it
cost-effective? Additional evidence for these important
questions is necessary and will ideally stem from prospectively designed studies.
ACTIVE SURVEILLANCE: PRESENT AND FUTURE

An abundance of data indicate that the use of AS is rising
rapidly and that it is a safe option to minimize overtreatment
in men with favorable-risk prostate cancer.8 Nonetheless,
based on current methods of patient selection and monitoring, more than one in five men eligible for AS will have
evidence of more aggressive disease on prostatectomy.96
Conversely, some men who fail to meet conventional AS
criteria may in fact harbor indolent cancers that will not

threaten their quantity or quality of life. Thus, there is
substantial need for more accurate diagnostic and prognostic tools to identify both those patients who are appropriate candidates for AS and those who are not.
We have herein outlined the updated grading system for
prostate cancer as determined according to the ISUP consensus conference, most recently in 2014. As in the past,
prognostic grade groups were determined based on longterm oncological outcomes. Perhaps most clinically important is the conversion from a complex Gleason score scale,
traditionally measured from 2 to 10, to a straightforward 15
scale. This modification should allow that the information
conveyed by pathologic grade is more clearly understood by
both patients and physicians. In the realm of AS, the most
substantial change in pathologic grading is the recommendation for uniform recording of the percent pattern 4 in
Gleason score 7 cancers. Although the safety of AS in higherrisk populations is not well established, the prognostic risk
associated with Gleason score 3 + 4 = 7 cancer varies substantially between samples with 1% Gleason pattern 4 and
50% Gleason pattern 4.97,98 Measuring such differences will
enable better understanding of these risks, and, when
clinically appropriate, consideration of careful expansion of
AS to populations with minimal Gleason pattern 4 disease.
We are at a critical point in the evolution of AS. Emerging
technologies such as mpMRI and molecular testing undoubtedly add to the bevy of tools available for patient
selection and monitoring. These tools, however, currently
lack a template for use. Now more than ever, prostate cancer
is diagnosed using TRUS/MRI fusion-guided biopsy, yet there
exists no standard by which cancers diagnosed on targeted
biopsy can be incorporated into conventional risk classification schemes such as that of the NCCN. Similarly, more
complex issues remain unaddressed, such as when and how
to incorporate mpMRI and genomic tests into the AS paradigm. A body of retrospective data exists to begin shaping
these approaches, but more evidence is needed. With
greater experience, the paradigm for incorporating these
tests into current management approaches should continue
to take shape.
A future in which AS protocols are largely based on mpMRI
and molecular testing is easy to envision, but many questions
lie between here and there. How soundly these questions are
answered will dictate ifand whenwe arrive.
References
1. Choo R, Klotz L, Danjoux C, et al. Feasibility study: watchful waiting
for localized low to intermediate grade prostate carcinoma with
selective delayed intervention based on prostate specific antigen, histological and/or clinical progression. J Urol. 2002;167:
1664-1669.
2. Carter HB, Walsh PC, Landis P, et al. Expectant management of nonpalpable prostate cancer with curative intent: preliminary results.
J Urol. 2002;167:1231-1234.
e242
3. Heidenreich A, Bastian PJ, Bellmunt J, et al; European Association of

Urology. EAU guidelines on prostate cancer. part 1: screening, diagnosis, and local treatment with curative intent-update 2013. Eur Urol.
2014;65:124-137.
4. Mohler JL, Armstrong AJ, Bahnson RR, et al; National Comprehensive
Cancer Network. NCCN Clinical Practice Guidelines in Oncology (NCCN
Guidelines): Prostate Cancer, version 1. Fort Washington, PA: National
Comprehensive Cancer Network Inc; 2015.
5. Klotz L, Vesprini D, Sethukavalan P, et al. Long-term follow-up of a large

active surveillance cohort of patients with prostate cancer. J Clin Oncol.
2015;33:272-277.
6. Tosoian JJ, Mamawala M, Epstein JI, et al. Intermediate and longer-term
outcomes from a prospective active-surveillance program for favorablerisk prostate cancer. J Clin Oncol. Epub 2015 Aug 31.
7. Bergman J, Litwin MS. Quality of life in men undergoing active surveillance for localized prostate cancer. J Natl Cancer Inst Monogr. 2012;
2012:242-249.
8. Tosoian JJ, Carter HB, Lepor A, et al. Active surveillance for prostate
cancer: current evidence and contemporary state of practice. Nat Rev
Urol. 2016. In press.
9. Loeb S, Vellekoop A, Ahmed HU, et al. Systematic review of complications of prostate biopsy. Eur Urol. 2013;64:876-892.
10. Bokhorst LP, Lepisto I, Kakehi Y, et al. Complications after prostate
biopsies in men on active surveillance and its effect on receiving further
biopsies in the Prostate Cancer Research International: Active Surveillance (PRIAS) study. BJU Int. Epub 2016 Jan 14.
11. Filippou P, Welty CJ, Cowan JE, et al. Immediate versus delayed radical
prostatectomy: updated outcomes following active surveillance of
prostate cancer. Eur Urol. 2015;68:458-463.
12. Tosoian JJ, Sundi D, Trock BJ, et al. Pathologic outcomes in favorable-risk
prostate cancer: comparative analysis of men electing active surveillance and immediate surgery. Eur Urol. Epub 2015 Oct 5.
13. Stephenson SK, Chang EK, Marks LS. Screening and detection advances
in magnetic resonance image-guided prostate biopsy. Urol Clin North
Am. 2014;41:315-326.
14. Vargas HA, Hotker AM, Goldman DA, et al. Updated prostate imaging
reporting and data system (PIRADS v2) recommendations for the detection of clinically significant prostate cancer using multiparametric
MRI: critical evaluation using whole-mount pathology as standard of
reference. Eur Radiol. Epub 2015 Sep 22.
15. Barbieri CE, Bangma CH, Bjartell A, et al. The mutational landscape of
16. van den Bergh RCN, Ahmed HU, Bangma CH, et al. Novel tools to
improve patient selection and monitoring on active surveillance for
low-risk prostate cancer: a systematic review. Eur Urol. 2014;65:
1023-1031.
17. Turkbey B, Brown AM, Sankineni S, et al. Multiparametric prostate
magnetic resonance imaging in the evaluation of prostate cancer. CA
Cancer J Clin. Epub 2015 Nov 23.
18. Ross AE, DAmico AV, Freedland SJ. Which, when and why? Rational use
of tissue-based molecular testing in localized prostate cancer. Prostate
Cancer Prostatic Dis. 2016;19:1-6.
19. Pierorazio PM, Walsh PC, Partin AW, et al. Prognostic Gleason grade
grouping: data based on the modified Gleason scoring system. BJU Int.
2013;111:753-760.
20. Epstein JI, Zelefsky MJ, Sjoberg DD, et al. A contemporary prostate
cancer grading system: a validated alternative to the Gleason score. Eur
Urol. 2016;69:428-435.
21. Cooperberg MR, Broering JM, Carroll PR. Time trends and local variation
in primary treatment of localized prostate cancer. J Clin Oncol. 2010;28:
1117-1123.
22. Carter HB. Active surveillance for prostate cancer: an underutilized
opportunity for reducing harm. J Natl Cancer Inst Monogr. 2012;2012:
175-183.
23. Cooperberg MR, Broering JM, Kantoff PW, et al. Contemporary trends in
low risk prostate cancer: risk assessment and treatment. J Urol. 2007;
178:S14-S19.
24. Ingimarsson JP, Celaya MO, Laviolette M, et al. Trends in initial management of prostate cancer in New Hampshire. Cancer Causes Control.
2015;26:923-929.
25. Womble PR, Montie JE, Ye Z, et al; Michigan Urological Surgery Improvement Collaborative. Contemporary use of initial active surveillance among men in Michigan with low-risk prostate cancer. Eur Urol.
2015;67:44-50.
26. Cooperberg MR, Carroll PR. Trends in management for patients with
localized prostate cancer, 1990-2013. JAMA. 2015;314:80-82.
27. Louis AS, Kalnin R, Maganti M, et al. Oncologic outcomes following
radical prostatectomy in the active surveillance era. Can Urol Assoc J.
2013;7:E475-E480.
28. Loeb S, Berglund A, Stattin P. Population based study of use and determinants of active surveillance and watchful waiting for low and
intermediate risk prostate cancer. J Urol. 2013;190:1742-1749.
29. Huland H, Graefen M. Changing trends in surgical management of
prostate cancer: the end of overtreatment? Eur Urol. 2015;68:175-178.
30. Weerakoon M, Papa N, Lawrentschuk N, et al. The current use of active
surveillance in an Australian cohort of men: a pattern of care analysis
from the Victorian Prostate Cancer Registry. BJU Int. 2015;115:50-56.
31. Filson CP, Schroeck FR, Ye Z, et al. Variation in use of active surveillance
among men undergoing expectant treatment for early stage prostate
cancer. J Urol. 2014;192:75-81.
32. Mitsuzuka K, Koga H, Sugimoto M, et al. Current use of active surveillance for localized prostate cancer: a nationwide survey in Japan. Int
J Urol. 2015;22:754-759.
33. Mohler JL, Armstrong AJ, Bahnson RR, et al. Prostate cancer, version
1.2016. J Natl Compr Canc Netw. 2016;14:19-30.
34. Welty CJ, Cowan JE, Nguyen H, et al. Extended followup and risk factors
for disease reclassification in a large active surveillance cohort for
localized prostate cancer. J Urol. 2015;193:807-811.
35. Selvadurai ED, Singhera M, Thomas K, et al. Medium-term outcomes of
active surveillance for localised prostate cancer. Eur Urol. 2013;64:
981-987.
36. Godtman RA, Holmberg E, Khatami A, et al. Outcome following active
surveillance of men with screen-detected prostate cancer. Results from
the Goteborg randomised population-based prostate cancer screening
trial. Eur Urol. 2013;63:101-107.
37. Gleason DF, Mellinger GT. Prediction of prognosis for prostatic adenocarcinoma by combined histological grading and clinical staging.
J Urol. 1974;111:58-64.
38. Epstein JI, Allsbrook WC Jr, Amin MB, et al; ISUP Grading Committee.
The 2005 International Society of Urological Pathology (ISUP) Consensus Conference on Gleason Grading of Prostatic Carcinoma. Am J
Surg Pathol. 2005;29:1228-1242.
39. Epstein JI, Egevad L, Amin MB, et al; Grading Committee. The 2014
International Society of Urological Pathology (ISUP) Consensus Conference on Gleason Grading of Prostatic Carcinoma: definition of
grading patterns and proposal for a new grading system. Am J Surg
Pathol. 2016;40:244-252.
40. Gleason DF, Tannenbaum M (eds). Histological Grading and Staging
of Prostatic Carcinoma: Urologic Pathology. Philadelphia, PA: Lea &
Feibiger; 1977.
41. Ghani KR, Grigor K, Tulloch DN, et al. Trends in reporting Gleason score
1991 to 2001: changes in the pathologists practice. Eur Urol. 2005;47:
196-201.
42. Fine SW, Epstein JI. A contemporary study correlating prostate needle
biopsy and radical prostatectomy Gleason score. J Urol. 2008;179:
1335-1338, discussion 1338-1339.
43. Steinberg DM, Sauvageot J, Piantadosi S, et al. Correlation of prostate
needle biopsy and radical prostatectomy Gleason grade in academic
and community settings. Am J Surg Pathol. 1997;21:566-576.
44. Helpap B, Egevad L. Correlation of modified Gleason grading of prostate
carcinoma with age, serum prostate specific antigen and tumor extent
in needle biopsy specimens. Anal Quant Cytol Histol. 2008;30:133-138.
e243
TOSOIAN ET AL
45. Gleason DF. Histologic grading of prostate cancer: a perspective. Hum

Pathol. 1992;23:273-279.
46. McNeal JE, Yemoto CE. Spread of adenocarcinoma within prostatic
ducts and acini. Morphologic and clinical correlations. Am J Surg Pathol.
1996;20:802-814.
47. Ross HM, Kryvenko ON, Cowan JE, et al. Do adenocarcinomas of the
prostate with Gleason score (GS) #6 have the potential to metastasize
to lymph nodes? Am J Surg Pathol. 2012;36:1346-1352.
48. Nakabayashi M, Hayes J, Taplin M-E, et al. Clinical predictors of survival
in men with castration-resistant prostate cancer: evidence that Gleason
score 6 cancer can evolve to lethal disease. Cancer. 2013;119:
2990-2998.
49. Kryvenko ON, Epstein JI. Re: nationwide prevalence of lymph node
metastases in Gleason score 3 + 3 = 6 prostate cancer. Pathology.
2015;47:394.
50. Haffner MC, De Marzo AM, Yegnasubramanian S, et al. Diagnostic
challenges of clonal heterogeneity in prostate cancer. J Clin Oncol. 2015;
33:e38-e40.
51. Esserman LJ, Thompson IM, Reid B, et al. Addressing overdiagnosis and
overtreatment in cancer: a prescription for change. Lancet Oncol. 2014;
15:e234-e242.
52. Carter HB, Partin AW, Walsh PC, et al. Gleason score 6 adenocarcinoma:
should it be labeled as cancer? J Clin Oncol. 2012;30:4294-4296.
53. Loeb S, Folkvaljon Y, Robinson D, et al. Evaluation of the 2015 Gleason
grade groups in a nationwide population-based cohort. Eur Urol. Epub
2015 Dec 17.
54. Eble JN, Sauter G, Epstein JI, et al (eds). Pathology and Genetics of
Tumours of the Urinary System and Male Genital Organs. Lyon, France:
International Agency for Research on Cancer; 2004.
55. Turkbey B, Mani H, Aras O, et al. Prostate cancer: can multiparametric
MR imaging help identify patients who are candidates for active surveillance? Radiology. 2013;268:144-152.
56. Siddiqui MM, Rais-Bahrami S, Turkbey B, et al. Comparison of
MR/ultrasound fusion-guided biopsy with ultrasound-guided biopsy for
the diagnosis of prostate cancer. JAMA. 2015;313:390-397.
57. Panebianco V, Barchetti F, Sciarra A, et al. Multiparametric magnetic
resonance imaging vs. standard care in men being evaluated for
prostate cancer: a randomized study. Urol Oncol. 2015;33:17.e1-17.e7.
58. Turkbey B, Shah VP, Pang Y, et al. Is apparent diffusion coefficient
associated with clinical risk scores for prostate cancers that are visible
on 3-T MR images? Radiology. 2011;258:488-495.
59. Vargas HA, Akin O, Franiel T, et al. Diffusion-weighted endorectal MR
imaging at 3 T for prostate cancer: tumor detection and assessment of
aggressiveness. Radiology. 2011;259:775-784.
60. Shukla-Dave A, Hricak H, Kattan MW, et al. The utility of magnetic
resonance imaging and spectroscopy for predicting insignificant
prostate cancer: an initial analysis. BJU Int. 2007;99:786-793.
61. Shukla-Dave A, Hricak H, Akin O, et al. Preoperative nomograms
incorporating magnetic resonance imaging and spectroscopy for
prediction of insignificant prostate cancer. BJU Int. 2012;109:
1315-1322.
62. Stamatakis L, Siddiqui MM, Nix JW, et al. Accuracy of multiparametric
magnetic resonance imaging in confirming eligibility for active surveillance for men with prostate cancer. Cancer. 2013;119:3359-3366.
63. Serrao EM, Barrett T, Wadhwa K, et al. Investigating the ability of
multiparametric MRI to exclude significant prostate cancer prior to
transperineal biopsy. Can Urol Assoc J. 2015;9:E853-E858.
64. Walton Diaz A, Shakir NA, George AK, et al. Use of serial multiparametric
magnetic resonance imaging in the management of patients with prostate
cancer on active surveillance. Urol Oncol. 2015;33:202.e1-202.e7.
65. Felker ER, Wu J, Natarajan S, et al. Serial MRI in active surveillance of
prostate cancer: incremental value. J Urol. Epub 2015 Dec 7.
e244
66. Knezevic D, Goddard AD, Natraj N, et al. Analytical validation of the

Oncotype DX prostate cancer assay - a clinical RT-PCR assay optimized
for prostate needle biopsies. BMC Genomics. 2013;14:690.
67. Klein EA, Cooperberg MR, Magi-Galluzzi C, et al. A 17-gene assay to
predict prostate cancer aggressiveness in the context of Gleason grade
heterogeneity, tumor multifocality, and biopsy undersampling. Eur
Urol. 2014;66:550-560.
68. Cullen J, Rosner IL, Brand TC, et al. A biopsy-based 17-gene genomic
prostate score predicts recurrence after radical prostatectomy and
adverse surgical pathology in a racially diverse population of men with
clinically low- and intermediate-risk prostate cancer. Eur Urol. 2015;68:
123-131.
69. Badani KK, Kemeter MJ, Febbo PG, et al. The impact of a biopsy based
17-gene genomic prostate score on treatment recommendations in
men with newly diagnosed clinically prostate cancer who are candidates
for active surveillance. Urol Pract. 2015;2:181-189.
70. Cuzick J, Swanson GP, Fisher G, et al; Transatlantic Prostate Group.
Prognostic value of an RNA expression signature derived from cell cycle
proliferation genes in patients with prostate cancer: a retrospective
71. Cuzick J, Berney DM, Fisher G, et al; Transatlantic Prostate Group.
Prognostic value of a cell cycle progression signature for prostate cancer
death in a conservatively managed needle biopsy cohort. Br J Cancer.
2012;106:1095-1099.
72. Bishoff JT, Freedland SJ, Gerber L, et al. Prognostic utility of the cell cycle
progression score generated from biopsy in men treated with prostatectomy. J Urol. 2014;192:409-414.
73. Freedland SJ, Gerber L, Reid J, et al. Prognostic utility of cell cycle
progression score in men with prostate cancer after primary external
beam radiation therapy. Int J Radiat Oncol Biol Phys. 2013;86:848-853.
74. Crawford ED, Scholz MC, Kar AJ, et al. Cell cycle progression score and
treatment decisions in prostate cancer: results from an ongoing registry. Curr Med Res Opin. 2014;30:1025-1031.
75. Shore N, Concepcion R, Saltzstein D, et al. Clinical utility of a biopsybased cell cycle gene expression assay in localized prostate cancer. Curr
Med Res Opin. 2014;30:547-553.
76. Shipitsin M, Small C, Giladi E, et al. Automated quantitative multiplex
immunofluorescence in situ imaging identifies phospho-S6 and
phospho-PRAS40 as predictive protein biomarkers for prostate cancer
lethality. Proteome Sci. 2014;12:40.
77. Shipitsin M, Small C, Choudhury S, et al. Identification of proteomic
biomarkers predicting prostate cancer aggressiveness and lethality
despite biopsy-sampling error. Br J Cancer. 2014;111:1201-1212.
78. Blume-Jensen P, Berman DM, Rimm DL, et al. Development and clinical
validation of an in situ biopsy-based multimarker assay for risk stratification in prostate cancer. Clin Cancer Res. 2015;21:2591-2600.
79. Leslie NR, Downes CP. PTEN function: how normal cells control it and
tumour cells lose it. Biochem J. 2004;382:1-11.
80. Lotan TL, Gurel B, Sutcliffe S, et al. PTEN protein loss by immunostaining: analytic validation and prognostic indicator for a high risk
surgical cohort of prostate cancer patients. Clin Cancer Res. 2011;17:
6563-6573.
81. Krohn A, Diedler T, Burkhardt L, et al. Genomic deletion of PTEN is
associated with tumor progression and early PSA recurrence in ERG
fusion-positive and fusion-negative prostate cancer. Am J Pathol. 2012;
181:401-412.
82. Liu W, Xie CC, Thomas CY, et al. Genetic markers associated with early
cancer-specific mortality following prostatectomy. Cancer. 2013;119:
2405-2412.
83. Cuzick J, Yang ZH, Fisher G, et al; Transatlantic Prostate Group. Prognostic value of PTEN loss in men with conservatively managed localised
prostate cancer. Br J Cancer. 2013;108:2582-2589.
84. Lotan TL, Carvalho FL, Peskoe SB, et al. PTEN loss is associated with
upgrading of prostate cancer from biopsy to radical prostatectomy.
Mod Pathol. 2015;28:128-137.
85. Mithal P, Allott E, Gerber L, et al. PTEN loss in biopsy tissue predicts poor
clinical outcomes in prostate cancer. Int J Urol. 2014;21:1209-1214.
86. Amin MB, Lin DW, Gore JL, et al. The critical role of the pathologist in
determining eligibility for active surveillance as a management option in
patients with prostate cancer: consensus statement with recommendations supported by the College of American Pathologists, International Society of Urological Pathology, Association of Directors of
Anatomic and Surgical Pathology, the New Zealand Society of Pathologists, and the Prostate Cancer Foundation. Arch Pathol Lab Med.
2014;138:1387-1405.
87. Long Q, Xu J, Osunkoya AO, et al. Global transcriptome analysis of
formalin-fixed prostate cancer specimens identifies biomarkers of
disease recurrence. Cancer Res. 2014;74:3228-3237.
88. Peng Z, Andersson K, Lindholm J, et al. Operator dependent choice of
prostate cancer biopsy has limited impact on a gene signature analysis
for the highly expressed genes IGFBP3 and F3 in prostate cancer epithelial cells. PLoS One. 2014;9:e109610.
89. Reichard CA, Stephenson AJ, Klein EA. Applying precision medicine to
the active surveillance of prostate cancer. Cancer. 2015;121:3403-3411.
90. Erho N, Crisan A, Vergara IA, et al. Discovery and validation of a prostate
cancer genomic classifier that predicts early metastasis following radical
prostatectomy. PLoS One. 2013;8:e66855.
91. Karnes RJ, Bergstralh EJ, Davicioni E, et al. Validation of a genomic

classifier that predicts metastasis following radical prostatectomy in an
at risk patient population. J Urol. 2013;190:2047-2053.
92. Den RB, Feng FY, Showalter TN, et al. Genomic prostate cancer classifier
predicts biochemical failure and metastases in patients after postoperative
radiation therapy. Int J Radiat Oncol Biol Phys. 2014;89:1038-1046.
93. Klein EA, Haddad Z, Yousefi K, et al. Decipher genomic classifier measured on
prostate biopsy predicts metastasis risk. Urology. Epub 2016 Jan 22.
94. Tosoian JJ, Loeb S, Feng Z, et al. Association of [-2]proPSA with biopsy
reclassification during active surveillance for prostate cancer. J Urol.
2012;188:1131-1136.
95. Hirama H, Sugimoto M, Ito K, et al. The impact of baseline [-2]proPSArelated indices on the prediction of pathological reclassification at
1 year during active surveillance for low-risk prostate cancer: the
Japanese multicenter study cohort. J Cancer Res Clin Oncol. 2014;140:
257-263.
96. Tosoian JJ, JohnBull E, Trock BJ, et al. Pathological outcomes in men with
low risk and very low risk prostate cancer: implications on the practice
of active surveillance. J Urol. 2013;190:1218-1223.
97. Cooperberg MR, Cowan JE, Hilton JF, et al. Outcomes of active surveillance for men with intermediate-risk prostate cancer. J Clin Oncol.
2011;29:228-234.
98. Han M, Partin AW, Zahurak M, et al. Biochemical (prostate specific
antigen) recurrence probability following radical prostatectomy for
clinically localized prostate cancer. J Urol. 2003;169:517-523.
e245
Oligometastatic Disease in
Prostate Cancer: Treating the
Patient or the Scan?
CHAIR
Neha Vapiwala, MD
University of Pennsylvania
Philadelphia, PA
SPEAKERS
Christopher J. Sweeney, MBBS
Dana-Farber Cancer Institute/Harvard Cancer Center
Boston, MA
R. Jeffrey Karnes, MD
Mayo Clinic
Rochester, MN
APPROACH TO OLIGOMETASTATIC PROSTATE CANCER
Approach to Oligometastatic Prostate Cancer

Brandon Bernard, MD, Boris Gershman, MD, R. Jeffrey Karnes, MD, Christopher J. Sweeney, MBBS, and
Neha Vapiwala, MD
OVERVIEW
Oligometastatic prostate cancer has increasingly been recognized as a unique clinical state with therapeutic implications. It
has been proposed that patients with oligometastases may have a more indolent course and that outcome may be further
improved with metastasis-directed local ablative therapy. In addition, there are differing schools of thoughts regarding
whether oligometastases represent isolated lesionswhere targeted therapy may render a patient disease freeor
whether they coexist with micrometastases, where targeted therapy in addition to systemic therapy is required for maximal
clinical impact. As such, the approach to the patient with oligometastatic prostate cancer requires multidisciplinary
consideration, with surgery, radiotherapy, and systemic therapy potentially of benefit either singularly or in combination.
Indeed, mounting evidence suggests durable disease-free intervals and, in some cases, possibly cure, may be achieved with
such a multimodal strategy. However, selecting patients that may benefit most from treatment of oligometastases is an
ongoing challenge. Moreover, with the advent of new, highly sensitive imaging technologies, the spectrum based on CT of
the abdomen and pelvis and technetium bone scan of localized to oligometastatic to widespread disease has become
increasingly blurred. As such, new MRI- and PET-based modalities require validation. As some clinical guidelines advise
against routine prostate-specific antigen screening, the possibility of more men presenting with locally advanced or de novo
oligometastatic prostate cancer exists; thus, knowing how best to treat these patients may become more relevant at a
population level. Ultimately, the arrival of prospective clinical data and better understanding of biology will hopefully further
inform how best to treat men with this disease.
ligometastatic prostate cancer is increasingly being

considered as a unique clinical situation and challenge.
Historically, distinctions with respect to lesser or greater
extent of metastatic disease burden were not typically
made, as patients were all treated with androgen deprivation therapy (ADT). Men with oligometastatic prostate
cancer were thus deemed incurable and treated systemically
with ADT alongside those with widespread metastases.
However, recent advances in imaging are identifying men
with potentially isolated metastases, and mounting evidence suggests that durable control is attainable with
treatment modalities targeting oligometastases, either with
or without the use of systemic therapy.1-3 Such data support
the hypothesis that oligometastatic prostate cancer may
represent a unique biologic state with its own natural history
requiring a distinct therapeutic approach.4 That said, there
are currently no clear-cut data or validated tools to guide
optimal therapy for an individual patient. One school of
thought is to consider isolated lesions on imaging as representing the only sites of disease, where local therapy is
sufficient. The other viewpoint is that oligometastatic
disease is most likely also associated with micrometastatic

disease; therefore, concurrent systemic therapy with localized therapy should be considered the optimal treatment.
In the latter case, disease at this stage is considered potentially curable as it is with adjuvant systemic therapy (in
the form of ADT) to radiotherapy (RT) for high-risk localized
disease or microscopic lymph nodepositive disease following radical prostatectomy.5-8
With this uncertainty, there is provider and patient bias
toward certain treatmentsoptions include ADT alone,
focused ablative therapy of radiographic disease alone, or
both. Moreover, the role of docetaxel with ADT is also now a
matter for consideration. Specifically, some providers and
patients will prefer less therapy with possible risk of
undertreatment and poorer cancer control, whereas others
will prefer more therapy with risk of overtreatment and
consequent adverse events. Furthermore, with greater use
of highly sensitive imaging technologies (e.g., prostatespecific membrane antigen [PSMA], choline, or sodium
fluoride [NaF] PET and total body MRI), patients are likely
being identified earlier in the natural history of the disease
From the Dana-Farber Cancer Institute, Boston, MA; Harvard Medical School, Boston, MA; Mayo Clinic, Rochester, MN; University of Pennsylvania, Philadelphia, PA.
Corresponding author: Neha Vapiwala, MD, Department of Radiation Oncology, Perelman Center for Advanced Medicine, 3400 Civic Center Blvd., TRC 2 West, Philadelphia, PA 19104;
email: vapiwala@uphs.upenn.edu.
119
BERNARD ET AL
and would have been considered micrometastatic on

standard imaging (CT abdomen and pelvis and technetium
bone scan).1 It is clear in light of this latter observation that
incorporating these technologies into the staging and
monitoring of men with prostate cancer requires rigorous
testing to define how best to integrate them into standard
clinical care.
In 2012, the U.S. Preventive Services Task Force updated
their guidelines on prostate cancer screening to recommend
against prostate-specific antigen (PSA) screening in all men9;
since then, there has been a documented decrease in PSA
screening and in incidence of early-stage prostate cancer.10
It is yet unknown whether this change has increased the
number of men presenting with more advanced disease and
consequent higher risk of relapse following local therapy,
however it has been estimated that up to an extra 1,241 men
may die of prostate cancer in the United States per year from lack
of screening.11 Moreover, it is unknown whether there is an
absolute increase in the number of men presenting with de novo
metastatic prostate cancer. As a proportion of such men will have
oligometastatic disease at presentation, having a standardized
approach to its assessment and management is necessary.
To design a systematic approach to oligometastatic
prostate cancer, first and foremost a universally accepted
definition is required. To that end, variability exists regarding
what constitutes oligometastases, with cutoffs of both
three or fewer and five or fewer lesions used in the literature, and others in actual practice.3,12 Importantly, an
accepted definition should require that targeted treatment
of all metastatic lesions with surgery or RT is technically
feasible. In addition, standardization of imaging is paramount; given the uncertainty of the role of newer technologies in routine care, conventional imaging with bone
KEY POINTS
120
Oligometastatic prostate cancer is a unique clinical state

with an inherently more indolent tumor biology
susceptible to metastasis-directed focal ablative
therapy.
Oligometastatic prostate cancer may represent isolated
sites of disease or coexist with micrometastases.
Modern, highly sensitive imaging technology is
potentially identifying patients with oligometastatic
prostate cancer who would be deemed to have localized
disease or no evidence of radiographic disease by
standard imaging of CT of the abdomen and pelvis and
bone scan.
A multimodal approach to patients with oligometastatic
disease is needed, with evidence for surgery,
radiotherapy, and systemic therapy, alone or in
combination, improving patient outcomes.
Further prospective data are needed to best select
patients with oligometastatic prostate cancer most
likely to benefit from a given therapeutic approach.
scan and CT of the abdomen and pelvis remains the gold

standard to define oligometastatic prostate cancer.
To help set the stage, the following five cases detail the
spectrum of patients presenting with oligometastatic disease and different management strategies available:
1. A 63-year-old man diagnosed with a Gleason 9 adenocarcinoma of the prostate, PSA 11, and an isolated
biopsy-positive lesion at T11. He was started on ADT
with a PSA decline to 2.5 and received a radical prostatectomy 2 months thereafter with negative margins.
Stereotactic radiosurgery was subsequently delivered to
T11, and ADT was continued, with a PSA nadir of 0.02
12 months after radical prostatectomy. At that time, a CT
urogram for nephrolithiasis showed multiple small
sclerotic bone lesions within the vertebral column
consistent with metastases. PSA has risen to 0.3 with a
testosterone level of less than 7.0 ng/dL (lower limit of
normal: 240 ng/dL) 16 months after ADT initiation.
2. A 66-year-old man presented with localized left hip pain
and a PSA of 1,244. CT showed an abnormal prostate
concerning for tumor, as well as an expansile lytic lesion
in the left pubic bone. Prostate biopsy confirmed Gleason
8 adenocarcinoma, and he was treated with concurrent
ADT plus RT to the prostate and isolated bone metastasis. Six months after starting ADT, his PSA had fallen to
0.01. With his PSA still undetectable after 24 months of
therapy, ADT was stopped and he entered surveillance.
As of last follow-up, PSA was 0.04 with a testosterone
level of 168 3 years after ADT cessation.
3. A 60-year-old man with a PSA of 6.6 underwent a
prostate biopsy identifying Gleason 9 adenocarcinoma. Subsequent radical prostatectomy found locally
advanced disease (pT3bN1M0) with evidence of
extracapsular extension, seminal vesicle invasion, and a
single positive lymph node. Postoperative PSA was 2.6,
and the patient received salvage treatment with
7 months of ADT and concurrent intensity-modulated
radiation therapy (5,040 cGy to the whole pelvis; 7,020
cGy to prostatic bed). His PSA became undetectable but
12 months later rose to 2.5, prompting reinitiation of
ADT and the eventual emergence of castration resistance. Imaging demonstrated a single metastasis in
the left scapula. He was treated with seven cycles of
docetaxel; however, PSA continued to rise to 6.5. At that
time, left scapular pain and progression of the lesion on
bone scan led to treatment with 250 cGy of palliative RT
in 20 fractions (5,000-cGy total dose). His PSA has subsequently dropped to 0.75 with resolution of symptoms.
4. A 50-year-old man underwent robotic-assisted radical
prostatectomy with pathology demonstrating pT3a NX
prostate adenocarcinoma. His received radiation therapy to the prostatic fossa for a persistently elevated
PSA postoperatively. Following radiation, PSA remained
elevated. He was placed on ADT but developed
castration resistance with a rising PSA. When the

PSA rose to 3, 11C-choline PET/CT demonstrated
node-only disease in the pelvis and retroperitoneum
(Fig 1). He elected to undergo combined salvage
extended pelvic lymph node dissection and retroperitoneal lymph node dissection. Pathology demonstrated 9 of 62 lymph nodes involved with prostate
adenocarcinoma. His PSA was less than 0.2 at 6 months
after surgery.
5. A 49-year-old man presented with an elevated PSA to
49. Prostate biopsy demonstrated Gleason 9 prostate
adenocarcinoma. CT scan revealed pelvic lymphadenopathy (Fig 2). Bone scan was without evidence of
osseous metastatic disease. He elected to proceed
with radical retropubic prostatectomy and extended
bilateral pelvic lymphadenectomy as part of a multimodal management plan. Surgical pathology demonstrated Gleason 9 pT3b N1 (7 of 27 lymph nodes
positive) prostate adenocarcinoma. He is receiving 2
years of adjuvant ADT and underwent a course of
adjuvant pelvic radiation therapy at 6 months after
surgery.
This review will outline the role of surgery, RT, and systemic therapy in the management of oligometastatic prostate cancer. Table 1 shows data from studies of focal therapy
with surgery or RT alone, systemic therapy with ADT alone,
or systemic therapy with ADT plus chemotherapy in men
with oligometastatic prostate cancer. Current evidence and
existing controversies will be discussed.
THE ROLE OF SURGERY IN OLIGOMETASTATIC

PROSTATE CANCER
The role of surgery in the management of oligometastatic
prostate cancer has received increased attention in recent years. However, the concept of surgery for advanced
prostate cancer is not new. For instance, observational
studies from the 1990s reported improved oncologic outcomes for men with pathologic lymph nodepositive (pN+)
prostate cancer treated with radical prostatectomy and ADT
compared with ADT alone, and prior radical prostatectomy
has been associated with improved overall survival (OS)
for men with metastatic disease.6-8,13 Given that aggressive prostate cancer can persist even after 1 year of systemic therapy, it is not surprising that surgery may improve
FIGURE 1. Case 4
121
BERNARD ET AL
FIGURE 1. Case 4 (Contd)
A 50-year-old man that received salvage extended pelvic and retroperitoneal lymph node dissection for choline PET-positive disease post-RP and adjuvant RT with subsequent
PSA decline.
oncologic outcomes.14 Potential mechanisms include reducing tumor burden (i.e., resetting the clock), preventing
continued metastatic spread from the primary, and relieving
tumor immune suppression.15
Surgery for Locally Advanced (Including Pelvic Lymph

Node) Disease
For men that have not had prior local therapy, radical
prostatectomy has been associated with improved oncologic
outcomes for men with pN+ disease in a number of observational studies. For example, in a matched comparison
of 79 men treated with radical prostatectomy and adjuvant orchiectomy versus 79 men treated with orchiectomy
alone, 10-year OS was 66% versus 28% in favor of radical
prostatectomy.8 Similarly, an analysis of the Munich Cancer
Registry reported 10-year OS of 64% for completed versus
28% for aborted radical prostatectomy in men with pN+
disease.16 Although there is inherent potential for selection
bias toward radical prostatectomy for patients with more
favorable disease, there is remarkable consistency in reported
oncologic outcomes among observational studies with 10-year
cancer-specific survival (CSS) and OS ranging from 70% to 85%
122
and 50% to 75%, respectively, for men treated with radical

prostatectomy plus ADT, compared with 50% to 65% and 25%
to 30%, respectively, for ADT alone.17,18 Moreover, randomized data reinforce these figures. Among men treated with
radical prostatectomy and immediate ADT in ECOG 3886, OS
was 64% at a median of 11.9 years versus 45% in those
managed with radical prostatectomy alone.19 Notably among
those managed with ADT alone in EORTC 30846, 10-year OS
was approximately 30%, although approximately 40% of the
deaths were due to nonprostate cancer.20
Risk stratification is essential for men with pN+ disease,
as some will have durable oncologic control with surgical
resection alone, whereas others will require multimodal
therapy. One multi-institutional analysis reported that men
with two or fewer positive nodes had significantly better
15-year CSS than those with more than two nodes (84% vs. 62%;
p , .001) when treated with radical prostatectomy and
adjuvant ADT.21 Another study similarly identified more
than two positive nodes as an adverse prognostic feature,
along with Gleason score greater than 7.22 Most recently,
Moschini et al23 developed a risk score for cancer-specific
mortality (CSM) among pN+ men after radical prostatectomy
with 88% receiving adjuvant ADT. Points are added for three
FIGURE 2. Case 5
A 49-year-old man that received radical prostatectomy and up-front bilateral

extended pelvic lymphadenectomy for clinically lymph nodepositive disease
followed by adjuvant RT and long course ADT.
or more positive lymph nodes, Gleason 710 disease, positive

surgical margins, and absence of adjuvant RT to categorize
patients into risk groups with 10-year CSM ranging from 19%
to 46%.
There is increasing evidence that pN+ prostate cancer
does not uniformly signify systemic disease. In one study of
clinical recurrence patterns in 1,011 pN+ men after radical
prostatectomy, 57% of clinical recurrences were identified
at a single location, and of these, 51% were local or node-only at
the time of relapse.24 In light of such data, management of clinical
lymphadenopathy (cN+), which has traditionally been distinguished from cN0pN+ disease, deserves reconsideration. For
example, the National Comprehensive Cancer Network (NCCN)
recommends ADT with or without radiation therapy for men
presenting with cN1M0 disease.25 However, in a recent study of
302 men with pN+ disease, of whom 17% were cN+, the authors
noted no difference in CSS or OS between cN+ and cN0 patients.26 Such data suggest that appropriately selected men with
clinical lymphadenopathy may be treated with surgical resection
as an alternative to pelvic radiation as for cN0pN+ disease.
Surgery in Presence of Distant Metastatic Disease

In the absence of prior local treatment, there is considerably
less data on the role of surgery in the setting of distant
metastases.27 In a multi-institutional report of 106 men who

underwent radical prostatectomy/extended pelvic lymph
node dissection (ePLND) for M1a/M1b disease, CSS was
89% at a median of 22.8 months, and perioperative morbidity was overall similar to that reported for radical
prostatectomy/ePLND for M0 disease.28 Another study
compared 23 patients with low-volume M1b disease (three
or fewer skeletal lesions) who underwent radical
prostatectomy/ePLND with 38 controls treated with ADT
alone.29 Cytoreductive radical prostatectomy was associated
with longer time to castration resistance (40 vs. 29 months),
improved clinical progression-free survival (PFS) (39 vs.
28 months), and improved CSS (96% vs. 84%). Populationbased data likewise support an oncologic benefit to local
therapy with surgery or radiation.30 A randomized trial evaluating best systemic therapy with or without local therapy in
the form of surgery or radiation for men with metastatic
prostate cancer should provide important prospective data
(NCT01751438).
Risk stratification to guide patient selection for surgery in
the setting of M1 disease is even more critical than for N+M0
disease. To this end, Fossati et al31 conducted a populationbased analysis of men with metastatic prostate cancer and
reported that baseline CSM risk significantly (p , .0001)
modified the oncologic benefit of local therapy (either
surgery or radiation), with no benefit to local therapy
beyond a specific threshold. Further studies are necessary to
determine which patient populations may derive greatest
benefit from cytoreductive radical prostatectomy in the
presence of M1 disease.
In the setting of clinical nodal recurrence following definitive local therapy, salvage lymph node dissection (sLND)
has gained increased utilization, driven in part by improved
imaging modalities such as 11C-choline PET/CT. Principles of
surgery for oligometastatic nodal disease include availability
of accurate preoperative imaging, complete resection of all
radiographic disease, and acceptable morbidity. Although
randomized data are lacking, observational studies are
encouraging. One U.S. series reported that, of 52 men who
underwent sLND for nodal recurrence after radical prostatectomy, 73% achieved undetectable PSA after surgery.32
In a follow-up study of 117 patients, the largest single series
to date, 5-year biochemical recurrence (BCR)free, radiologic recurrence-free, and CSS were 39%, 51%, and 97%,
respectively, after sLND with adjuvant ADT being given to
62.7% of patients.33 In a European series of 59 men who
underwent sLND, 59% achieved undetectable PSA, and
among these, the 8-year BCR-free survival was 23%.34 For the
overall cohort, 8-year clinical recurrence-free and CSS were
38% and 81%, respectively. A systematic review recently
synthesized the available data on sLND with or without adjuvant ADT from 12 observational studies, 5-year BCR-free
survival of 9% to 22%, and 5-year OS of approximately 75%.35
Favorable prognostic factors included complete PSA response
to surgery alone, fewer positive nodes, absence of retroperitoneal involvement, and lower preoperative PSA. What is not
known is the optimal utilization and duration of adjuvant
123
BERNARD ET AL
TABLE 1. Summary of Key Studies Evaluating Surgery, Radiotherapy, and Systemic Therapy in Oligometastatic
Prostate Cancer in the Post-PSA Era
Study
No. of Patients
Study Type
Imaging
Outcome
Engel et al16
938
Retrospective cohort
N/A (pN+)
10-year OS of 64% with RP vs. 28% without RP
ECOG 388619
98
RCT
N/A (pN+)
11.9-year OS of 64% with RP + immediate

ADT vs. 45% with RP + deferred ADT
234
RCT
N/A (pN+)
10-year CSS of 48% for immediate ADT vs.

44% for delayed ADT (none with RP)
Zattoni et al33
117
Case series
11
5-year 69% BCR-free, 49% radiographic

recurrence-free, and 97% CSS
Suardi et al34
59
Case series
11
8-year 23% BCR-free, 38% clinical

recurrence-free, and 81% CSS
106
Case series
Bone scan, CT
89% CSS at median of 22.8 months with

cytoreductive RP for M1 disease
61
Retrospective cohort
Bone scan, CT
CSS of 96% with RP + ADT vs. 84% with

ADT alone, at median of 34.5 months
Jereczek-Fossa et al59
19
Case series
PET/CT
30-month PFS, 64%
Schick et al60
50
Case series
PET/CT; bone scan
3-year OS, 92%
Decaestecker et al61
50
Case series
PET/CT
Median PFS, 19 months
Picchio et al62
83
Case series
PET/CT
PSA PR/CR, 83%
Surgery Alone
pN+ Disease
EORTC 3084620
sLND
C-Choline PET/CT
C-Choline PET/CT
M1 Disease
Sooriakumaran et al28
Heidenreich et al29
RT Alone
ADT Alone
Millikan et al47
84
Phase III RCT
Bone scan
Median OS, 94 months
S934663
798
Phase III RCT
Bone scan; CXR
GETUG 1553
102
Phase III RCT
Bone scan; CT
CHAARTED41
143
Phase III RCT
Bone scan; CT
4-year OS, 60%
CHT
Millikan et al47
76
Phase III RCT
Bone scan
Median OS, 93.6 months
GETUG 1553
100
Phase III RCT
Bone scan; CT
4-year OS, 70%
CHAARTED41
134
Phase III RCT
Bone scan; CT
4-year OS, 70%
Abbreviations: N/A, not applicable; RCT, randomized controlled trial; RP, radical prostatectomy; BCR, biochemical recurrence; ADT, androgen deprivation therapy; PR, partial
response; CR, complete response; CXR, chest x-ray.
systemic therapy with ADT, and whether concurrent docetaxel

may possibly further improve these outcomes.
Studies on Resection of Distant Metastases After

Local Therapy
Data reporting outcomes of surgery in this setting have been
limited to retroperitoneal lymph nodes as part of sLND,
whereas stereotactic body radiation therapy (SBRT) has
generally been used for bone or visceral oligometastatic
disease.35,36 Although retroperitoneal disease portends a
worse prognosis, data suggest that a subset of patients
may achieve durable clinical recurrence-free survival with
sLND with or without adjuvant ADT. One study of sLND
reported 8-year clinical recurrence-free survival of 22%
for retroperitoneal nodal involvement versus 69% without
retroperitoneal disease.34 Another study noted 5-year
clinical recurrence-free survival of 11% with retroperitoneal
124
involvement versus 53% without retroperitoneal disease.37

The phase II randomized trial STOMP will evaluate the
role of metastasis-directed therapy with either surgery or
SBRT in oligometastatic recurrence following local therapy
(NCT01558427) compared with active surveillance with ADT
to be deferred in both arms until more than three metastatic lesions develop.38 This should provide important data
regarding the potential oncologic benefit of both metastastectomy and sLND as monotherapy.
Overall, emerging data suggest that surgery may represent
an important therapeutic modality in the management of
oligometastatic prostate cancer. Ultimately, optimal patient
selection is critical. Radical prostatectomy/ePLND is associated with durable oncologic control in the setting of N+M0
disease, with emerging data for its use as local therapy in M1
disease. sLND is likewise associated with favorable oncologic
outcomes for oligometastatic nodal recurrence after local
therapy. Prospective studies will further inform these issues

and will also inform whether it can be used alone or in
combination with systemic therapy.
THE ROLE OF RADIATION THERAPY IN

OLIGOMETASTATIC PROSTATE CANCER
Radiation therapy is a well-established curative therapy for
intact localized and postoperative locally recurrent prostate
cancer, and an effective palliative therapy for symptomatic
metastatic disease. Conversely, its role in oligometastatic
disease in an effort to improve survival is unknown. Currently, the NCCN guidelines recommend considering RT for
metastases if symptomatic or residing in a weight-bearing
bone; otherwise, the mainstay for M1 disease remains
systemic with either medical or surgical castration, or
chemohormonal therapy (CHT).39 However, ADT is associated with a number of side effects and potential complications, so being able to delay ADT start and/or allow for
longer treatment breaks if given intermittently without
compromising survival is a worthy pursuit. Moreover, a
shorter course of ADT with concurrent RT to maximize
control may be another strategy to limit potential toxicity
associated with ADT. Plausibly, early detection of oligometastases with sensitive, accurate imaging may allow for
more proactive therapy and eradication of locally measurable disease, delayed time to ADT, and, perhaps, cure if there
is no associated micrometastatic disease and radiation is
able to eradicate the isolated metastatic lesion(s). In the
setting of postprostatectomy biochemical failure, the median time from PSA recurrence to metastases may be long
(8 years in one study)40; regular monitoring and prompt
utilization of metastases-directed focal therapy upon their
development may further improve outcomes and avoid the
need for systemic therapy.
Recently, Ost et al36 conducted a review of studies
employing metastases-directed therapy for relapsed prostate cancer. Of note, oligometastases were identified using
PET/CT in 98% of patients, and 78% of treated metastases
were nodal in origin. Although great heterogeneity among
patient populations and variable use of ADT and prophylactic
nodal irradiation made comparison among studies challenging, the authors found that around one-half of men were
progression-free 13 years after metastases-directed therapy; however, over 60% had adjuvant ADT. Caution should
be taken when evaluating PFS between these various
studies, however, as different definitions were used among
them. In addition, although promising, the quality of the
evidence generated to date is insufficient to recommend this
approach as routine standard care.
The impact of RT, specifically, on PFS when used to treat
recurrent, treatment-naive oligometastatic prostate cancer
was recently the focus of a multi-institutional analysis.3 The
study included 163 patients with three or fewer metastases
each. The authors assessed for both local PFS and distant PFS,
and identified a RT dose of less than or equal to 100 Gy as being
associated with a lower local PFS at 3 years (79% vs. 99%;
p = .01). Also, a nonsignificant trend was observed for improved

median distant PFS with adjuvant ADT (25 vs. 18 months;
p = .09). As in the review by Ost et al, most of the patients were
staged using PET/CT36; however, as this technique is not
currently considered standard of care, extrapolating and applying these data to clinical practice is difficult.
An important consideration when applying RT in any disease
setting is the accurate deposition of dose to the target while
avoiding dose-limiting toxicity to surrounding tissue. There are
several factors in the evolution of modern RT that are helping to
mitigate this concern. Enhanced imaging modalities and image
registration have led to improved accuracy in RT target definition, while advances in both external patient and internal
organ immobilization and tracking permit more precise radiation delivery. Furthermore, major technical evolutions in
hardware and software capabilities have enabled safe delivery of larger doses of radiation per treatment fraction
(i.e., hypofractionation); such dosing regimens are not only a
desirable from a patient convenience standpoint but can carry
greater radiobiological bang for your buck, delivering a potentially higher biologically effective dose over a shorter time
period.2 Consequently, various hypofractionated radiation
therapy dosing schedules have been proposed for treatment in
the oligometastatic setting, including one fraction of 20 Gy,
three fractions of 10 Gy, five to six fractions of 5 or 6 Gy, and 10
fractions of 5 Gyall deemed to be of equivalent efficacy.2
In summary, a growing body of retrospective literature
suggests RT for oligometastatic prostate cancer is feasible,
effective, and without significant toxicity. Although unproven, its role as a solitary therapy or in conjunction with
systemic and/or surgical strategies is evolving, as is the goal
and ultimate intent (i.e., aggressive palliation vs. durable
control vs. eradication and cure) of its use. Whether the
primary function of radiation in the oligometastatic setting
to primary (if intact) and/or metastasis is to delay time to
ADT, to consolidate potentially curative multimodal approaches, or both, is to be determined. Numerous clinical
trials are currently in development that will provide prospective data to better answer this question (NCT01859221;
NCT01777802; NCT02192788; NCT01558427).
THE ROLE OF SYSTEMIC THERAPY IN

OLIGOMETASTATIC PROSTATE CANCER
The cornerstone of therapy for metastatic prostate cancer
is ADT, either continuous or intermittent. Additionally,
chemotherapy has been shown to improve OS, both in
the hormone-sensitive and castration-resistant state.41-45 Although not curative, goals of systemic therapy include disease
control, symptom reduction, reduction of morbidity, and life
prolongation. However, its role in oligometastatic prostate
cancer, specifically, is less clear. There is speculation that
control of micrometastatic disease, in conjunction with targeted treatment to oligometastases, may improve outcome
and, in some, may possibly lead to cure. Chemotherapy may aid
by theoretically eliminating less androgen-sensitive tumor cells
and augment what can be achieved with ADT.
125
BERNARD ET AL
Prognostically, it has been shown that patients with highvolume disease (visceral metastases and multiple bone
metastases [e.g., four or more]) have worse outcomes.46,47
Furthermore, it was shown that those with five or fewer
bone metastases fared better than those with more than five
at time of relapse following RT for localized disease, with
distinct natural histories apparent for these patient subsets.48 Thus, patients with low-volume metastatic prostate
cancer at time of relapse appear to innately have more
indolent biology and more favorable prognoses compared
with those with more widespread disease. Attempting to
improve their survival further with metastases-directed
therapy, with or without systemic therapy, is an area of
great interest.
Notably, it has been shown that delayed ADT for metastatic prostate cancer does not negatively impact outcome
of patients with oligometastatic prostate cancer, with one
study reporting a median time from metastases to prostate
cancerspecific death of 82 months.49 Moreover, in men
with PSA-only relapse following radical prostatectomy and
deferred ADT, it has been demonstrated that metastasisfree survival may serve as a surrogate for OS.50 This study
also identified number of metastases (three or fewer vs. four
or more, hazard ratio [HR] 0.50; 95% CI, 0.290.85; p = .012)
as an independent prognostic variable for OS, further implying potential biologic differences between patients with
oligometastases and those with diffuse disease.50 It is
postulated that an improvement in outcome for those
with oligometastatic disease treated with focal ablative
therapy of their radiographic evident disease may be augmented with systemic therapy in an attempt to eradicate
micrometastases.
Evidence for improved survival with early use of CHT
consisting of docetaxel chemotherapy plus ADT comes from
three randomized, controlled trials. The first to demonstrate
the benefit, E3805 (CHAARTED), documented a 13.6-month
OS advantage with ADT plus six cycles of docetaxel versus
ADT alone (HR for death 0.61; 95% CI, 0.470.80; p , .001).41
This benefit was even more pronounced in those with highvolume disease (defined as visceral metastases, four or more
bone metastases with at least one beyond the vertebral
column and pelvis, or both), with an OS gain of 17.0 months
with CHT (HR 0.60; 95% CI, 0.450.81; p , .001). Furthermore, CHT conferred an 8.5-month improvement in median time to symptomatic, PSA, or radiographic progression
(HR 0.61; 95% CI, 0.510.72; p , .001). When patients with
protocol-defined low-volume disease were assessed as a
separate subgroup, there was also a 39% decrease risk of
deathbut there were far fewer events, and at time of the
first report, this was not statistically significant. This indicates that these patients have a longer natural history with
ADT alone with 4-year OS of approximately 60%. Moreover,
some of these patients with a shorter life expectancy because of age and/or comorbidities may die of a competing
cause.
These data were supported by those from the STAMPEDE
trial, which used a multiarm, multistage platform design to
126
assess the value of adding docetaxel, zoledronic acid, or both

drugs combined to standard of care in men with high-risk,
locally advanced, metastatic or relapsed prostate cancer.42
Standard of care consisted of ADT for all subgroups, with RT
optional but later required for those with localized disease.
This study used a novel, flexible trial design to pragmatically
ascertain the impact of therapy along the spectrum of advanced prostate cancer, with 2,265 participants (76%)
having node-positive, node-unknown, or distant metastases.
The addition of docetaxel to standard of care resulted in a
10-month improvement in OS (HR 0.78; 95% CI, 0.660.93;
p = .006). Importantly, the effect of docetaxel plus standard
of care (ADT for M1 and ADT plus radiation for most of those
with M0) was consistent across all subgroups, suggesting
that locally advanced disease (with high risk of micrometastases) and node-positive patients also derived a
benefit from early CHT. However, the OS benefit with CHT
was strongest and significant (p = .005) for those with
metastatic disease (note that there was no breakdown by
volume of disease), with nonmetastatic patients deriving a
failure-free survival (FFS) benefit, although not enough
events have occurred for an OS benefit at this time. Consistent with the benefit of CHT in localized high-risk setting is
the survival benefit seen with the phase III RTOG 0521 trial,
demonstrating a 32% reduction in risk of death at 4 years
with adjuvant CHT following concurrent ADT plus RT. In
addition, men that received docetaxel derived an improved
5-year FFS (HR 0.76; 95% CI, 0.571.00; p = .05). Likewise, the
phase III GETUG 12 trial showed CHT plus RT conferred a
reduction in risk of relapse or death of 29% at 8 years over
ADT plus RT alone51; longer follow-up is required to see
whether this benefit extends to OS as well.
The third trial, GETUG 15, was the first to explore CHT in
metastatic hormone-sensitive prostate cancer.52 Although
the early results did not suggest a benefit from CHT, longer
follow-up and harmonizing volume of disease definitions
with those of CHAARTED revealed more similar results to
those seen in CHAARTED, especially for those with highvolume disease.53 Specifically, it was found that CHT resulted
in a 13.5-month improvement in median OS at 7 years
(HR 0.88; 95% CI, 0.681.14; p = .3), with a post hoc analysis
identifying a 22% reduction in risk of death in men with highvolume disease, as per the CHAARTED definition.53 Furthermore, radiographic PFS was greater in those that received CHT in the total population (HR 0.69;95% CI,
0.550.87; p = .002) and in those with high-volume disease
(HR 0.61; 95% CI, 0.440.83; p = .002).
The utility of CHT in the subgroup of low-volume metastatic hormone-sensitive prostate cancer is less well defined
at early follow-up of the CHAARTED trial. However, when
one considers the totality of the data of adjuvant therapy in
high-risk localized disease and benefit in the overall population of metastatic disease (STAMPEDE and CHAARTED),
it is quite possible that, for patients with oligometastatic
prostate cancer who are at risk for dying of their disease,
there is a benefit from CHT. At the other end of the
spectrum, studies for patients with low-volume disease
aimed at avoiding cytotoxic chemotherapy and its associated adverse events are currently under investigation, with
a different hormone-based approach using the competitive
androgen receptor antagonist, ARN-509, in combination
with ADT versus ADT alone in chemotherapy-naive patients with low-volume mHSPC (NCT02489318). Of note,
docetaxel can be prescribed per investigator choice in this
study.
Overall, collective data from GETUG 15, CHAARTED, and
STAMPEDE support the hypothesis that an aggressive approach with systemic therapy early in the metastatic setting
improves outcome. Indeed, a recent meta-analyses of data
from these studies confirmed the improved survival of CHT
over ADT (HR, 0.77; 95% CI, 0.680.87; p , .0001).54 It is
therefore logical to apply the principle of systemic therapy
with localized/focal ablative therapy in the setting of PSA
relapse and high-risk localized disease.
The large phase III RTOG 9601 trial tested whether
2 years of anti-androgen therapy plus salvage RT was
superior to RT alone in men with PSA relapse following radical prostatectomy.55 Combined therapy resulted
in a 9% reduction in absolute incidence of metastases (14%
vs. 23%; p , .001) and 5% reduction in incidence of prostate
cancerspecific death (2.3% vs. 7.5%; p , .001) at 12 years.
Another retrospective study found that patients that received salvage RT plus concurrent ADT had a decreased risk
of relapse when treated at a PSA relapse of less than or equal
to 0.5 ng/mL following prostatectomy, although the effect
appeared limited to those with negative surgical margins.56
This is further evidence that focal ablative therapy with
systemic therapy improves outcome in the relapsed setting
and supports the role of combined modality therapy in lowvolume recurrent disease.
In attempting to select those with oligometastatic disease
for further metastases-directed therapy, both deep PSA
nadir and absence of PSA progression on ADT may predict for
improved survival.57,58 Such patients may be the ones who
benefit the most from targeted treatments to select lesions.
In total, there is growing evidence that more aggressive
therapy early in the course of disease leads to improved
outcome, and therefore utilizing systemic therapy as part
of a multimodal approach with focal ablative therapy to
radiographic evident disease for patients with oligometastatic prostate cancer should be considered.
DISCUSSION
A number of unanswered questions remain relevant to the
discussion of oligometastatic prostate cancer. It clearly has a
longer natural history and presumably has a more indolent
biology, most of the time. That said, it is unknown at this time
whether there will be a higher incidence of more advanced and
de novo metastatic disease with decreased PSA screening in
the United States. If this is the case, there will be more patients
with metastatic disease and an even greater need at the public
health level to improve therapy for metastatic hormonesensitive prostate cancer. Related to diagnosis is what, if
any, impact the new imaging modalities will have on the course
of disease. For example, for the patients identified with oligometastases by such highly sensitive tests, is RT to the
metastatic lesion alone, with less systemic effects, sufficient or
is the patient best treated with ADT plus RT? Moreover, there
is a paucity of data regarding whether the total number and
location of oligometastases (by standard CT and technetium
bone scan imaging) is prognostically important. Identifying the
patients most likely to benefit from treatment is needed.
As detailed above, treatment of oligometastases with focal
ablative therapies appears promising but requires further validation in prospective studies before it can be recommended
as standard of care.36 Data supporting treatment of lymph
nodeonly recurrence is also limited, although a recent review
suggested it may confer a therapeutic benefit.35 When designing
such prospective studies, there is differing opinion whether delay
in requiring ADT, OS, or both should be the primary endpoint.
Ultimately, the decision regarding when, and how, to treat
men with oligometastatic prostate cancer requires an assessment of the competing risks of burden of therapy versus
those of progressive disease and subsequent therapies. Such an
evaluation requires incorporating factors such as symptoms,
comorbidities, and life expectancy. Specifically, competing risks
of death are very relevant for patients with oligometastatic
hormone-sensitive prostate cancer. Moreover, actively engaging patients in the therapeutic discussion, defining goals of
care, and enabling shared decision making are all required.
Further prospective, randomized data are necessary to better
characterize the nature of oligometastatic prostate cancer and
define the role and value of therapies in this state. These trials
should take into account the gaps in knowledge described
herein, and help to address the most pressing quandaries
facing clinicians and patients today.
References
1. Reyes DK, Pienta KJ. The biology and treatment of oligometastatic
cancer. Oncotarget. 2015;6:8491-8524.
2. Conde Moreno AJ, Ferrer Albiach C, Muelas Soria R, et al. Oligometastases in prostate cancer: restaging stage IV cancers and new radiotherapy options. Radiat Oncol. 2014;9:258.
3. Ost P, Jereczek-Fossa BA, As NV, et al. Progression-free survival
following stereotactic body radiotherapy for oligometastatic prostate
cancer treatment-naive recurrence: a multi-institutional analysis. Eur
Urol. 2016;69:9-12.
4. Weichselbaum RR, Hellman S. Oligometastases revisited. Nat Rev Clin

Oncol. 2011;8:378-382.
5. Bolla M, Van Tienhoven G, Warde P, et al. External irradiation with or
without long-term androgen suppression for prostate cancer with high
metastatic risk: 10-year results of an EORTC randomised study. Lancet
Oncol. 2010;11:1066-1073.
6. Cheng CW, Bergstralh EJ, Zincke H. Stage D1 prostate cancer. A nonrandomized comparison of conservative treatment options versus
radical prostatectomy. Cancer. 1993;71(3, Suppl):996-1004.
127
BERNARD ET AL
7. Frohmuller HG, Theiss M, Manseck A, et al. Survival and quality of life of

patients with stage D1 (T1-3 pN1-2 M0) prostate cancer. Radical
prostatectomy plus androgen deprivation versus androgen deprivation
alone. Eur Urol. 1995;27:202-206.
8. Ghavamian R, Bergstralh EJ, Blute ML, et al. Radical retropubic prostatectomy plus orchiectomy versus orchiectomy alone for pTxN+
prostate cancer: a matched comparison. J Urol. 1999;161:1223-1227,
discussion 1227-1228.
9. U.S. Preventive Services Task Force. Final Update Summary: Prostate Cancer:
Screening. U.S. Preventive Services Task Force. May 2012. http://www.
uspreventiveservicestaskforce.org/Page/Document/UpdateSummaryFinal/
prostate-cancer-screening. Accessed January 15, 2016.
10. Jemal A, Fedewa SA, Ma J, et al. Prostate cancer incidence and PSA
testing patterns in relation to USPSTF screening recommendations.
JAMA. 2015;314:2054-2061.
11. Penson DF. The pendulum of prostate cancer screening. JAMA. 2015;
314:2031-2033.
12. Niibe Y, Chang JY. Novel insights of oligometastases and oligorecurrence and review of the literature. Pulm Med. 2012;2012:261096.
13. Thompson IM, Tangen C, Basler J, et al. Impact of previous local
treatment for prostate cancer on subsequent metastatic disease. J Urol.
2002;168:1008-1012.
14. Tzelepi V, Efstathiou E, Wen S, et al. Persistent, biologically meaningful
prostate cancer after 1 year of androgen ablation and docetaxel
treatment. J Clin Oncol. 2011;29:2574-2581.
15. Swanson G, Thompson I, Basler J, et al. Metastatic prostate cancerdoes treatment of the primary tumor matter? J Urol. 2006;176:
1292-1298.
16. Engel J, Bastian PJ, Baur H, et al. Survival benefit of radical prostatectomy in lymph node-positive patients with prostate cancer. Eur Urol.
2010;57:754-761.
17. Briganti A, Blute ML, Eastham JH, et al. Pelvic lymph node dissection in
18. Gakis G, Boorjian SA, Briganti A, et al. The role of radical prostatectomy
and lymph node dissection in lymph node-positive prostate cancer:
a systematic review of the literature. Eur Urol. 2014;66:191-199.
19. Messing EM, Manola J, Yao J, et al; Eastern Cooperative Oncology Group
study EST 3886. Immediate versus deferred androgen deprivation
treatment in patients with node-positive prostate cancer after radical
prostatectomy and pelvic lymphadenectomy. Lancet Oncol. 2006;7:
472-479.
20. Schroder FH, Kurth KH, Fossa SD, et al. Early versus delayed endocrine
treatment of T2-T3 pN1-3 M0 prostate cancer without local treatment
of the primary tumour: final results of European Organisation for the
Research and Treatment of Cancer Protocol 30846 after 13 years of
follow-up (a randomised controlled trial). Eur Urol. 2009;55:14-22.
21. Briganti A, Karnes JR, Da Pozzo LF, et al. Two positive nodes represent a
significant cut-off value for cancer specific survival in patients with node
positive prostate cancer. A new proposal based on a two-institution
experience on 703 consecutive N+ patients treated with radical
prostatectomy, extended pelvic lymph node dissection and adjuvant
therapy. Eur Urol. 2009;55:261-270.
22. Touijer KA, Mazzola CR, Sjoberg DD, et al. Long-term outcomes of
patients with lymph node metastasis treated with radical prostatectomy without adjuvant androgen-deprivation therapy. Eur Urol. 2014;
65:20-25.
23. Moschini M, Sharma V, Zattoni F, et al. Risk stratification of pN+ prostate
cancer after radical prostatectomy from a large institutional series with
long-term follow-up. J Urol. Epub 2015 Dec 23.
24. Moschini M, Sharma V, Zattoni F, et al. Natural history of clinical recurrence patterns of lymph node-positive prostate cancer after radical
prostatectomy. Eur Urol. 2016;69:135-142.
128
25. Mohler JL, Kantoff PW, Armstrong AJ, et al; National Comprehensive
Cancer Network. Prostate cancer, version 2.2014. J Natl Compr Canc
Netw. 2014;12:686-718.
26. Moschini M, Briganti A, Murphy CR, et al. Outcomes for patients with
clinical lymphadenopathy treated with radical prostatectomy. Eur Urol.
2016;69:193-196.
27. Bayne CE, Williams SB, Cooperberg MR, et al. Treatment of the primary
tumor in metastatic prostate cancer: current concepts and future
perspectives. Eur Urol. Epub 2015 May 20.
28. Sooriakumaran P, Karnes J, Stief C, et al. A multi-institutional analysis of
perioperative outcomes in 106 men who underwent radical prostatectomy for distant metastatic prostate cancer at presentation. Eur
Urol. Epub 2015 May 30.
29. Heidenreich A, Pfister D, Porres D. Cytoreductive radical prostatectomy
in patients with prostate cancer and low volume skeletal metastases:
results of a feasibility and case-control study. J Urol. 2015;193:832-838.
30. Culp SH, Schellhammer PF, Williams MB. Might men diagnosed with
metastatic prostate cancer benefit from definitive treatment of the
primary tumor? A SEER-based study. Eur Urol. 2014;65:1058-1066.
31. Fossati N, Trinh QD, Sammon J, et al. Identifying optimal candidates for
local treatment of the primary tumor among patients diagnosed with
metastatic prostate cancer: a SEER-based study. Eur Urol. 2015;67:3-6.
32. Karnes RJ, Murphy CR, Bergstralh EJ, et al. Salvage lymph node dissection for prostate cancer nodal recurrence detected by 11C-choline
positron emission tomography/computerized tomography. J Urol.
2015;193:111-116.
33. Zattoni F, Nehra A, Murphy CR, et al. Mid-term outcomes following
salvage lymph node dissection for prostate cancer nodal recurrence
status post-radical prostatectomy. Eur Urol Focus. Epub 2016 Feb 10.
34. Suardi N, Gandaglia G, Gallina A, et al. Long-term outcomes of salvage
lymph node dissection for clinically recurrent prostate cancer: results
of a single-institution series with a minimum follow-up of 5 years. Eur
Urol. 2015;67:299-309.
35. Ploussard G, Almeras C, Briganti A, et al. Management of node only
recurrence after primary local treatment for prostate cancer: a systematic review of the literature. J Urol. 2015;194:983-988.
36. Ost P, Bossi A, Decaestecker K, et al. Metastasis-directed therapy of
regional and distant recurrences after curative treatment of prostate
cancer: a systematic review of the literature. Eur Urol. 2015;67:852-863.
37. Rigatti P, Suardi N, Briganti A, et al. Pelvic/retroperitoneal salvage lymph
node dissection for patients treated with radical prostatectomy with
biochemical recurrence and nodal recurrence detected by [11C]choline
positron emission tomography/computed tomography. Eur Urol. 2011;
60:935-943.
38. Decaestecker K, De Meerleer G, Ameye F, et al. Surveillance or
metastasis-directed therapy for oligometastatic prostate cancer recurrence (STOMP): study protocol for a randomized phase II trial. BMC
Cancer. 2014;14:671.
39. National Comprehensive Cancer Network. Prostate Cancer (Version
1.2016). http://www.nccn.org/professionals/physician_gls/pdf/prostate.
40. Pound CR, Partin AW, Eisenberger MA, et al. Natural history of progression after PSA elevation following radical prostatectomy. JAMA.
1999;281:1591-1597.
41. Sweeney CJ, Chen YH, Carducci M, et al. Chemohormonal therapy in
metastatic hormone-sensitive prostate cancer. N Engl J Med. 2015;373:
737-746.
42. James ND, Sydes MR, Clarke NW, et al; STAMPEDE investigators. Addition of docetaxel, zoledronic acid, or both to first-line long-term
hormone therapy in prostate cancer (STAMPEDE): survival results from
an adaptive, multiarm, multistage, platform randomised controlled
trial. Lancet. Epub 2015 Dec 21.
43. Tannock IF, de Wit R, Berry WR, et al; TAX 327 Investigators. Docetaxel
plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med. 2004;351:1502-1512.
44. Petrylak DP, Tangen CM, Hussain MH, et al. Docetaxel and estramustine
compared with mitoxantrone and prednisone for advanced refractory
prostate cancer. N Engl J Med. 2004;351:1513-1520.
45. de Bono JS, Oudard S, Ozguroglu M, et al; TROPIC Investigators.
Prednisone plus cabazitaxel or mitoxantrone for metastatic castrationresistant prostate cancer progressing after docetaxel treatment:
a randomised open-label trial. Lancet. 2010;376:1147-1154.
46. Eisenberger MA, Blumenstein BA, Crawford ED, et al. Bilateral orchiectomy with or without flutamide for metastatic prostate cancer. N Engl
J Med. 1998;339:1036-1042.
47. Millikan RE, Wen S, Pagliaro LC, et al. Phase III trial of androgen ablation
with or without three cycles of systemic chemotherapy for advanced
prostate cancer. J Clin Oncol. 2008;26:5936-5942.
48. Singh D, Yi WS, Brasacchio RA, et al. Is there a favorable subset of
patients with prostate cancer who develop oligometastases? Int J
49. Makarov DV, Humphreys EB, Mangold LA, et al. The natural history of
men treated with deferred androgen deprivation therapy in whom
metastatic prostate cancer developed following radical prostatectomy.
J Urol. 2008;179:156-161, discussion 161-162.
50. Schweizer MT, Zhou XC, Wang H, et al. Metastasis-free survival is associated with overall survival in men with PSA-recurrent prostate
cancer treated with deferred androgen deprivation therapy. Ann Oncol.
2013;24:2881-2886.
51. Fizazi K, Faivre L, Lesaunier F, et al. Androgen deprivation therapy plus
docetaxel and estramustine versus androgen deprivation therapy alone
for high-risk localised prostate cancer (GETUG 12): a phase 3 randomised controlled trial. Lancet Oncol. 2015;16:787-794.
52. Gravis G, Fizazi K, Joly F, et al. Androgen-deprivation therapy alone or with
docetaxel in non-castrate metastatic prostate cancer (GETUG-AFU 15):
a randomised, open-label, phase 3 trial. Lancet Oncol. 2013;14:149-158.
53. Gravis G, Boher JM, Joly F, et al; GETUG. Androgen deprivation therapy
(ADT) plus docetaxel versus ADT alone in metastatic non castrate
prostate cancer: impact of metastatic burden and long-term survival
analysis of the randomized phase 3 GETUG-AFU15 trial. Eur Urol. Epub
2015 Nov 21.
54. Vale CL, Burdett S, Rydzewska LH, et al; STOpCaP Steering Group. Addition
of docetaxel or bisphosphonates to standard of care in men with localised
or metastatic, hormone-sensitive prostate cancer: a systematic review
and meta-analyses of aggregate data. Lancet Oncol. 2016;17:243-256.
55. Shipley WU, Seiferheld W, Lukka H, et al. Report of NRG Oncology/RTOG
9601, a phase 3 trial in prostate cancer: anti-androgen therapy (AAT)
with bicalutamide during and after radiation therapy (RT) in patients
following radical prostatectomy (RP) with pT2-3pN0 disease and an
elevated PSA. Int J Radiat Oncol Biol Phys. 2016;94:3.
56. Parekh A, Chen MH, Graham P, et al. Role of androgen deprivation
therapy in early salvage radiation among patients with prostatespecific antigen level of 0.5 or less. Clin Genitourin Cancer. 2015;
13:e1-e6.
57. Hussain M, Tangen CM, Higano C, et al; Southwest Oncology Group Trial
9346 (INT-0162). Absolute prostate-specific antigen value after androgen deprivation is a strong independent predictor of survival in new
metastatic prostate cancer: data from Southwest Oncology Group Trial
9346 (INT-0162). J Clin Oncol. 2006;24:3984-3990.
58. Hussain M, Goldman B, Tangen C, et al. Prostate-specific antigen
progression predicts overall survival in patients with metastatic prostate
cancer: data from Southwest Oncology Group Trials 9346 (Intergroup
Study 0162) and 9916. J Clin Oncol. 2009;27:2450-2456.
59. Jereczek-Fossa BA, Beltramo G, Fariselli L, et al. Robotic image-guided
stereotactic radiotherapy, for isolated recurrent primary, lymph node
or metastatic prostate cancer. Int J Radiat Oncol Biol Phys. 2012;82:
889-897.
60. Schick U, Jorcano S, Nouet P, et al. Androgen deprivation and high-dose
radiotherapy for oligometastatic prostate cancer patients with less
than five regional and/or distant metastases. Acta Oncol. 2013;52:
1622-1628.
61. Decaestecker K, De Meerleer G, Lambert B, et al. Repeated stereotactic
body radiotherapy for oligometastatic prostate cancer recurrence.
Radiat Oncol. 2014;9:135.
62. Picchio M, Berardi G, Fodor A, et al. 11C-Choline PET/CT as a guide to
radiation treatment planning of lymph-node relapses in prostate cancer
patients. Eur J Nucl Med Mol Imaging. 2014;41:1270-1279.
63. Hussain M, Tangen CM, Berry DL, et al. Intermittent versus continuous
androgen deprivation in prostate cancer. N Engl J Med. 2013;368:
1314-1325.
129
Precision Medicine in Advanced

Prostate Cancer: Understanding
Genomics, Androgen Receptor
Splice Variants, and Imaging
Biomarkers
CHAIR
Himisha Beltran, MD
New York, NY
SPEAKERS
Emmanuel S. Antonarakis, MD
The Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins
Baltimore, MD
Gerhardt Attard, MD, PhD
The Institute of Cancer Research and The Royal Marsden Hospital
Michael J. Morris, MD
Memorial Sloan Kettering Cancer Center and Weill Cornell Medical College
New York, NY
TISSUE, LIQUID, AND IMAGING BIOMARKERS IN CRPC
Emerging Molecular Biomarkers in Advanced Prostate Cancer:

Translation to the Clinic
Himisha Beltran, MD, Emmanuel S. Antonarakis, MD, Michael J. Morris, MD, and
Gerhardt Attard, MD, PhD
OVERVIEW
Recent clinical and preclinical studies focused on understanding the molecular landscape of castration-resistant prostate
cancer (CRPC) have provided insights into mechanisms of treatment resistance, disease heterogeneity, and potential
therapeutic targets. This work has served as a framework for several ongoing clinical studies focused on bringing novel
observations into the clinic in the form of tissue, liquid, and imaging biomarkers. Resistance in CRPC typically is driven
through reactivation of androgen receptor (AR) signaling, which can occur through AR-activating point mutations, amplification, splice variants (such as AR-V7), or other bypass mechanisms. Detection of AR aberrations in the circulation
negatively impacts response to subsequent AR-directed therapies such as abiraterone and enzalutamide. Other potentially
clinically relevant alterations in CRPC include defects in DNA damage repair (at either the somatic or germline level) in up to
20% of patients (with implications for PARP1 inhibitor therapy), PI3K/PTEN/Akt pathway activation, WNT signaling pathway
alterations, cell cycle gene alterations, and less common but potentially targetable alterations involving RAF and FGFR2.
Imaging biomarkers that include those focused on incorporating overexpressed androgen-regulated genes/proteins, such as
prostate-specific membrane antigen (PSMA) and dihydrotestosterone (DHT) in combination with CT, can noninvasively
identify patterns of AR-driven distribution of CRPC tumor cells, monitor early metastatic lesions, and potentially capture
heterogeneity of response to AR-directed therapies and other therapeutics. This article focuses on the current state of
clinical biomarker development and future directions for how they might be implemented into the clinic in the near term to
improve risk stratification and treatment selection for patients.
everal drugs are either approved or in clinical development for men with metastatic CRPC with varied mechanisms of action. Although many drugs have demonstrated
antitumor activity, and even an overall survival benefit in an
unselected population, the identification of biomarkers to
help select the best next-line systemic therapy for an individual patient is an unmet clinical need. Molecular biomarkers have the potential to help clinicians with risk
stratification, inform patient selection for therapy, and assist
patient and family counseling (Fig. 1).
Primary prostate cancer is commonly multifocal at the
time of diagnosis. It has a long natural history with often
5 years or more before distant metastases are detected and
castration resistance ensues. In the monoclonal origin model
of clonal evolution supported by several studies,1,2 one
tumor clone/nodule present in a multifocal primary is responsible for metastasis. However, identification of that
dominant lesion at diagnosis is challenging because it may
not always be the lesion with the highest Gleason grade or of
the largest size.2 Further, molecular changes can occur in the

tumor with disease progression and treatment resistance,
and there is recent evidence of polyclonal metastasis-tometastasis seeding.3 Therefore, assessment of the archival
primary tumor in making clinical decisions regarding systemic therapies for patients with CRPC may not be adequate
or fully representative of the tumors current state.
Metastatic biopsies in advanced prostate cancer have not
been considered standard practice. To date, they have been
performed mainly in the context of clinical trials. Metastatic
biopsies can be challenging to perform because of the high
frequency of sclerotic bone-only metastases in CRPC, which
often require special biopsy procedures and tissue handling
for molecular profiling. Further, single-site metastatic biopsies do not necessarily represent the entire metastatic
tumor burden of the patient because of intrapatient heterogeneity. Therefore, alternative approaches, such as
blood-based assays (e.g., circulating tumor cell, circulating
tumor DNA) and molecular imaging, may be more feasible
From Weill Cornell Medicine, New York, NY, The Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, Baltimore, MD; Memorial Sloan Kettering Cancer Center, Weill Cornell
Medicine, New York, NY; The Institute of Cancer Research, London, United Kingdom, The Royal Marsden Hospital, London, United Kingdom.
Corresponding Author: Himisha Beltran, MD, Weill Cornell Medicine, 413 East 69th St., New York, NY 10021; email: hip9004@med.cornell.edu.
131
BELTRAN ET AL
with the given advantage of performing serial assessments

during the course of a patients disease to track tumor
dynamics and treatment response.
THE GENOMIC LANDSCAPE OF CASTRATIONRESISTANT PROSTATE CANCER:

CHARACTERIZATION OF METASTATIC
TISSUE BIOPSIES
Several recent studies have independently reported molecular data of tumors obtained from large cohorts of
patients with clinically localized prostate cancer4-6 and
metastatic CRPC.7-9 Certain genomic alterations, such as ERG
gene rearrangements and SPOP mutations, are present at
similar frequency in both localized and metastatic disease,
suggesting that these lesions predate metastatic spread.
Other alterations, such as AR point mutations or amplification, are present primarily in CRPC (50%60%), which
suggests they are acquired following androgen deprivation.10,11 The acquisition of activating alterations involving
AR (or AR cofactors) seems logical, as multiple studies have
confirmed that most castration-resistant prostate tumors
remain dependent on AR signaling.12 Mechanisms associated with reactivation of AR signaling in CRPC include acquisition of AR genomic alterations, AR mRNA splice variant
expression (such as AR-V7), cofactor activation, AR bypass
(such as glucocorticoid receptor expression), or AR crosstalk
KEY POINTS
132
The study of biopsies from metastatic sites from patients

with advanced prostate cancer has revealed potentially
actionable or prognostic genomic alterations. Common
genomic alterations observed in CRPC are: ERG gene
fusion (40%50%), AR gene point mutation or
amplification (50%60%), TP53 mutation or deletion
(40%50%), PTEN deletion (40%50%), RB1 deletion
(20%), and alterations in DNA repair genes (20%).
Circulating tumor DNA analyses in patients with CRPC
have revealed intrapatient molecular heterogeneity and
the potential to track dynamic changes during the
course of therapeutic response and resistance in
patients with advanced prostate cancer.
Detection of the AR-V7 splice variant in circulating
tumor cells or AR genomic aberrations in circulating
tumor DNA has been associated with decreased clinical
response to abiraterone or enzalutamide. Prospective
trials underway are evaluating the prognostic and
predictive utility of AR-V7 and exploring novel agents
with potential activity against AR-Vexpressing CRPC.
Bi-allelic loss of DNA repair genes (e.g., BRCA1/2, ATM)
has been associated with clinical response to the PARP
inhibitor olaparib.
Advances in molecular imaging and assessment of bone
lesions have shown promise for early detection of
metastasis and detection of biologic heterogeneity
between sites.
pathways.13 What remains unclear is why some, but not all,

tumors in patients develop AR alterations and when they
occur during the course of AR-directed therapies. As described below, liquid biopsies have shown promise as a
means to study AR dynamics during the course of treatment.
Clinical and preclinical studies investigating therapeutic
approaches to better target the AR (and AR splice variants) in
the face of AR-driven resistance are underway; these include
combined CYP17A1 inhibition and AR antagonism, targeting
the AR N-terminal or DNA binding domain, or AR degradation. A subset of CRPC also may lose AR independence because of loss of AR expression and/or signaling. Identifying
patients who will no longer benefit from AR-directed therapies is also an area of biomarker investigation. One proposed mechanism of AR independence includes the
development of epithelial plasticity associated with pathologic and molecular features of neuroendocrine cancer.14,15
There are several other genomic aberrations reported or
enriched in CRPC, some of which also occur at a lesser
frequency in primary disease, many predicted to be actionable (such as PI3K/Akt/PTEN, RAF, FGFR2, WNT pathway,
cell cycle, and DNA repair alterations; recently described in
Robinson et al8). This knowledge has served as a framework
for preclinical testing of novel agents and cotargeting approaches and the design of biomarker-driven clinical studies.
Present at similar frequency in approximately 40% to 50% of
primary prostate cancer and CRPC, the TMPRSS2-ERG gene
fusion is prostate cancerspecific and androgen-regulated,
making it an attractive drug target. TMPRSS2-ERG does not
encode for a chimeric protein; instead, it results in overexpression of a truncated ERG protein in the setting of active
androgen signaling, driving a transcriptional program linked
to DNA damage and invasion.16 It is challenging to develop
drugs that target transcription factors, but short peptidomimetics that bind to ERG, bromodomain inhibitors, and
PARP inhibitors have been explored preclinically, and olaparib currently is being investigated in combination with
abiraterone in an ERG-fusion biomarkerstratified phase II
clinical trial (NCT01972217). Alteration of genes leading to
aberrant activation of the PI3K/Akt pathway, most commonly loss of PTEN, is implicated in 40% to 50% of CRPC cases
and is associated with signaling crosstalk with AR through
reciprocal negative feedback.17,18 Based on this, there are
several ongoing phase II trials investigating cotargeting of
both pathways (NCT01251861, NCT02525068, NCT02215096,
NCT02407054).
Somatic aberrations involving DNA defect repair genes
(such as deletions/mutations involving BRCA1/2, ataxiatelangiectasia mutated [ATM], CHEK2, Fanconi [FANC]
genes) are present in up to 20% of CRPC tumors. Somatic loss
of the BRCA2 gene, for instance, occurs in approximately
3% of localized prostate cancer cases (TCGA)6 and 13% of
CRPC cases (SU2C-PCF).8 Notably germline mutations, uncommonly present in patients diagnosed with localized
disease, are present at higher frequency in patients with
CRPC (in keeping with their association with more aggressive
disease at diagnosis).8,19,20 Tumors with defects in DNA
FIGURE 1. Schematic of Various Types of Molecular Biomarkers
This schematic illustrates various types of molecular biomarkers currently being tested in trials for patients with advanced prostate cancer, with a goal of ultimately
improving clinical decision making. Image credit: L.S. Beltran.
repair would be predicted to have disrupted homologous

recombination; therefore, they would be more sensitive to
inhibition of PARP1 or DNA cross-linking agents (such as
platinum chemotherapy) through a mechanism of synthetic
lethality. The PARP inhibitor olaparib has shown antitumor
activity in germline carriers of BRCA1/2 gene alterations
across cancer types, including prostate,21 and it is approved
for clinical use in patients with BRCA-mutated ovarian
cancer.
To investigate whether there are additional patients with
CRPC who may also benefit from PARP inhibition, such as
those with somatic alterations in DNA repair, Mateo et al
developed an adaptive design phase II trial of olaparib where
pretreatment tissue and blood biomarkers were investigated prospectively.19 In the 49 evaluable patients
enrolled in this study, overall response rate (defined by
either radiographic, prostate-specific antigen [PSA], or circulating tumor cell [CTC] conversion) was 33% with 12
patients receiving olaparib for more than 6 months and four
patients receiving the drug for more than 12 months. The
majority of responders on the study were found to have
either germline or somatic alterations in DNA repair genes;
14 of 16 DNA-repair altered cases (88%) had a response to
olaparib, which was associated with prolonged overall
survival (median, 13.8 vs. 7.5 months in those without DNA
repair alterations). Based on these promising results, olaparib recently received a U.S. Food and Drug Administration
(FDA) breakthrough designation for patients with BRCA1/2or ATM-mutated metastatic CRPC who have received a prior
taxane-based chemotherapy and at least either enzalutamide or abiraterone. A phase III study is planned. Reports
linking BRCA2 and FANCA alterations and platinum
133
BELTRAN ET AL
sensitivity in patients with CRPC also have been reported

recently.22,23
A question now is whether metastatic biopsies will soon be
warranted to look for DNA repair defects and, if so, when
should this be performed? Can these be detected reliably
using circulating tumor DNA or other noninvasive means?
Where will drugs such as PARP inhibitor or platinum chemotherapy ultimately be used within the growing clinical
armamentarium of approved drugs? A higher frequency of
germline alterations involving DNA repair genes in CRPC has
not only therapeutic considerations but also familial implications. Who and when to screen for germline alterations
and how genetic counseling and testing will be performed
and reimbursed are areas of active discussion.
Defects in DNA mismatch repair genes such as MLH1 and
MSH2 and microsatellite instability also occur at the somatic
level in up to 5% of prostate cancers8,24 and in the germline
DNA of Lynch syndrome families.25 This results in a hypermutated phenotype with a mutation rate typically greater
than 10-fold higher than other CRPC tumors. Given observations in other tumor types linking mutation rates with
response to immune checkpoint inhibitors,26 identifying this
small subset of cases may have clinical implications in the
context of immunotherapies. The design and recruitment for
biomarker-enriched trials involving low-frequency subsets,
such as those patients with mismatch repair defects, are
challenging and rely on multi-institutional collaboration.
Although informative, most of the genomic studies performed to date in CRPC have represented a snapshot in time.
How early events (such as SPOP, PTEN, or ERG) might cooperate with potentially later events to drive tumor growth,
and how these affect response to subsequent therapies, the
tumor microenvironment, and the epigenetic landscape are
areas of active investigation. Translation of genomics into the
clinic also will require a better understanding of tumor
evolution and intrapatient heterogeneity. Longitudinal
studies, including the two SU2C-PCF Dream Teams, and
multisample cases (such as rapid autopsies) have started to
provide new insights that are having an impact on the clinical
interpretation of single-site biopsies in the context of routine
clinical care and insights into mechanisms of tumor evolution.
DEVELOPMENT OF PROGNOSTIC AND

PREDICTIVE CIRCULATING BIOMARKERS
Because serial metastatic biopsies are not always feasible or
safe for most men with CRPC, noninvasive assays including
CTCs and/or plasma DNA may be an effective means to
assess biomarker status and capture heterogeneity. Prospective studies using the CellSearch system confirmed the
association between CTC count ($ in 7.5 mL) in patients with
advanced prostate cancer and worse clinical outcome.27 An
association with improved outcome for patients who had a
CTC conversion (from $ 5 to , 5) compared with those who
did not suggested that this approach could be used as an
early indicator of response and as a surrogate endpoint in
clinical trials. This led to the prospective evaluation of CTC
count in the phase III COU-301 trial of abiraterone and
134
prednisone in docetaxel-treated metastatic CRPC, which

showed that a biomarker panel containing CTC count and
lactate dehydrogenase level was a surrogate for survival at
the individual-patient level.28 Validation in additional positive phase III trials is now required.
Isolating and characterizing rare CTCs in the blood compartment composed of millions of white blood cells introduces technologic and logistical challenges. Several more
recent platforms are showing promise in obtaining important multiplex phenotypic data,29,30 and planned prospective studies will investigate associations with clinical
behavior and outcome. Additionally, as described below, the
AdnaTest platform allows the capture of CTCs with preservation of mRNA quality, which can then be used to study
expression of candidate biomarkers such as AR-V7. Overall,
the number of CTCs is associated with number (and site) of
metastases.31,32 Using currently available platforms, the
majority of patients with low-volume metastases have fewer
than 510 CTCs in 10 mL of blood. Newer approaches may
improve on this detection rate or allow processing of larger
volumes of blood. An alternative approach to CRPC molecular characterization is to study tumor-specific aberrations in cell-free DNA from plasma (i.e., the upper
compartment from blood centrifuged to precipitate the
cellular component).
Monitoring Plasma DNA to Modify Patient Treatment

Using next-generation sequencing approaches applied to
plasma DNA, tumor aberrations are detected in more than
95% of patients with progressing metastatic CRPC.33,34 These
are admixed with normal DNA at variable quantities, and
their relative abundance can provide information on tumor
behavior. Studies in prostate cancer and several other tumor
types have reported that changes in circulating tumor DNA
levels are associated with response to treatment and improvements in outcome.35-37 By monitoring sequential
plasma samples from patients with CRPC treated with
abiraterone, emergence or selection of an AR mutation
resulting in either a T878A or an L702H amino acid change
was observed in 15% to 20% of patients, often several months
before manifest clinical or radiologic progression.34,36 Preclinical studies confirm that these two mutations are activated by progesterone or prednisolone respectively, both
present at increased levels in patients treated with abiraterone and supporting their causative association with
resistance.36,38 This was followed by a study of pre- and posttreatment plasma samples from 100 patients treated with
abiraterone, the majority of whom were postdocetaxel and/or
had previously received potent AR-targeting agents. A minimum circulating tumor content (7.5%, present in 80 of 97
pretreatment samples), defined as the abundance or allelic
frequency of tumor aberrations measured using a custom
panel including six informative prostate cancer lesions, was
mandated to allow evaluation of AR copy number gain. Data
reported include a strong association between pretreatment
AR copy number gain and/or T878A or L702H AR mutations
and PSA decline (4.9 and 7.8 times less likely to have a $ 50
or $ 90% decline) and overall survival (HR 7.33; 95% CI,

3.5115.34, p = 1.3 3 1029) and progression-free survival
(PFS; HR 3.73; 95% CI, 2.176.41, p = 5.6 3 1027).34 Similarly,
an association between AR copy number or an exon 8 AR
mutation and resistance to enzalutamide in 39 patients with
CRPC has been reported.39 Increased AR copy number is
associated with higher AR mRNA transcripts (in tumor tissue),40 which drive resistance to androgen suppression
in vitro. Ongoing studies, as described below, will shed light on
the association between AR copy number and AR splice
variants and their association with other aberrations that
drive treatment resistance in CRPC. Overall, these data support the potential clinical applicability of circulating tumor
DNA studies in CRPC. Prospective evaluation of plasma AR
gene aberrations in randomized trials with novel and standard
AR targeting agents, taxanes, and radiopharmaceuticals are
now planned. Moreover, plasma DNA studies appear to
provide additional or complementary molecular data to tumor
biopsies,36,41 introducing the possibility of patient selection
for single-agent and combination therapeutic strategies targeting additional targets, including DNA repair defects and
PI3K/Akt.
AR-V7 as a Clinical Biomarker

Androgen receptor splice variants are abnormally truncated
isoforms of the AR that are capable of activating transcription of androgen-regulated genes without the requirement of ligand.42-45 Although at least 22 AR splice
variants have been reported,8 AR-V7 is the most frequently
observed and the most abundant AR-V detected at the
mRNA and protein level in clinical specimens from patients
with CRPC. A number of retrospective studies have suggested that the presence of AR-V7 is associated with more
rapid disease progression and shorter survival in men with
metastatic CRPC.46-49 The first prospective study to evaluate
the prognostic impact of AR-V7 was conducted by Efstathiou
et al.50 In that study, the detection of AR-V7 by immunohistochemistry in bone marrow biopsy specimens of patients
with CRPC was associated with resistance to enzalutamide;
AR-V7 protein was detected in 57% of men who developed
disease progression within 4 months of starting enzalutamide and was not detected in any patient who responded to
enzalutamide for longer than 6 months (p = .02). These
findings were further corroborated in a separate study by
the same investigators evaluating the combination of
abiraterone and enzalutamide.51
More recently, Antonarakis et al52 reported a prospective
study assessing the prognostic role of AR-V7 detected at the
mRNA level by reverse-transcriptase (RT)-PCR from CTCs of
patients receiving abiraterone or enzalutamide. They found
that the presence of AR-V7 in CTCs was associated with
lower PSA response rate, shorter PFS, and shorter overall
survival compared with those patients without detectable
circulating AR-V7.52 Notably, the prevalence of AR-V7 was
higher in men treated with enzalutamide who previously had
received abiraterone and in men treated with abiraterone
who had previously received enzalutamide; AR-V7 prevalence
was lowest in men who had not received either agent. Additionally, when assessing serial CTC samples over time, the
authors reported that all men with baseline detection of
AR-V7 remained AR-V7positive during the course of
therapy with abiraterone and enzalutamide, whereas 14%
of men with negative AR-V7 status at baseline converted
to AR-V7positive during treatment; these patients had
intermediate clinical outcomes.52
Another question is whether the presence of AR-V7 is
relevant in the setting of taxane chemotherapy, especially
because two preclinical studies have produced conflicting
results.48,53 Accordingly, Antonarakis et al54 performed a
second prospective study using their CTC-based AR-V7 assay
in 30 patients beginning treatment with docetaxel and seven
patients beginning treatment with cabazitaxel. The prevalence of AR-V7 in these patients, most of whom had previously received abiraterone and/or enzalutamide, was 46%.
PSA responses were observed in both AR-V7positive and
AR-V7negative men (41% vs. 65%; p = .19). Similarly, PFS
was not statistically different in patients who were ARV7positive and negative (p = .11). As a hypothesisgenerating exercise, Antonarakis et al incorporated data
from their prior study of 62 patients treated with abiraterone and enzalutamide and showed that clinical outcomes appeared to be better with taxanes compared with
enzalutamide or abiraterone in men who were AR-V7
positive, although outcomes did not appear to differ by
treatment type in men who were AR-V7negative. More
specifically, in patients who were AR-V7 positive, PSA responses were higher in men treated with taxane compared
with enzalutamide or men treated with abiraterone (41% vs.
0%, p , .001), and PFS was longer in men treated with
taxane (HR 0.21, p = .003). Intriguingly, 58% of patients with
baseline AR-V7positive status converted to AR-V7
negative status during treatment with docetaxel or cabazitaxel. Whether such transitions in AR-V7 status may
resensitize such patients to further AR-directed therapies is
unknown.55
Taken together, the clinical data to date suggest that the
presence of AR-V7 may be a marker of resistance to ARdirected therapy but not taxane chemotherapy; therefore, it
may represent a treatment selection biomarker in CRPC.
These findings now warrant further prospective validation
and clinical qualification before CTC-based AR-V7 testing
should be used in clinical practice to inform treatment
decisions. A number of AR-V7 biomarker validation studies
currently being conducted are summarized in a recent review.56 The Sanofi-sponsored PRIMCAB trial (NCT02379390)
is investigating cabazitaxel chemotherapy versus ARdirected therapy in men with metastatic CRPC who have
developed disease progression within 6 months of starting
abiraterone or enzalutamide. As a secondary endpoint, that
trial will prospectively evaluate baseline AR-V7 status from
CTCs as a putative predictive biomarker in this setting, in
which the prevalence of AR-V7 is expected to be about 33%.
The Johns Hopkins AR-V7 assay will serve as the central
laboratory for AR-V7 testing. Exploratory analyses will also
135
BELTRAN ET AL
evaluate transitions in AR-V7 status at the time of progression. In a Prostate Cancer Foundation (PCF)-sponsored
prospective biomarker trial (NCT02269982) coordinated by
the Duke Cancer Institute, three different CTC-based AR-V7
assays will be evaluated in 120 men with taxane-naive
metastatic CRPC and high-risk clinical features. In this
noninterventional trial, patients will receive standard-ofcare abiraterone or enzalutamide and then may also receive standard-of-care taxane at progression. AR-V7 testing
will be performed prior to AR-directed therapy, at progression on AR-directed therapy, and also at progression on
chemotherapy. Each patient will undergo AR-V7 testing with
three assays at each time point: the Johns Hopkins mRNAbased assay, the Weill-Cornell mRNA-based assay (which
also evaluates other AR variants), and the EPIC Sciences
protein-based assay.
Although no approved agents currently directly target
AR-V7, a number of compounds are now in clinical development; these are summarized in a recent review.56 Galeterone (Tokai Pharmaceuticals) is an oral drug with three
proposed mechanisms of action, including inhibition of
CYP17 lyase, antagonism of the AR ligand-binding domain,
and degradation of both full length AR (AR-FL) and AR splice
variants through a proteasome-dependent pathway.56,57
In a post hoc analysis of the phase II ARMOR2 trial, among
seven patients with AR C-terminal loss (as determined using
an immunohistochemistry-based CTC assay), six men
achieved more than a 50% PSA reduction with galeterone
treatment. Based on these preliminary data, a registration
phase III trial (ARMOR3-SV) was launched in 2015
(NCT02438007). Eligible patients are those with AR-V7
positive metastatic CRPC without prior treatment with novel
AR-directed therapies or taxane chemotherapies. AR-V7
testing will be conducted using a CLIA-certified assay developed by Qiagen. Patients who are CTC-positive and ARV7positive (148 total patients) will be randomly selected to
receive either galeterone or enzalutamide. Notably,
ARMOR3-SV is the first registration trial in prostate cancer to
use a biomarker-selection trial design.
EPI-506 (ESSA Pharma), an oral prodrug of EPI-002, is the
first drug reported as capable of targeting the AR N-terminal
domain.58 Specifically, EPI-506 binds covalently to the Tau5
region of the AF1 domain of the N-terminus, a region that is
responsible for 99% of the transcriptional activity of AR.
Because the N-terminus is present in both the AR-FL and in
all of the AR splice variants (including AR-V7), treatment with
EPI-506 would be expected to inhibit all forms of AR signaling. Preclinical studies with EPI-002 have shown that this
compound has activity in several AR-V7expressing cell lines
and xenograft models.59 A phase I trial (including a subsequent phase II expansion) using EPI-506 opened in 2016
(NCT02606123). Eligible patients are those with metastatic
CRPC who have previously received either abiraterone or
enzalutamide; one prior taxane is also permitted but not
required.60 Exploratory analyses of AR-V7 and plasma AR
gene aberrations also will be conducted, but this information
will not be used for patient selection or stratification.
136
In a drug library screen aimed at identifying FDA-approved

drugs capable of targeting AR-V7, the antihelminthic drug
niclosamide emerged as an unexpected success.61 Additional mechanistic studies suggested that this agent functions by degrading AR-V7 through a proteasome-dependent
pathway and that AR-V7 degradation occurred more rapidly
than AR-FL degradation. Niclosamide demonstrated substantial antitumor activity in a number of AR-V7expressing
CRPC cell lines and xenograft models resistant to enzalutamide, and the combination of niclosamide and enzalutamide produced maximal tumor inhibition. Based on these
preclinical data, a phase I clinical trial was launched in 2015
(NCT02532114) for men with abiraterone-pretreated CRPC
who test positive for AR-V7 using an AR-V7 assay performed
at the University of Washington. In this trial, patients will
receive enzalutamide plus escalating doses of oral niclosamide. Exploratory analyses will evaluate changes in AR-V7
status during the course of niclosamide treatment and at the
time of progression.
IMAGING METASTATIC DISEASE: CURRENT AND

FUTURE PERSPECTIVES
The tissue and blood-based technologies described above
may accelerate drug development and personalize clinical
decision making by yielding biomarkers for drug sensitivity
and response. Advances in imaging technology can hasten
the pace of such progress. Imaging is capable of not only
assessing disease extent and distribution, but it can also
identify biologic features of a patients totality of lesions
rather than those of a single biopsy or blood draw. Yet, of all
of the biomarkers used for prostate cancer staging
prognostication, prediction, response assessment, and biologic characterizationimaging has been one of the most
challenging to develop.
A substantial basis of these difficulties is that prostate
cancer is a bone-tropic solid tumor, which is difficult to
assess, much less measure. The Tc-99 MDP tracer is incorporated into reactive bone rather than the tumor.
Consequently, bone scintigraphy is more sensitive to progression than response. Importantly, in 20% to 50% of patients, the bone scan will paradoxically worsen shortly after
the patient is placed on effective therapy, which can thereby
mislead the inattentive clinician to interpret a worsening
scan as disease progression rather than response.62-64 Despite these limitations, bone scintigraphy is ubiquitously
available, inexpensive, and the existing standard for imaging
bone. Data collection by the Prostate Cancer Working Group
2s definition for disease progression was standardized,65
incorporated into case report forms, and then the definition
and data collection methods were prospectively validated
for reproducibility and tested for an association with overall
survival in three prospective clinical trials involving abiraterone (NCT00887198), enzalutamide (NCT01212991),
and orteronel (NCT01193244). These trials demonstrated
excellent concordance between central and investigator
reviewers. 66,67 In the first trial assessed, a significant
association between radiographic PFS and overall survival
was found (Spearmans coefficient, 0.72), and the endpoint

was used for regulatory approval of abiraterone for patients
with chemotherapy-naive CRPC.65,68,69 Newer methods of
using bone scintigraphy involve quantitative descriptors of
disease burden, such as the bone scan index, which is the
fraction of skeletal mass represented by tracer uptake70 and
can be used to describe both response and progression. The
bone scan index has been shown to be prognostic of overall
survival in a preliminary retrospective analysis,71 and automated systems are undergoing analytic validation.72-74
Bone-directed PET/CT imaging using 18F sodium fluoride
(NaF) may more accurately localize disease and produce a
quantitative output amenable to biomarker development
and modeling. However, it is still subject to flare phenomena
and does not directly demonstrate disease. Early studies
suggested that the modality was superior to both fluorodeoxyglucose (FDG) PET and 99mTc-MDP single photon
emission computed tomography (SPECT) in detection of
bone metastases.75-77 These findings were supported by
subsequent meta-analyses, such as one analyzing 11 studies
involving 613 patients that concluded, on a per-patient basis,
NaF PET had a sensitivity and specificity of 96% (93%98%)
and 91% (88%94%) versus 88% (84%96%) and 80%
(75%84%) for bone scintigraphy.78 The National Oncology
PET Registry (NOPR) is examining the use of NaF PET for
clinical decision making. Data on 1,940 patients with prostate cancer demonstrated that NaF PET findings prompted a
change in treatment in 41.8% of patients. However, NOPR
did not seek to, nor did it define, a standard definition of
either response or progression, nor assess whether the
change in treatment was justified or associated with patient
benefit. A preliminary multicenter study (NCT01516866)
recently has been completed that delineates the reproducibility of NaF across centers and examines posttreatment changes in patients treated with AR-directed
therapy and taxane chemotherapy. As it stands now, it is
not clear that NaF PET/CT confers clinically useful information as a response biomarker, and per the Prostate
Cancer Working Group 2 (and the soon-to-be published
Prostate Cancer Working Group 3), should be considered
investigational as an endpoint in clinical trials.79
Assessing the actual disease burden in a patient, as opposed to statistical risk assessments of either distant or
localized disease as determined by mathematical
modeling,80-82 is key to personalized decision making and
clinical trial design and risk stratification. Accurate staging is
particularly important for men with high-risk localized disease and for those whose disease has biochemically relapsed
following local therapy. In these cases, decision making
frequently occurs without the benefit of imaging findings
that declare the patients true distribution of disease. New
molecular imaging techniques appear to be superior to
standard cross-sectional imaging and bone scintigraphy,
although it is unclear how to best use this information for
clinical care.
Prostate-specific membrane antigen is a transmembrane
carboxypeptidase that is expressed in prostatic tissue,
upregulated in prostate cancer (especially CRPC), and

present regardless of disease site.83 In 1996, the FDA approved Indium (111In) capromab pendetide,84 based on an
antibody that targets the internal domain of PSMA. Newer
tracers, such as the Zr-89 labeled anti-PSMA antibody J591,
target the external domain of the molecule. A detailed
lesional analysis of 89Zr-J591 in men with CRPC, verified by
pathology, revealed 89Zr-J591 detected 491 osseous sites
compared with 339 by MDP in 50 patients.85 The overall
accuracy of 89Zr-J591 by pathology was 95.2% for osseous
lesions; however, the performance in soft tissue was less
accurate, at an accuracy of 60%. A major drawback of intact
antibodies is a long half-life, resulting in delays between
injection and imaging, which can be overcome by the use of
minibodies86 or small molecules that inhibit the enzymatic
site of PSMA. An illustrative example of a Zr-89 radiolabeled
minibody PET scan (IAB2M, ImaginAb, Inglewood, CA) with
detection of bone lesions in a patient with CRPC is shown in
Fig. 2. Although molecular imaging has the capacity to
demonstrate a greater extent of disease earlier in the disease course, what to do with that information clinically is
unknown.
The most commonly used anti-PSMA small molecule
ligand is Glu-NH-CO-NH-Lys-(Ahx)-[68Ga(HBED-CC)](68GaDKFZ-PSMA-11), which has been extensively studied in
Europe. In a recent retrospective analysis, 319 patients with
progressive disease across a variety of clinical states were
scanned.87 On a lesional level, sensitivity and specificity were
76.6% and 100%, respectively. A patient-based analysis
revealed a sensitivity of 88.1%. Especially relevant for the
patient who experienced biochemical relapse was early
evidence that disease could be detected in patients with very
low PSA values. Lesions were detected in approximately 47%
(8/17) of patients with a PSA of less than or equal to 0.2, 50%
(5/10) of cases with PSA were 0.21 to less than or equal to
0.5, 58% (14/24) of cases with a PSA between 0.51 and 1,
72% (28/31) in the range of 0.5 to less than 1, and greater
than 90% with higher PSA values (where the number of
patients was much greater). In another large retrospective
analysis of 248 patients, detection rates were 96.8%, (PSA
value 2 ng/mL or higher), 93.0% (PSA value 1-2 ng/mL),
72.7% (PSA value 0.5-1 ng/mL), and 57.9% (PSA value 0.2-0.5
ng/mL).88 These analyses suggest that PSMA-based imaging
might detect disease at PSA ranges in which decisions about
salvage radiotherapy are indeed highly relevant, and would
appear to be more informative than standard imaging
modalities.89
11
C choline is chemically identical to physiologic choline
and is taken up by cancer cells requiring phospholipids to
create cell membranes. The detection of prostate cancer by
11
C choline was described in a recent pooled analysis of 12
studies involving 1,055 patients with biochemical relapse
who underwent either 11C choline or 18F fluorocholine. The
pooled sensitivity, specificity, and diagnostic odds ratio
was 85% (95% CI, 79%89%), 88% (95% CI, 73%95%), and
41.4 (95% CI, 19.7%86.8%), respectively.90 In another
meta-analysis involving 19 studies and 1,555 patients,91 a
137
BELTRAN ET AL
FIGURE 2. Imaging of Bone Lesions in a Patient With Castration-Resistant Prostate Cancer
Imaging of bone lesions of a patient with metastatic CRPC with bone scintigraphy (99mTc-MDP), FDG MIP, and a novel PSMA-directed PET scan using a Zr-89 radiolabeled
minibody (IAB2M, ImaginAb, Inglewood, CA). Although significantly more lesions are identified by IAB2M, whether the informative biomarker of novel imaging modalities, be it some
property of the SUV, lesion number, lesion location, change over time, or some other property, is as yet undefined.
Abbreviations: CRPC, castration-resistant prostate cancer; FDG MIP, fluorodeoxyglucose maximum intensity production; PSMA, prostate-specific membrane antigen; SUV,
standard uptake value.
pooled sensitivity of 85.6% (95% CI, 82.9%88.1%) and

specificity of 92.6% (95% CI, 90.1%94.6%) for all sites of
disease (prostatic fossa, lymph nodes, and bone) was reported. However, 11C choline was relatively insensitive for
patients with PSA values less than 2 ng/mL, with a 5% detection rate for PSA levels of less than 1 ng/mL, 15% for PSA
levels of 1 to 2 ng/mL, and 28% for PSA levels of greater than
2 ng/mL.92 As PSA increases, and particularly as PSA doubling
time decreases, 11C-choline PET is more likely to be positive.93,94 In the context of these European data and a small
number of U.S. studies, the FDA approved 11C choline for
patients with biochemical relapse after primary therapy
when conventional imaging does not indicate a definitive
sign of cancer. However, the modality has only limited
availability in the United States, and the indication makes a
particular point that the PET is not a replacement for tissue
sampling and testing.95
Numerous other tracers are being used or developed for
disease detection. However, there are limited prospective
studies to date that have an underlying hypothesis, adequate
power, and a preplanned strategy for confirming that imaging findings are true or false positives or negatives. Further, the actionable findings of these imaging modalities, and
the clinical pathways that such findings should generate,
remain undefined. Detection of early distant disease provokes improvised surgical plans for local primary disease,
lymph node dissections in the salvage setting, irradiation of
individual metastatic foci for the oligometastatic setting, or
the institution of systemic therapy, in the absence of any data
that such maneuvers are beneficial, much less not harmful.
Trials to define these treatment algorithms must be performed and currently are being planned and opened now.
As the age of molecular profiling in prostate cancer dawns,
imaging could allow for predictive biomarkers and treatment
selection by global assessments of a patients bony, nodal,
138
and soft-tissue disease. The image and the therapy could

also conceivably target the same molecule, such as PSMA.
An early example of the power of such technology is 16b18
F-Fluoro-5a-Dihydrotestosterone (FDHT), a fluorinated ligand
of the AR.96 FDHT is able to detect the presence of overexpressed AR in CRPC, and can therefore be used to establish
drug targeting, as it was in the early clinical development of
enzalutamide.97 The tracer also was used to establish the
phase II dose of the AR antagonist ARN509.98 Although one
tracer can establish the targeting of the drug under study,
another could be used to detect post-treatment signaling, as
has been shown preclinically. For example, PSMA imaging
can be used to detect changes in AR signaling as a result of
treatment with enzalutamide.99 Ultimately, molecular imaging would not only be used as an early drug development
tool but as biomarkers of a patients susceptibility or resistance to a given therapy, a means to capture differences
between metastatic lesions, and/or as an indicator of early
response.
CONCLUSION
Recent studies focused on integration of preclinical observations with clinical information, including specimens and
imaging, and patient response data have started to reveal
the intricacies underlying inter- and intrapatient heterogeneity in CRPC, possibly driving differences in clinical response to systemic agents. Tissue, liquid, and imaging-based
biomarkers in prostate cancer hold great potential as biomarkers for response assessments, staging and early detection of disease, and biologic characterization and drug
development (Fig. 1). However, translation of such observations into the clinic to eventually impact care requires
prospective studies demonstrating clinical value of biomarkers for prognostication or prediction of response. It is
crucial that development be performed with no less rigor
than a drug would be developed. For this to happen, more

prospective analytic and clinical validation studies must be
performed that are controlled for clinical states and have
prespecified endpoints, predefined interventions, and adequate statistical power. The development of clinically
qualified biomarkers is compounded by the regulatory
framework for biomarker development, 100,101 which
requires a rigorous analytic validation process in which the
performance characteristics of the modality are defined,

followed by clinical qualification with hypothesis-driven
retrospectively or prospectively collected evidence of an
association between the biomarker and a clinically meaningful event. Although the technologies are ever more
available on a routine basis, the appropriate trials to define
how to use these tools to promote how patients feel,
function, or survive must be performed.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
Liu W, Laitinen S, Khan S, et al. Copy number analysis indicates

monoclonal origin of lethal metastatic prostate cancer. Nat Med.
2009;15:559-565.
Haffner MC, Mosbruger T, Esopi DM, et al. Tracking the clonal origin of
lethal prostate cancer. J Clin Invest. 2013;123:4918-4922.
Gundem G, Van Loo P, Kremeyer B, et al; ICGC Prostate UK Group. The
evolutionary history of lethal metastatic prostate cancer. Nature.
2015;520:353-357.
Baca SC, Prandi D, Lawrence MS, et al. Punctuated evolution of
prostate cancer genomes. Cell. 2013;153:666-677.
Barbieri CE, Baca SC, Lawrence MS, et al. Exome sequencing identifies
recurrent SPOP, FOXA1 and MED12 mutations in prostate cancer. Nat
Genet. 2012;44:685-689.
Cancer Genome Atlas Research Network. The molecular taxonomy of
primary prostate cancer. Cell. 2015;163:1011-1025.
Grasso CS, Wu YM, Robinson DR, et al. The mutational landscape of
lethal castration-resistant prostate cancer. Nature. 2012;487:239-243.
Robinson D, Van Allen EM, Wu YM, et al. Integrative clinical genomics
of advanced prostate cancer. Cell. 2015;161:1215-1228.
Beltran H, Yelensky R, Frampton GM, et al. Targeted next-generation
sequencing of advanced prostate cancer identifies potential therapeutic targets and disease heterogeneity. Eur Urol. 2013;63:920-926.
Taplin ME, Bubley GJ, Shuster TD, et al. Mutation of the androgenreceptor gene in metastatic androgen-independent prostate cancer.
N Engl J Med. 1995;332:1393-1398.
Visakorpi T, Hyytinen E, Koivisto P, et al. In vivo amplification of the
androgen receptor gene and progression of human prostate cancer.
Nat Genet. 1995;9:401-406.
Chen CD, Welsbie DS, Tran C, et al. Molecular determinants of resistance to antiandrogen therapy. Nat Med. 2004;10:33-39.
Watson PA, Arora VK, Sawyers CL. Emerging mechanisms of resistance
to androgen receptor inhibitors in prostate cancer. Nat Rev Cancer.
2015;15:701-711.
Beltran H, Rickman DS, Park K, et al. Molecular characterization of
neuroendocrine prostate cancer and identification of new drug targets. Cancer Discov. 2011;1:487-495.
Beltran H, Prandi D, Mosquera JM, et al. Divergent clonal evolution of
castration-resistant neuroendocrine prostate cancer. Nat Med. 2016.
In press.
Yu J, Yu J, Mani RS, et al. An integrated network of androgen receptor,
polycomb, and TMPRSS2-ERG gene fusions in prostate cancer progression. Cancer Cell. 2010;17:443-454.
Carver BS, Chapinski C, Wongvipat J, et al. Reciprocal feedback regulation of PI3K and androgen receptor signaling in PTEN-deficient
prostate cancer. Cancer Cell. 2011;19:575-586.
Mulholland DJ, Tran LM, Li Y, et al. Cell autonomous role of PTEN in
regulating castration-resistant prostate cancer growth. Cancer Cell.
2011;19:792-804.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.

Castro E, Goh C, Olmos D, et al. Germline BRCA mutations are associated
with higher risk of nodal involvement, distant metastasis, and poor
survival outcomes in prostate cancer. J Clin Oncol. 2013;31:1748-1757.
Fong PC, Boss DS, Yap TA, et al. Inhibition of poly(ADP-ribose) polymerase in tumors from BRCA mutation carriers. N Engl J Med. 2009;
361:123-134.
Beltran H, Eng K, Mosquera JM, et al. Whole-exome sequencing of
metastatic cancer and biomarkers of treatment response. JAMA
Oncol. 2015;1:466-474.
Cheng HH, Pritchard CC, Boyd T, et al. Biallelic inactivation of BRCA2 in
platinum-sensitive metastatic castration-resistant prostate cancer.
Eur Urol. 2015;S0302-2838(15)01177-X.
Pritchard CC, Morrissey C, Kumar A, et al. Complex MSH2 and MSH6
mutations in hypermutated microsatellite unstable advanced prostate cancer. Nat Commun. 2014;5:4988.
Beltran H. DNA mismatch repair in prostate cancer. J Clin Oncol. 2013;
31:1782-1784.
Le DT, Uram JN, Wang H, et al. PD-1 blockade in tumors with
mismatch-repair deficiency. N Engl J Med. 2015;372:2509-2520.
de Bono JS, Scher HI, Montgomery RB, et al. Circulating tumor cells
predict survival benefit from treatment in metastatic castrationresistant prostate cancer. Clin Cancer Res. 2008;14:6302-6309.
Scher HI, Heller G, Molina A, et al. Circulating tumor cell biomarker
panel as an individual-level surrogate for survival in metastatic
castration-resistant prostate cancer. J Clin Oncol. 2015;33:1348-1355.
Beltran H, Jendrisak A, Landers M, et al. The initial detection and
partial characterization of circulating tumor cells in neuroendocrine
prostate cancer. Clin Cancer Res. Epub 2015 Dec 15.
Stott SL, Lee RJ, Nagrath S, et al. Isolation and characterization of
circulating tumor cells from patients with localized and metastatic
prostate cancer. Sci Transl Med. 2010;2:25ra23.
Danila DC, Heller G, Gignac GA, et al. Circulating tumor cell number
and prognosis in progressive castration-resistant prostate cancer. Clin
Cancer Res. 2007;13:7053-7058.
Olmos, D, Arkenau HT, Ang JE, et al. Circulating tumour cell (CTC)
counts as intermediate end points in castration-resistant prostate
cancer (CRPC): a single-centre experience. Ann Oncol. 2009;20:27-33.
Heitzer E, Ulz P, Belic J, et al. Tumor-associated copy number changes
in the circulation of patients with prostate cancer identified through
whole-genome sequencing. Genome Med. 2013;5:30.
Romanel A, Tandefelt DG, Conteduca V, et al. Plasma AR and
abiraterone-resistant prostate cancer. Sci Transl Med. 2015;7:
312re10.
Bettegowda C, Sausen M, Leary RJ, et al. Detection of circulating
tumor DNA in early- and late-stage human malignancies. Sci Transl
Med. 2014;6:224ra24.
139
BELTRAN ET AL
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
140
Carreira S, Romanel A, Goodall J, et al. Tumor clone dynamics in lethal

prostate cancer. Sci Transl Med. 2014;6:254ra125.
Dawson SJ, Tsui DW, Murtaza M, et al. Analysis of circulating tumor
DNA to monitor metastatic breast cancer. N Engl J Med. 2013;368:
1199-1209.
Chen EJ, Sowalsky AG, Gao S, et al. Abiraterone treatment in
castration-resistant prostate cancer selects for progesterone responsive mutant androgen receptors. Clin Cancer Res. 2015;21:
1273-1280.
Azad AA, Volik SV, Wyatt AW, et al. Androgen receptor gene aberrations in circulating cell-free DNA: biomarkers of therapeutic resistance in castration-resistant prostate cancer. Clin Cancer Res. 2015;
21:2315-2324.
Linja MJ, Savinainen KJ, Saramaki OR, et al. Amplification and overexpression of androgen receptor gene in hormone-refractory prostate
Murtaza M, Dawson SJ, Pogrebniak K, et al. Multifocal clonal evolution
characterized using circulating tumour DNA in a case of metastatic
breast cancer. Nat Commun. 2015;6:8760.
Dehm SM, Schmidt LJ, Heemers HV, et al. Splicing of a novel androgen
receptor exon generates a constitutively active androgen receptor
that mediates prostate cancer therapy resistance. Cancer Res. 2008;
68:5469-5477.
Hu DG, Hickey TE, Irvine C, et al. Identification of androgen receptor
splice variant transcripts in breast cancer cell lines and human tissues.
Horm Cancer. 2014;5:61-71.
Lu J, Lonergan PE, Nacusi LP, et al. The cistrome and gene signature of
androgen receptor splice variants in castration resistant prostate
cancer cells. J Urol. 2015;193:690-698.
Ware KE, Garcia-Blanco MA, Armstrong AJ, et al. Biologic and clinical
significance of androgen receptor variants in castration resistant
prostate cancer. Endocr Relat Cancer. 2014;21:T87-T103.
Guo Z, Yang X, Sun F, et al. A novel androgen receptor splice variant is
up-regulated during prostate cancer progression and promotes androgen depletion-resistant growth. Cancer Res. 2009;69:2305-2313.
Hornberg E, Ylitalo EB, Crnalic S, et al. Expression of androgen receptor
splice variants in prostate cancer bone metastases is associated with
castration-resistance and short survival. PLoS One. 2011;6:e19059.
Zhang X, Morrissey C, Sun S, et al. Androgen receptor variants occur
frequently in castration resistant prostate cancer metastases. PLoS
One. 2011;6:e27970.
Qu Y, Dai B, Ye D, et al. Constitutively active AR-V7 plays an essential
role in the development and progression of castration-resistant
prostate cancer. Sci Rep. 2015;5:7654.
Efstathiou E, Titus M, Wen S, et al. Molecular characterization of
enzalutamide-treated bone metastatic castration-resistant prostate
cancer. Eur Urol. 2015;67:53-60.
Efstathiou E, Wen S, San Miguel A, et al. Enzalutamide in combination
with abiraterone acetate in bone-metastatic castration-resistant
prostate cancer. J Clin Oncol. 2014;32:5s (suppl; abstr 5000).
Antonarakis ES, Lu C, Wang H, et al. AR-V7 and resistance to enzalutamide and abiraterone in prostate cancer. N Engl J Med. 2014;371:
1028-1038.
Thadani-Mulero M, Portella L, Sun S, et al. Androgen receptor splice
variants determine taxane sensitivity in prostate cancer. Cancer Res.
2014;74:2270-2282.
Antonarakis ES, Lu C, Luber B, et al. Androgen receptor splice Variant 7
and efficacy of taxane chemotherapy in patients with metastatic
castration-resistant prostate cancer. JAMA Oncol. 2015;1:582-591.
Nakazawa M, Lu C, Chen Y, et al. Serial blood-based analysis of AR-V7
in men with advanced prostate cancer. Ann Oncol. 2015;26:
1859-1865.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
Maughan BL, Antonarakis ES. Clinical relevance of androgen receptor

splice variants in castration-resistant prostate cancer. Curr Treat
Options Oncol. 2015;16:57.
Kwegyir-Afful AK, Ramalingam S, Purushottamachar P, et al. Galeterone and VNPT55 induce proteasomal degradation of AR/AR-V7,
induce significant apoptosis via cytochrome C release and suppress
growth of castration resistant prostate cancer xenografts in vivo.
Oncotarget. 2015;6:27440-27460.
Andersen RJ, Mawji NR, Wang J, et al. Regression of castrate-recurrent
prostate cancer by a small-molecule inhibitor of the amino-terminus
domain of the androgen receptor. Cancer Cell. 2010;17:535-546.
Myung JK, Banuelos CA, Fernandez JG, et al. An androgen receptor Nterminal domain antagonist for treating prostate cancer. J Clin Invest.
2013;123:2948-2960.
Montgomery R, Antonarakis ES, Hussain M, et al. A phase 1/2 openlabel study of safety and antitumor activity of EPI-506, a novel AR Nterminal domain inhibitor, in men with metastatic castration-resistant
prostate cancer (mCPRC) with progression after enzalutamide or
abiraterone. J Clin Oncol. 2015;33 (suppl; abstr TPS5072).
Liu C, Lou W, Zhu Y, et al. Niclosamide inhibits androgen receptor
variants expression and overcomes enzalutamide resistance in
castration-resistant prostate cancer. Clin Cancer Res. 2014;20:
3198-3210.
Johns WD, Garnick MB, Kaplan WD. Leuprolide therapy for prostate
cancer. An association with scintigraphic flare on bone scan. Clin
Nucl Med. 1990;15:485-487.
Pollen JJ, Witztum KF, Ashburn WL. The flare phenomenon on radionuclide bone scan in metastatic prostate cancer. AJR Am J
Roentgenol. 1984;142:773-776.
Ryan CJ, Shah S, Efstathiou E, et al. Phase II study of abiraterone
acetate in chemotherapy-naive metastatic castration-resistant
prostate cancer displaying bone flare discordant with serologic response. Clin Cancer Res. 2011;17:4854-4861.
Morris MJ, Farrelly JS, Fox JJ, et al. The Prostate Cancer Clinical Trials
Consortium (PCCTC) bone scan data capture tool for clinical trials
using Prostate Cancer Working Group 2 (PCWG2) criteria: effect on
data accuracy and workload. J Clin Oncol. 2011;29 (suppl 7; abstr 121).
Ryan CJ, Smith MR, de Bono JS, et al; COU-AA-302 Investigators.
Abiraterone in metastatic prostate cancer without previous chemotherapy. N Engl J Med. 2013;368:138-148.
Beer TM, Armstrong AJ, Rathkopf DE, et al; PREVAIL Investigators.
Enzalutamide in metastatic prostate cancer before chemotherapy.
N Engl J Med. 2014;371:424-433.
Morris MJ, Molina A, Small EJ, et al. Radiographic progression-free
survival as a response biomarker in metastatic castration-resistant
prostate cancer: COU-AA-302 results. J Clin Oncol. 2015;33:
1356-1363.
Kluetz PG, Ning YM, Maher VE, et al. Abiraterone acetate in combination with prednisone for the treatment of patients with metastatic
castration-resistant prostate cancer: U.S. Food and Drug Administration drug approval summary. Clin Cancer Res. 2013;19:6650-6656.
Imbriaco M, Larson SM, Yeung HW, et al. A new parameter for
measuring metastatic bone involvement by prostate cancer: the Bone
Scan Index. Clin Cancer Res. 1998;4:1765-1772.
Morris MJ, et al. Post-treatment serial bone scan index (BSI) as an
outcome measure predicting for survival. Genitourinary Cancers
Symposium Proceedings (2008).
Ulmert D, Kaboteh R, Fox JJ, et al. A novel automated platform for
quantifying the extent of skeletal tumour involvement in prostate
cancer patients using the Bone Scan Index. Eur Urol. 2012;62:78-84.
Anand A, Morris MJ, Kaboteh R, et al. Analytic validation of the automated Bone Scan Index as an imaging biomarker to standardize
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
quantitative changes in bone scans of patients with metastatic

prostate cancer. J Nucl Med. 2016;57:41-45.
Armstrong AJ, Kaboteh R, Carducci MA, et al. Assessment of the bone
scan index in a randomized placebo-controlled trial of tasquinimod in
men with metastatic castration-resistant prostate cancer (mCRPC).
Urol Oncol. 2014;32:1308-1316.
Grant, FD, Fahey FH, Packard AB, et al. Skeletal PET with 18F-fluoride:
applying new technology to an old tracer. J Nucl Med. 2008;49:68-78.
Schirrmeister, H, Glatting G, Hetzel J, et al. Prospective evaluation of
the clinical value of planar bone scans, SPECT, and (18)F-labeled NaF
PET in newly diagnosed lung cancer. J Nucl Med. 2001;42:1800-1804.
Hetzel M, Arslandemir C, Konig HH, et al. F-18 NaF PET for detection of
bone metastases in lung cancer: accuracy, cost-effectiveness, and
impact on patient management. J Bone Miner Res. 2003;18:
2206-2214.
Shen CT, Qiu ZL, Han TT, et al. Performance of 18F-fluoride PET or PET/CT
for the detection of bone metastases: a meta-analysis. Clin Nucl Med.
2015;40:103-110.
Scher HI, Halabi S, Tannock I, et al. Trial design and objectives for
castration-resistant prostate cancer: updated recommendations from
the Prostate Cancer Clinical Trials Working Group 3. J Clin Oncol. 2016;
34:1402-1418.
Stephenson AJ, Scardino PT, Kattan MW, et al. Predicting the outcome
of salvage radiation therapy for recurrent prostate cancer after radical
prostatectomy. J Clin Oncol. 2007;25:2035-2041.
Stephenson AJ, Scardino PT, Eastham JA, et al. Postoperative nomogram predicting the 10-year probability of prostate cancer recurrence after radical prostatectomy. J Clin Oncol. 2005;23:
7005-7012.
Kattan MW, Zelefsky MJ, Kupelian PA, et al. Pretreatment nomogram
for predicting the outcome of three-dimensional conformal radiotherapy in prostate cancer. J Clin Oncol. 2000;18:3352-3359.
Foss CA, Mease RC, Cho SY, et al. GCPII imaging and cancer. Curr Med
Chem. 2012;19:1346-1359.
Schuster DM, Nieh PT, Jani AB, et al. Anti-3-[(18)F]FACBC positron emission
tomography-computerized tomography and (111)In-capromab pendetide
single photon emission computerized tomography-computerized tomography for recurrent prostate carcinoma: results of a prospective clinical
trial. J Urol. 2014;191:1446-1453.
Pandit-Taskar N, ODonoghue JA, Durack JC, et al. A phase I/II study for
analytic validation of 89Zr-J591 ImmunoPET as a molecular imaging
agent for metastatic prostate cancer. Clin Cancer Res. 2015;21:
5277-5285.
Pandit-Taskar N, ODonoghue J, Martin D, et al. First in human 89Zr-DfIAB2M anti-prostate specific membrane antigen (PSMA) minibody in
patients with metastatic prostate cancer. J Nucl Med. 2015;56:400.
Afshar-Oromieh A, Zechmann CM, Malcher A, et al. Comparison of PET
imaging with a (68)Ga-labelled PSMA ligand and (18)F-choline-based
PET/CT for the diagnosis of recurrent prostate cancer. Eur J Nucl Med
Mol Imaging. 2014;41:11-20.
88.
Eiber, M, Maurer T, Souvatzoglou M, et al. Evaluation of hybrid (6)(8)

Ga-PSMA ligand PET/CT in 248 patients with biochemical recurrence
after radical prostatectomy. J Nucl Med. 2015;56:668-674.
89. Gotto GT, Yu C, Bernstein M, et al. Development of a nomogram model
predicting current bone scan positivity in patients treated with
androgen-deprivation therapy for prostate cancer. Front Oncol. 2014;
4:296.
90. Umbehr MH, Muntener M, Hany T, et al. The role of 11C-choline and
18
F-fluorocholine positron emission tomography (PET) and PET/CT in
prostate cancer: a systematic review and meta-analysis. Eur Urol.
2013;64:106-117.
91. Evangelista L, Zattoni F, Guttilla A, et al. Choline PET or PET/CT and
biochemical relapse of prostate cancer: a systematic review and metaanalysis. Clin Nucl Med. 2013;38:305-314.
92. Giovacchini G, Picchio M, Briganti A, et al. [11C]choline positron
emission tomography/computerized tomography to restage prostate
cancer cases with biochemical failure after radical prostatectomy and
no disease evidence on conventional imaging. J Urol. 2010;184:
938-943.
93. Giovacchini G, Picchio M, Scattoni V, et al. PSA doubling time for
prediction of [(11)C]choline PET/CT findings in prostate cancer patients with biochemical failure after radical prostatectomy. Eur J Nucl
Med Mol Imaging. 2010;37:1106-1116.
94. Krause BJ, Souvatzoglou M, Tuncel M, et al. The detection rate of [11C]
choline-PET/CT depends on the serum PSA-value in patients with
biochemical recurrence of prostate cancer. Eur J Nucl Med Mol Imaging. 2008;35:18-23.
95. U.S. Food and Drug Administration. FDA Approves Production of
Imaging Agent That Helps Detect Prostate Cancer. http://www.fda.
gov/NewsEvents/Newsroom/PressAnnouncements/ucm319201.htm.
Accessed April 27, 2016.
96. Scher HI, Sawyers CL. Biology of progressive, castration-resistant
prostate cancer: directed therapies targeting the androgenreceptor signaling axis. J Clin Oncol. 2005;23:8253-8261.
97. Scher HI, Beer TM, Higano CS, et al; Prostate Cancer Foundation/
Department of Defense Prostate Cancer Clinical Trials Consortium.
Antitumour activity of MDV3100 in castration-resistant prostate
cancer: a phase 1-2 study. Lancet. 2010;375:1437-1446.
98. Rathkopf DE, Morris MJ, Fox JJ, et al. Phase I study of ARN-509, a novel
antiandrogen, in the treatment of castration-resistant prostate cancer. J Clin Oncol. 2013;31:3525-3530.
99. Evans MJ, Smith-Jones PM, Wongvipat J, et al. Noninvasive measurement of androgen receptor signaling with a positron-emitting
radiopharmaceutical that targets prostate-specific membrane antigen. Proc Natl Acad Sci USA. 2011;108:9578-9582.
100. U.S. Department of Health and Human Services. FDA Guidance:
Clinical Trial Imaging Endpoints Process Standards, Draft Guidance.
2011.
101. U.S. Food and Drug Administration. Guidance for Industry and FDA
Staff: Qualification Process for Drug Development Tools. January 2014.
www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatory
Information/Guidances/UCM230597.pdf. Accessed April 17, 2012.
141
GYNECOLOGIC CANCER
Paradigm Shift in the

Management Strategy for
Epithelial Ovarian Cancer
CHAIR
Keiichi Fujiwara, MD, PhD
Saitama Medical University International Medical Center
Hidaka, Japan
SPEAKERS
Amit M. Oza, MD
Toronto, ON, Canada
Jessica N. McAlpine, MD
University of British Columbia
Vancouver, BC, Canada
PARADIGM SHIFT IN OVARIAN CANCER STRATEGY
Paradigm Shift in the Management Strategy for Epithelial

Ovarian Cancer
Keiichi Fujiwara, MD, PhD, Jessica N. McAlpine, MD, Stephanie Lheureux, MD, PhD, Noriomi Matsumura,
MD, PhD, and Amit M. Oza, BSc, MD, MBBS, FRCPC
OVERVIEW
The hypothesis on the pathogenesis of epithelial ovarian cancer continues to evolve. Although epithelial ovarian cancer had
been assumed to arise from the coelomic epithelium of the ovarian surface, it is now becoming clearer that the majority of
serous carcinomas arise from epithelium of the distal fallopian tube, whereas clear cell and endometrioid cancers arise from
endometriosis. Molecular and genomic characteristics of epithelial ovarian cancer have been extensively investigated. Our
understanding of pathogenesis of the various histologic types of ovarian cancer have begun to inform changes to the
strategies for management of epithelial ovarian cancer, which represent a paradigm shift not only for treatment but also for
prevention, which previously had not been considered achievable. In this article, we will discuss novel attempts at the
prevention of high-grade serous ovarian cancer and treatment strategies for two distinct entities in epithelial ovarian cancer:
low-grade serous and clear cell ovarian carcinomas, which are relatively rare and resistant to conventional chemotherapy.
e have seen numerous advances in the field of ovarian

cancer in the past decade, yet overall survival statistics for this disease are essentially unchanged. Despite the
excellent design and execution of multiple screening trials,
the relative impact in terms of number of cases of early
ovarian cancer detected has been small and, unfortunately,
tempered by the number of unnecessary surgeries performed
as a result of abnormal screening results.1-4 New methods of
disease control are desperately needed, and this has prompted a
push to explore methods of primary prevention.
PROPHYLACTIC SALGINGECTOMY: SHOULD ALL

TUBES BE REMOVED?
In BRCA1/2-mutation carriers, risk-reducing bilateral salpingooophorectomy has been shown to be highly protective for
ovarian cancer, with a reduced risk of at least 80% after surgery.5,6 Most high-risk women have reportedly experienced a
high quality of physical and mental well-being after risk-reducing
bilateral salpingo-oophorectomy,7 and all-cause mortality is
significantly reduced by this procedure.5,8 However, risk-reducing
bilateral salpingo-oophorectomy is not recommended for the
general/low-risk population, because removal of the ovaries is
reportedly associated with increased mortality, coronary heart
disease, stroke, osteoporosis, and colorectal cancer.9,10
Alternative opportunities for prevention were inspired by

the appreciation of the role of the fallopian tube in ovarian
cancer (Fig. 1). The distal fallopian tube is the site of origin of
the majority of the most common histotype of epithelial
ovarian cancerhigh-grade serousin women with hereditary predisposition (e.g., BRCA1/2 mutation) and in
sporadic occurrences in the general population.11,12 In addition, the fallopian tube serves as a conduit for the passage
of endometrial inflammatory cytokines, infections, and/or
irritants to the ovary and peritoneal cavity.13 Ectopic endometrium may undergo malignant transformation and is
the purported site of origin of the next two most common
histotypes of epithelial ovarian cancer: clear cell and
endometrioid cancers.14 Long-substantiated risk factors for
ovarian cancer make sense in the context of the critical role
that the fallopian tube plays in ovarian cancer. Inflammatory
conditions, such as pelvic inflammatory disease, incessant
ovulation, or irritants such as talc that have ascended from
the lower genital tract via the fallopian tube to the tuboovarian junction, increase mutagenesis and increase the risk
of ovarian cancer. Interventions that decrease these parameters
(e.g., tubal ligation, ovarian quiescence during pregnancy, or
breast feeding) are associated with a decreased risk of ovarian
cancer.13 Large studies with international cohorts of women
who have undergone tubal ligation have shown a decrease in
From the Department of Gynecologic Oncology, Saitama Medical University International Medical Center, Saitama, Japan; Department of Gynecology and Obstetrics, University of British
Columbia, Vancouver, Canada; Division of Medical Oncology and Hematology, Bras Family Drug Development Program, Princess Margaret Cancer Centre, Toronto, Canada; Department
of Gynecology and Obstetrics, Kyoto University, Kyoto, Japan.
Corresponding author: Keiichi Fujiwara, MD, PhD, Saitama Medical University International Medical Center, 1397-1 Yamane, Hidaka 350-1241, Japan;
email: fujiwara@saitama-med.ac.jp.
e247
FUJIWARA ET AL
risk of developing ovarian carcinoma by 29% overall, with

histotype-specific variation in protective effects. The greatest
risk reduction was observed in endometrioid (52%) and clear
cell (48%) histologies.15 Risk reduction was only 20% in highgrade serous carcinoma; however, given that this histotype
encompasses 70% of all epithelial ovarian cancers and accounts
for 90% of deaths, this impact is still substantial. A recent metaanalysis of published series with tubal ligation data reported
similar findings,16 and, importantly, the protective effect (60%
risk reduction) of tubal ligation has also been observed in highrisk populations (BRCA1/2-mutation carriers).17 Although not
the focus of this article, oral contraception pills have arguably
had the greatest success in ovarian cancer risk reduction globally
and across all histotypes; however, the mechanism for the
protective effect of the oral contraception pills is not fully
understood. Past and current hypotheses again suggest an
important role of the fallopian tube, and the influence of the
progesterone component of oral contraception pills on tubal
epithelial cell proliferation (differentiated and/or stem cell) is
under active investigation.
The Intervention: Is There an Opportunity to Remove

Fallopian Tubes?
Beginning in 2008, and then as a formal initiative in 2010, we,
as members of the British Columbia Ovarian Cancer Research Team (OVCARE) that is based in Vancouver, have
suggested to gynecologic surgeons that they should consider
the following: perform bilateral salpingectomy in all women
at the time of hysterectomy (even when the ovaries are
being preserved) and perform bilateral salpingectomy in
place of tubal ligation for sterilization. In September 2010,
we circulated a DVD to all gynecologic surgeons in British
KEY POINTS
e248
Appreciation of the changing landscape of epithelial

ovarian cancer enables the development of novel
strategies in treatment and prevention.
Opportunistic salpingectomy should be considered in
women undergoing gynecologic procedures with access
to the peritoneal cavity, at hysterectomy, or for
permanent sterilization. For women at increased risk of
ovarian carcinoma development, risk-reducing bilateral
salpingo-oophorectomy remains standard of care.
Accruing data support the safety, acceptability,
feasibility, and cost effectiveness of this procedure.
Low-grade serous ovarian cancer, a rare subtype of
ovarian cancer, is relatively chemotherapy resistant,
and current development focus is on targeted therapy
that is based on its distinct biology.
Clear cell ovarian cancer is also a distinct pathological
and clinical entity of rare epithelial ovarian cancer. Its
carcinogenesis from endometriosis is now becoming
clearer, and research on molecular characteristics of
clear cell ovarian cancer has led to the new trials of
target agents.
Columbia, which included a review of the rationale for

salpingectomy. This initiative has had a profound effect on
clinical practice in British Columbia: the proportion of
hysterectomies that had associated salpingectomies (excluding hysterectomies in which ovaries were removed, the
rate of which remains unchanged) increased from 8% in
2008 to 63% in 2011 and to 75% in 2013, and the proportion
of sterilizations by bilateral salpingectomy increased from
0.5% in 2008 to 33% in 2011 and to 48% in 2013.18 The
Society of Gynecologic Oncologists of Canada, U.S. Society
for Gynecologic Oncology, American College of Obstetrics
and Gynecology, and Royal Australian and New Zealand
College of Obstetricians and Gynecologists have all since
published statements that recommend consideration of
salpingectomy for all women at the time of gynecologic
surgery,19-22 and physician surveys show willingness to undertake practice change.23-26 It is essential to clarify that this
campaign is aimed at women in the general (low-risk) population
who have a lifetime risk of developing ovarian cancer of approximately 1.7%; importantly, risk-reducing bilateral salpingooophorectomy at age 40 or after completion of childbearing27 is
still recommended and represents the standard of care in highrisk populations, such as in BRCA1/2-mutation carriers. A twostep procedure of risk-reducing salpingectomy followed by
oophorectomy in high-risk women, although feasible, carries
increased surgical risks, and there is insufficient data to support
optimal timing or protective effect (if any) for breast and ovarian
cancers; thus, the two-step procedure is reserved for women
who are unwilling to undergo risk-reducing bilateral salpingooophorectomy.28,29
We have termed this intervention in the low-risk population opportunistic salpingectomy, which suggests that
surgeons should take the opportunity to remove the fallopian tubes if there is access to these anatomic structures
during other scheduled gynecologic procedures but should
not perform a procedure solely for the purpose of tubal
removal for ovarian cancer prevention. We have published
data from a time period 2 years prior and 2 years after the
formal introduction of this campaign, including data from
procedures performed on 43,931 women. We observed no
increase in major perioperative complications or events.
Blood loss, length of hospital stay, and readmission rates
were the same in patients who underwent opportunistic
salpingectomy and in patients who underwent hysterectomy alone or tubal ligation procedures. 18 Potential
long-term hormonal consequences of opportunistic salpingectomy have been investigated through measurement
of ovarian sonographic parameters and hormonal assays;
data are reassuring but have relatively short follow-up.30-32
Given that an earlier age at menopause has been associated
with increased mortality, it is imperative that this does not
offset any projected protective effect in ovarian cancer risk
reduction. Hysterectomy is known to affect ovarian reserve33,34
and the relative impact of opportunistic salpingectomy performed with hysterectomy is likely small or immeasurable. To
answer this definitively we are studying our British Columbia
cohort of women who have undergone these procedures,
FIGURE 1. Role of the Fallopian Tube in Ovarian Cancer
comparing onset of menopause (defined as cessation of

menses for 1 year) in individuals who have undergone bilateral
salpingectomy versus tubal ligation for permanent sterilization.
In addition, there is a planned randomized clinical trial in the
United Kingdom in women who will undergo bilateral salpingectomy either at hysterectomy or as a stand-alone procedure for contraception.35
Is the increased uptake of opportunistic salpingectomy
responsible in terms of use of resources and health economics? Our group and others have shown that this procedure can be undertaken by different surgical routes
(vaginal, minimally invasive, laparotomy)18,36-38 so it does
not need to be limited to high-resource countries. Additional
operating room time needed for opportunistic salpingectomy with hysterectomy is approximately 16 minutes, and,
in lieu of tubal ligation, the procedure requires approximately 10 minutes,18 a time that arguably is of no clinical
impact to the majority of women undergoing these procedures nor to the health systems where the procedures are
being performed. Cost analysis modeling, which considers
perioperative risks, impact on ovarian cancer risk reduction,
and morbidities associated with premature menopause
secondary to oophorectomy, found that opportunistic salpingectomy with hysterectomy was less costly and more
effective than hysterectomy alone, reduced the number of
ovarian cancer occurrences, and prolonged the average life
expectancy. Opportunistic salpingectomy for sterilization
would be considered more costly than tubal ligation in terms
of operative time and complication risk; however, opportunistic salpingectomy was more effective at reducing risk of
ovarian cancer. The calculated number needed to treat to

prevent a single occurrence of ovarian cancer was acceptable for both procedures.39
Measuring the Impact of Opportunistic

Salpingectomy
Finally, and most importantly, will this change in surgical
paradigm translate to a decreased incidence of ovarian
cancer? Proof of success of this initiative will be an observed
reduction in new occurrences of ovarian cancer, and we
anticipate there will be a shift in the distribution of histotypes of epithelial ovarian cancer in the population of
women exposed to this procedure. The age of women undergoing this procedure as part of hysterectomy or for
permanent sterilization is younger than the age of onset of
ovarian cancer in the general population, so we anticipate
that it will be at least 10 years and up to 20 years from the
start of our campaign before we will be able to discern a
difference that would provide definitive support for this
intervention. Pooling of our data with data from other
geographic areas that have adopted this practice may help
power the analysis and hasten results. Although nay-sayers
may be frustrated or unsatisfied with this lack of immediate
measurable impact, we hold to this long view and are very
encouraged by emerging evidence from other historical
cohorts that supports this intervention. In addition to the
strong protective effect demonstrated in women who have
undergone tubal ligation (outlined previously), there are
data from a small number of studies on women who have
e249
FUJIWARA ET AL
undergone excisional tubal surgery (defined as complete

salpingectomy, distal fimbriectomy, or partial salpingectomy).
Researchers from the Rochester Epidemiology Project reported a
64% reduction in the risk of ovarian cancer after excisional tubal
sterilization compared with those without sterilization or with
nonexcisional tubal sterilization (odds ratio [OR] 0.36; 95% CI,
0.131.02).40 Danish researchers, who used a national database,
reported that bilateral salpingectomy reduced risk for ovarian
cancer by 42% (OR 0.58; 95% CI, 0.360.95).41 Most recently, a
retrospective population-based study that used Swedish health
registers reported that bilateral salpingectomy was associated
with a 65% reduction in risk (hazard ratio 0.35; 95% CI,
0.170.73).42 Both the Danish and Swedish studies examined
retrospective data and included women who underwent salpingectomies for pathologic reasons (e.g., hydrosalpinx, pelvic
inflammatory disease, ectopic pregnancy), which makes them
significantly different from the women undergoing opportunistic
salpingectomy in British Columbia for the purposes of ovarian
cancer prophylaxis. We hypothesize that salpingectomy performed for risk-reduction purposes may confer more protection
than those performed for other indications, because surgeons
will be more careful to remove the entire distal end of the
fallopian tube.
Published data reassure us of the safety of the procedure and
suggest that opportunistic salpingectomy should be easily within
the skill set of any pelvic/abdominal surgeon and can be performed by multiple surgical routes; there is health economic
support, and the evidence of high uptake suggests feasibility. In
addition, there are numerous nononcologic arguments for tubal
removal, including prevention of hydrosalpinges, pyosalpinx, or
tubal prolapse that may require subsequent surgeries.43 However,
as we wait to definitively measure the impact of this procedure on
ovarian cancer risk, whether a woman undergoes opportunistic
salpingectomy will remain a decision between her and her
treating physician, and the decision requires good judgment and
common sense. Our first dictum as physicians is to do no harm.
LOW-GRADE SEROUS OVARIAN CANCER

Current Clinical Practice
The standard strategy for advanced epithelial ovarian cancer
remains the combination of surgery and platinum/taxanebased chemotherapy,44 which is primarily driven by activity
in unselected populations enrolled in clinical trials; the
populations are predominantly composed of women with
high-grade serous ovarian cancer. One of five histologic subtypes of epithelial ovarian cancer, low-grade serous ovarian
cancer is a rare subentity with a specific genomic landscape, as
evidenced by a different natural history and pattern of response to therapy.45 From available evidence, low-grade serous ovarian cancer may arise after an initial diagnosis of serous
tumor of low malignant potential (Fig. 2) or de novo and does
not seem to be part of the hereditary breast-ovarian cancer
syndrome related to the BRCA1/2 gene mutation.46
The dogma of cytotoxic chemotherapy for all epithelial
ovarian cancers is evolving on the basis of the recent clinical
and molecular classification of the different subtypes of
epithelial ovarian cancer.47 To date, no prospective randomized clinical trial data are available to provide guidance
about the optimal postoperative treatment of low-grade
serous ovarian cancer.48 Although available data are mostly
based on small retrospective, single-institution studies, chemotherapy resistance is consistently reported in low-grade
serous ovarian cancer.48,49 From a single-institution database,
112 women with stages II to IV low-grade serous ovarian cancer
who underwent primary surgery followed by platinum-based
chemotherapy were retrospectively identified. Of these, 42
patients underwent second-look surgery, and 2, 13, and 24
had evidence of no residual disease, microscopically positive
disease, or macroscopically positive disease, respectively.50
Similar findings were reported with low-grade primary peritoneal cancer, with high rates of persistent disease at the
completion of adjuvant chemotherapy.51 This relative chemotherapy resistance, possibly related to the nature of
FIGURE 2. Carcinogenesis Pathway in Low-Grade Serous Ovarian Cancer
Pictures courtesy of Dr. Patricia A. Shaw, University Health Network of Toronto.
e250
low-grade serous ovarian cancer, was also observed in a

separate retrospective study that included patients treated
with neoadjuvant chemotherapy.52 Despite the receipt of a
taxane and platinum combination in the majority of the 25
patients with advanced low-grade serous ovarian cancer, only
one patient had a complete response (CR); an additional 21
women had stable disease, but two experienced progression
after neoadjuvant chemotherapy. Although CA-125 levels were
reduced by greater than 50% with chemotherapy in half of the
patients, radiologic response remained low.
Chemotherapy resistance also has been demonstrated in
the setting of recurrent disease. For example, in a retrospective study of 58 women with recurrent low-grade serous
ovarian cancer who received 108 separate chemotherapy
regimens,49 only four responses were seen (one CR and
three partial responses; overall response rate [ORR], 3.7%).
The ORR was 4.9% for the platinum-sensitive cohort and
2.1% for the platinum-resistant cohort. Stable disease that
was observed in 65 (60.2%) of 108 chemotherapy regimens
may have been related to tumor biology or may have been a
bias of assessmentmeasureable lesions can often appear
stable by RECIST criteria on CT scans53and not a true
therapeutic effect. As such, given the relative chemotherapy
resistance, we should consider making clinical trials the new
standard. Targeted biologically directed therapy is an interesting area of investigation in low-grade serous ovarian
cancer, given the molecular characteristics of disease.
Compared with high-grade serous ovarian cancer, low-grade
serous ovarian cancer has lower expression of p53, WT1,
c-KIT, Ki-67, and MMP-954 but a higher expression of estrogen
receptor, progesterone receptor, and E-cadherin.55
Hormonal Therapy
Low-grade serous ovarian cancer often is diagnosed at a
younger age (43 to 55 years), so a higher proportion of
patients with low-grade serous ovarian cancer are premenopausal at diagnosis; as such, hormonal status may
be implicated in pathogenesis. 56 Indeed, among the five
different subtypes of ovarian cancer, low-grade serous
ovarian cancer has the higher proportion of hormone
receptorpositive (progesterone receptor and/or estrogen
receptorpositive) tumors.57 In this large retrospective
study, strong progesterone but not estrogen receptor expression appeared to be associated with improved survival
after accounting for site, age, stage, and grade (p = .019 for
progesterone and p = .78 for estrogen receptor). But this
association was not statistically significant after adjusting for
residual disease in this subset of 64 patients with low-grade
serous ovarian cancer (p = .27 and .90, respectively).57 The
agents used as hormonal therapy, which are targeted
therapies against the progesterone receptor and estrogen
receptor, are mainly tamoxifen and aromatase inhibitors
(anastrozole and letrozole). A retrospective study in 64
patients with recurrent low-grade serous ovarian cancer
who received 89 separate hormonal regimens showed an
ORR of 9% (six CRs and two partial responses).58 In total, 61%
of the regimens resulted in a progression-free survival (PFS)

duration of at least 6 months. Regimens that involved treatment of estrogen receptorpositive/progesterone receptor
positive disease produced a longer median time to progression
(8.9 months) than regimens that involved treatment of estrogen receptorpositive/progesterone receptor2negative
disease (time to progression, 6.2 months; p = .053). Given the
relatively favorable adverse effect profile and demonstrated
efficacy, hormonal therapy represents an interesting treatment
alternative.
The Mitogen-Activated Protein Kinase Pathway

The mitogen-activated protein kinase (MAPK) pathway is
activated and appears to play a prominent role in the
pathogenesis of low-grade serous ovarian cancer.59 Approximately 20% to 40% of low-grade serous carcinomas
have a KRAS mutation, whereas BRAF mutations are rare
(approximately 5%).60,61 KRAS is a frequent mutation detected in advanced low-grade serous ovarian cancer, and the
BRAF V600E mutation is associated with serous borderline
tumors and early-stage low-grade serous ovarian cancer.62
As a result, the V600E mutation is associated with a better
clinical outcome. In a study of 23 patients with an original
diagnosis of serous tumor of low malignant potential who
subsequently experienced recurrence with a diagnosis of
low-grade serous ovarian cancer, patients with KRAS G12V
mutations experienced shorter survival times than those
with either KRAS G12D, wild-type, or rare KRAS variants
(hazard ratio 4.77; p = .023).63 Another retrospective study
showed the potential impact of mutational status; patients
with KRAS and BRAF mutations appeared to have better
overall survival than those with wild-type KRAS or BRAF.64
Additional investigations are warranted to elucidate the role
of the mutation type in low-grade serous ovarian cancer,
because this may have important clinical implications, such
as introduction of the analysis of BRAF/KRAS status in the
postdebulking setting to predict the risk of recurrence.65
The MAPK cascade is triggered by the binding of a ligand
that ultimately leads to phosphorylation of ERK.66,67 Thus,
MEK is a good candidate for targeted therapy, and a number
of MEK inhibitors (MEKi) have been developed.
The GOG-0239 open-label phase II study of patients with
recurrent low-grade serous ovarian cancer, by prospective
pathologic evaluation, received selumetinib (AZD6244)
a MEK1/2 inhibitor that targets the downstream effect of
activating BRAF and KRAS mutationsat a dosage of 50 mg
twice daily.68 The majority of patients had received three or
more prior chemotherapy regimens. Fifty-two women with
recurrent low-grade serous ovarian cancer were enrolled,
and the ORR was 15% (one CR and seven partial responses).
Another 65% of patients in the trial had stable disease, and
the median PFS was 11.0 months. The most common toxicities were gastrointestinal, dermatologic, and metabolic.
Three patients experienced grade 4 toxicitiesone each of
cardiac, pain, and pulmonary toxicity. Mutational analysis
was conducted on formalin-fixed, paraffin-embedded tumor
e251
FUJIWARA ET AL
samples from 34 patients enrolled in this trial, and the

primary tumor accounted for 82% of the cases. In these 34
cases, there were two BRAF mutations (6%) and 14 KRAS
mutations (41%), and 15% of tumors had NRAS mutations. In
this study, there was no correlation between mutations
of BRAF or KRAS and ORR; however, only hotspot mutations were assessed. Tissue analysis of an extraordinary
responder who experienced a complete, durable, and ongoing
(. 5 years) response to selumetinib identified a previously
uncharacterized MAP2K1 deletion (Q56_V60del). Additional
investigations support that this novel deletion is a driver
alteration in this exceptional responder and serves as the
molecular basis for her dramatic, sustained response to the
MEKi.69 This finding may explain why targeted genotyping of
only the most common hotspot alterations in BRAF and KRAS
failed to ascertain the molecular basis for a subset of the
responses observed in the GOG-0239 trial and justify the
incorporation of broader profiling methods. After these
promising results, several studies have been designed, such
as the ongoing GOG-281 trial. GOG-281 is a randomized
phase II study between investigator choice (of letrozole,
tamoxifen, weekly paclitaxel, pegylated liposomal doxorubicin, or weekly topotecan) and the MEKi agent trametinib
for recurrent low-grade serous ovarian cancer. The study
incorporates prospective pathology review, pretissue biopsy, and blood collection for correlative analyses, such as
next-generation sequencing and proteomics. The primary
endpoint is PFS, the estimated enrolment is 250 patients,
and crossover at disease progression is allowed.
Several additional targeted studies are also ongoing in
this space. NCT01936363 was a randomized phase II trial
for recurrent low-grade serous ovarian cancer that compared pimasertib (MEKi) plus or minus SAR245409a
phosphatidylinositol-4,5-biphosphate 3-kinase and mTOR.
This randomized phase II trial started, and patients were
recruited, but the trial subsequently was stopped because of
toxicity with the combination. No results have been presented or published as yet. The MILO trial (NCT01849874)
is a randomized, open-label phase III study between physician choice (paclitaxel, liposomal doxorubicin, topotecan)
and binimetinib, or MEK162, for recurrent low-grade serous
ovarian cancer. In addition, the RTM 1313 study is a randomized phase II trial with patients with newly diagnosed
stages II to IV low-grade serous ovarian cancer that is investigating trametinib/GSK 214170550 every 3 weeks for six
cycles versus standard adjuvant chemotherapy carboplatin/
paclitaxel for six cycles. Taken together, biomarkers that predict
MEKi activity and the identification of MEKi-independent
compensatory pathways are needed.
Antiangiogenesis
The angiogenesis pathway may also be a therapeutic target
in patients with low-grade serous ovarian cancer. An initial
report of three patients with recurrent low-grade serous
ovarian cancer treated with bevacizumab, the monoclonal
antibody against the vascular endothelial growth factor
e252
A (VEGF-A), showed sustained responses of 15-, 15-, and

22-month durations.70 From a retrospective cohort of 17
patients with low-grade serous ovarian cancer and serous
borderline tumors, two patients were treated with singleagent bevacizumab, and the others were treated with a
combination of bevacizumab and chemotherapy.71 Fifteen
patients were evaluable for response; six had a partial response, and five had stable disease that lasted 3 months or
longer. The response rate for the low-grade serous ovarian
cancer group was 55%. Therefore, additional studies to
evaluate the possible role of antiangiogenics in low-grade
serous ovarian cancer are warranted, potentially in combination with MAPK inhibitors.
Future Directions
The insulin-like growth factor (IGF) pathway is another
pathway overexpressed in low-grade serous ovarian cancer,
and related effectors, such as PI3K/Akt/mTOR, have been
also described to play a role in disease pathogenesis.72
Activating mutations of PI3KCA are observed in approximately 40% of tumors, whereas inactivating PTEN mutations
are present in 3% to 8% of tumors.73 AMG-479, a fully human
antiinsulin-like growth factor receptor type I monoclonal
antibody, had been assessed in frontline and recurrent
settings in ovarian cancer but not specifically in low-grade
serous ovarian cancer (NCT00718523 and NCT00719212).
OSI-906, a tyrosine kinase inhibitor of both IGF-1R and the
insulin receptor was assessed in recurrent ovarian cancer
(NCT00889382).
On the basis of retrospective studies and ad hoc cooperative group studies, low-grade serous ovarian cancer is
described as relatively chemotherapy resistant, which has
led to the investigation of novel therapies that are based on
the specific molecular pathways of low-grade serous ovarian
cancer. Although low-grade serous ovarian cancer is associated with superior overall survival compared with the
other histologic subgroups of ovarian cancer, more than 70%
of low-grade serous ovarian cancer patients experience
relapse and die of their disease. There is an urgent clinical
need to develop additional trials of combination targeted
agents that are tolerable for the patients and that do not
significantly impinge upon quality of life. Because low-grade
serous ovarian cancer is a rare entity, the cooperation between different institutes is essential alongside a common
effort to reduce heterogeneity of grading and biases in
treatment and response evaluation.65
CLEAR CELL OVARIAN CANCER: TIME FOR A NEW

PARADIGM?
Clear cell ovarian cancer is a unique entity of adenocarcinoma of the ovary. Sugiyama et al74 first reported that
advanced-stage clear cell ovarian cancer was less sensitive
than the serous counterpart to conventional chemotherapy
and provoked the discussion about its uniqueness.74
Recently, a marked ethnic difference in the incidence of
clear cell ovarian cancer has been recognized, although the
reason is not clear. The incidence of clear cell ovarian cancer

is less than 10% in Europe and North America.75 However, in
Japan, the prevalence of clear cell ovarian cancer is increasing, and now approximately 25% of epithelial ovarian
cancer is clear cell, according to the Japanese Society of
Obstetrics and Gynecology tumor registry data from 2014.74
A hypothesis for the reason of increasing incidence of this
disease in Japan is the increasing incidence of endometriosis.
An, association between clear cell ovarian cancer and
endometriosis has been reported (relative risk, 12.4).76 The
risk increased significantly when the patients were diagnosed at older ages (. 50), which suggests that the
malignant change of endometriosis occurs near menopause
stage. Pathogenesis of clear cell ovarian cancer from endometriosis will be discussed later.
Clinical Features of Clear Cell Ovarian Cancer

Most clear cell ovarian cancer tumors are unilateral at diagnosis, and most are diagnosed at an early stage.77 In a
prospective randomized trial (JGOG 3017), the proportion of
patients with stage I disease was 67%.78 The incidence of
lymph node metastasis was 9.1% in apparent stage IA tumors, 7.1% in stage IC tumors, and was 10.8% in pT2 tumors.77 This stands in contrast with serous ovarian cancers,
which were associated with a higher incidence of lymph
node metastasis than nonserous tumors.79 Although it is
controversial, evaluation of lymph node status by surgical
staging is recommended, because lymph node involvement
in patients with clinical stage I clear cell ovarian cancer was
identified as a stronger prognostic factor.77
Management of clear cell ovarian cancer with positive
peritoneal cytology or surgical rupture remains unclear.
Disease progression was observed in 11% of stage IC
intraoperative rupture tumors but in only 3% of stage IA
tumors.77 Progression-free survival of the patients who had
stage IC tumors with ascites/malignant washing or ovarian
surface involvement was significantly worse than patients
who had stage IC disease with capsule rupture during surgery (p = .04),77 which implies that positive peritoneal
cytology could suggest microscopic implantation of clear cell

ovarian cancer cells.
Another important recent observation is the incidence of
thromboembolic complication in clear cell ovarian cancer. It
has been reported to be higher than in other epithelial
ovarian cancers (16.9%27.3% vs. 0%6.8%),77 and it is also
associated with an elevated interleukin 6 level and with
worse prognosis.80
Chemotherapy Resistance
Several studies that retrospectively analyzed the prognosis
of patients in large randomized clinical trials have been
reported in terms of difference of prognosis between
pathologic subtypes.81,82 The survival was significantly
worse in clear cell than in the serous counterpart in advanced stages, which suggests that clear cell ovarian cancer
was resistant to conventional chemotherapy regimens. New
cytotoxic regimens, dose-dense paclitaxel regimen,83 and
CPT-11 plus cisplatin78 did not demonstrate a benefit for
clear cell ovarian cancer. Chemotherapy administered intraperitoneally using cisplatin and paclitaxel improved the
OS in optimally debulked stage III ovarian cancer, but a
survival benefit with intraperitoneal chemotherapy was not
observed in clear cell ovarian cancer and mucinous adenocarcinoma.84 Therefore, finding new strategies for advanced or relapsed clear cell ovarian cancer is urgent.
Finding molecular pathologic characteristics of this disease
to direct improvement of targeted therapies is critical.
Pathogenesis of Clear Cell Ovarian Cancer From

Endometriosis
Investigators from Kyoto University have conducted a series
of excellent studies to delineate the pathogenesis of clear
cell ovarian cancer from endometriosis (Fig. 3).
They first analyzed the content of endometriotic cysts and
found that the cysts contained huge amounts of iron, oxidative stress marker LPO, and 8-0HdG, a marker of DNA
damage caused by oxidative stress. When the content of an
endometriotic cyst or iron was added to immortalized
FIGURE 3. Hypothetical Carcinogenic Pathway of Clear Cell Ovarian Cancer From Endometriotic Cyst
Abbreviations: IL, interleukin; HNF1b, hepatocyte nuclear factor 1-beta; CCOC, clear cell ovarian cancer.
e253
FUJIWARA ET AL
ovarian surface epithelial cells, intracellular reactive oxygen

species was elevated, and the contents of endometriotic cyst
or iron increased DNA mutations. Therefore, researchers
hypothesized that iron-mediated reactive oxygen species
may cause DNA mutations and carcinogenesis.85
Next, researchers conducted a gene expression microarray
analysis and identified clear cellspecific genes, including
HNF1B, SOD2, ANXA4 HIF1A, IL6, and STAT3. They enriched
gene ontology terms related to oxidative stress and glucose
metabolism. Hepatocyte nuclear factor (HNF) 1-beta binds
to DNA as either a homodimer or a heterodimer with the
related protein HNF 1-alpha. HNF1 alpha and beta share
a common biding motif. Genes having the HNF1 binding
motif in promoter regions are candidate downstream genes
of HNF 1 beta. Interestingly, these genes were upregulated
in immortalized ovarian surface epithelial cells by adding the
content of endometriotic cysts or iron.86 Researchers also
conducted a methylation DNA microarray analysis, which
revealed that clear cell is distinct from other subtypes in terms
of methylation profile. Estrogen receptor pathway genes were
hypermethylated and downregulated, whereas HNF1 pathway
genes were hypermethylated and upregulated.87 Therefore,
clear cell ovarian cancerspecific gene expression seems to be
stabilized via epigenetic mechanisms.
The researchers also investigated the roles of HNF1B in
metabolism of clear cell ovarian cancer cells and found that
HNF1B increases glucose uptake by increasing expression
of a glucose transporter, GLUT1.88 A subsequent metabolome analysis was done and revealed that upregulated
HNF1B expression enhances anaerobic glucose metabolism.
Known as the Warburg effect, the enhanced anaerobic
glucose metabolism has been reported to cause resistance
to oxidative stress. Investigators also analyzed the relationship between HNF1B and oxidative stress in clear cell
ovarian cancer. Knockdown of HNF1B decreased the amount
of glutathione, a redox substance. This was due to decreased
intracellular cystine, a substrate for the biosynthesis of
glutathione, via the decreased expression of rBAT, a cystine
transporter. Then, the investigators found that HNF1B
knockdown increased intracellular reactive oxygen species
and cytotoxicity by iron-induced oxidative stress. Furthermore, in hypoxia, suppression of HNF1B increased sensitivity to cisplatin. Collectively, HNF1B in clear cell carcinoma
causes resistance to oxidative stress and platinum.89
Finally, researchers conducted an exome sequencing
analysis of clear cell ovarian cancer. ARID1A was the topmutated gene, and PIK3CA was the second one. Through an
integrated analysis of gene mutations and copy number
variations, researchers found that the KRAS-PI3K pathway,
SWI/SNF complex, and MYC-RB pathway were the most
frequently altered pathways (written communication with
N. Matsumura, February 2016).
The researchers hypothesize that iron-induced oxidative
stress may cause DNA damage and mutations of PIK3CA and
ARID1A, which may lead to carcinogenesis of clear cell
ovarian cancer. HNF1B plays an important role in the
Warburg effect and in resistance to oxidative stress. This
e254
may be important for progression in the stressful condition

of endometriotic cysts and platinum resistance. Epigenetic
changes, gene mutations, and copy number alterations may
cause stabilization of clear cell ovarian cancerspecific gene
expression and biologic features, including chemotherapy
resistance.
Molecular and Genomic Features of Clear Cell Ovarian

Cancer
Compared with other histologic types of epithelial ovarian
cancer, mutation in p53 is less frequent90 in clear cell ovarian
cancer, and mutation of BRCA1 and BRCA2 is also lower.91
Wilms tumor suppressor 1 gene (WT1) and the WT1antisense promoter were significantly methylated in clear
cell ovarian cancer (88.2%) compared with serous adenocarcinoma (24%).92 PTEN mutation is also frequently observed (27.3%) in this disease.93 Multidrug resistance
protein 3 (MRP3)94 and HNF1B, which has antiapoptotic
effects, are highly expressed.95 More recently, frequent
alteration of ARID1A96 and PIK3CA97 has been reported.
Mabuchi et al showed that the Akt/mTOR pathway is
thought to be the most important passage for tumor growth
in clear cell ovarian cancer.98,99
Recently, Uehara et al100 conducted genomic analyses on
clear cell ovarian cancer and found that, by using microarray analysis, this disease could be classified into three
clusters: one that showed fewer alterations of PIK3CA and
ARID1A and a better prognosis compared with the two
other clusters, which had more alterations.100 This work
suggests that the genomic pattern may be a strong
prognostic factor.
Recently Closed or Ongoing Clinical Trials and Future

Directions
As mentioned earlier, the Japanese Gynecologic Oncology
Group conducted a randomized phase III trial to compare
conventional paclitaxel with carboplatin regimen versus an
irinotecan with cisplatin regimen for patients with newly
diagnosed stage I to IV clear cell ovarian cancer. This is the
first randomized trial specifically on this disease. Unfortunately, there was no survival benefit between the
regimens.78
On the basis of the molecular profile pattern of clear cell
ovarian cancer, the therapeutic target of interest involves
the Akt/mTOR pathway. The GOG-268 study tested one of
the mTOR inhibitors, temsirolimus, administered in combination with paclitaxel and carboplatin for six cycles and
then given for 11 more cycles as a maintenance therapy.
Although the results will be presented this year at the
2016 American Society of Clinical Oncology Annual Meeting,
we await comparison of results in the United States and
South Korea with those obtained in Japan. In addition, the
translational component of this trial is underway to evaluate
the possible difference on molecular characteristics of advanced clear cell ovarian cancer between Japanese and nonJapanese populations.
Antiangiogenetic agents are also of interest as therapeutic

targets for clear cell ovarian cancer. The GOG-254 study
evaluated the efficacy and safety of sunitinib and showed that
sunitinib was minimally active in the second- and third-line
treatments of persistent or recurrent clear cell ovarian cancer.
Another antiangiogenetic agent being tested is nintedanib,
and a Scottish group is conducting a randomized phase II study
to compare nintedanib to other chemotherapeutic agents.
Finally, immunotherapy that uses checkpoint inhibitors is
of interest as a treatment strategy. One recent publication
on an antibody to PD-1, nivolumab, showed promising signs
of efficacy for clear cell ovarian cancer.101
CONCLUSION
The treatment of epithelial ovarian cancer has been informed by an improved understanding of the pathogenesis
of the disease coupled with a heightened awareness of the

genomic variability of this cancer, with recognition that
ovarian cancer is not one disease but comprises many types
of histologies, each with its own genomic landscape. We
are now looking at opportunistic salpingectomy as a means
of prevention, which builds on the growing evidence that
implicates the fallopian tube as the origin of high-grade
serous ovarian cancer. In addition, we are seeing a separation of treatment paradigms, particularly when it comes
to the treatment of high-grade serous ovarian cancer
versus low-grade serous ovarian cancer. We hope that
continued progress in translational research will yield
important findings that can be used therapeutically to
redefine our approach to epithelial ovarian cancer, including the rarer histologies, such as clear cell and lowgrade serous cancers.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Buys SS, Partridge E, Black A, et al; PLCO Project Team. Effect of

screening on ovarian cancer mortality: the Prostate, Lung, Colorectal
and Ovarian (PLCO) cancer screening randomized controlled trial.
JAMA. 2011;305:2295-2303.
Kobayashi H, Yamada Y, Sado T, et al. A randomized study of screening
for ovarian cancer: a multicenter study in Japan. Int J Gynecol Cancer.
2008;18:414-420.
Menon U, Gentry-Maharaj A, Hallett R, et al. Sensitivity and specificity
of multimodal and ultrasound screening for ovarian cancer, and stage
distribution of detected cancers: results of the prevalence screen of
the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS).
Lancet Oncol. 2009;10:327-340.
Jacobs IJ, Menon U, Ryan A, et al. Ovarian cancer screening and
mortality in the UK Collaborative Trial of Ovarian Cancer Screening
(UKCTOCS): a randomised controlled trial. Lancet. 2015;S0140-6736
(15)01224-6.
Domchek SM, Friebel TM, Singer CF, et al. Association of risk-reducing
mortality. JAMA. 2010;304:967-975.
Marchetti C, De Felice F, Palaia I, et al. Risk-reducing salpingooophorectomy: a meta-analysis on impact on ovarian cancer risk
and all cause mortality in BRCA1 and BRCA2 mutation carriers. BMC
Womens Health. 2014;14:150.
Finch A, Metcalfe KA, Chiang J, et al. The impact of prophylactic
salpingo-oophorectomy on quality of life and psychological distress in women with a BRCA mutation. Psychooncology. 2013;22:
212-219.
Finch AP, Lubinski J, Mller P, et al. Impact of oophorectomy on cancer
incidence and mortality in women with a BRCA1 or BRCA2 mutation.
J Clin Oncol. 2014;32:1547-1553.
Parker WH, Broder MS, Chang E, et al. Ovarian conservation at the
time of hysterectomy and long-term health outcomes in the nurses
health study. Obstet Gynecol. 2009;113:1027-1037.
Parker WH, Feskanich D, Broder MS, et al. Long-term mortality associated
with oophorectomy compared with ovarian conservation in the nurses
health study. Obstet Gynecol. 2013;121:709-716.
Kindelberger DW, Lee Y, Miron A, et al. Intraepithelial carcinoma of
the fimbria and pelvic serous carcinoma: evidence for a causal relationship. Am J Surg Pathol. 2007;31:161-169.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
Przybycin CG, Kurman RJ, Ronnett BM, et al. Are all pelvic (nonuterine)
serous carcinomas of tubal origin? Am J Surg Pathol. 2010;34:1407-1416.
Tone AA, Salvador S, Finlayson SJ, et al. The role of the fallopian tube in
ovarian cancer. Clin Adv Hematol Oncol. 2012;10:296-306.
Kurman RJ, Shih IeM. Molecular pathogenesis and extraovarian origin
of epithelial ovarian cancer: shifting the paradigm. Hum Pathol. 2011;
42:918-931.
Sieh W, Salvador S, McGuire V, et al; Australian Cancer Study (Ovarian
Cancer); Australian Ovarian Cancer Study Group; Ovarian Cancer Association Consortium. Tubal ligation and risk of ovarian cancer subtypes:
a pooled analysis of case-control studies. Int J Epidemiol. 2013;42:579-589.
Rice MS, Murphy MA, Tworoger SS. Tubal ligation, hysterectomy and
ovarian cancer: a meta-analysis. J Ovarian Res. 2012;5:13.
Narod SA, Sun P, Ghadirian P, et al. Tubal ligation and risk of ovarian
cancer in carriers of BRCA1 or BRCA2 mutations: a case-control study.
Lancet. 2001;357:1467-1470.
McAlpine JN, Hanley GE, Woo MM, et al. Opportunistic salpingectomy: uptake, risks, and complications of a regional initiative for
ovarian cancer prevention. Am J Obstet Gynecol. 2014;210:471 e1-11.
The Society of Gynecologic Oncology of Canada (GOC). GOC statement
regarding salpingectomy and ovarian cancer prevention, 2011. http://
www.g-o-c.org/uploads/11sept15_gocevidentiarystatement_final_
en.pdf. Accessed March 12, 2016.
Society of Gynecologic Oncology (SGO). SGO clinical practice statement: salpingectomy for ovarian cancer, 2013. https://www.sgo.org/
clinical-practice/guidelines/sgo-clinical-practice-statement-salpingectomyfor-ovarian-cancer-prevention/. Accessed March 12, 2016.
Committee on Gynecologic Practice. Committee opinion no. 620:
salpingectomy for ovarian cancer prevention. Obstet Gynecol. 2015;
125:279-281.
Kapurubandara S, Qin V, Gurram D, et al. Opportunistic bilateral
salpingectomy during gynaecological surgery for benign disease:
a survey of current Australian practice. Aust N Z J Obstet Gynaecol.
2015;55:606-611.
Venturella R, Rocca M, Lico D, et al. Prophylactic bilateral salpingectomy for the prevention of ovarian cancers: what is happening
in Italy? Eur J Cancer Prev. Epub 2015 Aug 13.
Parker W. Oophorectomy versus salpingectomy: a convergence of
ideas. Menopause. 2014;21:323-324.
e255
FUJIWARA ET AL
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
e256
Reade CJ, Finlayson S, McAlpine J, et al. Risk-reducing salpingectomy

in Canada: a survey of obstetrician-gynaecologists. J Obstet Gynaecol
Can. 2013;35:627-634.
Gill SE, Mills BB. Physician opinions regarding elective bilateral salpingectomy with hysterectomy and for sterilization. J Minim Invasive
Gynecol. 2013;20:517-521.
American College of Obstetritians and Gynecologists (ACOG). ACOG
practice bulletin No. 89: elective and risk-reducing salpingo-oophorectomy. Obstet Gynecol. 2008;111:231-241.
Chandrasekaran D, Menon U, Evans G, et al. Risk reducing salpingectomy and delayed oophorectomy in high-risk women: views of
cancer geneticists, genetic counselors and gynecological oncologists
in the UK. Fam Cancer. 2015;14:521-530.
Kwon JS, Tinker A, Pansegrau G, et al. Prophylactic salpingectomy and
delayed oophorectomy as an alternative for BRCA mutation carriers.
Obstet Gynecol. 2013;121:14-24.
Morelli M, Venturella R, Mocciaro R, et al. Prophylactic salpingectomy
in premenopausal low-risk women for ovarian cancer: primum non
nocere. Gynecol Oncol. 2013;129:448-451.
Venturella R, Morelli M, Lico D, et al. Wide excision of soft tissues adjacent
to the ovary and fallopian tube does not impair the ovarian reserve in
women undergoing prophylactic bilateral salpingectomy: results from a
randomized, controlled trial. Fertil Steril. 2015;104:1332-1339.
Findley AD, Siedhoff MT, Hobbs KA, et al. Short-term effects of salpingectomy during laparoscopic hysterectomy on ovarian reserve:
a pilot randomized controlled trial. Fertil Steril. 2013;100:1704-1708.
Yuan H, Wang C, Wang D, et al. Comparing the effect of laparoscopic
supracervical and total hysterectomy for uterine fibroids on ovarian
reserve by assessing serum anti-mullerian hormone levels: a prospective cohort study. J Minim Invasive Gynecol. 2015;22:637-641.
Wang HY, Quan S, Zhang RL, et al. Comparison of serum anti-mullerian
hormone levels following hysterectomy and myomectomy for benign gynecological conditions. Eur J Obstet Gynecol Reprod Biol. 2013;171:368-371.
Manchanda R, Chandrasekaran D, Saridogan E, et al. Should opportunistic bilateral salpingectomy (OBS) for prevention of ovarian cancer
be incorporated into routine care or offered in the context of a clinical
trial? Int J Gynecol Cancer. 2016;26:31-33.
Robert M, Cenaiko D, Sepandj J, et al. Success and complications of
salpingectomy at the time of vaginal hysterectomy. J Minim Invasive
Gynecol. 2015;22:864-869.
Vorwergk J, Radosa MP, Nicolaus K, et al. Prophylactic bilateral salpingectomy (PBS) to reduce ovarian cancer risk incorporated in
standard premenopausal hysterectomy: complications and reoperation rate. J Cancer Res Clin Oncol. 2014;140:859-865.
Kamran MW, Vaughan D, Crosby D, et al. Opportunistic and interventional
salpingectomy in women at risk: a strategy for preventing pelvic serous
cancer (PSC). Eur J Obstet Gynecol Reprod Biol. 2013;170:251-254.
Kwon JS, McAlpine JN, Hanley GE, et al. Costs and benefits of opportunistic salpingectomy as an ovarian cancer prevention strategy.
Obstet Gynecol. 2014;125:338-345.
Lessard-Anderson CR, Handlogten KS, Molitor RJ, et al. Effect of tubal
sterilization technique on risk of serous epithelial ovarian and primary
peritoneal carcinoma. Gynecol Oncol. 2014;135:423-427.
Madsen C, Baandrup L, Dehlendorff C, et al. Tubal ligation and salpingectomy and the risk of epithelial ovarian cancer and borderline
ovarian tumors: a nationwide case-control study. Acta Obstet Gynecol
Scand. 2015;94:86-94.
Falconer H, Yin L, Gronberg H, et al. Ovarian cancer risk after salpingectomy: a nationwide population-based study. J Natl Cancer Inst.
2015;107.
Dietl J, Wischhusen J, Hausler SF. The post-reproductive fallopian
tube: better removed? Hum Reprod. 2011;26:2918-2924.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
Jayson GC, Kohn EC, Kitchener HC, et al. Ovarian cancer. Lancet. 2014;
384:1376-1388.
Sung PL, Chang YH, Chao KC, et al; Task Force on Systematic Review
and Meta-analysis of Ovarian Cancer. Global distribution pattern of
histological subtypes of epithelial ovarian cancer: a database analysis
and systematic review. Gynecol Oncol. 2014;133:147-154.
Gershenson DM. Molecular targeting of low-grade serous and mucinous carcinomas of the ovary or peritoneum. Transl Cancer Res.
2015;4:29-39.
Bookman MA, Gilks CB, Kohn EC, et al. Better therapeutic trials in
ovarian cancer. J Natl Cancer Inst. 2014;106:dju029.
Gourley C, Farley J, Provencher DM, et al. Gynecologic Cancer
InterGroup (GCIG) consensus review for ovarian and primary
peritoneal low-grade serous carcinomas. Int J Gynecol Cancer.
2014;24(9 Suppl 3)S9-S13.
Gershenson DM, Sun CC, Bodurka D, et al. Recurrent low-grade serous
ovarian carcinoma is relatively chemoresistant. Gynecol Oncol. 2009;
114:48-52.
Gershenson DM, Sun CC, Lu KH, et al. Clinical behavior of stage II-IV lowgrade serous carcinoma of the ovary. Obstet Gynecol. 2006;108:361-368.
Schmeler KM, Sun CC, Malpica A, et al. Low-grade serous primary
peritoneal carcinoma. Gynecol Oncol. 2011;121:482-486.
Schmeler KM, Sun CC, Bodurka DC, et al. Neoadjuvant chemotherapy
for low-grade serous carcinoma of the ovary or peritoneum. Gynecol
Oncol. 2008;108:510-514.
Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation
criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J
Cancer. 2009;45:228-247.
ONeill CJ, Deavers MT, Malpica A, et al. An immunohistochemical
comparison between low-grade and high-grade ovarian serous
carcinomas: significantly higher expression of p53, MIB1, BCL2,
HER-2/neu, and C-KIT in high-grade neoplasms. Am J Surg Pathol.
2005;29:1034-1041.
Wong KK, Lu KH, Malpica A, et al. Significantly greater expression of
ER, PR, and ECAD in advanced-stage low-grade ovarian serous carcinoma as revealed by immunohistochemical analysis. Int J Gynecol
Pathol. 2007;26:404-409.
Plaxe SC. Epidemiology of low-grade serous ovarian cancer. Am J
Obstet Gynecol. 2008;198:459 e1-8.
Sieh W, Kobel M, Longacre TA, et al. Hormone-receptor expression
and ovarian cancer survival: an Ovarian Tumor Tissue Analysis consortium study. Lancet Oncol. 2013;14:853-862.
Gershenson DM, Sun CC, Iyer RB, et al. Hormonal therapy for recurrent
low-grade serous carcinoma of the ovary or peritoneum. Gynecol
Oncol. 2012;125:661-666.
Romero I, Sun CC, Wong KK, et al. Low-grade serous carcinoma: new
concepts and emerging therapies. Gynecol Oncol. 2013;130:660-666.
Singer G, Oldt R III, Cohen Y, et al. Mutations in BRAF and KRAS
characterize the development of low-grade ovarian serous carcinoma.
Wong KK, Tsang YT, Deavers MT, et al. BRAF mutation is rare in
advanced-stage low-grade ovarian serous carcinomas. Am J Pathol.
2010;177:1611-1617.
Grisham RN, Iyer G, Garg K, et al. BRAF mutation is associated with
early-stage disease and improved outcome in patients with low-grade
serous ovarian cancer. Cancer. 2013;119:548-554.
Tsang YT, Deavers MT, Sun CC, et al. KRAS (but not BRAF) mutations in
ovarian serous borderline tumour are associated with recurrent lowgrade serous carcinoma. J Pathol. 2013;231:449-456.
Gershenson DM, Sun CC, Wong KK. Impact of mutational status on
survival in low-grade serous carcinoma of the ovary or peritoneum. Br
J Cancer. 2015;113:1254-1258.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
Della Pepa C, Tonini G, Santini D, et al. Low-grade serous ovarian

carcinoma: from the molecular characterization to the best therapeutic strategy. Cancer Treat Rev. 2015;41:136-143.
Friday BB, Adjei AA. Advances in targeting the Ras/Raf/MEK/Erk
mitogen-activated protein kinase cascade with MEK inhibitors for
cancer therapy. Clin Cancer Res. 2008;14:342-346.
Jing J, Greshock J, Holbrook JD, et al. Comprehensive predictive
biomarker analysis for MEK inhibitor GSK1120212. Mol Cancer Ther.
2012;11:720-729.
Farley J, Brady WE, Vathipadiekal V, et al. Selumetinib in women with
recurrent low-grade serous carcinoma of the ovary or peritoneum: an
open-label, single-arm, phase 2 study. Lancet Oncol. 2013;14:134-140.
Grisham RN, Sylvester BE, Won H, et al. Extreme outlier analysis
identifies occult mitogen-activated protein kinase pathway mutations
in patients with low-grade serous ovarian cancer. J Clin Oncol. 2015;
33:4099-4105.
Bidus MA, Webb JC, Seidman JD, et al. Sustained response to bevacizumab in refractory well-differentiated ovarian neoplasms.
Gynecol Oncol. 2006;102:5-7.
Grisham RN, Iyer G, Sala E, et al. Bevacizumab shows activity in patients with low-grade serous ovarian and primary peritoneal cancer.
Int J Gynecol Cancer. 2014;24:1010-1014.
Groen RS, Gershenson DM, Fader AN. Updates and emerging therapies for rare epithelial ovarian cancers: one size no longer fits all.
Gynecol Oncol. 2015;136:373-383.
Angarita AM, Cholakian D, Fader AN. Low-grade serous carcinoma:
molecular features and contemporary treatment strategies. Expert
Rev Anticancer Ther. 2015;15:893-899.
Sugiyama T, Kamura T, Kigawa J, et al. Clinical characteristics of clear
cell carcinoma of the ovary: a distinct histologic type with poor
prognosis and resistance to platinum-based chemotherapy. Cancer.
2000;88:2584-2589.
del Carmen MG, Birrer M, Schorge JO. Clear cell carcinoma of the
ovary: a review of the literature. Gynecol Oncol. 2012;126:481-490.
Kobayashi H, Sumimoto K, Moniwa N, et al. Risk of developing ovarian
cancer among women with ovarian endometrioma: a cohort study in
Shizuoka, Japan. Int J Gynecol Cancer. 2007;17:37-43.
Takano M, Kikuchi Y, Yaegashi N, et al. Clear cell carcinoma of the
ovary: a retrospective multicentre experience of 254 patients with
complete surgical staging. Br J Cancer. 2006;94:1369-1374.
Okamoto A, Sugiyama T, Hamano T, et al. Randomized phase III trial of
paclitaxel/carboplatin (PC) versus cisplatin/irinotecan (CPT-P) as firstline chemotherapy in patients with clear cell carcinoma (CCC) of the
ovary: A Japanese Gynecologic Oncology Group (JGOG)/GCIG study. J
Clin Oncol. 2014;32:5s (suppl; abstr 5507).
Takeshima N, Hirai Y, Umayahara K, et al. Lymph node metastasis in
ovarian cancer: difference between serous and non-serous primary
tumors. Gynecol Oncol. 2005;99:427-431.
Matsuo K, Hasegawa K, Yoshino K, et al. Venous thromboembolism,
interleukin-6 and survival outcomes in patients with advanced ovarian
clear cell carcinoma. Eur J Cancer. 2015;51:1978-1988.
Winter WE III, Maxwell GL, Tian C, et al; Gynecologic Oncology Group
Study. Prognostic factors for stage III epithelial ovarian cancer: a Gynecologic Oncology Group Study. J Clin Oncol. 2007;25:3621-3627.
Mackay HJ, Brady MF, Oza AM, et al; Gynecologic Cancer InterGroup.
Prognostic relevance of uncommon ovarian histology in women with stage
III/IV epithelial ovarian cancer. Int J Gynecol Cancer. 2010;20:945-952.
Katsumata N, Yasuda M, Isonishi S, et al; Japanese Gynecologic Oncology Group. Long-term results of dose-dense paclitaxel and carboplatin versus conventional paclitaxel and carboplatin for treatment
of advanced epithelial ovarian, fallopian tube, or primary peritoneal
cancer (JGOG 3016): a randomised, controlled, open-label trial. Lancet

Oncol. 2013;14:1020-1026.
84. Tewari D, Java JJ, Salani R, et al. Long-term survival advantage and
prognostic factors associated with intraperitoneal chemotherapy
treatment in advanced ovarian cancer: a gynecologic oncology group
85. Yamaguchi K, Mandai M, Toyokuni S, et al. Contents of endometriotic
cysts, especially the high concentration of free iron, are a possible
cause of carcinogenesis in the cysts through the iron-induced persistent oxidative stress. Clin Cancer Res. 2008;14:32-40.
86. Yamaguchi K, Mandai M, Oura T, et al. Identification of an ovarian
clear cell carcinoma gene signature that reflects inherent disease
biology and the carcinogenic processes. Oncogene. 2010;29:
1741-1752.
87. Yamaguchi K, Huang Z, Matsumura N, et al. Epigenetic determinants of
ovarian clear cell carcinoma biology. Int J Cancer. 2014;135:585-597.
88. Okamoto T, Mandai M, Matsumura N, et al. Hepatocyte nuclear
factor-1b (HNF-1b) promotes glucose uptake and glycolytic activity in
ovarian clear cell carcinoma. Mol Carcinog. 2015;54:35-49.
89. Amano Y, Mandai M, Yamaguchi K, et al. Metabolic alterations caused
by HNF1b expression in ovarian clear cell carcinoma contribute to cell
survival. Oncotarget. 2015;6:26002-26017.
90. Ho ES, Lai CR, Hsieh YT, et al. p53 mutation is infrequent in clear cell
carcinoma of the ovary. Gynecol Oncol. 2001;80:189-193.
91. Kobel M, Kalloger SE, Boyd N, et al. Ovarian carcinoma subtypes are
different diseases: implications for biomarker studies. PLoS Med.
2008;5:e232.
92. Kaneuchi M, Sasaki M, Tanaka Y, et al. WT1 and WT1-AS genes are
inactivated by promoter methylation in ovarian clear cell adenocarcinoma. Cancer. 2005;104:1924-1930.
93. Sato N, Tsunoda H, Nishida M, et al. Loss of heterozygosity on 10q23.3
and mutation of the tumor suppressor gene PTEN in benign endometrial cyst of the ovary: possible sequence progression from benign
endometrial cyst to endometrioid carcinoma and clear cell carcinoma
of the ovary. Cancer Res. 2000;60:7052-7056.
94. Ohishi Y, Oda Y, Uchiumi T, et al. ATP-binding cassette superfamily
transporter gene expression in human primary ovarian carcinoma.
Clin Cancer Res. 2002;8:3767-3775.
95. Tsuchiya A, Sakamoto M, Yasuda J, et al. Expression profiling in ovarian
clear cell carcinoma: identification of hepatocyte nuclear factor-1 beta
as a molecular marker and a possible molecular target for therapy of
ovarian clear cell carcinoma. Am J Pathol. 2003;163:2503-2512.
96. Wiegand KC, Shah SP, Al-Agha OM, et al. ARID1A mutations in
endometriosis-associated ovarian carcinomas. N Engl J Med. 2010;
363:1532-1543.
97. Kuo KT, Mao tubal ligation, Jones S, et al. Frequent activating mutations of PIK3CA in ovarian clear cell carcinoma. Am J Pathol. 2009;
174:1597-1601.
98. Hisamatsu T, Mabuchi S, Matsumoto Y, et al. Potential role of mTORC2
as a therapeutic target in clear cell carcinoma of the ovary. Mol Cancer
Ther. 2013;12:1367-1377.
99. Mabuchi S, Kawase C, Altomare DA, et al. mTOR is a promising
therapeutic target both in cisplatin-sensitive and cisplatin-resistant
clear cell carcinoma of the ovary. Clin Cancer Res. 2009;15:5404-5413.
100. Uehara Y, Oda K, Ikeda Y, et al. Integrated copy number and expression
analysis identifies profiles of whole-arm chromosomal alterations and
subgroups with favorable outcome in ovarian clear cell carcinomas.
PLoS One. 2015;10:e0128066.
101. Hamanishi J, Mandai M, Ikeda T, et al. Safety and antitumor activity of
anti-pd-1 antibody, nivolumab, in patients with platinum-resistant
ovarian cancer. J Clin Oncol. 2015;33:4015-4022.
e257
GYNECOLOGIC CANCER
Divide and Conquer: Epithelial

Gynecologic Cancers Beyond
BRCA
CHAIR
Elise C. Kohn, MD
National Cancer Institute at the National Institutes of Health
Bethesda, MD
SPEAKERS
Paul J. Goodfellow, PhD
The Ohio State University College of Medicine
Columbus, OH
Rebecca S. Kristeleit, MD, PhD
UCL Cancer Institute
TARGETING DNA REPAIR DEFICIENCY IN GYNECOLOGIC CANCERS
Gynecologic Cancers: Emerging Novel Strategies for Targeting

DNA Repair Deficiency
Rebecca S. Kristeleit, MD, PhD, Rowan E. Miller, MD, PhD, and Elise C. Kohn, MD
OVERVIEW
The presence of a BRCA mutation, somatic or germline, is now established as a standard of care for selecting patients with
ovarian cancer for treatment with a PARP inhibitor. During the clinical development of the PARP inhibitor class of agents, a
subset of women without BRCA mutations were shown to respond to these drugs (termed BRCAness). It was hypothesized
that other genetic abnormalities causing a homologous recombinant deficiency (HRD) were sensitizing the BRCA wild-type
cancers to PARP inhibition. The molecular basis for these other causes of HRD are being defined. They include individual gene
defects (e.g., RAD51 mutation, CHEK2 mutation), homozygous somatic loss, and whole genome properties such as genomic
scarring. Testing this knowledge is possible when selecting patients to receive molecular therapy targeting DNA repair, not
only for patients with ovarian cancer but also endometrial and cervical cancers. The validity of HRD assays and multiple gene
sequencing panels to select a broader population of patients for treatment with PARP inhibitor therapy is under evaluation.
Other non-HRD targets for exploiting DNA repair defects in gynecologic cancers include mismatch repair (MMR), checkpoint
signaling, and nonhomologous end-joining (NHEJ) DNA repair. This article describes recent evidence supporting strategies in
addition to BRCA mutation for selecting patients for treatment with PARP inhibitor therapy. Additionally, the challenges and
opportunities of exploiting DNA repair pathways other than homologous recombination for molecular therapy in gynecologic cancers is discussed.
ur increased understanding of cancer biology coupled

with increasingly refined technology to examine the cancer
genome has offered a rich supply of opportunities applicable
to improving outcomes in patients with gynecologic cancer.1
We now recognize at least six major interactive pathways
involved in DNA damage and repair.2 The most recent example of molecularly targeted drug success in patients with
ovarian cancer is the development of PARP inhibitors as therapeutics. The PARP inhibitor olaparib was the first drug
worldwide to be licensed in 2014 for a molecularly defined
population of patients with BRCA-mutated ovarian cancer.
The rapidity and success of identifying this molecularly directed and defined therapeutic has catalyzed the gynecologic
cancer community to further explore the application of DNA
repair inhibitors as a class and question other ways DNA repair
defects might be harnessed for novel treatment approaches.
Targeting tumors with defective DNA repair exploits the
molecular differences between tumor and normal cells. This
mechanism is the basis for tumor-specific cell death induced
by PARP inhibitors in patients with BRCA-mutated ovarian
cancer.3,4 BRCA1 and BRCA2 are essential for maintaining
genomic stability through the error-free repair of DNA
double-strand breaks via the highly conserved homologousrecombination repair (HRR) pathway.5 Although the synthetic lethality between deleterious BRCA mutations and
PARP inhibition is well established,4,5 it is becoming increasingly evident that gynecologic tumors have other molecular features, germline or somatic, which portend an HRD
or BRCA-like susceptibility to platinums and DNA repair inhibitors. Substantial efforts are underway to categorize the
breadth of molecular causes of HRD. Individual genetic mutations and whole genome features, expressed as genomic
scarring, have been identified as HRD-causing and correlate
with potential responsiveness to DNA repair inhibitors.6-9 In
addition, defects in other DNA repair pathways, such as the
MMR pathway (common in patients with endometrial cancer)
and cell cycle checkpoint proteins, cause potential vulnerabilities that offer therapeutic possibilities (Fig. 1).10-13 Going
beyond BRCA-targeting in gynecologic cancers highlights the
potential for expanded therapeutic strategies. The phenotypic and genotypic consequences of HRD are a particular
vulnerability in patients with epithelial ovarian cancer, and
new insight shows how other events in endometrial, cervical,
and ovarian cancers also may yield a HRD phenotype. An
From the Department of Medical Oncology, University College London Hospital, London, United Kingdom; UCL Cancer Institute, University College London, London, United Kingdom;
Clinical Investigations Branch, Cancer Therapy Evaluation Program, National Cancer Institute, Bethesda, MD.
Corresponding author: Rebecca S. Kristeleit, MD, PhD, UCL Cancer Institute, 72 Huntley St., London WC1E 6BT United Kingdom; email: r.kristeleit@ucl.ac.uk.
e259
KRISTELEIT, MILLER, AND KOHN
immediate challenge is learning how best to navigate and

apply the new information at both the scientific and clinical
levels, where genetically defined treatment decisions are now
being made with patients.
FIGURE 1. Representation of the Main DNA Repair

Pathways and Interaction
HOMOLOGOUS-RECOMBINATION REPAIR AND

BRCA1/2
DNA is constantly subjected to damage by environmental
exposures and endogenous activities such as DNA replication and cellular free radical generation. These cause a variety
of DNA lesions, including base modifications, double-strand
breaks, and single-strand breaks.14 DNA repair is critical to
maintain genomic integrity by allowing cells to progress
through the cell cycle and complete replication without
errors.15 Homologous-recombination repair is the principle
mechanism by which double-strand breaks are repaired.
The BRCA1/2 genes, along with other genes in the Fanconi
anemia (FA) pathway, encode essential proteins for this
process. Homologous-recombination repair is a conservative form of DNA repair that restores the DNA to its original
sequence using the homologous normal DNA template during S and G2/M phases of the cell cycle. When either BRCA1
or BRCA2 are defective, homologous recombination is dysfunctional and double-strand break repair proceeds using
error-prone nonconservative repair mechanisms such as NHEJ
and single-strand repair.5 Nonhomologous end-joining does
not use a DNA template and occurs in G0 or G1, propagating
error rather than repairing it.16 Any two free DNA break ends
are directly ligated during repair of double-strand breaks by
NHEJ.
A variety of mechanisms exist for repairing single-strand
breaks. These include base excision repair, nucleotide excision repair, and MMR, processes that are modulated by
KEY POINTS
e260
The synthetic lethality and validity of using BRCA1/2

mutation as a biomarker predictive of response to PARP
inhibition has been confirmed in clinical trials of patients
with ovarian cancer.
Homologous recombinant deficiency (HRD) represents a
key vulnerability in patients with high-grade ovarian
cancer and possibly other gynecologic cancers, which
can be exploited using PARP inhibitors.
Methods to identify HRD cancers and broaden the
applicability of DNA repair inhibitors are being
evaluated in the clinic.
Nonhomologous recombinant DNA repair pathways,
such as mismatch repair and nonhomologous
end-joining, represent important targets for novel
therapeutic strategies in gynecologic cancer.
Targeting cell cycle checkpoint proteins such as CHK1,
CHK2, and WEE-1 that regulate DNA damage and
repair is a promising therapeutic approach, particularly
in molecular subsets such as p53 mutant and
ARID1A-mutated gynecologic cancer.
This figure represents the six main DNA repair pathways and their close but
discrete interaction. Targetable proteins associated with each pathway for which
pharmacological agents exist appear in red.
Abbreviations: MMR, mismatch repair; BER, base excision repair; NHEJ,
nonhomologous end-joining; HRR, homologous recombination repair; NER,
nucleotide excision repair; TLJ, translesional joining.
PARP. PARP senses and binds to DNA-break sites, which

results in catalytic activation and the recruitment of other
components of the DNA repair complex.17 If a cell fails to
repair single-strand breaks before attempting replication, a
double-strand break will then form.
Inherited mutations in the tumor suppressor genes BRCA1
and BRCA2 account for the majority of familial ovarian
cancers.18 The BRCA1 and BRCA2 protein products function
in multiple cellular pathways, including cell cycle regulation
and maintenance of genome integrity.19 Cells with defective
HRR must rely on alternative pathways for DNA repair to
survive, thereby providing potential therapeutic targets.
Patients with epithelial ovarian cancer and germline or somatic BRCA1 or BRCA2 mutations demonstrate impaired
ability to repair double-strand breaks through HRR, which
likely explains the increased sensitivity to platinum and the
potentially more favorable outcome compared with patients
who are wild-type.20,21
HOMOLOGOUS-RECOMBINATION REPAIR
INHIBITION
PARP Inhibition
PARP inhibitors were developed for BRCA1/2 mutant epithelial ovarian cancer following observation that BRCA1/2
mutations greatly increased the in vitro sensitivity to PARP
inhibition, exploiting a concept known as synthetic lethality.3,4 Synthetic lethality arises when a combination of defects in two or more genes or proteins leads to cell death,
whereas a single defect is compatible with cell viability.
BRCA1/2 defective cells are dependent on non-HR DNA
repair and they are sensitive to any induction in doublestrand breaks. PARP inhibition produces stalled replication
forks, which increases the number of double-strand breaks
and leads to genetic chaos and cell death by apoptosis or
senescence.2 The synthetic lethality between BRCA1/2
mutations and PARP inhibition has been confirmed in
clinical trials.22-25 Multiple PARP inhibitors, including olaparib
(AZD2281), rucaparib (CO-338), veliparib (ABT888), and
niraparib (MK4827), are in clinical development either
as single agents or in combination therapy for the
management of patients with epithelial ovarian cancer
(Table 1).
Olaparib, (Lynparza/AZD2281) is the first licensed PARPi
for the treatment of BRCA-mutated epithelial ovarian cancer.26,27 The initial olaparib phase I study provided clinical
proof-of-concept of synthetic lethality between BRCA1/2
mutant tumors and PARP inhibition. Of the expansion phase
patients, 40% attained either RECIST partial or complete response, CA-125 responses by Gynecological Cancer Intergroup
criteria, or both. Subsequent phase II studies evaluating olaparib
monotherapy in patients with relapsed epithelial ovarian cancer
have shown response rates of 31% to 41% in BRCA1/2-mutation
carriers and up to 21% in BRCA1/2 wild-type patients.28,29
A phase II trial investigating olaparib maintenance therapy
following an initial response to platinum therapy showed a
progression-free survival (PFS) extension from 4.3 months
with placebo to 11.2 months with olaparib (hazard ratio [HR]
0.18; 95% CI, 0.100.31) in tumors harboring BRCA1/2 mutations. A benefit for olaparib maintenance in patients with
BRCA wild-type tumors was observed, although the magnitude was smaller (7.4 vs. 5.5 months; HR 0.54, 95% CI,
0.340.85).24,30 No overall survival (OS) benefit has been
observed at the current time of reporting, the reasons for
which are likely multifactorial.
Defects in Non-BRCA Homologous-Recombination

Repair Genes That Modulate Genomic Stability and
May Promote Sensitivity to DNA Repair Inhibitors
Genomic instability is an important therapeutic target in
gynecologic cancers, not just because of the advent of PARPi
but also because of the key roles of radiation and platinum
therapies in managing them. Platinum analogs induce intraand interstrand purine base cross-links (ICL), which form
covalent bonds and stress DNA repair. Repair of ICLs depends on nucleotide excision repair and, secondarily, upon
double-strand break formation.14 The marked sensitivity of
epithelial ovarian cancer to platinum agents is thought to
be related to the high frequency of underlying HRR defects.
Germline or homozygous somatic mutations in other
members of the FA family, such as RAD51C, RAD51D, and
BRIP1, increase susceptibility to ovarian cancer.31-33 In vitro
studies have demonstrated that deficiency in these genes

and in other HRR-associated proteins, such as ATM, CHEK1,
CHEK2 and CDK12, also confer sensitivity to DNA damage
and DNA repair inhibition.31,34-36 These have been found
in sporadic epithelial ovarian cancer and other cancers in
which they appear to function to create a BRCA mutationlike phenocopy.37 Understanding other DNA damage mechanisms and potential targets across gynecologic cancers
can further extend the success of DNA repair inhibitors,
exemplified by PARP inhibitors.
The Cancer Genome Atlas (TCGA) identified mutations
now recognized to be related to the HRR pathway in approximately 30% of high-grade serous ovarian cancers.38
This included somatic mutations in BRCA1/2 (3%), ATM and
ATR (2%), the FANC family (5%), and hypermethylation of
RAD51C (3%), as well as germline mutations in BRCA1 (9%)
or BRCA2 (8%). EMSY amplification (13%), which is proposed
to inactive BRCA2, has not been validated in patients yet.
Pennington and colleagues used targeted capture and
massively parallel genomic sequencing to examine germline
and somatic loss-of-function mutations in 30 genes, including 13 HRR genes in 390 epithelial ovarian cancers.36
Thirty-one percent of ovarian cancers had a deleterious
germline (24%) and/or somatic (9%) mutation in one or more
of the 13 HRR genes, with similar incidence in serous (31%)
and nonserous ovarian cancers (28%, p = .06). The germline
or somatic HRR gene mutations predicted platinum sensitivity (p = .0002) and improved OS (p = .0006; Table 2). The
majority of germline and somatic HRR gene mutations were
in BRCA1/2, and 26% occurred in other HRR genes. A similar
frequency of mutations was observed in patients with nonserous epithelial ovarian cancer but with a different spectrum
of targeted genes. The functionality of these additional
mutations has been observed with rucaparib in patients
with both germline and somatic RAD51C mutations within
the ongoing phase II ARIEL2 study (NCT01891344).39 It is
now being recognized in epithelial ovarian cancer and in
other tumor types that the presence of HRD might be a
viable strategy for selection for DNA repair inhibitor trials.
For example, 88% (14 of 16) of men with metastatic
castration-resistant prostate cancer in the TO-PARP phase II
olaparib study with a somatic mutation in an HRR gene,
including BRCA1, BRCA2, ATM, the FANC genes, and CHEK2,
responded to olaparib compared with only two of 33 (6%) of
patients who were wild-type.40
New Opportunities Leveraging Genomic Instability

Recent translational data show BRCA1/2-mutated epithelial
ovarian cancer has a greater immune infiltration.41,42 It has
been suggested that these cancers may have sensitivity to
immune checkpoint inhibitors targeting the PD-1/PD-L1
pathway. It is hypothesized that PD-1/PD-L1 targeting agents
may preferentially benefit these patients because BRCA1/2mutated and other HRR-deficient tumors have higher numbers
of neoantigens.41 This hypothesis remains to be tested in clinical
trials.
e261
e262

III
ICON9
II
NCT02354586
Single Agent
Maintenance
Combination
First-line
Treatment
NA
NA
Veliparib and temozolamide
Veliparib in combination with

carboplatin and paclitaxel
and as maintenance therapy
Placebo
NA
Olaparib with cediranib as

maintenance therapy
following platinum-based
chemotherapy with cediranib
Olaparib and cediranib
Olaparib and cediranib
Olaparib in combination
with platinum-taxane
and bevacizumab and as
maintenance therapy
NA
NA
Combination
Single-agent and combination phase I studies in multiple tumor types
III
II
NCT01113957
NCT01847274
III
NCT02470585
Maintenance
Maintenance
First relapse,
maintenance
Combination
Combination
First-line
treatment
Maintenance
Maintenance
Study Type
NA
Placebo
Liposomal
Doxorubicin
Placebo
Placebo
NA
Cediranib and placebo

as maintenance
therapy
Chemotherapy, olaparib
alone or cediranib
alone
Platinum-doublet
chemotherapy
Placebo
Placebo
Placebo
Comparator
Abbreviations: PR, platinum resistant ovarian cancer; PS, platinum sensitive ovarian cancer; HGSOC, High-grade serous ovarian cancer.
Talazoparib (BMN673)
Niraparib (MK4827)
Veliparib (ABT-888)
III
III
NCT02502266
NRG GY005
II
III
NCT02446600
NRG GY004
NCT01968213
(ARIEL3)
III
NCT02477644
(PAOLA-1)
NCT01482715
III
NCT01844986
(SOLO1)
Rucaparib (CO338)
III
NCT02392676
Olaparib (AZD2281)
Phase
NCT Trial Number
PARP Inhibitor
TABLE 1. Phase II/III Trials of PARP Inhibitors in Ovarian Cancer
PS
PS
PR
NA
PS
PS
Any BRCA
HGSOC or
endometrioid
HGSOC, stage III-IV
HGSOC
HGSOC gBRCA and
any high-grade
histology
HGSOC
$ 3 chemotherapy
$ 2 platinum
Treatment nave
$ 1 platinum and # 3
cytotoxic regimes
$ 2 platinum
$ 2 chemotherapy
HGSOC
1 platinum
HGSOC/
endometrioid
or gBRCA and
any high-grade
histology
# 2 in platR setting
PR
PS
HGSOC/
endometrioid
or gBRCA and
any high-grade
histology
Any number of
platinum regimes
HGSOC/
endometrioid,
stage IIIB-IV
Treatment nave
sBRCA, sHRR
gBRCA
$ 2 platinum
1 platinum
Inclusion Criteria
Previous Lines of Treatment
PS
NA
PS
PS
Platinum Status
TABLE 2. Known Deleterious Homologous Recombinant Deficiency Gene Frequencies in Ovarian Cancer
HR-Pathway Gene
Observed Frequency All Epithelial Observed Frequency High-Grade

Ovarian Cancer (%)
Ovarian Cancer (%)
References
RAD51C
0.412.9
1.9
Walsh et al8, Pennington et al36, Minion et al83, Cunningham et al84,

Song et al85
RAD51D
0.351.1
0.95
Pennington et al36, Cancer Genome Atlas Research Network38, Song

et al85
RAD51B
0.06
0.95
Cancer Genome Atlas Research Network38, Song et al85
RAD50
0.21.0
Walsh et al8, Minion et al83
RAD54L
0.5
Kristeleit et al86
ATM
0.80.86
0.321.0
Pennington et al36, Cancer Genome Atlas Research Network38,

Minion et al83
BRIP1
0.94.0
0.321.0
Walsh et al8, Pennington et al36, Cancer Genome Atlas Research

Network38, Ramus et al87
CHEK2
0.45.0
0.321.0

Network38, Minion et al83
FANCA
0.5
Kristeleit et al86
FANCI
0.5
Kristeleit et al86
NBN
0.21.0
0.631.0

Network38, Candido-dos-Reis71, Minion et al83
PALB2
0.22.0
0.63

Network38, Ramus et al87
The Need for Biomarkers: Identifying the Homologous

Recombinant Deficiency Signature
The ability to perform rapid whole-genome sequencing
makes it feasible to classify cancers according to their underlying mutational spectrum. A series of in vitro studies has
confirmed the presence of large subchromosomal deletions
and other genomic changes that cause allelic imbalance
and confer an HRD phenotype.6,43,44 This has yielded several assays or weighted signatures now being examined
as companion diagnostic predictive biomarkers. This HRRdeficient mutational signature or mutational scar indicates reliance on error-prone DNA repair pathways.9
Array-based technology using single nucleotide polymorphisms genotyping and comparative genomic hybridization has demonstrated that the genomes of high-grade
serous ovarian cancer harbor common loss of single parental
alleles. Wang and colleagues examined loss of heterozygosity (LOH) and copy number changes in patients with highgrade serous ovarian cancer and divided patients into two
clusters of LOH high and LOH low.44 High-levels of LOH were
associated with platinum sensitivity and improved PFS.
BRCA1/2-mutant tumors fell within the LOH high group; an
increased sensitivity to platinum was seen in the LOH high
group even after exclusion of these patients. An LOH-based
score has been developed that is strongly associated with
functional defects in BRCA1/2 and other genes implicated in
the HRR pathway.6 Prospective validation of a genomic scar
LOH assay is ongoing within the ARIEL2 rucaparib PARP
inhibitor phase II trial (NCT01891344) to dichotomize BRCA
wild-type patients who benefit from rucaparib. Initial results
suggest increased activity for rucaparib within the BRCA-like
population with high genomic LOH compared with the low
LOH population (PFS 7.1 vs. 3.7 months; HR 0.61), although

the benefit was not as great as in the BRCA1/2 mutant cohort
(PFS 9.4 months).39 Additional assays to identify structural
changes associated with HRD include the large-scale transitions assay quantifying chromosomal breaks of at least
10 Mb45 and the telomeric alleic imbalance score,43 both of
which correlate with alterations in BRCA1/2 and other HRR
pathway genes in patients with ovarian cancer. Although each
of these assays offers the exciting prospect of identifying a
subset of BRCA-like tumors, which respond to HRR-directed
therapy, the relevance of each assay requires prospective
validation in clinical trials. As more data emerge from post hoc
analyses of tumor samples from completed studies, it is likely
that HRD mutational signatures for consideration of DNA
repair inhibition therapy will become more established.
DNA REPAIR INHIBITION BEYOND

HOMOLOGOUS-RECOMBINATION REPAIR
GENES
Nature has recognized the need to maintain genomic
integrity as demonstrated by at least six majorand
interactiveDNA repair pathways (Fig. 1). Dissection of HRR
has shown this interaction and identified new potential
targets for therapeutic intervention. Likewise, dissection of
developmental processes in patients with endometrial and
colon cancers, and now cervical cancers, has led to recognition of other key DNA repair cancer risk genes. Some of
these risk genes are involved in other DNA repair or biologic
pathways critical to cellular survival.
Early results implicate deficiency in ARID1A, a key component of the chromatin-remodeling complex, as sensitizing
e263
tumor cells to PARP inhibition in vitro and in animal

models.46 ARID1A is a suppressor gene. It is recruited by the
homologous-recombinant protein ATR, upstream of the cell
cycle G2/M regulator, CHK1, to sites of DNA double-strand
breaks where it facilitates efficient processing of doublestrand breaks and sustains DNA damage signaling. Mutations in ARID1A are common in clear cell and endometrioid
epithelial ovarian cancer occurring in up to 57% and 30%
of cases, respectively.47,48 Loss of ARID1A expression is also
common in patients with endometrioid endometrial cancer
(approximately 40%49), and similarly in patients with cervical cancer. These results imply that DNA repair inhibitors
must be tested more broadly in patients with gynecologic
cancers.
DNA repair pathways have cell cycle specificity.50 Accordingly, many proteins of the cell cycle, especially within
the G2/M checkpoint, also regulate DNA damage and repair.
CHK1/2 and WEE-1 kinases are examples.51-53 Agents targeting these kinases have minimal single-agent activity in
the absence of p53, or more likely, p53 and second (so
far undefined) mutations. Efficacy is much greater when
examined in a defined p53 mutant background in which
the G1/S checkpoint is aberrant and also in combination.
There is a subset of patients with ovarian cancer with wildtype BRCA1/2 function and cyclin E amplification and
overexpression.38,54-56 Emerging data indicate that blockade of the G1/S checkpoint, such as with pertinent CDK
inhibitors, may unmask an unexpected sensitivity to DNA
repair inhibitors such as PARP inhibitors.57 Furthermore,
pharmacologic augmentation of hypoxia, such as in the use
of cediranib, can reduce expression of many key homologousrecombinant proteins. This may underpin the activity of the
cediranib/olaparib combination in women who have wildtype BRCA.58
Targeting Nonhomologous-Recombination Repair

DNA Repair Pathways
Mismatch repair deficiency. Mismatch repair deficiency is a
single-strand DNA repair mechanism. It maintains genomic
integrity by correcting base substitution mismatches and
small insertion-deletion mismatches generated by errors in
base pairing during DNA replication. Mismatch repair deficiency is critical to maintaining genomic stability. Failure
to recognize and repair DNA mismatches results in microsatellite instability and a mutator phenotype, with mutation
rates 100- to 1,000-fold higher than in MMR proficient
cells.13 Loss of one of the MMR proteins (MSH2, MSH6,
MSH3, MLH1, and PMS2) is associated with an increased risk
of cancers, including endometrial, ovarian, colorectal, and
gastric cancers.59 Germline mutations in MMR genes give
rise to Lynch syndrome, which is associated with a 60% and
25% lifetime risk of developing endometrial and colon
cancers, and ovarian cancer, respectively.59,60 Somatic loss
of MMR genes can occur either by mutation or methylation,
such that MMR deficiency is associated with up to 30% of all
endometrial cancers.61
e264
No direct therapy exists to target MMR-deficient tumors;

however, in vitro data suggest a number of potential directions, including inhibition of select DNA polymerases
through an accumulation of oxidative DNA damage. Specifically, MSH2 deficiency is synthetically lethal with inhibition of DNA polymerase b (POLB), the DNA polymerase
that catalyzes nuclear base excision repair. MLH1 deficiency
is synthetically lethal with inhibition of DNA polymerase g
(POLG), the only polymerase specific to mitochondrial DNA.62
Methotrexate has been shown to be lethal to MSH2-deficient
cells through the accumulation of nuclear oxidative DNA
damage,63 leading to an ongoing phase II trial in colorectal
cancer (NCT00952016). Perhaps more promising is the
ability to target secondary mutations that arise as a result of
MMR deficiency believed to drive the tumorigenic phenotype. Secondary mutations in the double-strand break DNA
repair gene MRE11 are commonly associated with MMRdeficient colorectal cancer and lead to PARP inhibitor sensitivity in vitro.64 The role of PARP inhibition in patients with
endometrial cancer is ready for assessing in clinical trial with
accompanying tissue analysis for microsatellite instability,
MRE11, and PTEN status.
The TCGA endometrial cancer group61 identified a second
subgroup of endometrial cancers characterized by mutations in POLE, a catalytic subunit DNA polymerase epsilon
involved in nuclear DNA replication and repair. Mutations in POLE resulted in an ultramutated phenotype with a
mutation frequency approximately 10-fold greater than
seen in microsatellite instability tumors. The POLE-mutated
subgroup also had extraordinarily good survival. Very preliminary data suggest that the microsatellite instability/POLE
phenotype may result in marked increased frequency
in neoantigen production, rendering the tumors more
immunogenic and, therefore, vulnerable to immunotherapy.65 Early preclinical data support the use of antiPDL1 therapy in POLE-hypermutated and microsatellite
instabilitypositive endometrial cancers because they
have greater expression of PD-1, PD-L1, and tumor infiltrating lymphocytes.66 Several clinical trials currently
underway are evaluating the role of antiPD-1/PD-L1 therapy in endometrial cancer either as a single agent (e.g.,
NCT02628067) or in combination with carboplatin and
paclitaxel (NCT02549209).
Base excision repair. Many other potential targets have
been identified in other DNA repair pathways. The potential
therapeutic importance of DNA-PKcs in base excision repair
was identified during the demonstration that PARP was not
the base excision repair rate-limiting step.16 A normal role
of PARP is to maintain inactive DNA-PKcs, which keeps
NHEJ quiet. Agents now in clinical development are targeting
DNA-PKcs. DNA-PKcs has been implicated with adverse
outcome in patients with epithelial ovarian cancer, suggesting they may be a clinically relevant target in gynecologic
cancer.67 TRC-102 is an experimental agent targeting base
excision repair that also can be considered for patients with
gynecologic cancers.
Cervical Cancer and DNA Repair Defects

The role of defective DNA repair in cervical cancer is less well
established. HPV infection and its associated production of
the oncoviral proteins E6 and E7 causes inactivation and
degradation of the p53 and pRB tumor-suppressor genes.
This results in cell cycle dysfunction and altered DNA repair
capacity.68 Because of defective DNA damage response,
cervical tumor cells are increasingly dependent on residual
repair pathways to cope with certain types of DNA damage.
In support of this, a correlation between response to DNA
damaging therapy and activation of DNA repair pathways
has been noted in clinical series, albeit involving relatively
small numbers of patients. Patients treated with chemoradiotherapy were found to have high expression of the
nucleotide excision repair protein ERCC1 associated with
a decreased PFS and worse OS 69 and activation of the
FA/BRCA pathway correlated with treatment failure; conversely, impaired NHEJ repair was related to increased OS in
patients treated primarily with radiotherapy.70 These observations suggest that therapeutic exploitation of DNA
repair pathways may be useful in potentiating chemoradiotherapy, which is the mainstay of treatment for cervical
cancer. Numerous early phase trials incorporating modulators of DNA repair such as PARP, ATM and ATR inhibitors,
and triapine in combination with standard chemotherapy or
radiotherapy currently are underway in patients with advanced cervical cancer (NCT01281852, NCT02223923,
NCT02595879, NCT02466971).
implications of the potential findings or the treatment plan

that the results should inform.74 Using a targeted panel of
HRR genes may allow the identification of HRD cancers,
notwithstanding a number of limitations. With the exception
of BRCA1/2, each individual HRR gene defect is present at
very low frequency in patients with ovarian cancer, and
the overall numbers of additional HRD patients identified
by testing this way is small (Table 2).8,74,75 Homologous
recombinant deficiency arises via heterogenous mechanisms, which include epigenetic changes, gene amplifications, and chromosomal translocations. Therefore, a suitable
and validated genomic platform is required to capture all
patients with potentially actionable HRR aberrations. A third
concern is that not all genes are equally important in determining therapeutic response, and the number of variants
of uncertain significance is massive, leaving many patients
and physicians without guidance.
Addressing whether there is sufficient overall risk to
mandate germline testing for all women across all 11 or
more genes (panel testing) or to focus on validated biomarker(s) is critical. The latter would allow us to address the
more complicated question of the role of haplo-insufficiency
complementation in genomic instability of more than one
of the homologous-recombinant genes or homologousrecombinant plus another DNA repair pathway gene. Financial, personal, and family costs make navigating this new
information more complicated.
Applying the New Information
NAVIGATING THE NEW INFORMATION

Picture the cartoon: a patient with the balloon above her
head full of question marks. The caption reads What do I
do? The most difficult question for providers and patients is
how to weigh progress and new agents in the context of
population data when addressing individual patient decisions. As our understanding of the meaning and roles of
HRD mutation and dysfunction progresses, we complicate
informing patients.
We always start with whats simple. Germline deleterious
BRCA mutation in patients with ovarian cancer is the most
common and actionable finding that generally directs simple decisions as described above. BRCA mutation or genomic instability as a result of HRD, in general, allows greater
confidence in the use of DNA damaging agents, DNA repair
inhibitors, cell cycle inhibitors, and novel combinations that
leverage these dysfunctional pathways. We know that BRCAmutation patients with ovarian cancer are more responsive
and have better PFS, at least in the first decade after
diagnosis.1,71-73 Newer data of small, unpublished numbers
suggest that any homozygous loss of BRCA1/2 may result in a
nearly identical phenotype to germline or somatic homozygous loss, an HRD phenotype for which new directions
have been outlined.
Germline evaluations of BRCA and other recently identified HRD genes and broader gene panel testing are being
done for many patients, often without considering the
BRCA1 or BRCA2 germline or homozygous somatic mutations are now considered predictive biomarkers for platinum- and PARP inhibitorsensitivity.36 A woman with a
BRCA1 or BRCA2 mutation who progresses on a platinumbased therapy often will respond subsequently. Data are
limited to guide application of the platinum-resistant
moniker to such women. How to apply the growing data
related to complex DNA repair pathways and new agents
also is becoming more complex. The first question we need
to address urgently in the clinic is how to interpret loss of
sensitivity to such agents as the patient moves along the
treatment line. Acquisition of secondary mutations in BRCA1
or BRCA2 may result in resumption of BRCA1 or BRCA2
function.76,77 Estimates of frequency of these secondary
mutations range from single digits to nearly 40% of cases in
the small case numbers examined to date. Other potential
molecular events include loss of 53bp1, a regulator of the
poor fidelity NHEJ DNA repair pathway,78,79 and regulation of phosphoproteins, such as DNA-PKcs,16 and others in
the DNA repair pathways, such as ATM and ATR.80-82 The
cause of true platinum resistance in these women is still a
conundrum.
WHERE TO GO NEXT?
DNA repair is a very complex series of parallel and interacting
pathways. There are several druggable targets within these
pathways. Progress in understanding these pathways and
e265
TABLE 3. Defining the Decisions?

Event
Implication in Cancer
Opportunities
Germline Deleterious gBRCA
Presumed homozygous deletion, especially breast and

ovarian cancers
DNA damaging agent and PARP inhibitor

susceptibility
Somatic BRCA Mutation
Role of haplo-insufficiency is unclear
Sensitivity requires homozygous loss of function
BRCA1 Promoter Methylation
Does not appear to confer stable gene downregulation

in ovarian cancer
Unclear any benefit
BRCA1/2 Downregulation
Ambient micro-environmental events may affect protein

production or stability
Correlative and preclinical studies suggest this may

yield a sensitive phenotype
Cyclin E Overexpression
Cell cycle dysregulation is associated with altered

balance in DNA damage repair
Targeting CDKs alone and in combination with DNA

repair inhibitors
Unknown and Wild-Type
Work still to do
Wide-open opportunities for patient care and

learning
identifying areas of commonality, such as the role of CHK1

in both the G2/M cell cycle checkpoint and downstream of
ATM and ATR in DNA repair, has led to new therapeutic
approaches. This is an example of the concept that targeting
sites of biochemical convergence may yield more than the
sum of the inhibition of single independent targets. Chemotherapy is targeted therapy. Most chemotherapies target
DNA damage. We now add the therapeutic category of
DNA repair inhibitors. How to exploit this new and growing
therapeutic category in the context of the treatment of all
women with ovarian cancers is a challenge and a great
opportunity (Table 3). This growth provides great opportunities beyond BRCA.
CONCLUSION AND FUTURE DIRECTIONS

Targeting defective DNA repair represents a viable treatment option for patients with gynecologic malignancies.
Defective HRR is a key vulnerability for patients with highgrade serous epithelial ovarian cancer, which is present in up
to 50% of patients. The use of PARP inhibitors allows the
exploitation of molecular differences between tumor normal tissues. To maximize the benefit from PARP inhibition, it
is important to identify those tumors not only characterized
by BRCA1/2 mutations but also those with HRD as a result of
other mechanisms. Many questions remain unanswered.

These include what determines the best predictor of response to PARP inhibition and how best to use them in
the management of epithelial ovarian cancer (i.e., as single
agents, in combination with chemotherapy, or as maintenance therapy)? Should we rechallenge with PARP inhibitors for patients in which PARP-directed therapy has
failed?
Several trials in progress will likely help answer these
questions. It is becoming increasingly apparent that targeting defective DNA repair in patients with nonserous epithelial ovarian cancer, endometrial cancer, and cervical
cancer also may have therapeutic potential. This could be
achieved either through the identification of non-HRR genes,
which may modulate HRR pathways, or by targeting alternative DNA repair pathways such as MMR. Options likely will
increase in tandem with our understanding as long as we
keep asking relevant questions. The overall aim is, as always,
to increase benefit for our patients.
ACKNOWLEDGMENT
Rebecca Kristeleit, MD, PhD, is supported in part by the
UCH/UCL Biomedical Research Centre and UCL Experimental
Cancer Medicine Centre.
References
1. Jayson GC, Kohn EC, Kitchener HC, et al. Ovarian cancer. Lancet. 2014;
384:1376-1388.
2. Dietlein F, Reinhardt HC. Molecular pathways: exploiting tumor-specific
molecular defects in DNA repair pathways for precision cancer therapy.
Clin Cancer Res. 2014;20:5882-5887.
3. Bryant HE, Schultz N, Thomas HD, et al. Specific killing of BRCA2-deficient
tumours with inhibitors of poly(ADP-ribose) polymerase. Nature. 2005;
434:913-917.
4. Farmer H, McCabe N, Lord CJ, et al. Targeting the DNA repair defect in
BRCA mutant cells as a therapeutic strategy. Nature. 2005;434:917-921.
5. Prakash R, Zhang Y, Feng W, et al. Homologous recombination and
human health: the roles of BRCA1, BRCA2, and associated proteins. Cold
Spring Harb Perspect Biol. 2015;7:a016600.
e266
6. Abkevich V, Timms KM, Hennessy BT, et al. Patterns of genomic loss

of heterozygosity predict homologous recombination repair defects in
epithelial ovarian cancer. Br J Cancer. 2012;107:1776-1782.
7. Konstantinopoulos PA, Spentzos D, Karlan BY, et al. Gene expression
profile of BRCAness that correlates with responsiveness to chemotherapy and with outcome in patients with epithelial ovarian cancer.
J Clin Oncol. 2010;28:3555-3561.
8. Walsh T, Casadei S, Lee MK, et al. Mutations in 12 genes for inherited
ovarian, fallopian tube, and peritoneal carcinoma identified by massively
parallel sequencing. Proc Natl Acad Sci USA. 2011;108:18032-18037.
9. Watkins JA, Irshad S, Grigoriadis A, et al. Genomic scars as biomarkers of
homologous recombination deficiency and drug response in breast and
ovarian cancers. Breast Cancer Res. 2014;16:211.
10. Do K, Doroshow JH, Kummar S. Wee1 kinase as a target for cancer

therapy. Cell Cycle. 2013;12:3159-3164.
11. Kobayashi H, Shigetomi H, Yoshimoto C. Checkpoint kinase 1 inhibitors
as targeted molecular agents for clear cell carcinoma of the ovary. Oncol
Lett. 2015;10:571-576.
12. Krajewska M, Fehrmann RS, Schoonen PM, et al. ATR inhibition preferentially targets homologous recombination-deficient tumor cells.
Oncogene. 2015;34:3474-3481.
13. Martin SA, Lord CJ, Ashworth A. Therapeutic targeting of the DNA
mismatch repair pathway. Clin Cancer Res. 2010;16:5107-5113.
14. Hoeijmakers JH. Genome maintenance mechanisms for preventing
cancer. Nature. 2001;411:366-374.
15. Bouwman P, Jonkers J. The effects of deregulated DNA damage signaling on cancer chemotherapy response and resistance. Nat Rev
Cancer. 2012;12:587-598.
16. Patel AG, Sarkaria JN, Kaufmann SH. Nonhomologous end joining drives
poly(ADP-ribose) polymerase (PARP) inhibitor lethality in homologous recombination-deficient cells. Proc Natl Acad Sci USA. 2011;108:
3406-3411.
17. Murai J, Huang SY, Das BB, et al. Trapping of PARP1 and PARP2 by clinical
PARP inhibitors. Cancer Res. 2012;72:5588-5599.
18. Tutt A, Ashworth A. The relationship between the roles of BRCA genes
in DNA repair and cancer predisposition. Trends Mol Med. 2002;8:
571-576.
19. Gudmundsdottir K, Ashworth A. The roles of BRCA1 and BRCA2 and
associated proteins in the maintenance of genomic stability. Oncogene.
2006;25:5864-5874.
20. Cass I, Baldwin RL, Varkey T, et al. Improved survival in women with
BRCA-associated ovarian carcinoma. Cancer. 2003;97:2187-2195.
21. Tan DSP, Rothermundt C, Thomas K, et al. BRCAness syndrome in
ovarian cancer: a case-control study describing the clinical features and
outcome of patients with epithelial ovarian cancer associated with
BRCA1 and BRCA2 mutations. J Clin Oncol. 2008;26:5530-5536.
22. Fong PC, Boss DS, Yap TA, et al. Inhibition of poly(ADP-ribose) polymerase in tumors from BRCA mutation carriers. N Engl J Med. 2009;361:
123-134.
23. Fong PC, Yap TA, Boss DS, et al. Poly(ADP)-ribose polymerase inhibition:
frequent durable responses in BRCA carrier ovarian cancer correlating
with platinum-free interval. J Clin Oncol. 2010;28:2512-2519.
24. Ledermann J, Harter P, Gourley C, et al. Olaparib maintenance therapy
in platinum-sensitive relapsed ovarian cancer. N Engl J Med. 2012;366:
1382-1392.
25. Lord CJ, Tutt AN, Ashworth A. Synthetic lethality and cancer therapy:
lessons learned from the development of PARP inhibitors. Annu Rev
Med. 2015;66:455-470.
26. European Medicines Agency. 2015, EMA Approval of Olaparib. http://www.
ema.europa.eu/docs/en_GB/document_library/ EPAR_-_Summary_
for_the_public/human/003726/WC500180153.pdf. Accessed February 1,
2016.
27. Kim G, Ison G, McKee AE, et al. FDA approval summary: olaparib
monotherapy in patients with deleterious germline BRCA-mutated
advanced ovarian cancer treated with three or more lines of chemotherapy. Clin Cancer Res. 2015;21:4257-4261.
28. Gelmon KA, Tischkowitz M, Mackay H, et al. Olaparib in patients with
recurrent high-grade serous or poorly differentiated ovarian carcinoma
or triple-negative breast cancer: a phase 2, multicentre, open-label,
non-randomised study. Lancet Oncol. 2011;12:852-861.
in patients with platinum-sensitive relapsed serous ovarian cancer:
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
a preplanned retrospective analysis of outcomes by BRCA status in a

Loveday C, Turnbull C, Ramsay E, et al; Breast Cancer Susceptibility
Collaboration (UK). Germline mutations in RAD51D confer susceptibility
to ovarian cancer. Nat Genet. 2011;43:879-882.
Meindl A, Hellebrand H, Wiek C, et al. Germline mutations in breast
and ovarian cancer pedigrees establish RAD51C as a human cancer
susceptibility gene. Nat Genet. 2010;42:410-414.
Rafnar T, Gudbjartsson DF, Sulem P, et al. Mutations in BRIP1 confer
high risk of ovarian cancer. Nat Genet. 2011;43:1104-1107.
Bajrami I, Frankum JR, Konde A, et al. Genome-wide profiling of genetic
synthetic lethality identifies CDK12 as a novel determinant of PARP1/2
inhibitor sensitivity. Cancer Res. 2014;74:287-297.
McCabe N, Turner NC, Lord CJ, et al. Deficiency in the repair of DNA
damage by homologous recombination and sensitivity to poly(ADPribose) polymerase inhibition. Cancer Res. 2006;66:8109-8115.
Pennington KP, Walsh T, Harrell MI, et al. Germline and somatic mutations in homologous recombination genes predict platinum response
and survival in ovarian, fallopian tube, and peritoneal carcinomas. Clin
Cancer Res. 2014;20:764-775.
Lord CJ, Ashworth A. BRCAness revisited. Nat Rev Cancer. 2016;16:
110-120.
Cancer Genome Atlas Research Network. Integrated genomic analyses
of ovarian carcinoma. Nature. 2011;474:609-615.
McNeish IA, Oza AM, Coleman RL, et al. Results of ARIEL2: a phase 2 trial
to prospectively identify ovarian cancer patients likely to respond to
rucaparib using tumor genetic analysis. J Clin Oncol. 2015;33s (suppl;
abstr 5508).
Birkbak NJ, Kochupurakkal B, Izarzugaza JM, et al. Tumor mutation
burden forecasts outcome in ovarian cancer with BRCA1 or BRCA2
mutations. PLoS One. 2013;8:e80023.
Strickland K, Howitt BE, Rodig SJ, et al. Tumor infiltrating and peritumoral T cells and expression of PD-L1 in BRCA1/2-mutated high grade
serous ovarian cancers. J Clin Oncol. 2015;33s (suppl: abstr 5512).
Birkbak NJ, Wang ZC, Kim JY, et al. Telomeric allelic imbalance indicates
defective DNA repair and sensitivity to DNA-damaging agents. Cancer
Discov. 2012;2:366-375.
Wang ZC, Birkbak NJ, Culhane AC, et al; Australian Ovarian Cancer
Study Group. Profiles of genomic instability in high-grade serous
ovarian cancer predict treatment outcome. Clin Cancer Res. 2012;18:
5806-5815.
Popova T, Manie E, Rieunier G, et al. Ploidy and large-scale genomic
instability consistently identify basal-like breast carcinomas with BRCA1/2
inactivation. Cancer Res. 2012;72:5454-5462.
Shen J, Peng Y, Wei L, et al. ARID1A deficiency impairs the DNA damage
checkpoint and sensitizes cells to PARP inhibitors. Cancer Discov. 2015;
5:752-767.
Jones S, Wang TL, Shih IeM, et al. Frequent mutations of chromatin
remodeling gene ARID1A in ovarian clear cell carcinoma. Science. 2010;
330:228-231.
Wiegand KC, Shah SP, Al-Agha OM, et al. ARID1A mutations in
endometriosis-associated ovarian carcinomas. N Engl J Med. 2010;363:
1532-1543.
Wiegand KC, Lee AF, Al-Agha OM, et al. Loss of BAF250a (ARID1A) is
frequent in high-grade endometrial carcinomas. J Pathol. 2011;224:
328-333.
Morgan MA, Parsels LA, Maybaum J, et al. Improving the efficacy of
chemoradiation with targeted agents. Cancer Discov. 2014;4:280-291.
Niida H, Nakanishi M. DNA damage checkpoints in mammals. Mutagenesis. 2006;21:3-9.
e267
52. Stathis A, Oza A. Targeting Wee1-like protein kinase to treat cancer.

Drug News Perspect. 2010;23:425-429.
53. Zhang Y, Hunter T. Roles of Chk1 in cell biology and cancer therapy. Int J
Cancer. 2014;134:1013-1023.
54. Etemadmoghadam D, Au-Yeung G, Wall M, et al. Resistance to CDK2
inhibitors is associated with selection of polyploid cells in CCNE1amplified ovarian cancer. Clin Cancer Res. 2013;19:5960-5971.
55. Etemadmoghadam D, Weir BA, Au-Yeung G, et al; Australian Ovarian
Cancer Study Group. Synthetic lethality between CCNE1 amplification
and loss of BRCA1. Proc Natl Acad Sci USA. 2013;110:19489-19494.
56. Karst AM, Jones PM, Vena N, et al. Cyclin E1 deregulation occurs early
in secretory cell transformation to promote formation of fallopian
tube-derived high-grade serous ovarian cancers. Cancer Res. 2014;74:
1141-1152.
57. Alagpulinsa DA, Ayyadevara S, Yaccoby S, et al. A cyclin-dependent
kinase inhibitor, dinaciclib, impairs homologous recombination and
sensitizes multiple myeloma cells to PARP inhibition. Mol Cancer Ther.
2016;15:241-250.
58. Liu JF, Barry WT, Birrer M, et al. Combination cediranib and olaparib
versus olaparib alone for women with recurrent platinum-sensitive
ovarian cancer: a randomised phase 2 study. Lancet Oncol. 2014;15:
1207-1214.
59. Guillotin D, Martin SA. Exploiting DNA mismatch repair deficiency as a
therapeutic strategy. Exp Cell Res. 2014;329:110-115.
60. Aarnio M, Mecklin JP, Aaltonen LA, et al. Life-time risk of different
cancers in hereditary non-polyposis colorectal cancer (HNPCC) syndrome. Int J Cancer. 1995;64:430-433.
61. Kandoth C, Schultz N, Cherniack AD, et al; Cancer Genome Atlas Research Network. Integrated genomic characterization of endometrial
carcinoma. Nature. 2013;497:67-73.
62. Martin SA, McCabe N, Mullarkey M, et al. DNA polymerases as potential
therapeutic targets for cancers deficient in the DNA mismatch repair
proteins MSH2 or MLH1. Cancer Cell. 2010;17:235-248.
63. Martin SA, McCarthy A, Barber LJ, et al. Methotrexate induces oxidative
DNA damage and is selectively lethal to tumour cells with defects in the
DNA mismatch repair gene MSH2. EMBO Mol Med. 2009;1:323-337.
64. Vilar E, Bartnik CM, Stenzel SL, et al. MRE11 deficiency increases
sensitivity to poly(ADP-ribose) polymerase inhibition in microsatellite
unstable colorectal cancers. Cancer Res. 2011;71:2632-2642.
65. Alexandrov LB, Nik-Zainal S, Wedge DC, et al; Australian Pancreatic
Cancer Genome Initiative; ICGC Breast Cancer Consortium; ICGC MMMLSeq Consortium; ICGC PedBrain. Signatures of mutational processes in
human cancer. Nature. 2013;500:415-421.
66. Howitt BE, Sholl LM, Ritterhouse L, et al. Association of POLE-mutated
and MSI endometrial cancers with an elevated number of tumorinfiltrating and peritumoral lymphocytes and higher expression of
PD-L1. J Clin Oncol. 2015;33s (suppl: abstr 5511).
67. Abdel-Fatah TM, Arora A, Moseley P, et al. ATM, ATR and DNA-PKcs
expressions correlate to adverse clinical outcomes in epithelial ovarian
cancers. BBA Clin. 2014;2:10-17.
68. Duensing S, Munger K. The human papillomavirus type 16 E6 and
E7 oncoproteins independently induce numerical and structural chromosome instability. Cancer Res. 2002;62:7075-7082.
69. Balacescu O, Balacescu L, Tudoran O, et al. Gene expression profiling
reveals activation of the FA/BRCA pathway in advanced squamous
e268
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
cervical cancer with intrinsic resistance and therapy failure. BMC

Cancer. 2014;14:246.
Wilson CR, Davidson SE, Margison GP, et al. Expression of Ku70 correlates with survival in carcinoma of the cervix. Br J Cancer. 2000;83:
1702-1706.
Candido-dos-Reis FJ, Song H, Goode EL, et al; for EMBRACE; kConFab
Investigators; Australian Ovarian Cancer Study Group. Germline
mutation in BRCA1 or BRCA2 and ten-year survival for women diagnosed with epithelial ovarian cancer. Clin Cancer Res. 2015;21:
652-657.
Jain RK. Normalization of tumor vasculature: an emerging concept in
antiangiogenic therapy. Science. 2005;307:58-62.
Kotsopoulos J, Rosen B, Fan I, et al. Ten-year survival after epithelial
ovarian cancer is not associated with BRCA mutation status. Gynecol
Oncol. 2016;140:42-47.
Norquist BM, Pennington KP, Agnew KJ, et al. Characteristics of women
with ovarian carcinoma who have BRCA1 and BRCA2 mutations not
identified by clinical testing. Gynecol Oncol. 2013;128:483-487.
Norquist BM, Harrell MI, Brady MF, et al. Inherited mutations in women
with ovarian carcinoma. JAMA Oncol. Epub 2015 Dec 30.
Edwards SL, Brough R, Lord CJ, et al. Resistance to therapy caused by
intragenic deletion in BRCA2. Nature. 2008;451:1111-1115.
Norquist B, Wurz KA, Pennil CC, et al. Secondary somatic mutations
restoring BRCA1/2 predict chemotherapy resistance in hereditary
ovarian carcinomas. J Clin Oncol. 2011;29:3008-3015.
Bunting SF, Callen E, Wong N, et al. 53BP1 inhibits homologous recombination in Brca1-deficient cells by blocking resection of DNA breaks.
Cell. 2010;141:243-254.
Pennington KP, Wickramanayake A, Norquist BM, et al. 53BP1 expression in sporadic and inherited ovarian carcinoma: relationship
to genetic status and clinical outcomes. Gynecol Oncol. 2013;128:
493-499.
Guleria A, Chandna S. ATM kinase: much more than a DNA damage
responsive protein. DNA Repair (Amst). Epub 2015 Dec 29.
Khanna A. DNA damage in cancer therapeutics: a boon or a curse?
Cancer Res. 2015;75:2133-2138.
Roos WP, Thomas AD, Kaina B. DNA damage and the balance between
survival and death in cancer biology. Nat Rev Cancer. 2016;16:20-33.
Minion LE, Dolinsky JS, Chase DM, et al. Hereditary predisposition to
ovarian cancer, looking beyond BRCA1/BRCA2. Gynecol Oncol. 2015;
137:86-92.
Cunningham JM, Cicek MS, Larson NB, et al. Clinical characteristics of
ovarian cancer classified by BRCA1, BRCA2, and RAD51C status. Sci Rep.
2014;4:4026.
Song H, Dicks E, Ramus SJ, et al. Contribution of germline mutations in
the RAD51B, RAD51C, and RAD51D genes to ovarian cancer in the
population. J Clin Oncol. 2015;33:2901-2907.
Kristeleit R, Swisher E, Oza A, et al. Final results of ARIEL2 (Part 1):
a phase 2 trial to prospectively identify ovarian cancer (OC) responders
to rucaparib using tumor genetic analysis. Eur J Cancer. 2015;51s (suppl;
S531).
Ramus SJ, Song H, Dicks E, et al; AOCS Study Group; Ovarian Cancer
Association Consortium. Germline mutations in the BRIP1, BARD1,
PALB2, and NBN genes in women with ovarian cancer. J Natl Cancer Inst.
2015;107:11.
GYNECOLOGIC CANCER
Intraperitoneal Chemotherapy for

Ovarian Cancer: Trials and
Tribulations
CHAIR
Helen J. Mackay, MD
Toronto, ON
SPEAKERS
Charlie Gourley, PhD, FRCPE
University of Edinburgh Cancer Research UK Centre
Edinburgh, United Kingdom
Joan L. Walker, MD
The University of Oklahoma Health Sciences Center
Oklahoma City, OK
INTRAPERITONEAL CHEMOTHERAPY IN THE TREATMENT OF OVARIAN CANCER
Update on Intraperitoneal Chemotherapy for the Treatment

of Epithelial Ovarian Cancer
Charlie Gourley, BSC, MBChB, PhD, FRCP, Joan L. Walker, MD, and Helen J. Mackay, BSc, MBCh, MD, MRCP
OVERVIEW
Surgical treatment and chemotherapy administration in women with epithelial ovarian cancer is more controversial today
than at any point in the last 3 decades. The use of chemotherapy administered intraperitoneally has been particularly
contentious. Three large randomized phase III studies, multiple meta-analyses, and now real-world data have demonstrated
substantial outcome benefit for the use of chemotherapy administered intraperitoneally versus intravenously for first-line
postoperative treatment of optimally debulked advanced ovarian cancer. Unfortunately, for each of these randomized
studies, there was scope to either criticize the design or otherwise refute adoption of this route of administration. As a result,
the uptake has been variable in North America, although in Europe it has been practically nonexistent. Reasons for this
include unquestionable additional toxicity, more inconvenience, and extra cost. However, 10-year follow up of these studies
demonstrates unprecedented survival in the intraperitoneal arm (median survival 110 months in patients with completely
debulked stage III), raising the possibility that by combining maximal debulking surgery with postoperative intraperitoneal
chemotherapy it may be possible to bring about a step change in the outcomes for these patients. In this review, we discuss
the rationale for administering chemotherapy intraperitoneally, the merits of the main randomized clinical trials, the
evidence regarding optimal regimes, issues of toxicity, port considerations, and reasons for lack of universal adoption. We
also explore potential clinical and biologic factors that may be useful for patient selection in the future.
e have witnessed improvements in epithelial ovarian

cancer survival over the last 3 decades without seeing
significant improvements in disease-specific mortality rates
(Fig. 1). Epithelial ovarian cancer remains the leading cause
of death from gynecologic malignancy in North America with
the majority of women presenting with stage III or IV disease.1,2 Advances in genetic testing, counseling, and prevention with risk-reducing salpingo-oophorectomy (and
salpingectomy) have the potential to produce further
modest decreases in ovarian cancer mortality in the future.3
However, we currently do not have a reliable population
screening test for epithelial ovarian cancer, and, given the
often nonspecific symptoms with which epithelial ovarian
cancer presents, it is likely the majority of patients will
continue to present with late-stage disease. Therefore,
optimizing treatment is critical if we are to improve outcome.
Retrospective analyses suggest that overall survival (OS) is
associated with younger age, good performance status,
lower stage of disease, and lower comorbidity scores.4,5
Although histologic subtype and tumor grade previously
were regarded as simply prognostic, it has now become clear
that they represent pathologic markers of what are essentially discrete disease entities. These differ in terms of their
tissue of origin, stage of presentation, driver molecular
mutations, sensitivity to chemotherapy, and prognosis
(Fig. 2). Ultimately, it is very likely that these histologic
subtypes will require different treatment strategies.6
In clinical practice, only a few risk factors remain that can
be modified based on the decisions of patients and their
physicians. The surgical decision making and chemotherapy
administration choices, particularly for women with stage III
or IV epithelial ovarian cancer, can potentially affect survival.
Informed selection of the best treatment (including clinical
trial options) for any individual patient is most likely to be
achieved with enthusiastically committed multidisciplinary
teams working in high-volume institutions.7,8 Discussions
around chemotherapy administered intraperitoneally for
individual patients require this type of environment to allow
the patient to make an informed choice about care.
This article summarizes the history and role of chemotherapy administered intraperitoneally in epithelial ovarian
cancer, focuses on the practical choices that patients and
From the Edinburgh Cancer Research Centre, Medical Research Council, Institute of Genetics and Molecular Medicine, University of Edinburgh, Edinburgh, United Kingdom; Stephenson
Cancer Center, University of Oklahoma, Health Sciences Center, Oklahoma City, OK; Faculty of Medicine, University of Toronto, Sunnybrook Odette Cancer Centre, Toronto, Canada.
Corresponding author: Helen J. Mackay, BSc, MBCh, MD, MRCP, University of Toronto, Sunnybrook Odette Cancer Centre, 2075 Bayview Ave., Toronto, ON M4N 3M5, Canada; email:
helen.mackay@sunnybrook.ca.
143
GOURLEY, WALKER, AND MACKAY
FIGURE 1. Increase in Overall Survival Over Time for

Women Diagnosed With Ovarian Cancer
clinicians face (with an emphasis on emerging data), and

stresses the importance of building skilled multidisciplinary
teams for treating women undergoing intraperitoneal
treatment. Finally, we explore how emerging data on epithelial ovarian cancer biology may be able to guide the
future of intraperitoneal therapy in this disease.
RATIONALE FOR INTRAPERITONEAL

CHEMOTHERAPY
The peritoneal cavity is the principle site of spread and
recurrence in women with epithelial ovarian cancer. Administration of chemotherapy intraperitoneally is a means of
increasing the dose intensity delivered to the tumor and
KEY POINTS
144
Patients with epithelial ovarian cancer should be treated

by high-volume multidisciplinary teams.
Randomized phase III trial, meta-analysis, and realworld data support the benefit of chemotherapy
administered intraperitoneally in the treatment of
select groups of women with epithelial ovarian cancer
following up-front optimal cytoreductive surgery.
The optimal intraperitoneal/intravenous chemotherapy
regimen has yet to be defined, and emerging data from
the randomized studies GOG 252, OV21/PETROC, and
JGOG iPocc will help clarify whether cisplatin
administered intraperitoneally can be replaced by
carboplatin administered intraperitoneally.
Discussion of chemotherapy administered
intraperitoneally to treat women diagnosed with
ovarian cancer is essential.
Understanding the biology underlying the success of
chemotherapy administered intraperitoneally is a
priority; initial data suggest exploration of tumors
deficient in DNA repair are of particular interest.
minimizes systemic toxicity, therefore, it is an attractive

therapeutic approach.9 Advantages of intraperitoneal administration include high intraperitoneal concentration of
the drug, as well as a longer half-life of the drug in the
peritoneal cavity, compared with that observed with administration intravenously alone. For cisplatin, historically,
the most extensively studied agent administered by the
intraperitoneal route in epithelial ovarian cancer translates
into a 10- to 20-fold greater exposure over that which is
achievable with the intravenously administrated route.10-12
Furthermore, preclinical studies suggested that cisplatin is
capable of penetrating small volume tumors (13 mm).
Hence, the hypothesis arose that the maximum benefit from
administration of the drug intraperitoneally was likely to be
demonstrated in patients with microscopic or low-volume
macroscopic disease.13 However, our understanding of why
the intraperitoneal chemotherapy route is more effective
may be oversimplified, and it has been challenged. Measurement of drug in peritoneal fluid probably does not
represent actual tumor drug penetration in patients.14 There
may be added barriers in tumor implants, notably disordered
capillary architecture, fibrosis, and adhesions. Furthermore,
dose intensification of platinum administered intravenously
has failed to show a benefit in multiple randomized studies.15,16 However, these dose intensification studies did not
select neither patients with high-grade serous ovarian
cancer nor the subgroups with disease that we know to be
most sensitive to platinum (those exhibiting homologous
recombination deficiency, discussed below). As such, the
impact in the dose-dense arms of these studies (which were
twice the density of the control arm at most) in any patients
with disease sensitive to this approach would be diluted out
by the majority of patients whose tumor biology would make
them unlikely to benefit from this approach. Although the
intraperitoneal studies were similarly unselected for
immunohistologic or molecular subtype, the local dose intensification in the intraperitoneal arm is potentially an
order of magnitude higher than in the intravenously administrated arm. Under these circumstances, it may be
possible that the signal would be evident even without
preselection on the basis of histology or biology.
Other factors such as drug recirculation following peritoneal absorption may play a role in efficacy. Despite the fact
that chemotherapy administered intraperitoneally has been
studied for decades, we are still not fully aware of the key
biologic factors that determine its success.15,16 Moving
forward, a greater understanding of how tumor biology and
the immune and micro-environments are affected by
treatment administered intraperitoneally may help us understand how this treatment can best be deployed and
combined with the newer generation of targeted agents.
INTRAPERITONEAL CHEMOTHERAPY TRIALS IN

WOMEN WITH EPITHELIAL OVARIAN CANCER
Intraperitoneal chemotherapy is not a recent concept in the
treatment of epithelial ovarian cancer; it was originally used
FIGURE 2. Schema for GOG 252
in the 1950s to control ascites. Interest in chemotherapy

administered intraperitoneally as a strategy for reducing
the risk of disease recurrence and prolonging survival
emerged approximately 30 years ago. This resulted in a
number of randomized phase III studies that demonstrated
an improvement in survival for the combination of delivery
of chemotherapy intraperitoneally and intravenously over
chemotherapy administrated intravenously alone for
select patients following primary cytoreductive surgery

(Table 1).17-23 GOG 172, a randomized phase III study of
cisplatin administered intraperitoneally combined with
both delivery of paclitaxel intraperitoneally and intravenously,
published in 2006, demonstrated a 16-month improvement
in median OS over intravenous administration of the same
drugs alone.18 This prompted the National Cancer Institute
(NCI) to issue a rare clinical announcement regarding the
clinical utility of cisplatin-based chemotherapy administered intraperitoneally in the treatment of patients with
small volume (, 1 cm), advanced-stage (stage III) epithelial
ovarian cancer following an attempt at maximal cytoreductive surgery. On average, intraperitoneal/intravenous
chemotherapy was associated with a 21.6% decrease in
risk of death (hazard ratio (HR) 0.78; 95% CI, 0.690.89,
the original clinical announcement can be viewed at ctep.
cancer.gov).
An update published in 2015 with a median follow-up
of 10.7 years showed that women who underwent
intraperitoneal/intravenous chemotherapy in GOG 172
continued to derive benefit with a median survival of
61.8 months (95% CI, 55.569.5 months) compared with
51.4 months (95% CI, 4658.2 months) for chemotherapy
administered intravenously alone.24 The recent Cochrane
Review, restricted to newly diagnosed patients receiving
treatment after primary cytoreductive surgery, accepted
data from eight randomized studies on 2,026 women and
TABLE 1. Summary of Randomized Clinical Trials of Intraperitoneal Chemotherapy for Up-Front Primary
Cytoreductive Surgery
Study/Reference
Control Regimen
Experimental Regimen
Eligible Patients
Kirmani et al20
Cisplatin 100 mg/m2 IV cyclophosphamide

600 mg/m2
Cisplatin 200 mg/m2 IP; etoposide 350 mg/m2 IP
Stage IIC-IV
Every 3 weeks x 6
Every 4 weeks x 6
Cisplatin 100 mg/m2 IV; cyclophosphamide

600 mg/m2 IV
Cisplatin 100 mg/m2 IP; cyclophosphamide

600 mg/m2 IV
Every 3 weeks x 6
Every 3 weeks x 6
Carboplatin 350 mg/m2 IV; cyclophosphamide

600 mg/m2 IV
Carboplatin 350 mg/m2 IP; cyclophosphamide

600 mg/ m2 IV
Every 3 weeks x 6
Every 3 weeks x 6

600 mg/m2 IV; epidoxorubicin 60 mg/m2 IV

600 mg/ m2 IV; epidoxorubicin 60 mg/m2 IV
Every 4 weeks x 6
Every 4 weeks x 6
Cisplatin 75 mg/m2 IV; paclitaxel

135 mg/m2 (24-hr) IV
Carboplatin (AUC 9) IV every 28 days x 2;

cisplatin 100 mg/m2 IP; paclitaxel
135 mg/m2 (24-hr) IV
Every 3 weeks x 6
Every 3 weeks x 6

50 mg/m2 IV; epidoroxorubin/
doxorubicin 50 mg/m2 IV

500 mg/m2 IV; epidoxorubicin/
doxorubicin 50 mg/m2 IV
Every 3 weeks x 6
Every 3 weeks x 6
Cisplatin 75 mg/m2 IV; paclitaxel

135 mg/m2 (24-hr) IV
Paclitaxel 135 mg/m2 (24-hr) IV; cisplatin

100 mg/m2 IP; paclitaxel 60 mg/m2 IP on day 8
Every 3 weeks x 6
Every 3 weeks x 6
SWOG 8501/
GOG 10417
Polyzos et al22
GONO19
GOG 114/
SWOG 922721
Yen et al23
GOG 17218
Stage III,
# 2 cm residual
No. of
Patients
62
546
Stage III
90
Stage II-IV,
, 2 cm residual
113
Stage III,
# 1 cm residual
462
Stage III,
# 1 cm residual
118
Stage III,
# 1 cm residual
415
Abbreviations: IP, intraperitoneally; IV, intravenously.
145
concluded that women experienced increased survival if

they received intraperitoneal/intravenous chemotherapy
(HR 0.81; 95% CI, 0.700.9) and that intraperitoneal/
intravenous chemotherapy also prolonged the diseasefree interval (five studies, 1,311 women; HR 0.78; 95% CI,
0.70.86),25 thus potentially affecting quality of life going
forward.
Wright et al recently reported data on the real-world
uptake of intraperitoneal/intravenous chemotherapy in a
prospective cohort of 823 women with stage III optimally
cytoreduced epithelial ovarian cancer treated in six National
Comprehensive Cancer Network (NCCN) institutions.8 Despite the trial-based evidence and although adoption of
intraperitoneal/intravenous chemotherapy increased between 2007 and 2008, it plateaued with fewer than 50% of
eligible patients receiving intraperitoneal/intravenous
treatment. They also observed marked variation in uptake between institutions from 4% to 67%, suggesting
underutilization of this effective treatment approach in
this subgroup of women.
WHY HAS INTRAPERITONEAL/INTRAVENOUS

CHEMOTHERAPY NOT BEEN UNIVERSALLY
ADOPTED?
Despite a proven survival benefit, clearly intraperitoneal/
intravenous chemotherapy has not been universally adopted for the treatment of epithelial ovarian cancer. The
reasons behind this are numerous. The publication of each
positive intraperitoneal/intravenous randomized study has
been met with considerable debate over the interpretation
of the trial data.26 Arguments around the validity of the
conclusions drawn from GOG 172 have included statistical
analysis queries (intention-to-treat analysis, number of
patients lost to follow-up), and questions over second-line
treatment and scheduling effects.26 The most relevant
criticism of the pivotal intraperitoneal studies is the inequality of dose intensity between the treatment arms in
both GOG11421 and GOG17218 (Table 1). Given the fact that
paclitaxel administered intravenously weekly has been
shown to be superior to paclitaxel administered intravenously tri-weekly in one large randomized phase III
study,27 it could be argued that the administration of
an additional dose of paclitaxel on day 8 in the Armstrong
study was solely responsible for the benefit demonstrated.
However, a second randomized phase III study of weekly
versus triweekly paclitaxel administered intravenously did
not show superiority.28 In addition, the Alberts (GOG104)
study was a clean comparison of the same doses of cisplatin
administered intravenously or intraperitoneally, which
demonstrated substantial progression-free and OS advantages,17 but it was overlooked in many areas of the world
because paclitaxel came to prominence and the comparator
arm in GOG104 was regarded as outdated. It does, however,
serve as a useful proof of principle regarding the advantages
that can be attributed purely to the route of administration.
In addition, when the two studies with unequal dose
146
intensities between the arms (GOG 114 and GOG 172)

were excluded from the analysis of a Cochrane systematic
review in 2011, the survival benefit remained in favor of
the intraperitoneal route of administration.25 In the realworld analysis by Wright et al, 43% of patients received
modified intraperitoneal/intravenous regimens over time
(i.e., regimens differing from trial-specified protocols). Despite these modifications, women receiving intraperitoneal/
intravenous chemotherapy continued to derive benefit over
those who only received treatment intravenously (3-year OS
81% vs. 71%; HR 0.68; 95% CI, 0.470.99).8
Undoubtedly, delivery of intraperitoneal/intravenous
chemotherapy requires increased resources in terms of
both space and time to deliver compared with therapy
administrated intravenously. Because of placement of
catheters and regional delivery of drug, intraperitoneal/
intravenous chemotherapy is potentially associated with
greater toxicity, including catheter-related complications,
gastrointestinal toxicity, pain, and infection. These have
been reported across trials and supported in meta-analyses.25,29 However, real-world reports suggest many of
these potential issues can be overcome with time and development of expertise.30,31 Furthermore, randomized data
and the NCI alert support the use of cisplatin administered
intraperitoneally.17-21,23 Cisplatin is known to be more toxic
than the standard of care carboplatin, which is administered
intravenously for epithelial ovarian cancer.32 To date, we do
not have randomized trial data to determine whether carboplatin administered intraperitoneally is equivalent to
cisplatin administered intraperitoneally in terms of its impact on survival. However, preclinical and some clinical data
suggest that it might be equivalent in efficacy and less
toxic than cisplatin administered intraperitoneally.33 Results
from GOG 252 (NCT00951496) and OV21/PETROC (NCT00993655)
are awaited. These trials include direct comparison of
regimens, including cisplatin and carboplatin administered
intraperitoneally.
As yet, we have not arrived at the optimal intraperitoneal/
intravenous chemotherapy regimen, which balances efficacy
with toxicity and quality of life. Finally, effective and safe
delivery of intraperitoneal/intravenous chemotherapy requires the multidisciplinary expertise of a skilled team, which
simply may not be available in smaller-volume centers.
Therefore, consideration for referral and management in
high-volume centers is appropriate. Even with all the concerns that exist, it is clear that women who are appropriate
for intraperitoneal/intravenous chemotherapy should be at
least offered this as an option, and clinicians should be
considering how best to make it available to them.31
SURGICAL CONSIDERATIONS FOR

INTRAPERITONEAL CHEMOTHERAPY
The goal of ovarian cancer surgery whenever it is performed
is no gross residual disease or R0, as defined by Chi and
Bristow.7,34 This has been shown to result in improved
progression-free survival (PFS) and OS. Intraoperative
treatment decisions that have been demonstrated to influence patient survival and maximize surgical effort must
include a willingness to perform diaphragm resection,
splenectomy, bowel resections, and thorough peritoneal
and retroperitoneal resections of tumor. This requires experience and potentially a team of surgeons.35 Although an
R0 resection improves OS, it remains somewhat unclear
whether microscopic versus visible disease has an impact on
intraperitoneal chemotherapy effectiveness, as the allowable residual volume at the end of surgery differed in
the randomized trials (, 2 or , 1 cm). Many have made
the assumption that patients with microscopic disease
would derive the greatest benefit from intraperitoneal/
intravenous chemotherapy. However, in a subgroup analysis
of the GOG172 patient population, the 64% of women in
GOG172 who had macroscopic (gross) residual disease less
than or equal to 1 cm (which was the upper limit allowed by
study eligibility) had a significant improvement in OS (HR
0.75; 95% CI, 0.620.92).24 Further exploration of the effect
on larger-volume residual disease should emerge from GOG0252 and the Japanese iPocc trial (NCT01506856), both of
which enrolled a proportion of patients with larger-volume
residual disease.
Landrum looked into the effects of lymphadenectomy and
nodal metastasis on the benefit of administering chemotherapy intraperitoneally using data from both GOG 114 and
GOG 172. In these studies, despite undergoing cytoreductive
surgery to less than 1 cm, only 59% of women had lymph
nodes sampled or excised. Of the 254 women who had lymph
node evaluation, intraperitoneal benefit in terms of PFS and
OS was independent of nodal status. This suggests that
chemotherapy administered intraperitoneally may be equally
effective for patients with both intraperitoneal and retroperitoneal disease. Interestingly, the patients without lymphadenectomy did worse than patients with metastatic
tumor removed from their lymph nodes, although the decision not to perform the lymphadenectomy may have been
secondary to some poor prognostic factor perceived by the
surgeon. An important long-term quality of life finding by
Landrum was a decrease in recurrence in the abdominal cavity
after intraperitoneal chemotherapy. Thus intraperitoneal/
intravenous treatment may spare patients from suffering
from ascites and an inability to eat when they recur.4
A key and controversial area around surgical decision
making in epithelial ovarian cancer is the choice of neoadjuvant chemotherapy followed by a definitive cytoreductive surgical attempt versus primary cytoreductive
surgery. Initially, neoadjuvant chemotherapy was considered only for those women who were medically unfit for
aggressive surgery or for women with a high tumor burden
(especially those with stage IV disease).36,37 In recent years,
the use of neoadjuvant chemotherapy has gained in popularity. This followed the publication of two studies: the
European Organization for Research and Treatment of
Cancer Gynecologic Cancer Group (EORTC GCG) 5597138 and
the CHORUS study,39 which demonstrated equivalence in
outcome with some reduction in morbidity for patients
receiving neoadjuvant treatment. Neoadjuvant chemotherapy is usually platinum-based and administered intravenously; there is no role for intraperitoneal/intravenous
therapy in the preoperative patient. A recent study by Rosen
et al40 showed that the long-term survival for patients
undergoing primary cytoreduction was far superior to those
receiving neoadjuvant chemotherapy and delayed primary
surgery (9% vs. 41%, p , .0001). Although selection bias may
account for some of this difference, the percentage of longterm survivors is strikingly low in neoadjuvant chemotherapy studies.41 Patients undergoing optimal cytoreductive
surgery following administration of neoadjuvant chemotherapy were not included in the previous intraperitoneal/
intravenous randomized trials. Theoretically, they may
derive a similar level of benefit to women undergoing
up-front cytoreductive surgery from intraperitoneal/
intravenous chemotherapy delivery. The combination of
intraperitoneal/intravenous chemotherapy following neoadjuvant chemotherapy and optimal cytoreductive surgery
is being studied in the Gynecologic Cancer Intergroup study
OV21/PETROC. Women who had initial (clinical/imaging)
stage IIB-IV (intravenously based on the presence of pleural effusion alone) epithelial ovarian cancer and who had
received three or four cycles of platinum-based neoadjuvant
chemotherapy before definitive optimal cytoreductive
surgery (# 1 cm residual disease) were eligible for this trial.
Women were enrolled either intra- or postoperatively. This
study has now closed and the primary analysis is expected
shortly.
INTRAPERITONEAL PORT PLACEMENT

Essential to the delivery of chemotherapy administered intraperitoneally is the placement of an intraperitoneal port.
Consideration of port placement should occur at the time of
surgery when the port can be placed under direct visualization by the surgeon. This is the most time-efficient approach and prevents delays in the initiation of chemotherapy
postoperatively. Patients should be treated as soon as they
have resumed a normal diet and bowel function and are
ambulatory at home, which should occur within 21 days
of the primary surgery. Patients electing to receive intraperitoneal chemotherapy who do not already have
peritoneal catheters can have devices implanted by interventional radiologists or surgeons familiar with laparoscopic
techniques using the right upper quadrant entry techniques.
Minilaparotomy in the right lower quadrant is also generally
successful if resection of the terminal ileum or right colon did
not occur. Careful review of the cytoreduction operative
report can improve outcomes with intraperitoneal port
placement to avoid complications.42,43
The preferred location for port placement is the right
lower costal grill on the midclavicular line, below the location
where the breast or the bra may be uncomfortable in
a standing position. This site allows for posterior rigid
support that facilitates access. Ports also are installed over
the left side or right lower quadrant for convenience.
147
Failures of intraperitoneal catheters can be corrected infrequently. Infected catheters should be removed and not
replaced. Blocked catheters can be replaced if the patient
has free intraperitoneal space remaining between bowel
loops and has not had peritonitis as a complication of having
undergone surgery or having received chemotherapy. Access problems attributable to a rotated port can be easily
corrected. Most patients, however, resume their chemotherapy with treatment administered intravenously alone
when catheter complications occur.
INTRAPERITONEAL CHEMOTHERAPY: WHICH

REGIMEN?
The 2013 update of the NCCN Guidelines suggested an intraperitoneal regimen based on the experimental arm of GOG
172: cisplatin administered intraperitoneally 75 to 100 mg/m2
and a 3-hour infusion of paclitaxel intravenously on day 1 with
60 mg/mg2 delivered intraperitoneally on day 8. The reduced
dose of cisplatin administered intraperitoneally (75 mg/m2)
was included because of concerns over the toxicity of 100 mg/
m2 with patients who did not complete the planned six cycles
of intraperitoneal/intravenous therapy postsurgery. This
dosing is also reflected in the experimental arms of both GOG
252 and OV21/PETROC.44
Given the toxicity associated with cisplatin and the
emerging nonrandomized data on carboplatin administered
intraperitoneally, some centers have adopted intraperitoneal
carboplatin-based regimens.45,46 Randomized data to better
inform the selection of an intraperitoneal/intravenous regimen will be available soon from GOG 252 and OV21/PETROC.
Given that GOG 172 demonstrated an OS benefit in an
intention-to-treat analysis when the median number of
cycles delivered was three (although the studies to date have
looked at six cycles of chemotherapy), a question remains
as to whether six is the optimal number of intraperitoneal
treatments.18
When discussing chemotherapy options for patients
whose disease is optimally cytoreduced, clinicians face the
challenge of discussing three basic choices: (1) carboplatin
and paclitaxel administered intravenously every 3 weeks
(GOG 158, GOG 182), (2) carboplatin administered intravenously every 3 weeks with paclitaxel administered
intravenously weekly (JGOG-3016, GOG 262), or (3) an
intraperitoneal/intravenous regimen.32,42,47,48 The subset of
women with stage III optimally cytoreduced disease in ICON
7 did not derive benefit from the addition of bevacizumab to
carboplatin and paclitaxel administered intravenously every
3 weeks.49 Data from OV21/PETROC, GOG 252, and JGOG
iPocc (NCT01506856) studies are eagerly awaited because
they will provide information on the comparison of a dosedense intravenous regimen with intraperitoneal/intravenous
chemotherapy. This will address the question of dose density
that remains from the GOG 172 data. The role of bevacizumab
in combination with intraperitoneal/intravenous chemotherapy is being explored in GOG 252. The tolerability of the
regimen was established in a previous phase II trial, although
148
three bowel obstructions (7%) were observed.50 Further

safety data on the combination of bevacizumab and
intraperitoneal/intravenous therapy will be available from
GOG 252. GOG 262 is a study that compared carboplatin/
paclitaxel administered intravenously every 3 weeks with
carboplatin administered intravenously on day 1 with
weekly dosing of paclitaxel intravenously. Bevacizumab
was allowed as an (nonrandomized) option in both arms of
the trial. This study demonstrated that dose-dense paclitaxel improve PFS over the dosing of paclitaxel every
3 weeks, but this was only true in women not receiving
bevacizumab. The addition of bevacizumab appeared to
eliminate the beneficial effects of dose-dense paclitaxel.47 In GOG 252, bevacizumab is included on cycles
222, and the study was designed with the assumption
that there would be no interaction between the various
chemotherapy arms and bevacizumab, which would hide
the effects of the individual chemotherapy regimens. The
surprising results of GOG 262 raise concern that the addition of bevacizumab to all patients on GOG 252 may
have obscured the effects of the individual chemotherapy
regimens alone. Insufficient data exist at this time to recommend bevacizumab in combination with intraperitoneal/
intravenous chemotherapy as a standard-of-care option.
Results for GOG 252 are expected to be reported at the
Society of Gynecologic Oncology meeting in March 2016.
Ultimately, the factors that patients and families should be
empowered to consider when making decisions about adjuvant treatment will include convenience, potential toxicities, and quality of life in addition to the data on efficacy.
PATIENT SELECTION FOR INTRAPERITONEAL

CHEMOTHERAPY
The last consideration is whether there are individual
patient findings that should be factored into decisions
regarding whether chemotherapy should be administered
intraperitoneally/intravenously or just intravenously to patients. It is now clear that at the immunohistochemic level,
epithelial ovarian cancer consists of at least five different
diseases.51-53 Although previous randomized studies of
intraperitoneal/intravenous chemotherapy versus intravenous chemotherapy did not select or stratify on the basis
of histologic subtype, it is now apparent that subtypes such as
low-grade serous, low-grade mucinous, and clear cell are less
sensitive to chemotherapy than high-grade serous or highgrade endometrioid ovarian cancer. The low likelihood of a
considerable chemotherapy dose-response relationship in
these less-sensitive subtypes makes it difficult to justify
subjecting these patients to the additional toxicity of intraperitoneal chemotherapy. Rather, the main subgroup of
patients who should be considered for intraperitoneal administration of chemotherapy are those with high-grade
serous ovarian cancer (patients with high-grade endometrioid cancer are an under-researched subgroup), but what
little evidence there is suggests they could be considered
similar to high-grade serous from a biologic perspective).
Currently, given the strong evidence of benefit from GOG

11421 and GOG17218 studies, there is insufficient evidence to
support any further patient selection based on biologic
subtype in patients with optimally cytoreduced disease who
have good performance status. However, the finding that
patients from the GOG 172 study with low-tumor BRCA1
expression (as assessed by immunohistochemistry) appeared
to benefit more from intraperitoneal/intravenous chemotherapy than intravenous chemotherapy (median survival 84 compared with 48 months, p = .0002) than those
with normal BRCA1 expression (median survival 58 months
for intraperitoneal chemotherapy compared with 50 months
for intravenous chemotherapy) suggests that there are
molecular subgroups that may benefit more from the intraperitoneal route of delivery.54 The main molecular
characteristics underlying high-grade serous ovarian cancer have recently been uncovered (Fig. 3).51,55 Given the
strong preclinical and clinical data supporting a high level of
platinum sensitivity in BRCA1- and BRCA2-deficient ovarian
cancer cells, it is not surprising that patients with low
BRCA1 protein expression may derive particular benefit
from administration of chemotherapy intraperitoneally.
The question is whether this benefit applies only to tumors
that carry germline or somatic BRCA1 mutations, or whether
those patients with epigenetic BRCA1 inactivation, BRCA2
mutations, PTEN loss, EMSY amplification, or mutations in
other homologous recombination deficiency genes also derive benefit. Although BRCA1 immunohistochemistry analysis
is notoriously inaccurate, the technology now exists to perform this other sequencing and copy number characterization
on archival formalin-fixed, paraffin-embedded material. This
retrospective analysis should be performed as a matter of
urgency using material from randomized studies of intraperitoneal chemotherapy to demonstrate whether particular molecular subgroups have more to gain from
intraperitoneal treatment. It is already clear that some
molecular subgroups have a high incidence of primary
platinum resistance (Fig. 3; homologous recombination
proficient),55,56 and these patients may not benefit from
intraperitoneal therapy; instead, perhaps they should be
considered for dose-dense paclitaxel and carboplatin administered intravenously, or trials of other novel therapies
in combination with their first-line chemotherapy, to optimize survival and avoid unnecessary toxicity.
Although cure is a word that most ovarian cancer oncologists try to avoid, this must be our aim. The survival that
has been demonstrated in patients with completely
cytoreduced disease and who received intraperitoneal
chemotherapy in the GOG172 study is a sea change in terms
of what we expect in this disease. It reinforces the argument
that in fit patients, maximal cytoreductive surgery followed
by intraperitoneal chemotherapy is the treatment likely to
produce the best possible outcome. However, the morbidity
induced by both these treatments makes it even more imperative that we are able to define whether it is only patients
with a particular biology (e.g., BRCA mutations or homologous recombination deficiency) that benefit. For these patients, full-on surgical/intraperitoneal chemotherapy with
or without PARP maintenance inhibitors (depending on the
outcome of currently running first-line PARP inhibitor
studies) could be adopted. For the patients with different
biology, alternative tailored approaches could be sought.
CONCLUSION
FIGURE 3. Molecular Subgroups of High-Grade Serous
Epithelial Ovarian Cancer
Abbreviations: HRD, homologous recombination deficient.
Large randomized phase III studies have been ubiquitously

positive, producing some of the best survival data ever seen
in the treatment of ovarian cancer. However, all of these
studies have had issues, either with the intravenous chemotherapy (control) arm being perceived as no longer contemporary or the intraperitoneal (test) arm having higher
dose intensity than the control arm and, therefore, not
being a trial solely of route of administration. These apparent
shortcomings combined with undoubted toxicity of the approach have led to inconsistent uptake in North America and
Australia and negligible uptake in Europe. This seems a shame
given that the benefits in terms of outcome appear so
marked. Indeed, the benefits are well in excess of those
demonstrated in the first-line bevacizumab studies that led to
licensing and reimbursement of this agent across Europe.
Further work is required to improve the toxicity from
chemotherapy administered intraperitoneally and to identify the best regimen and determine the extent to which the
benefits witnessed also can be produced by dose-dense
approaches or the use of targeted agents. These latter
two questions may be answered to some extent by the
GOG252 study. Perhaps the most important question that
requires to be answered is exactly which molecular subgroups of ovarian cancer patients benefit most from
149
intraperitoneal treatment. If this can be established, these

patients may benefit greatly from an aggressive surgical and
intraperitoneal chemotherapy approach. The question is,

will this be enough to change practice?
References
1. National Cancer Institute. SEER Stat Fact Sheets: Ovary Cancer. http://
seer.cancer.gov/statfacts/html/ovary.html. Accessed February 5, 2016.
2. Cannistra SA. Cancer of the ovary. N Engl J Med. 2004;351:2519-2529.
3. McCann GA, Eisenhauer EL. Hereditary cancer syndromes with high risk
of endometrial and ovarian cancer: surgical options for personalized
care. J Surg Oncol. 2015;111:118-124.
4. Landrum LM, Java J, Mathews CA, et al. Prognostic factors for stage III
epithelial ovarian cancer treated with intraperitoneal chemotherapy:
a Gynecologic Oncology Group study. Gynecol Oncol. 2013;130:12-18.
5. Winter WE III, Maxwell GL, Tian C, et al; Gynecologic Oncology Group
Study. Prognostic factors for stage III epithelial ovarian cancer: a Gynecologic Oncology Group Study. J Clin Oncol. 2007;25:3621-3627.
6. Banerjee S, Kaye SB. New strategies in the treatment of ovarian cancer:
current clinical perspectives and future potential. Clin Cancer Res. 2013;
19:961-968.
7. Bristow RE, Tomacruz RS, Armstrong DK, et al. Survival effect of maximal
cytoreductive surgery for advanced ovarian carcinoma during the
platinum era: a meta-analysis. J Clin Oncol. 2002;20:1248-1259.
8. Wright AA, Cronin A, Milne DE, et al. Use and effectiveness of intraperitoneal chemotherapy for treatment of ovarian cancer. J Clin
Oncol. 2015;33:2841-2847.
9. Dedrick RL, Myers CE, Bungay PM, et al. Pharmacokinetic rationale for
peritoneal drug administration in the treatment of ovarian cancer.
Cancer Treat Rep. 1978;62:1-11.
10. Fujiwara K, Armstrong D, Morgan M, et al. Principles and practice of
intraperitoneal chemotherapy for ovarian cancer. Int J Gynecol Cancer.
2007;17:1-20.
11. Howell SB. Pharmacologic principles of intraperitoneal chemotherapy
for the treatment of ovarian cancer. Int J Gynecol Cancer. 2008;18(suppl
1):20-25.
12. Markman M. Intraperitoneal chemotherapy. Semin Oncol. 1991;18:
248-254.
13. Markman M. Intraperitoneal drug delivery of antineoplastics. Drugs.
2001;61:1057-1065.
14. Royer B, Jullien V, Guardiola E, et al. Population pharmacokinetics and
dosing recommendations for cisplatin during intraperitoneal peroperative administration: development of a limited sampling strategy
for toxicity risk assessment. Clin Pharmacokinet. 2009;48:169-180.
15. Bookman MA, Brady MF. Intraperitoneal chemotherapy: long-term
outcomes revive a long-running debate. J Clin Oncol. 2015;33:
1424-1426.
16. Kaye SB, Paul J, Cassidy J, et al; Scottish Gynecology Cancer Trials Group.
Mature results of a randomized trial of two doses of cisplatin for the
treatment of ovarian cancer. J Clin Oncol. 1996;14:2113-2119.
17. Alberts DS, Liu PY, Hannigan EV, et al. Intraperitoneal cisplatin plus
intravenous cyclophosphamide versus intravenous cisplatin plus intravenous cyclophosphamide for stage III ovarian cancer. N Engl J Med.
1996;335:1950-1955.
18. Armstrong DK, Bundy B, Wenzel L, et al; Gynecologic Oncology Group.
Intraperitoneal cisplatin and paclitaxel in ovarian cancer. N Engl J Med.
2006;354:34-43.
19. Gadducci A, Carnino F, Chiara S, et al. Intraperitoneal versus intravenous
cisplatin in combination with intravenous cyclophosphamide and
epidoxorubicin in optimally cytoreduced advanced epithelial ovarian
150
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
cancer: a randomized trial of the Gruppo Oncologico Nord-Ovest.

Gynecol Oncol. 2000;76:157-162.
Kirmani S, Braly PS, McClay EF, et al. A comparison of intravenous versus
intraperitoneal chemotherapy for the initial treatment of ovarian
cancer. Gynecol Oncol. 1994;54:338-344.
Markman M, Bundy BN, Alberts DS, et al. Phase III trial of standard-dose
intravenous cisplatin plus paclitaxel versus moderately high-dose carboplatin followed by intravenous paclitaxel and intraperitoneal cisplatin
in small-volume stage III ovarian carcinoma: an intergroup study of the
Gynecologic Oncology Group, Southwestern Oncology Group, and
Eastern Cooperative Oncology Group. J Clin Oncol. 2001;19:1001-1007.
Polyzos A, Tsavaris N, Kosmas C, et al. A comparative study of intraperitoneal carboplatin versus intravenous carboplatin with intravenous cyclophosphamide in both arms as initial chemotherapy for
stage III ovarian cancer. Oncology. 1999;56:291-296.
Yen MS, Juang CM, Lai CR, et al. Intraperitoneal cisplatin-based chemotherapy vs. intravenous cisplatin-based chemotherapy for stage III
optimally cytoreduced epithelial ovarian cancer. Int J Gynaecol Obstet.
2001;72:55-60.
Tewari D, Java JJ, Salani R, et al. Long-term survival advantage and
prognostic factors associated with intraperitoneal chemotherapy
treatment in advanced ovarian cancer: a Gynecologic Oncology Group
Jaaback K, Johnson N, Lawrie TA. Intraperitoneal chemotherapy for the
initial management of primary epithelial ovarian cancer. Cochrane
Database Syst Rev. 2016;1:CD005340.
Gore M, du Bois A, Vergote I. Intraperitoneal chemotherapy in ovarian
cancer remains experimental. J Clin Oncol. 2006;24:4528-4530.
Katsumata N, Yasuda M, Takahashi F, et al; Japanese Gynecologic
Oncology Group. Dose-dense paclitaxel once a week in combination
with carboplatin every 3 weeks for advanced ovarian cancer: a phase 3,
open-label, randomised controlled trial. Lancet. 2009;374:1331-1338.
Pignata S, Scambia G, Katsaros D, et al; Multicentre Italian Trials in
Ovarian cancer (MITO-7); Groupe dInvestigateurs Nationaux pour
lEtude des Cancers Ovariens et du sein (GINECO); Mario Negri Gynecologic Oncology (MaNGO); European Network of Gynaecological
Oncological Trial Groups (ENGOT-OV-10); Gynecologic Cancer InterGroup (GCIG) Investigators. Carboplatin plus paclitaxel once a week
versus every 3 weeks in patients with advanced ovarian cancer (MITO7): a randomised, multicentre, open-label, phase 3 trial. Lancet Oncol.
2014;15:396-405.
Elit L, Oliver TK, Covens A, et al. Intraperitoneal chemotherapy in the
first-line treatment of women with stage III epithelial ovarian cancer:
a systematic review with metaanalyses. Cancer. 2007;109:692-702.
Chin SN, Pinto V, Rosen B, et al. Evaluation of an intraperitoneal
chemotherapy program implemented at the Princess Margaret Hospital
for patients with epithelial ovarian carcinoma. Gynecol Oncol. 2009;
112:450-454.
Markman M. Chemotherapy: Limited use of the intraperitoneal route
for ovarian cancer-why? Nat Rev Clin Oncol. 2015;12:628-630.
Ozols RF, Bundy BN, Greer BE, et al; Gynecologic Oncology Group. Phase
III trial of carboplatin and paclitaxel compared with cisplatin and
paclitaxel in patients with optimally resected stage III ovarian cancer:
a Gynecologic Oncology Group study. J Clin Oncol. 2003;21:3194-3200.
33. Jandial DD, Messer K, Farshchi-Heydari S, et al. Tumor platinum concentration following intraperitoneal administration of cisplatin versus
carboplatin in an ovarian cancer model. Gynecol Oncol. 2009;115:
362-366.
34. Chi DS, Eisenhauer EL, Zivanovic O, et al. Improved progression-free and
overall survival in advanced ovarian cancer as a result of a change in
surgical paradigm. Gynecol Oncol. 2009;114:26-31.
35. Aletti GD, Dowdy SC, Gostout BS, et al. Aggressive surgical effort and
improved survival in advanced-stage ovarian cancer. Obstet Gynecol.
2006;107:77-85.
36. Ansquer Y, Leblanc E, Clough K, et al. Neoadjuvant chemotherapy for
unresectable ovarian carcinoma: a French multicenter study. Cancer.
2001;91:2329-2334.
37. Schwartz PE, Rutherford TJ, Chambers JT, et al. Neoadjuvant chemotherapy for advanced ovarian cancer: long-term survival. Gynecol Oncol.
1999;72:93-99.
38. Vergote I, Trope CG, Amant F, et al; European Organization for Research
and Treatment of Cancer-Gynaecological Cancer Group; NCIC Clinical
Trials Group. Neoadjuvant chemotherapy or primary surgery in stage
IIIC or IV ovarian cancer. N Engl J Med. 2010;363:943-953.
39. Kehoe S, Hook J, Nankivell M, et al. Primary chemotherapy versus
primary surgery for newly diagnosed advanced ovarian cancer (CHORUS): an open-label, randomised, controlled, non-inferiority trial.
Lancet. 2015;386:249-257.
40. Rosen B, Laframboise S, Ferguson S, et al. The impacts of neoadjuvant
chemotherapy and of debulking surgery on survival from advanced
ovarian cancer. Gynecol Oncol. 2014;134:462-467.
41. Colombo PE, Labaki M, Fabbro M, et al. Impact of neoadjuvant chemotherapy cycles prior to interval surgery in patients with advanced
epithelial ovarian cancer. Gynecol Oncol. 2014;135:223-230.
42. Bookman MA, Brady MF, McGuire WP, et al. Evaluation of new
platinum-based treatment regimens in advanced-stage ovarian cancer:
a phase III trial of the Gynecologic Cancer Intergroup. J Clin Oncol. 2009;
27:1419-1425.
43. Walker JL, Armstrong DK, Huang HQ, et al. Intraperitoneal catheter
outcomes in a phase III trial of intravenous versus intraperitoneal
chemotherapy in optimal stage III ovarian and primary peritoneal
cancer: a Gynecologic Oncology Group Study. Gynecol Oncol. 2006;100:
27-32.
44. Morgan RJ Jr, Alvarez RD, Armstrong DK, et al; National Comprehensive
Cancer Networks. Ovarian cancer, version 2.2013. J Natl Compr Canc
Netw. 2013;11:1199-1209.
45. Bouchard-Fortier G, Rosen B, Vyarvelska I, et al. A comparison of the
toxicity and tolerability of two intraperitoneal chemotherapy regimens
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
for advanced-stage epithelial ovarian cancer. Gynecol Oncol. 2016;140:

36-41.
Fujiwara K, Markman M, Morgan M, et al. Intraperitoneal carboplatinbased chemotherapy for epithelial ovarian cancer. Gynecol Oncol. 2005;
97:10-15.
Chan J, Brady M, Penson R, et al. Phase III trial of every-3-weeks
paclitaxel vs. dose dense weekly paclitaxel with carboplatin +/2
bevacizumab In epithelial ovarian, peritoneal, fallopian tube cancer:
GOG 262 (NCT01167712). Int J Gynecol Cancer. 2013;23:8(suppl 1, abstr
8).
Katsumata N, Yasuda M, Isonishi S, et al; Japanese Gynecologic Oncology Group. Long-term results of dose-dense paclitaxel and carboplatin versus conventional paclitaxel and carboplatin for treatment of
advanced epithelial ovarian, fallopian tube, or primary peritoneal
cancer (JGOG 3016): a randomised, controlled, open-label trial. Lancet
Oncol. 2013;14:1020-1026.
Oza AM, Cook AD, Pfisterer J, et al; ICON7 Trial Investigators. Standard
chemotherapy with or without bevacizumab for women with newly
diagnosed ovarian cancer (ICON7): overall survival results of a phase 3
randomised trial. Lancet Oncol. 2015;16:928-936.
Konner JA, Grabon DM, Gerst SR, et al. Phase II study of intraperitoneal
paclitaxel plus cisplatin and intravenous paclitaxel plus bevacizumab as
adjuvant treatment of optimal stage II/III epithelial ovarian cancer. J Clin
Oncol. 2011;29:4662-4668.
Bowtell DD, Bohm S, Ahmed AA, et al. Rethinking ovarian cancer II:
reducing mortality from high-grade serous ovarian cancer. Nat Rev
Cancer. 2015;15:668-679.
Kobel M, Kalloger SE, Boyd N, et al. Ovarian carcinoma subtypes are
different diseases: implications for biomarker studies. PLoS Med. 2008;
5:e232.
Vaughan S, Coward JI, Bast RC Jr, et al. Rethinking ovarian cancer:
recommendations for improving outcomes. Nat Rev Cancer. 2011;11:
719-725.
Lesnock JL, Darcy KM, Tian C, et al. BRCA1 expression and improved
survival in ovarian cancer patients treated with intraperitoneal cisplatin
and paclitaxel: a Gynecologic Oncology Group Study. Br J Cancer. 2013;
108:1231-1237.
Patch AM, Christie EL, Etemadmoghadam D, et al; Australian Ovarian
Cancer Study Group. Whole-genome characterization of chemoresistant ovarian cancer. Nature. 2015;521:489-494.
Etemadmoghadam D, George J, Cowin PA, et al; Australian Ovarian
Cancer Study Group. Amplicon-dependent CCNE1 expression is critical
for clonogenic survival after cisplatin treatment and is correlated with
20q11 gain in ovarian cancer. PLoS One. 2010;5:e15498.
151
GYNECOLOGIC CANCER
Neoadjuvant Chemotherapy:
Location, Location, Location
CHAIR
Alexandra Leary, MD, PhD
Gustave Roussy Cancer Center
Villejuif, France
SPEAKERS
Dennis Chi, MD
New York, NY
Sean Kehoe, MD
University of Birmingham
Birmingham, United Kingdom
KEEPING THE PEACE AND BALANCING THE BIOLOGY
Primary Surgery or Neoadjuvant Chemotherapy in Advanced

Ovarian Cancer: The Debate Continues
Alexandra Leary, MD, PhD, Renee Cowan, MD, MPH, Dennis Chi, MD, Sean Kehoe, MD, DCH, FRCOG, and
Matthew Nankivell, MSc
OVERVIEW
Primary debulking surgery (PDS) followed by platinum-based chemotherapy has been the cornerstone of treatment for
advanced ovarian cancer for decades. Primary debulking surgery has been repeatedly identified as one of the key factors in
improving survival in patients with advanced ovarian cancer, especially when minimal or no residual disease is left behind.
Achieving these results sometimes requires extensive abdominal and pelvic surgical procedures and consultation with other
surgical teams. Some clinicians who propose a primary chemotherapy approach reported an increased likelihood of leaving
no macroscopic disease after surgery and improved patient-reported outcomes and quality-of-life (QOL) measures. Given
the ongoing debate regarding the relative benefit of PDS versus neoadjuvant chemotherapy (NACT), tumor biology may aid
in patient selection for each approach. Neoadjuvant chemotherapy offers the opportunity for in vivo chemosensitivity
testing. Studies are needed to determine the best way to evaluate the impact of NACT in each individual patient with
advanced ovarian cancer. Indeed, the biggest utility of NACT may be in research, where this approach provides the opportunity for the investigation of predictive markers, mechanisms of resistance, and a forum to test novel therapies.
espite the fact that most patients with ovarian cancer

have stage III or IV disease, advances in surgical and
medical therapies over the last 4 decades have significantly
improved the 5-year and overall survival (OS) for women
diagnosed with ovarian cancer.1,2 What used to be a death
sentence can now be considered a chronic disease in many
cases.3 However, ovarian cancer is still the leading gynecologic malignancy cause of cancer-related deaths in highincome countries.4 Furthermore, there is much work to be
done in efforts to improve not only OS and progression-free
survival (PFS), but also QOL.
According to the National Comprehensive Cancer Network, the current primary treatment of advanced epithelial
ovarian cancer is optimal cytoreductive surgery followed by
six to eight cycles of dual-agent platinum- and taxanebased chemotherapy.5 Cytoreductive surgery (also called
debulking) was initially proposed by Meigs in 1934.6 Forty
years later, Griffiths showed that survival depended on the
maximum diameter of residual disease left after cytoreductive surgery.7 Primary debulking surgery has since been
repeatedly demonstrated as one of the key factors in improving survival and the cornerstone of ovarian cancer
treatment by many studies.8-10
There are some in the field of gynecologic oncology,

however, who propose that inherent tumor biology is more
important than surgical resection and question the value of
extensive cytoreductive surgeries as the first step of the
advanced ovarian cancer treatment plan. Within the last
decade, this issue has been widely debated. In 2008, Vergote
presented the data of the first randomized control phase III
trial comparing PDS versus NACT followed by interval
debulking surgery (IDS). The trial enrolled more than
600 women with bulky stage III or IV advanced ovarian
cancer and was conducted by the European Organisation
for Research and Treatment of CancerGynecologic Cancer
Group (EORTC-GCG) and the National Cancer Institute of
Canada (NCIC) Clinical Trials Group. They found no major
difference in the PFS and OS between the two treatment
arms and concluded that NACT with IDS was not inferior
and possibly safer than the current standard of care,
PDS.11,12 The recently published trial on primary chemotherapy versus primary surgery for newly diagnosed advanced ovarian cancer (CHORUS) supported the EORTC/NCIC
results. In the CHORUS trial, more than 550 women with
advanced ovarian cancer were randomly selected to receive
primary chemotherapy versus primary surgery, with similar
From the Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY; School of Cancer Sciences, University of Birmingham, Birmingham, United Kingdom; St. Peters
College, National Cancer Intelligence Network, Public Health England, National Health Service, Birmingham, United Kingdom; School of Cancer Sciences, University of Birmingham,
Birmingham, United Kingdom; Medical Research Council Clinical Trials Unit, University College London, London, United Kingdom; Gustave Roussy Cancer Centre, Translational Research
Laboratory, Gustave Roussy Cancer Centre, Villejuif, France.
Corresponding author: Alexandra Leary, MD, PhD, Gustave Roussy Cancer Center, 114 Rue Edouard Vaillant, Villejuif, France; email: leary.alexandra@gmail.com.
153
LEARY ET AL
conclusions that NACT/IDS yielded comparable survival

results and decreased surgical morbidity when compared
with PDS.13
THE ARGUMENT FOR PRIMARY DEBULKING

SURGERY
It Is Possible
It is well accepted that patients who undergo suboptimal
tumor cytoreductive surgeries are at risk for surgical morbidity without deriving the survival benefit of the debulking
procedure.14,15 However, when a patient is treated with the
appropriate surgery by an experienced surgeon or team of
surgeons at an experienced hospital, optimal and even complete gross resection (CGR) at the time of PDS is achievable
in a vast majority of cases. Reported optimal cytoreduction
rates in the literature range from 15% to greater than
85%.10,16 Many experts agree that this variation is not
caused by a difference in geography, case presentation,
and/or case selection, but is highly dependent on the surgical
training, expertise, commitment, and resources of the primary operating surgeon, team, and institution. Moreover,
greater than 40% of patients with advanced ovarian cancer
have bulky upper abdominal disease that is difficult to
resect, which is cited by many gynecologic oncologists as
the most common reason for a suboptimal outcome at the
time of PDS.17,18
To address, the widespread nature of advanced ovarian
cancer at presentation, two strategies have emerged. One
strategy, NACT for three or more cycles followed by the
cytoreductive procedure, IDS, followed by the completion of
up to six to eight total cycles of chemotherapy, has been
proposed as a viable method to decrease preoperative
disease burden, increase the rates of optimal cytoreduction,
and lower postoperative morbidity. The second strategy has
been to expand the surgical armamentarium of gynecologic
oncology surgeons to include extensive upper abdominal
KEY POINTS
154
When a patient is treated with the appropriate surgery

by an experienced surgeon or team of surgeons at an
experienced hospital, optimal and even complete gross
resection at the time of PDS is achievable in a vast
majority of cases.
With the median overall survival of 66 months for
patients with stage III, optimally debulked ovarian
cancer treated with intraperitoneal chemotherapy on
GOG 172 serves as the gold standard for quality care,
which all future studies should attempt to attain.
NACT may result in better patient-reported outcomes
and QOL measures with decreased morbidity when
compared with PDS.
The biggest utility of NACT may be in research, where
this approach provides the opportunity for the
investigation of predictive markers, mechanisms of
resistance, and a forum to test novel therapies.
surgical procedures as needed during their PDS. This approach also increases the rate of optimal cytoreduction,
without significantly increasing postoperative morbidity.
In 2009, Chi et al demonstrated a change in the surgical
paradigm at Memorial Sloan Kettering Cancer Center
(MSKCC) that incorporated splenectomy, distal pancreatectomy, cholecystectomy, liver wedge resection, dissection of the porta hepatis, and diaphragm resection/
peritonectomy when necessary to achieve optimal PDS.
This paradigm change led to an improvement in optimal
primary cytoreduction from 46% to 80%. The rate of CGR
also more than doubled.19 Multiple U.S. and international
studies have supported these findings, demonstrating not
only the feasibility, but also the safety, of this approach in
expert centers, with the morbidity, mortality, and time to
start of chemotherapy not being statistically different
from that of pelvic limited surgery.20-24 The complexity of
the surgery alone is not a predictor of PFS or OS. When
cancer in these patients is optimally debulked to less than
1 cm of residual disease, they are afforded the same
survival rates as patients who required only standard
pelvic surgery.25-27
Because the surgeries can be more complex, preoperative
preparation is crucial, sometimes requiring consultation
with general surgeons, surgical oncologists, or hepatobiliary
surgeons to assist in the event that disease encountered
at PDS is difficult or impossible for the gynecologic oncologist to remove. This can sometimes be difficult at medical
centers with less experienced practitioners. Recent analyses
have shown that optimal cytoreductive surgery rates are
higher among more experienced surgeons at higher-volume
hospitals, with gynecologic oncologists at expert centers
attaining optimal resection rates of greater than 75%.10,28,29
This is in stark contrast to the 41% to 42% optimal cytoreduction
rates reported in the PDS arms of the EORTC/NCIC and
CHORUS PDS versus NACT/IDS trials (Table 1).12,13 Moreover, the median operating times of 165 and 120 minutes,
respectively, in the PDS arms of the two studies seems to be
inadequate for a complete surgical evaluation and attempt
at achieving optimal or no gross residual disease status.12,13
These findings question whether the PDS attempted in
these two trials were what one would see in more experienced, expert centers. Indeed, it is the duty of the operating physician to give the patient the best chance
at achieving CGR, even if that means additional training, multispecialty consultation, or referral to another
surgeon.
It Decreases Chemoresistance
There are innate benefits to PDS. Large bulky tumors are
often necrotic and hypoxic. Their poor blood supply does not
often lend itself to maximal intravenous or intraperitoneal
chemotherapy delivery. Comparatively, smaller tumors are
well perfused and easier to target.30 At the time of diagnosis and the initiation of treatment, both chemotherapysensitive and chemotherapy-resistant cells are present in a
TABLE 1. Randomized Trials of PDS Versus NACT/IDS

EORTC/
NCIC PDS
EORTC/
NCIC NACT
CHORUS
PDS
CHORUS
NACT
Number of
Patients
361
357
255
219
Residual 1 cm
42%
81%
41%
73%
No Gross Residual
18%
45%
17%
39%
PFS (months)
12
12
11
12
OS (months)
29
30
23
24
FIGURE 1. MSKCC Primary Cytoreduction OS and CGR

Rates
Abbreviations: IDS, interval debulking surgery; EORTC, European Organization for

Research and Treatment of Cancer; NCIC, National Cancer Institute of Canada; PDS,
primary debulking surgery; NACT, neoadjuvant chemotherapy; PFS, progression-free
survival; OS, overall survival.
patient. Primary surgery decreases the tumor burdens of

both of these cell lines and decreases the quantity of cells
that can spontaneously mutate to drug-resistant phenotypes.31 Thus, surgical debulking aids in overcoming negative
tumor biology.32 Approximately 25% of patients have
platinum-resistant disease at the time of their first relapse,
and almost all patients who have disease recurrence will
eventually become chemotherapy-resistant.33 By initiating
chemical debulking first with the NACT approach, the tumor
cells have more time to build increased resistance. Breaking
up the chemotherapy by introducing IDS in the middle of the
six to eight cycles of chemotherapy may also have this effect.
It Yields Better Survival Outcomes

The median OS of 66 months in stage III ovarian cancer
patients who underwent optimal debulking and were
treated with intraperitoneal chemotherapy on GOG 172 was
stated to be the longest median survival ever reported in a
randomized phase III ovarian cancer GOG clinical trial.34
These results serve as a gold standard for quality care, which
future studies should try to attain. Although the PFS and OS
for the NACT arms of the EORTC/NCIC and CHORUS trials are
consistent with those described in other NACT studies, the
overall OS rates of 23 to 30 months in all four arms of these
studies, and, in fact, in essentially all studies advocating
the NACT/IDS approach, are very low (Table 1).12,13,35 In
contrast, a retrospective review of an identical sample
population of patients treated during the same time period
as the EORTC/NCIC trial at MSKCC, demonstrated a median
OS rate of 50 months for all patients treated with PDS.36 A
recent publication from the same institution reported a
median survival of 72 months for all patients who had
undergone PDS regardless of residual disease status (including those who had both optimal and suboptimal
cytoreduction; Fig. 1).37
CHEMOTHERAPY FIRST: QUICKER

IMPROVEMENT IN QUALITY OF LIFE?
Quality of Life Is an Essential Parameter
QOL is an important parameter in the evaluation of therapeutic interventions and is currently the main mechanism to derive patients perspective regarding the impact
Abbreviations: MSKCC, Memorial Sloan Kettering Cancer Center; OS, overall

survival; CGR, complete gross resection; mos, months. CGR, complete gross
resection; PDS, primary debulking surgery.
of any medical intervention. Though not an integral part

of standard care, QOL measures are regularly used in
prospective clinical trials. Increasingly popular is the use
of patient-reported outcomes (PROs), in which QOL measures serve as an inherent element in addition to reported
adverse sequelae.38 The increasing importance of QOL is
accentuated by data demonstrating that it is also a predictive
variable for survival.39 Furthermore, a recent cost-efficacy
analysis of the GOG 218 trial determining the benefit of the
addition of bevacizumab to platinum-based chemotherapy
revealed that the incorporation of QOL outcomes made a
meaningful contribution to the results, rendering the addition of bevacizumab even less favorable compared with
the Markov model, which excluded QOL outcomes.40 Including QOL outcomes in noninferiority studies as a measure
of the morbidity and overall impact of an intervention is
undoubtedly of major importance.
Although both surgery and chemotherapy are essential
interventions in treating patients with advanced ovarian
cancer, there is a relative paucity of trials addressing the
role of surgery. The surgically based, randomized prospective studies encompass the effect of IDS, the role
of secondary cytoreductive surgery at the time of recurrence, the value of lymphadenectomy in debulking, and
delayed PDS.12,13,30,41-43 The EORTC/NCIC and CHORUS
studies are the only two reported prospective randomized
trials addressing the use of NACT in ovarian cancer. In
both trials, similar QOL assessment tools were used: the
EORTC QLQ-C30 questionnaire version 3, which consists
of global health status/QOL multifunctional scales and
155
LEARY ET AL
symptom scales combined with a scale regarding financial

difficulties; and the ovarian cancerspecific QOL (QLQOv28).44 These are well-validated and reliable questionnaires used frequently in ovarian cancer studies. The
assessments were administered before patient randomization at the end of the third and the sixth cycles of
chemotherapy and at 6- and 12-month follow-up visits. In
both studies, the end of the third cycle of chemotherapy
was the presurgical assessment in the NACT arm of the
studies.
An Evaluation of Quality of Life in EORTC 55971

The EORTC 55971 trial included QOL analysis from more
than 400 women from the institutions with best compliance for QOL data collection. This provided sufficient
numbers to permit clinically meaningful differences to be
detected.44 Good compliance was defined as institutions
that had at least 50% compliance at baseline and 35%
compliance on average during follow-up over all enrolled
patients. The QLQ-C30 scales and single-item questions
were linearly transformed to a 0 to 100 scale and analyzed
according to the procedures recommended by the EORTC
QOL group. A higher score on the functional and global/QOL
scales indicated better QOL, whereas a higher score on the
symptoms scale indicated poorer QOL. At least a 10-point
alteration from baseline was necessary to indicate a major change in QOL. The analysis was also repeated
with stratification factors at randomization including
country, method of tumor biopsy, International Federation
of Gynecology and Obstetrics stage, and largest tumor
diameter.
There were no differences noted in the QOL functioning
or symptom scales between the two arms except for
fatigue, pain, and dyspnea. These changes were deemed
clinically relevant because there was a 10-point change
from baseline. The level of fatigue improved in the PDS
arm at the third cycle of chemotherapy and persisted to
12 months, whereas in the NACT arm, the same improvement occurred, though was noted only at 6 months
(i.e., no changes during treatment, and persisted to
12 months). The pain scale also improved in both arms to
subsequent points of measurement. Regarding dyspnea,
an improvement was noted at the sixth chemotherapy
cycle and at 6 months in the NACT arm, but not the PDS
arm. Though these differences were noted, the overall
conclusion was that QOL did not differ between the treatment
strategies. However, there was an unusual and notable
finding. There was a significant survival difference between
institutions with good compliance collecting QOL measures
versus those who were less compliant: the median OS was
32 months versus 23 months (p = .0006). Additionally, the
optimal debulking rates were nearly twice that in selected
versus excluded institutions at 40% and 20%, respectively
(p = .001). Although there may be other explanations and
possible confounding factors, QOL data collection compliance
may also be an indicator of an institutions excellence.
156
An Evaluation of Quality of Life in CHORUS

The CHORUS trial measured similar QOL parameters within
the same time frame as the EORTC study. CHORUS reported
an equivalent survival pattern in each arm, but with lower
median survival rates at 22 months for PDS and 23 months
for NACT.13 Notably, the median age of women enrolled was
older, at age 65, and 19% had a performance status (PS)
score of 2 or 3. The older age and poorer PS probably
contributed, in part, to the lower median survival patterns.
For the purposes of this article, the QOL data were collected
from all possible cases and not confined to the centers with
better QOL data collection rates. There was sufficient data
available from 102 patients in the PDS arm and 117 in the
NACT arm. The Global QOL remained unchanged in both
arms, with mean scores of 51.5 (95% CI, 45.856.2) at
baseline to 69.2 (95% CI, 56.164.3) at 3 months, and 53.6
(95% CI, 48.658.7) to 58.6 (95% CI, 54.562.8; p = .58,
respectively). Again, accepting a 10-point change in the mean
score as being clinically significant, improvements were noted
in nausea/vomiting and dyspnea at 3 months compared with
baseline in the PDS arm. In the same time frame, diarrhea
symptoms improved in the NACT arm. In both arms, pain and
appetite scores improved, whereas abdominal/gastric
symptoms and peripheral neuropathy worsened. Not surprisingly, chemotherapy-associated side effects worsened
in the NACT arm. These were the only parameters with
changes; however, they were not significant. However, if
the data are graphically demonstrated, as can be seen in
Fig. 2, over the 6-month period, the trend is that the NACT
arm showed an improving trend compared with the
PDS arm.
Hence, is NACT a quicker route to a better QOL compared
with the primary surgery approach? Both prospective randomized trials seem to indicate that there are no major
statistical differences in QOL between PDS and NACT.
However, it may be argued that the trends are of more
importance than the exact numbers. Combined with other
FIGURE 2. CHORUS Global Quality of Life Evolution

Over Time
Abbreviation: QOL, quality of life.
morbidity data, one could favor the neoadjuvant approach in

this patient population.
BIOLOGIC CONSIDERATIONS REGARDING THE

THERAPEUTIC SEQUENCE
Histologic and Genomic Factors to Select Patients
One argument proposed in support of NACT is to reduce
tumor bulk and increase the chance of a complete and
potentially less morbid surgical resection. This makes
biologic sense in chemotherapy-sensitive high-grade serous ovarian cancer (HGSOC), where response rates to
first-line platinum-based chemotherapy approach 75%.
However, rare subtypes such as low-grade serous ovarian
cancer (LGSOC) and mucinous ovarian cancer (mOC) are
much less responsive to standard first-line platinumbased therapy (Table 2).45-54 A small study on LGSOC
reported an objective response rate of 3.5% to platinumbased NACT, whereas only 20% of patients with advanced
mucinous ovarian cancer recruited to a randomized trial
showed an objective response to a first-line platinum
combination.45,50 Although numbers of patients included
in these reports were small, these data raise important
questions regarding the value of NACT in rare subtypes of
epithelial ovarian cancer. A major concern with NACT is
the risk of progression and loss of opportunity for patients
who might go from being candidates for a morbid, but
feasible, PDS to not being candidates for surgery at all. In
this regard, tumor biology should clearly influence the
therapeutic sequence and one could propose a histologically
driven algorithm for patients with bulky epithelial ovarian
cancer. Recent comprehensive genomic studies in HGSOC
have started to identify candidate molecular predictors
of platinum sensitivity (BRCA1/2 mutations or alterations
in other homologous recombination genes) or resistance
(CCNE1 amplifications).55-57 Future studies will be crucial
TABLE 2. Response to First-Line Platinum-Based

Chemotherapy in Rare Subtype of Epithelial Ovarian
Cancer
No. of Patients
With Evaluable
Disease
Activity
(oRR)
Reference(s)
, 5%
Schmeler et al 45
LGSOC
First-Line
24
CarboplatinPaclitaxel
CCC
First-Line
PlatinumBased
4 studies, 2368 22%41% Kita et al, Sugyama

et al, Ho et al, and
Takano et al 46-49
mOC
First-Line
Platinum
Based
5 studies, 950
13%60% Hess et al, Alexandre

et al, Pectasides
et al, Gore et al, and
Shimada et al 50-54
Abbreviations: oRR, objective response rate; LGSOC, low-grade serous ovarian cancer;
CCC, clear cell ovarian cancer; mOC, mucinous ovarian cancer.
to determine whether these genomic markers may influence the therapeutic sequence for patients with advanced HGSOC.
Evidence for NACT as a Driver of Chemotherapy

Resistance
As previously mentioned, a risk of the neoadjuvant approach
is that treating a greater volume of a heterogeneous tumor
with NACT may be driving platinum resistance.58 Retrospective studies have shown that NACT was associated
with a higher risk of platinum-resistant relapse, even when
IDS resulted in complete macroscopic resection.59,60 In the
case of platinum-sensitive progression, lower response rates
to a platinum combination have been described after NACT,
suggesting that NACT may be compromising responsiveness
to subsequent therapy.59,61 Although this remains a hypothetical risk, these studies are limited by their retrospective
nature and inherent biases that justified the decision to offer
NACT in the first place, which likely explain these patients
poor outcomes. Similarly, studies evaluating whether the
number of cycles of NACT influenced outcome have shown
worse survival with more than three or four cycles of NACT
even when complete macroscopic resection was achieved at
IDS.62,63 Again, because of the retrospective nature of these
studies, the poor outcome for patients receiving more than
four cycles may simply reflect inherent differences in tumor
biology and chemotherapy responsiveness rather than a
deleterious impact of more cycles of NACT. In fact, a more
recent meta-analysis of 21 studies failed to show that the
number of NACT cycles affected survival.64 The only available prospective data come from the two randomized
controlled trials of PDS versus NACT, and these failed to
show a difference in PFS or OS, making it unlikely that NACT
is a major driver of platinum resistance.12,24
In vivo evaluation of chemotherapy responsiveness during
NACT. As opposed to the adjuvant setting in which patients
are treated with no measurable disease, an advantage of
NACT is that it could offer the opportunity for an in vivo
evaluation of chemotherapy responsiveness. Careful monitoring of tumor response during chemotherapy could
identify nonresponders early and justify treatment change in
an effort to improve the efficacy of NACT and long-term
patient outcomes. Some prospective studies in locally advanced breast cancer have started to suggest a role for
response-adapted NACT where the neoadjuvant protocol is
modified according to early tumor response. One trial
showed an improvement in PFS and OS among patients with
locally advanced breast cancer randomly selected to an
adaptive NACT strategy with a chemotherapy switch in
case of poor response compared with standard NACT.65
Whether adaptive NACT could be beneficial in ovarian
cancer has never been investigated and may be limited by
the fact that measuring on-treatment response to NACT is
difficult in ovarian cancer.66 Neither improvement in CT
imaging nor biologic response by CA-125 are reliable for
157
LEARY ET AL
NACT response evaluation. 67 Other technologies such as

metabolic imaging by fluorodeoxyglucose-PET or circulating cell-free tumor DNA may offer early-response information during NACT for advanced ovarian cancer.
However, to date it is unclear how this information would be
clinically useful. A small subset of patients (15%) with HGSOC
are refractory and progress clinically and biologically during
NACT. The only value of identifying these early nonresponders would be if this could inform clinical decision
making. Unfortunately, ovarian cancer does not have an
effective alternative to standard neoadjuvant carboplatin
and paclitaxel.
In vivo evaluation of chemotherapy responsiveness
at IDS. Complete resection of all macroscopic disease at PDS
has been confirmed as a critical prognostic factor32; however, its value after NACT is less clear because visualization of
residual disease may be less reliable after chemotherapy.68
Studies have sought to evaluate whether the degree of
pathologic response to NACT could provide prognostic
information (Table 3).69-76 A number of studies have
confirmed the prognostic value of pathologic complete
response (pCR) after NACT. Among patients treated with
NACT and no residual disease at IDS, Ferron et al showed that
pCR with no viable tumor cells in the surgical specimen
occurred in 14% of patients and was predictive of PFS, but
the degree of histologic response was not.69 A larger study
(322 patients) confirmed pCR (defined as no residual tumor
cells in the surgical specimen) as predictive of both PFS and
OS, whereas a good response not amounting to pCR was
not prognostic.70 Several groups investigated composite
pathologic response scores, incorporating an assessment
of viable tumor cells, fibrosis, inflammatory changes, or
macrocytic infiltration, and have shown conflicting results using heterogeneous definitions for pathologic response (Table 3).71-75,77 A more recent study suggested that a
three-tier histopathologic chemotherapy response score was
reproducible and prognostically meaningful for patients with
HGSOC treated with NACT, regardless of debulking status.76
Complete or near-complete pathologic response (defined as
a chemotherapy response score [CRS] of 3) was associated
with significantly improved PFS (18 vs. 12 months, p , .001)
and decreased risk of platinum-resistant relapse (odds ratio
[OR] 0.08; p , .001) compared with those tumors showing
less in vivo chemosensitivity (CRS scores of 1 and 2).
Although the authors concluded that the three-tier scoring
system may be useful, in fact, outcomes were only different
when considering CRS 3 versus CRS 1/2, making it more of a
two-tier system. In addition, CRS 3 included near-complete
responses defined as residual viable cells, or nodules larger
than 2 mm, and did not predict OS. Although the pathologic
response score may need refining and will require prospective validation, taken together these data suggest that
an evaluation of in vivo response to NACT at the time of IDS
could offer useful prognostic information and may inform
postoperative management. The identification of a chemotherapy response score to risk-stratify patients would
158
allow for optimization of postoperative management, consideration of maintenance treatment, or participation in

clinical trials, with the aim of improving long-term outcome.
The question remains whether the degree of pathologic response is prognostic, or whether the only valid surrogate for
OS is pCR strictly defined as no residual cancer cells within the
entire surgical specimen. Studies in breast cancer have shown
that complete response in both primary site and lymph nodes
(occurring in 18% of patients with NACT) is required to
predict a major improvement in OS,78 the biologic explanation
being that pCR is the evidence that other micrometastatic
deposits may have also been eradicated by NACT. Interestingly, the correlation between pathologic response and
OS was the strongest for triple-negative breast cancer, a
highly proliferative subtype that shares genetic homology
with HGSOC.
The Neoadjuvant Setting as a Tool to Evaluate

New Regimens
The neoadjuvant design may be useful to test new therapies
in a faster way and with fewer patients than is required for
large adjuvant trials. Given the strong correlation between
pCR and OS in breast cancer, new drugs are being evaluated
in randomized neoadjuvant trials in which the primary
endpoint is an improvement in pCR at surgery.79 pCR rates
are somewhat lower in ovarian cancer than in other tumor
types (6% to 12%), but randomized studies designed to test
the benefit of adding a new agent to standard chemotherapy, where the primary objective would be a major
improvement in pCR, could provide a useful signal-finding
strategy for new drugs. The advantages of this approach
are a translational research opportunity to identify predictive biomarkers and obtaining a rapid answer because
results are available within a few months of the last patient
who was randomly placed. In addition, the neoadjuvant trial
design allows the testing of a novel agent in treatment-naive
patients, as opposed to metastatic studies, where prior
exposure to multiple lines of chemotherapy may distort drug
activity. The U.S. Food and Drug Administration has recently
issued a statement that they will strongly consider evidence
from randomized controlled studies using pCR as an endpoint to allow accelerated approval of novel therapies.80
Although this guidance was designed for trials conducted in
breast cancer, using pCR as an endpoint has the potential to
help address unmet needs in high-risk populations in a far
shorter time frame than would be required via the conventional approach to drug development.79
Translational Research Opportunities

Despite exquisite sensitivity to first-line platinum-based
chemotherapy, HGSOC almost invariably relapses. Yet the
mechanisms underpinning primary or acquired chemotherapy resistance are poorly understood (Fig. 3). HGSOC
is characterized by extreme genomic instability and
intratumoral heterogeneity (ITH).81 Inherent genetic instability and ITH provide the ideal setting for adaptation and
TABLE 3. Chemotherapy Pathologic Scores Evaluated in Retrospective Studies

Reference
Ferron et al69
(58 patients)
Categories of Pathological Response

Score
No. of
Patients Correlation of Pathological Score and Outcome
Group 1: No residual tumor cells
3-year PFS: 12% (not significant)
Group 3: Persistence of at least one site 36

with no chemotherapy effect
3-year PFS: 19% (not significant)
21
PFS: 36 months (p 5 .001 vs. micro/macroPR); OS: 72 months (p 5 .018 vs.

micro/macroPR)
MicroPR: Residual tumor foci # 3 mm
104
MacroPR . 3-mm foci
197
PFS microPR: 16 months vs. 13 months for macro PR (not significant); OS microPR:
38 months vs. 29 months for macroPR (not significant)
Score* 1-8: Some chemotherapy effect
NA
Petrillo et al70
pCR: No residual tumor cells
(322 patients)
Le et al71
(62 patients)
3-year PFS: 63% (p 5 .02 vs. groups 2/3)
Group 2: Persistent disease with marked 14

changes
Score 0: No chemotherapy effect
NA
Le et al72
(73 patients)
Marked/moderate necrosis
36
Little to no necrosis
37
Le et al73
(66 patients)
Chemotherapy effect on omental

deposits
58
No omental response
Composite score** G2/3
Muraji et al74
(124 patients)
Composite score** G0/1
OS: Scores 1-8 vs. score 0 (73 months vs. 39 months; p 5 .14)
Improved PFS; HR 0.51; p 5 .048
OS: 84 months
OS: 31 months
72
Improved OS; HR 0.61; p 5 .03
52
Sassen et al75
(49 patients)
No residual tumor cells or foci , 5 mm
OS: 46 months; p 5 .02
Tumor foci . 5 mm
42
OS: 27 months
Bohm et al76
(71 patients)
CRS 3: No residual tumor cells or

foci , 2 mm
19
Improved PFS: CRS 3 vs. CRS 1/2, (18 months vs. 12 months;
p , .001); OS: CRS 3 (45 months) vs. CRS 1/2 (28 months) not significant
CRS 2: Appreciable response
47
CRS 1: No or minimal response
Abbreviations: PFS, progression-free survival; OS, overall survival; IDS, interval debulking surgery; NA, not available; pCR, pathologic complete response; CRS, chemotherapy response
score; HR, hazard ratio; microPR, microscopic pathologic response; macroPR, macroscopic pathologic response.
*Score (0-8) of necrosis, fibrosis, macrophage infiltration, tumor inflammation.
**Composite score 5 degree of disappearance of tumor cells, displacement by necrosis and fibrosis, and inflammatory changes graded from 0 to 3, with G0 being no chemotherapy
effect and G3 being no viable tumor cells.
treatment escape and have been shown to drive the

accelerated acquisition of multidrug resistance.82 Evidence
in HGSOC cell lines suggests that relapse is not caused by
linear acquisition of genetic alterations, but rather to
treatment-induced selection and expansion of intrinsically
resistant clones.83 Most genomic studies to date in HGSOC
have focused on comprehensive analyses of the primary
tumor, but evidence is emerging that a comparison between matched tumor samples from diagnosis and IDS can
reveal the mechanisms accounting for the acquisition of
resistance in individual patients. For example, an enrichment of ALDH1-positive cells in post-NACT samples has
been demonstrated in some patients and shown to correlate with poor survival post-NACT, suggesting that a
cancer stem cell subpopulation may account for chemotherapy resistance.84 Studies among patients with germline
BRCA mutations treated with NACT have shown that, although initial response rates to platinum are high, a subset
shows restoration of BRCA heterozygosity in the post-NACT
sample, suggesting a rapid expansion of subclones without
somatic LOH for the BRCA wild-type allele during
treatment.85
Profiling HGSOC at diagnosis when the bulk of disease is chemotherapy-sensitive and potentially highly
heterogeneous may be relatively uninformative. However,

identifying chromosomal alterations or somatic mutational events that occur at low frequency at baseline and
are selected for during therapy could uncover new therapeutic strategies to overcome platinum resistance. Given
the intrinsic chemotherapy sensitivity of the majority of
HGSOC, platinum-based chemotherapy will likely always
provide an essential component of the treatment strategy; however, profiling the post-NACT tumor may identify
actionable targets and provide the rationale for subsequent clinical trials testing novel agents as maintenance
therapy in an effort to eradicate the minimal residual disease post-chemotherapy.
FUTURE DIRECTIONS AND CONCLUSION

The discussed considerations truly underscore the need
for further research on the most appropriate sequence of
care for women with advanced ovarian cancer. Despite
years of debate and two randomized controlled trials,
the question of whether surgery or chemotherapy should
be the primary treatment strategy has yet to be answered. Survival and QOL-related measures are both
of paramount importance because they pertain to
159
LEARY ET AL
FIGURE 3. The Neoadjuvant Setting: A Translational Research Opportunity
(A) Fifteen percent of HGSOC cases have refractory disease and progress on NACT. Prognosis is poor and studies are needed to characterize the profile of these primary
resistant tumors and develop effective alternatives to carboplatin and paclitaxel. However, HGSOC is very chemotherapy-sensitive and most show an initial response to
NACT and may progress to IDS. (B) A small subset (5% to 12%) show pCR with no residual tumor cells after NACT. Prognosis is excellent and studies should focus on
finding therapies that increase the proportion of patients achieving pCR. (C) Most HGSOC demonstrate initial sensitivity but despite complete debulking will relapse.
Identification of molecular alterations selectively enriched in these post-treatment tumors may uncover mechanisms of acquired resistance and identify new therapeutic
targets that could eradicate minimal residual disease.
Abbreviations: HGSOC, high-grade serous ovarian cancer; NACT, neoadjuvant chemotherapy; IDS, interval debulking surgery; pCR, pathologic complete response.
overall outcomes. Although we seek to answer these

questions, excellent clinical judgment is necessary to carefully and aptly select the appropriate patient for each
treatment regimen. Furthermore, there is a definitive
role for NACT in the laboratory and in clinical research.

Much effort must be made to determine novel approaches to the management of this complicated and
ever-evolving disease.
References
1. Holschneider CH, Berek JS. Ovarian cancer: epidemiology, biology, and
prognostic factors. Semin Surg Oncol. 2000;19:3-10.
2. Lowe KA, Chia VM, Taylor A, et al. An international assessment of
ovarian cancer incidence and mortality. Gynecol Oncol. 2013;130:
107-114.
3. Armstrong DK. Relapsed ovarian cancer: challenges and management
strategies for a chronic disease. Oncologist. 2002;7(suppl 5):20-28.
4. Torre LA, Bray F, Siegel RL, et al. Global cancer statistics, 2012. CA Cancer
J Clin. 2015;65:87-108.
5. Morgan RJ Jr, Alvarez RD, Armstrong DK, et al; National Comprehensive
Cancer Network. Epithelial ovarian cancer. J Natl Compr Canc Netw.
2011;9:82-113.
6. Meigs JV, Greenough RB. Tumors of the female pelvic organs. Am J Med
Sci. 1935;189:430.
7. Griffiths CT. Surgical resection of tumor bulk in the primary treatment of
ovarian carcinoma. Natl Cancer Inst Monogr. 1975;42:101-104.
8. Eisenkop SM, Friedman RL, Wang H-J. Complete cytoreductive surgery
is feasible and maximizes survival in patients with advanced epithelial
10. Bristow RE, Tomacruz RS, Armstrong DK, et al. Survival effect of maximal cytoreductive surgery for advanced ovarian carcinoma during the
platinum era: a meta-analysis. J Clin Oncol. 2002;20:1248-1259.
11. Vergote I, Trope CG, Amant F, et al. EORTC-GCG/NCIC-CTG randomised
trial comparing primary debulking surgery with neoadjuvant chemotherapy in stage IIIC-IV ovarian, fallopian tube and peritoneal cancer
(OVCA). Presented at: 12th Biennial Meeting of the International Gynecologic Cancer Society; October 2008; Bangkok, Thailand.
12. Vergote I, Trope CG, Amant F, et al; European Organization for Research
and Treatment of Cancer-Gynaecological Cancer Group; NCIC Clinical
Trials Group. Neoadjuvant chemotherapy or primary surgery in stage
IIIC or IV ovarian cancer. N Engl J Med. 2010;363:943-953.
13. Kehoe S, et al. Chemotherapy or upfront surgery for newly diagnosed
advanced ovarian cancer: Results from the MRC CHORUS trial. J Clin
Oncol. 2013;31:15s (suppl; abstr 5500).
14. Bristow RE, Duska LR, Lambrou NC, et al. A model for predicting surgical
ovarian cancer: a prospective study. Gynecol Oncol. 1998;69:103-108.

9. Allen DG, Heintz AP, Touw FW. A meta-analysis of residual disease and
outcome in patients with advanced ovarian carcinoma using computed

tomography. Cancer. 2000;89:1532-1540.
15. Hoskins WJ, McGuire WP, Brady MF, et al. The effect of diameter of
largest residual disease on survival after primary cytoreductive surgery
survival in stage III and IV carcinoma of the ovary. Eur J Gynaecol Oncol.
in patients with suboptimal residual epithelial ovarian carcinoma. Am J
1995;16:349-356.
Obstet Gynecol. 1994;170:974-979, discussion 979-980.
160
16. Dauplat J, Le Bouedec G, Pomel C, Scherer C. Cytoreductive surgery for

advanced stages of ovarian cancer. Semin Surg Oncol. 2000;19:42-48.
17. Zivanovic O, Eisenhauer EL, Zhou Q, et al. The impact of bulky upper
abdominal disease cephalad to the greater omentum on surgical
outcome for stage IIIC epithelial ovarian, fallopian tube, and primary
peritoneal cancer. Gynecol Oncol. 2008;108:287-292.
18. Eisenkop SM, Spirtos NM. What are the current surgical objectives,
strategies, and technical capabilities of gynecologic oncologists treating
advanced epithelial ovarian cancer? Gynecol Oncol. 2001;82:489-497.
19. Chi DS, Eisenhauer EL, Zivanovic O, et al. Improved progression-free and
overall survival in advanced ovarian cancer as a result of a change in
surgical paradigm. Gynecol Oncol. 2009;114:26-31.
20. Angioli R, Plotti F, Aloisi A, et al. Does extensive upper abdomen surgery
during primary cytoreduction impact on long-term quality of life? Int J
Gynecol Cancer. 2013;23:442-447.
21. Barlin JN, Long KC, Tanner EJ, et al. Optimal (#1 cm) but visible residual
disease: is extensive debulking warranted? Gynecol Oncol. 2013;130:
284-288.
22. Chi DS, Zivanovic O, Levinson KL, et al. The incidence of major complications after the performance of extensive upper abdominal surgical
procedures during primary cytoreduction of advanced ovarian, tubal,
and peritoneal carcinomas. Gynecol Oncol. 2010;119:38-42.
23. Fanfani F, Fagotti A, Gallotta V, et al. Upper abdominal surgery in
advanced and recurrent ovarian cancer: role of diaphragmatic surgery.
Gynecol Oncol. 2010;116:497-501.
24. Kehoe SM, Eisenhauer EL, Chi DS. Upper abdominal surgical procedures:
liver mobilization and diaphragm peritonectomy/resection, splenectomy,
and distal pancreatectomy. Gynecol Oncol. 2008;111:S51-S55.
25. Aletti GD, Dowdy SC, Gostout BS, et al. Aggressive surgical effort and
improved survival in advanced-stage ovarian cancer. Obstet Gynecol.
2006;107:77-85.
26. Eisenhauer EL, Abu-Rustum NR, Sonoda Y, et al. The addition of
extensive upper abdominal surgery to achieve optimal cytoreduction
improves survival in patients with stages IIIC-IV epithelial ovarian
cancer. Gynecol Oncol. 2006;103:1083-1090.
27. Horowitz N, Miller A, Rungruang B, et al. Reply to G.D. Aletti et al.
J Clin Oncol. 2015;33:3521-3522.
28. Vernooij F, Heintz AP, Coebergh JW, et al. Specialized and high-volume
care leads to better outcomes of ovarian cancer treatment in the
Netherlands. Gynecol Oncol. 2009;112:455-461.
29. Long B, Chang J, Ziogas A, et al. Impact of race, socioeconomic status,
and the health care system on the treatment of advanced-stage ovarian
cancer in California. Am J Obstet Gynecol. 2015;212:468.e1-468.e9.
30. van der Burg ME, van Lent M, Buyse M, et al; Gynecological Cancer
Cooperative Group of the European Organization for Research and
Treatment of Cancer. The effect of debulking surgery after induction
chemotherapy on the prognosis in advanced epithelial ovarian cancer.
N Engl J Med. 1995;332:629-634.
31. Goldie JH, Coldman AJ. The genetic origin of drug resistance in neoplasms: implications for systemic therapy. Cancer Res. 1984;44:
3643-3653.
32. du Bois A, Reuss A, Pujade-Lauraine E, et al. Role of surgical outcome as
prognostic factor in advanced epithelial ovarian cancer: a combined
exploratory analysis of 3 prospectively randomized phase 3 multicenter trials: by the Arbeitsgemeinschaft Gynaekologische Onkologie
Studiengruppe Ovarialkarzinom (AGO-OVAR) and the Groupe
dInvestigateurs Nationaux Pour les Etudes des Cancers de lOvaire
(GINECO). Cancer. 2009;115:1234-1244.
33. Pujade-Lauraine E, Hilpert F, Weber B, et al. Bevacizumab combined
with chemotherapy for platinum-resistant recurrent ovarian cancer:
The AURELIA open-label randomized phase III trial. J Clin Oncol. 2014;
32:1302-1308.
34. Armstrong DK, Bundy B, Wenzel L, et al; Gynecologic Oncology Group.

Intraperitoneal cisplatin and paclitaxel in ovarian cancer. N Engl J Med.
2006;354:34-43.
35. Bristow RE, Chi DS. Platinum-based neoadjuvant chemotherapy and
interval surgical cytoreduction for advanced ovarian cancer: a metaanalysis. Gynecol Oncol. 2006;103:1070-1076.
36. Chi DS, Musa F, Dao F, et al. An analysis of patients with bulky advanced
stage ovarian, tubal, and peritoneal carcinoma treated with primary
debulking surgery (PDS) during an identical time period as the randomized EORTC-NCIC trial of PDS vs neoadjuvant chemotherapy
(NACT). Gynecol Oncol. 2012;124:10-14.
37. Mueller JJ, Zhou QC, Iasonos A, et al. Neoadjuvant chemotherapy and
primary debulking surgery utilization for advanced-stage ovarian cancer
at a comprehensive cancer center. Gynecol Oncol. 2016;140:436-442.
38. Soo Hoo S, Marriott N, Houlton A, et al. Patient-reported outcomes
after extensive (ultraradical) surgery for ovarian cancer: results from a
prospective longitudinal feasibility study. Int J Gynecol Cancer. 2015;25:
1599-1607.
39. Carey MS, Bacon M, Tu D, et al. The prognostic effects of performance
status and quality of life scores on progression-free survival and overall
survival in advanced ovarian cancer. Gynecol Oncol. 2008;108:100-105.
40. Cohn DE, Barnett JC, Wenzel L, et al. A cost-utility analysis of NRG
Oncology/Gynecologic Oncology Group Protocol 218: incorporating
prospectively collected quality-of-life scores in an economic model of
treatment of ovarian cancer. Gynecol Oncol. 2015;136:293-299.
41. Rose PG, Nerenstone S, Brady MF, et al; Gynecologic Oncology Group.
Secondary surgical cytoreduction for advanced ovarian carcinoma.
N Engl J Med. 2004;351:2489-2497.
42. Maggioni A, Benedetti Panici P, DellAnna T, et al. Randomised study of
systematic lymphadenectomy in patients with epithelial ovarian cancer
macroscopically confined to the pelvis. Br J Cancer. 2006;95:699-704.
43. Gynecologic Cancer Intergroup. GCIG Clinical Trials By Lead Cooperative
Group. http://www.gcig.igcs.org/ClinicalTrials.html. Accessed February
1, 2016.
44. Greimel E, Kristensen GB, van der Burg ME, et al; European Organization
for Research and Treatment of Cancer - Gynaecological Cancer Group
and NCIC Clinical Trials Group. Quality of life of advanced ovarian cancer
patients in the randomized phase III study comparing primary debulking
surgery versus neo-adjuvant chemotherapy. Gynecol Oncol. 2013;131:
437-444.
45. Schmeler KM, Sun CC, Bodurka DC, et al. Neoadjuvant chemotherapy
for low-grade serous carcinoma of the ovary or peritoneum. Gynecol
Oncol. 2008;108:510-514.
46. Takano M, Kikuchi Y, Yaegashi N, et al. Clear cell carcinoma of the ovary:
a retrospective multicentre experience of 254 patients with complete
surgical staging. Br J Cancer. 2006;94:1369-1374.
47. Kita T, Kikuchi Y, Kudoh K, et al. Exploratory study of effective chemotherapy to clear cell carcinoma of the ovary. Oncol Rep. 2000;7:
327-331.
48. Sugiyama T, Kamura T, Kigawa J, et al. Clinical characteristics of clear cell
carcinoma of the ovary: a distinct histologic type with poor prognosis and
resistance to platinum-based chemotherapy. Cancer. 2000;88:2584-2589.
49. Ho CM, Huang YJ, Chen TC, et al. Pure-type clear cell carcinoma of the
ovary as a distinct histological type and improved survival in patients
treated with paclitaxel-platinum-based chemotherapy in pure-type
advanced disease. Gynecol Oncol. 2004;94:197-203.
50. Gore ME, Hackshaw A, Brady WE, Penson RT, et al. Multicenter trial of
carboplatin/paclitaxel versus oxaliplatin/capecitabine, each with/without
bevacizumab, as first line chemotherapy for patients with mucinous
epithelial ovarian cancer. J Clin Oncol. 2015;33 (suppl; abstr 5528).
161
LEARY ET AL
51. Hess V, AHern R, Nasiri N, et al. Mucinous epithelial ovarian cancer:

a separate entity requiring specific treatment. J Clin Oncol. 2004;22:
1040-1044.
52. Alexandre J, Ray-Coquard I, Selle F, et al; GINECO. Mucinous advanced
epithelial ovarian carcinoma: clinical presentation and sensitivity to
platinum-paclitaxel-based chemotherapy, the GINECO experience. Ann
Oncol. 2010;21:2377-2381.
53. Pectasides D, Fountzilas G, Aravantinos G, et al. Advanced stage mucinous epithelial ovarian cancer: the Hellenic Cooperative Oncology
Group experience. Gynecol Oncol. 2005;97:436-441.
54. Shimada M, Kigawa J, Ohishi Y, et al. Clinicopathological characteristics
of mucinous adenocarcinoma of the ovary. Gynecol Oncol. 2009;113:
331-334.
55. Tan DS, Rothermundt C, Thomas K, et al. BRCAness syndrome in
ovarian cancer: a case-control study describing the clinical features and
outcome of patients with epithelial ovarian cancer associated with
BRCA1 and BRCA2 mutations. J Clin Oncol. 2008;26:5530-5536.
56. Dann RB, DeLoia JA, Timms KM, et al. BRCA1/2 mutations and expression: response to platinum chemotherapy in patients with
advanced stage epithelial ovarian cancer. Gynecol Oncol. 2012;125:
677-682.
57. Patch AM, Christie EL, Etemadmoghadam D, et al; Australian Ovarian
Cancer Study Group. Whole-genome characterization of chemoresistant
ovarian cancer. Nature. 2015;521:489-494.
58. Goldie JH, Coldman AJ. Quantitative model for multiple levels of drug
resistance in clinical tumors. Cancer Treat Rep. 1983;67:923-931.
59. Rauh-Hain JA, Nitschmann CC, Worley MJ Jr, et al. Platinum resistance
after neoadjuvant chemotherapy compared to primary surgery in
patients with advanced epithelial ovarian carcinoma. Gynecol Oncol.
2013;129:63-68.
60. Petrillo M, Ferrandina G, Fagotti A, et al. Timing and pattern of recurrence in ovarian cancer patients with high tumor dissemination
treated with primary debulking surgery versus neoadjuvant chemotherapy. Ann Surg Oncol. 2013;20:3955-3960.
61. da Costa AA, Valadares CV, Baiocchi G, et al. Neoadjuvant chemotherapy followed by interval debulking surgery and the risk of platinum
resistance in epithelial ovarian cancer. Ann Surg Oncol. 2015;22(suppl
3):971-978.
62. Bristow RE, Nugent AC, Zahurak ML, et al. Impact of surgeon specialty
on ovarian-conserving surgery in young females with an adnexal mass.
J Adolesc Health. 2006;39:411-416.
63. Colombo PE, Labaki M, Fabbro M, et al. Impact of neoadjuvant chemotherapy cycles prior to interval surgery in patients with advanced
epithelial ovarian cancer. Gynecol Oncol. 2014;135:223-230.
64. Kang S, Nam BH. Does neoadjuvant chemotherapy increase optimal
cytoreduction rate in advanced ovarian cancer? Meta-analysis of 21
studies. Ann Surg Oncol. 2009;16:2315-2320.
65. von Minckwitz G, Blohmer JU, Costa SD, et al. Response-guided neoadjuvant chemotherapy for breast cancer. J Clin Oncol. 2013;31:
3623-3630.
66. Bellati F, Gasparri ML, Caccetta J, et al. Response criteria can be misleading when drawing conclusion regarding neoadjuvant chemotherapy in advanced ovarian cancer. J Surg Oncol. 2012;106:529, author
reply 527-528.
67. Menczer J, Usviatzov I, Ben-Shem E, et al. Neoadjuvant chemotherapy in
ovarian, primary peritoneal and tubal carcinoma: can imaging results
prior to interval debulking predict survival? J Gynecol Oncol. 2011;22:
183-187.
68. Hynninen J, Lavonius M, Oksa S, et al. Is perioperative visual estimation
of intra-abdominal tumor spread reliable in ovarian cancer surgery after
neoadjuvant chemotherapy? Gynecol Oncol. 2013;128:229-232.
162
69. Ferron JG, Uzan C, Rey A, et al. Histological response is not a prognostic
factor after neoadjuvant chemotherapy in advanced-stage ovarian
cancer with no residual disease. Eur J Obstet Gynecol Reprod Biol. 2009;
147:101-105.
70. Petrillo M, Zannoni GF, Tortorella L, et al. Prognostic role and predictors
of complete pathologic response to neoadjuvant chemotherapy in
primary unresectable ovarian cancer. AM J Obstet Gynecol. 2014; 211:
632.e1-8.
71. Le T, Williams K, Senterman M, et al. Histopathologic assessment of
chemotherapy effects in epithelial ovarian cancer patients treated with
neoadjuvant chemotherapy and delayed primary surgical debulking.
Gynecol Oncol. 2007;106:160-163.
72. Le T, Shahriari P, Hopkins L, et al. Prognostic significance of tumor
necrosis in ovarian cancer patients treated with neoadjuvant chemotherapy and interval surgical debulking. Int J Gynecol Cancer. 2006;16:
986-990.
73. Le T, Williams K, Senterman M, et al. Omental chemotherapy effects as a
prognostic factor in ovarian cancer patients treated with neoadjuvant
chemotherapy and delayed primary surgical debulking. Ann Surg Oncol.
2007;14:2649-2653.
74. Muraji M, Sudo T, Iwasaki S, et al. Histopathology predicts clinical
outcome in advanced epithelial ovarian cancer patients treated with
neoadjuvant chemotherapy and debulking surgery. Gynecol Oncol.
2013;131:531-534.
75. Sassen S, Schmalfeldt B, Avril N, et al. Histopathologic assessment of
tumor regression after neoadjuvant chemotherapy in advanced-stage
ovarian cancer. Hum Pathol. 2007;38:926-934.
76. Bohm S, Faruqi A, Said I, et al. Chemotherapy response score: development and validation of a system to quantify histopathologic response
to neoadjuvant chemotherapy in tubo-ovarian high-grade serous carcinoma. J Clin Oncol. 2015;33:2457-2463.
77. Le T, Alshaikh G, Hopkins L, et al. Prognostic significance of postoperative morbidities in patients with advanced epithelial ovarian
cancer treated with neoadjuvant chemotherapy and delayed primary
surgical debulking. Ann Surg Oncol. 2006;13:1711-1716.
78. Cortazar P, Zhang L, Untch M, et al. Pathological complete response and
79. Prowell TM, Pazdur R. Pathological complete response and accelerated
drug approval in early breast cancer. N Engl J Med. 2012;366:
2438-2441.
80. U.S. Food and Drug Administration. Guidance for Industry: Pathological
Complete Response in Neoadjuvant Treatment of High-Risk Early-Stage
Breast Cancer: Use as an Endpoint to Support accelerated Approval. http://
www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/
guidances/ucm305501.pdf. Accessed March 21, 2016.
81. Choi YP, Shim HS, Gao MQ, et al. Molecular portraits of intratumoral
heterogeneity in human ovarian cancer. Cancer Lett. 2011;307:62-71.
82. Duesberg P, Stindl R, Hehlmann R. Explaining the high mutation rates of
cancer cells to drug and multidrug resistance by chromosome reassortments that are catalyzed by aneuploidy. Proc Natl Acad Sci USA.
2000;97:14295-14300.
83. Cooke SL, Ng CK, Melnyk N, et al. Genomic analysis of genetic heterogeneity and evolution in high-grade serous ovarian carcinoma.
Oncogene. 2010;29:4905-4913.
84. Ayub TH, Keyver-Paik MD, Debald M, et al. Accumulation of ALDH1positive cells after neoadjuvant chemotherapy predicts treatment resistance and prognosticates poor outcome in ovarian cancer. Oncotarget.
2015;6:16437-16448.
85. Gorodnova TV, Sokolenko AP, Ivantsov AO, et al. High response rates to
neoadjuvant platinum-based therapy in ovarian cancer patients carrying germ-line BRCA mutation. Cancer Lett. 2015;369:363-367.
GYNECOLOGIC CANCER
Practical Challenges in
Endometrial Cancer
CHAIR
Linda Duska, MD
University of Virginia Health System
Charlottesville, VA
SPEAKERS
Armin Shahrokni, MD, MPH
New York, NY
Melanie Powell, MD
Barts Health NHS Trust
DUSKA, SHAHROKNI, AND POWELL
Treatment of Older Women With Endometrial Cancer:

Improving Outcomes With Personalized Care
Linda Duska, MD, MPH, Armin Shahrokni, MD, MPH, and Melanie Powell, MD, FRCR, FRCP
OVERVIEW
Endometrial cancer is the most common gynecologic cancer, and with a median age of 62 at diagnosis, it affects a significant
number of older women. With increasing age and obesity rates in the worlds population, there is an anticipated concomitant
increase in older women with endometrial cancer. Older women are more likely to die of endometrial cancer compared with
younger patients. Reasons for this include more aggressive tumor biology, less favorable clinicopathologic features, and
more advanced disease. Other factors, however, such as reluctance to offer surgical treatment to the older patient and
increased complications of treatment are likely to be important. Management of endometrial cancer requires multidisciplinary care (surgery, radiation therapy, and systemic therapy). For each treatment, the feasibility (related to technical
aspect of the procedure/treatment), side effects and safety (related to older-patient factors), and the overall benefit as it
pertains to older women with endometrial cancer should be assessed carefully with a multidisciplinary approach. Despite
the importance of these issues, the data are limited to answer these issues with clarity. In this article, we will review each
treatment modality for older women with endometrial cancer. We will introduce the components of comprehensive
geriatric assessment and their practical implication for older women with cancer in general and older women with endometrial cancer specifically.
ndometrial cancer is the most common gynecologic

cancer in the United States, with 54,870 cases expected in
2016.1 It is also a cancer of older women, with a median age
at diagnosis of 62.1 As the worlds population ages, we can
expect to see more cases of endometrial cancer in older
women. In particular, the oldest old, or those age 80 or
older, is the segment of the population that will see the
greatest growth in the next 2 decades. In the United States,
the number of residents over age 80 is expected to increase
from 3.3% in 2000 to 5.5% in 2030. At the global level, the
average annual growth rate of persons age 80 or older is
twice as high as the growth rate of the population over age
60 (3.8% vs. 1.9%).2 Obesity is also a considerable risk factor
for developing endometrial cancer, and obesity rates are
rising worldwide, with an associated increase in endometrial
cancer rates.3
Forty percent of women diagnosed with endometrial
cancer are age 65 or older (24%, age 6574; 12%, age 7584;
and 4%, age . 84).1 Notably, most deaths (67%) from endometrial cancer occur in women over 65, with a median age
at death of 70 (28%, age 6574; 24%, age 7584; and 15%,
age . 84). There are a number of explanations that have
been postulated for the increased cancer-specific death rate
in older women, including: more advanced stage at diagnosis, higher tumor grade, more aggressive tumor biology,
and a selection bias by providers for less aggressive surgery
and less adjuvant therapy.
The standard-of-care treatment of most patients with
endometrial cancer is surgical and begins with hysterectomy, including removal of the uterus, cervix, and bilateral
fallopian tubes and ovaries. Although the role of complete
lymphadenectomy (removal of bilateral pelvic and paraaortic retroperitoneal lymph nodes) remains controversial, most U.S. gynecologic oncologists follow the so-called
Mayo criteria with respect to staging and remove lymph
nodes for high-grade and/or deeply myoinvasive tumors.4
Surgery alone may be curative for early-stage, low-grade
disease, with radiation and/or chemotherapy recommended
for adjuvant therapy depending on stage of disease and
histology.
In the older woman, the management of endometrial
cancer may require a more multidisciplinary approach.
Older women are more likely to have high-grade disease,
poor histology, and advanced disease; require adjuvant
treatment; and experience disease recurrence. Gayar et al5
considered 121 patients age 75 and older with apparent
From the Division of Gynecologic Oncology, Obstetrics and Gynecology, University of Virginia School of Medicine, Charlottesville, VA; Memorial Sloan Kettering Cancer Center, Weill
Cornell Medicine, New York, NY; Department of Clinical Oncology, Barts Health NHS Trust, St. Bartholomews Hospital, London, United Kingdom.
Corresponding author: Linda Duska, MD, MPH, University of Virginia School of Medicine, P.O. Box 800712, Charlottesville, VA 22903; email: lduska@virginia.edu.
164
MULTIDISCIPLINARY CARE OF OLDER PATIENTS WITH ENDOMETRIAL CANCER
early-stage disease, specifically considering tumor recurrence and survival. When compared with patients
younger than age 75, the older patients had higher Federation Internationale de Gynecologie et dObstetrique
grade, higher stage tumors, more deep myometrial invasion,
and more lower uterine segment involvement. Tumor
recurrence rates were significantly higher in the older
patients (15% vs. 7%) and 5-year disease-specific survival
was 91% versus 96%. Interestingly, in this study, age alone
was not found to be a specific independent predictor of
recurrence or disease-specific survival. However, other
studies have demonstrated that cancer-specific mortality is higher in older women.6 In the robotic study from
Vaknin et al7 (in which all patients underwent hysterectomy and pelvic node dissection), the elderly group of
patients (age 70 and older) had a higher rate of advanced
disease (39% vs. 18.7%; p = .04); of the eight patients in
the series who had positive nodes, six were older than
age 80.
The limited available data suggest that age alone should
not be a sole decision factor in prescribing treatment,
surgical and otherwise, for older women with endometrial
cancer. Instead, older patients with cancer should be
assessed for their fitness to receive any type of endometrial
cancer treatments. For older patients with (or without)
cancer, a comprehensive geriatric assessment (CGA) can
evaluate their fitness for treatment in much more detail. The
model proposes that older patients (regardless of their age)
who are fit should be offered standard-of-care treatment of
their endometrial cancer, whereas those who are frail require more extensive discussion about the risks of standard
cancer treatment. Involvement of the patients primary care
provider and/or geriatrician may assist in obtaining a better
understanding of both her overall fitness for treatment and
her goals of care.
In this article, we will discuss the approach to the older
woman with endometrial cancer, taking into account the
complexities of caring for a geriatric population with cancer.
KEY POINTS
Endometrial cancer is the most common gynecologic

cancer among older patients with cancer.
Older women with endometrial cancer are more likely
to be diagnosed with poor prognosis disease (higher
stage or high-risk histology) and more likely to die of
disease compared with younger patients.
Older patients with endometrial cancer are less likely to
be offered surgical therapy, retroperitoneal node
dissection, and adjuvant therapy.
Older patients should not be denied treatment (surgery
or adjuvant therapy) based on age alone.
Involving geriatricians and/or performing
comprehensive geriatric assessment may aid in
treatment decisionmaking.
We will discuss the holistic approach toward older patients

with cancer (as a part of person-centered care), the role and
outcome of surgery in older patients with endometrial
cancer, radiation therapy considerations in these patients,
and, finally, how CGA can be a useful tool for risk stratification, management, and predicting outcomes of older
patients with cancer across the cancer care continuum.
Although a holistic approach such as the one described
below would likely benefit all patients considering cancer
care, it will clearly benefit the geriatric patient who may
suffer untoward effects of an overly aggressive treatment
approach.
HOLISTIC APPROACH TO OLDER PATIENTS WITH

CANCER
Person-centered care is a novel model of care that places
emphasis on the patient as a whole rather than just a
biomedical approach.8 Some of the major principles and
values of person-centered care include caring for the whole
person through a holistic approach, developing an individualized model of care, and understanding the patients
psychologic, social, and cultural complexity.9 To reach these
goals, person-centered care proposes to develop and enhance the connection between medical care and supportive
services and to provide multidisciplinary and team-based
care.
Older patients with cancer will benefit significantly from
person-centered care as a result of the increased complexity
of their care, their unmet needs, and the aging-related issues. The approach begins with CGA. Over the past 2 decades, investigators have developed CGA specific to the
cancer setting. Although there is no one standard instrument
to perform CGA,10 it is important for cancer care providers to
be familiar with the components of CGA (Fig. 1) and some of
the more common instruments, the likelihood of patients
having abnormal findings, the practical implications of such
findings, and the proper solutions (Table 1).
SURGERY IN THE OLDER WOMAN WITH

ENDOMETRIAL CANCER: SAFETY AND
FEASIBILITY
The goals of any medical or surgical intervention in the older
age group should be to maintain or maximize the potential
life span, maximize independent function, relieve suffering,
and maintain dignity. Although most modern series suggest
that surgery is feasible in the older woman, it is not without
morbidity. The older patient with endometrial cancer,
variably defined with respect to age parameters, appears to
suffer more postoperative morbidity than her younger
counterpart, regardless of surgical approach. There is also
morbidity from general anesthesia, required for all minimally invasive procedures (but potentially avoidable in the
vaginal approach, which may be performed in some cases
under regional anesthesia). Physiologic changes associated
with increasing age must be considered, including reduced
ability to increase cardiac output, decreased capacity of the
165
FIGURE 1. Components of a Comprehensive Geriatric Assessment
respiratory system with reduced lung elastic recoil and increased chest wall stiffness, decreased functional reserve,
and skin changes that may impact wound healing. These agerelated physiologic changes along with the medical
comorbidities that accompany aging must be carefully
considered when planning a surgical procedure.
In the relatively recent past, surgery for endometrial
cancer was traditionally performed via laparotomy, allowing
for complete removal of the uterus, cervix, and bilateral
tubes and ovaries along with a complete retroperitoneal
lymphadenectomy. With improvement in optics and
instrumentation, a minimally invasive approach became
more acceptable, and in fact became standard of care with
the publication of LAP2, a Gynecology Oncology Group
(GOG) study that compared oncologic outcomes between
laparotomy and laparoscopy in women with endometrial
cancer.11 The approval of the robotic system in the United
States in 2005, with its increased magnification, threedimensional optics, and wristed instruments, resulted in
an increased percentage of endometrial cancer procedures
performed via a minimally invasive approach. Finally, the
American Congress of Obstetrics and Gynecology has endorsed the vaginal approach as the optimal approach for
benign hysterectomy; although this method does not always
allow removal of the tubes and ovaries and never allows
assessment of the retroperitoneal lymph nodes, it may be an
appropriate alternative for a patient who is too frail for an
abdominal operation and general anesthesia.
Several retrospective studies have addressed the safety of
surgery for endometrial cancer in the older woman. DeMarzi
et al12 analyzed operability, perioperative morbidity, and
mortality in 124 women age 65 and older who were undergoing either open laparotomy or vaginal approach for
hysterectomy. For statistical purposes, the authors divided
the patients into younger than age 75 and 75 or older. Not
surprisingly, the vaginal approach was used more often in
the older patient group. Overall perioperative complication
166
rate was 13.1%, with the older group more likely to have
postoperative complications (23.1% vs. 8.5%). There were
no postoperative deaths. In a logistic regression model, age
75 or older, chronic lung disease, and the performance of
lymphadenectomy (discussed further below) correlated
with higher probability of perioperative morbidity. The
authors concluded that older patients should be offered
surgical treatment given the survival advantage associated
with surgery.
Other authors have also demonstrated the safety of
hysterectomy with or without lymphadenectomy in the
older population of patients. These studies are discussed
further below. It is notable, however, that in most of these
studies, the perioperative complication rate for the older
group of patients was higher than that of the younger group.
Perhaps more importantly, by their retrospective nature and
study design, these studies only report those women who
were felt to be good surgical candidates; they do not provide
information regarding patients who were not offered surgery as a primary approach to cancer treatment.
Laparoscopy
Laparoscopy is the preferred surgical technique to laparotomy for endometrial cancer for multiple reasons, including shorter hospital stay, shorter recovery period, less
discomfort (and therefore less use of postoperative narcotic
medications), and potentially improved quality of life. In
contrast to these benefits, laparoscopy requires longer
operating room (OR) time and steep Trendelenburg positioning, aspects of surgery that may be particularly difficult
for the older patient.
The LAP2 study was the first randomized study in the
United States to compare open laparotomy to laparoscopy
for the surgical treatment of endometrial cancer and
was a major impetus for the transition to minimally invasive
surgery for endometrial cancer in the United States.11 The
study was sponsored by the GOG, and patients were enrolled
TABLE 1. Instruments and Practice Implication of Comprehensive Geriatric Assessment Components

Likelihood of
Abnormality
Component
Instrument
Description of Instrument
Functional
Activity
Activities of Daily
Living
Assesses patients ability to do bathing, dressing, 15%20%

grooming, feeding, walking, and bladder and
bowel control
Practical Use
Solution
Could lead to malnutrition, difficulty to present to health

care provider office
Assessing caregiver ability to

take care of patients
needs
Involving social worker for
further support
Instrumental Activi- Assesses patients ability for telephone use, doing Up to 50%52
ties of Daily Living
laundry, shopping, preparing meals, doing
housework, handling own medications, handling money and finances, and transportation
Falls
Number of falls in the past 6 or 12 months
Up to 25%50
Could lead to malnutrition, lack Assessing caregiver ability to

take care of patients
of compliance with medicaneeds
tions, difficulty to present to
health care provider office
Involving social worker for
further support
May lead to bone fracture,
leading to interruption in
cancer care
Referral to physical therapy

Further exploration of the
context of falls
May need to avoid neurotoxic

chemotherapy agents if falls
are a result of neuropathy
Assessment of patients
medication list for
sedatives
Use of assistive
device
Not applicable
Up to 25%53
May be suggestive of patients Referral to physical therapy

history of falls, imbalance, or
fear of falling
Sensory deficit
(hearing, vision)
Observation, physical exam
Hearing deficit
$ 33%54
Could lead to decline in quality of Referral to audiologist and/or

ophthalmologist
life and cognitive function;
patient may not hear the
medical instructions properly
May need to avoid ototoxic
chemotherapy agents
Geriatric Depression
Emotional
Scale55,56
Well-being
(Depression)
Emotional
Well-being
(Distress)
It has 30-, 15-, and four-item versions
10%65%57
Patients with depression could Explore the reasons for

be at higher risk for receiving
depression
nondefinitive treatment and
worse overall survival58,59
Hospital Anxiety and Seven items related to depression and seven

Depression
items related to anxiety
Scale62
It increases the risk of being

nonadherent to medical
treatment60
Patient Health
It has nine-, four-, and two-item versions
Questionnaire63,64
Could be suggestive to multiple If depression is result of lack

other geriatrics deficits61
of proper social support,
refer to social worker
Distress
Thermometer65
Patient rates his/her distress level over the past About 40%66
two weeks from 010. Score of $ 4 is suggestive of high level of distress.
Refer to psychologist/
psychiatrist
It is usually suggestive of unmet More and better patientneeds66

physician communication
could lower the distress
level67
Referral to social worker,
mental health service, and/
or chaplaincy service68
Social Support
Medical Outcome
Study-Social Support Survey6971
Available in 19-, 12-, and four-item versions

The 19- and four-item versions measure four
social support domains: emotional, tangible,
affectionate, and positive social interaction.
Multidimensional
12-item questionnaire that categorizes patients
Scale of Perceived
social support into support he/she receives
76
Social Support
from family, friends, and significant other
Nutritional
Status
Mini-Nutritional
Assessment77
The exact likelihood of abnormality is

unknown
Could be associated with more Referral to social worker

psychosomatic symptoms72,73
might be needed
Could be associated with worse Empower the limited but

overall survival74,75
available social support for
the patient
Available in six-item questionnaire with the score Up to 65% could It could be multifactorial (disof 014
be at risk for
ease progression, fatigue, almalnutrition78
tered taste, inability to
prepare meals, etc.)
Score of 811 is suggestive of being at risk for
malnutrition, and score of # 7 is suggestive of
malnutrition. The longer version includes
additional questions and the score ranges from
030. Score of 1723.5 is suggestive of being
at risk for malnutrition, and score of # 17 is
suggestive of being malnourished.
Referral to nutritionist
Patients with malnutrition have Explore other reasons for

worse overall survival79
malnutrition besides cancer progression and/or
side effects from
chemotherapy
Continued
167
TABLE 1. Instruments and Practice Implication of Comprehensive Geriatric Assessment Components (Contd)
Component
Instrument
Polypharmacy/ Beers criteria80

Potentially
Inappropriate
Medication
Polypharmacy
Description of Instrument
Likelihood of
Abnormality
Practical Use
Solution
List of medications that could harmful for older Could be as

Could lead to prolonged
high as 50%81
patients. The recommendations for harmful
hospitalization82
medications are categorized into avoid, and
use with caution.
Referral to pharmacist
Taking $ 5 medications is considered

polypharmacy83
Review medication list
Could lead to drug-drug interaction, noncompliance, and

increase in cost51
Taking . 10 is considered excessive

polypharmacy47
Discontinue nonessential
medications, especially in
patients with limited life
expectancy
Check for drug-drug
interaction
Cognitive
Status
Mini-Cog84
Patients could be at higher risk

Short instrument that takes about 2 to 3 minutes. About 20%,78
for delirium during hospital
It includes three-word recall and clock
but it varies by
stay43,85
drawing.
age group
Mini-Mental Status
Examination
(MMSE)86
Administration of the test takes longer. It assesses cognitive domains of orientation, registration, attention and calculation, recall and
language.
Montreal Cognitive
Assessment87
Administration of the test takes longer than

MMSE. It assesses visuospatial, naming,
memory, attention, language, abstraction,
delayed recall, and orientation of patients.
Referral to geriatrician, neurologist, psychiatrist, or

neuropsychologist
Blessed Orientation- Takes 10 minutes to administer. It has 26 questions and tests orientation, long-term memMemory-Concenory, recall, and concentration.
tration test88
Comorbidity
Charlson Comorbidity Index89
It is a method of categorizing comorbidities (a

total of 22 conditions) based on the International Classification of Diseases, which can
predict the 10-year mortality
Adult Comorbidity
Evaluation 2793
It is a validated chart-based comorbidity

instrument
Varies by age
Impacts overall survival90
Referral to primary care provider/geriatrician for optimizing comorbid condition

management
It may negatively impact compliance with cancer

treatment91
Each individual comorbid condition may have negative outcome on survival92
from multiple sites across the United States. Notably, entry

into LAP2 required a laparoscopic lymphadenectomy as part
of the surgical staging procedure. Thirty-one percent of
patients in this study were age 70 or older, and 6.6% were
80 or older. Conversion to laparotomy occurred in 25.8%
of patients. Failure to successfully complete laparoscopy
was greater with increasing age (odds ratio 1.27; 95% CI,
1.141.42 for a 10-year increase in age), but the reasons for
this finding are not clear. Additionally, operating time was
significantly longer in the laparoscopy arm (204 vs. 103
minutes). The longer operating time is significant in the older
age group, in part because longer duration of anesthesia may
be associated with postoperative delirium, particularly in
patients with a pre-existing cognitive defect. There was
statistically significantly better quality of life across many
parameters in the laparoscopy arm at 6 weeks, but age was
not specifically considered as a factor for quality of life.13
Multiple single-institution studies have suggested that a
laparoscopic approach is safe in the older patient. Two
Italian studies found that laparoscopy was safe and feasible
and associated with lower blood loss, shorter hospital stay,
and lower complication rates. 14,15 It should be noted,
168
however, that in the study from Bogani et al,15 lymphadenectomy was not generally performed in the oldest patients.
In contrast, in the retrospective review of women age 65 and
older from Oklahoma, a more aggressive approach to
lymphadenectomy was undertaken. Sixty-seven women
underwent either laparoscopic hysterectomy or laparoscopically assisted vaginal hysterectomy with lymphadenectomy. Laparoscopy was successful in 78% percent of
cases, and a complete node dissection was successfully
performed in all patients. Of note, the laparoscopic group
had a longer OR time than the comparison laparotomy group
(236 vs. 148 minutes). Despite the longer OR time, the
authors note there was no subsequent increase in morbidity
and mortality.16
Robotic Surgery
Robotic surgery, approved in the United States in 2005,
provides advantages over traditional laparoscopy, but there
are also patient-related drawbacks: in particular, once the
robot is docked, the patient must remain in steep Trendelenburg position. Prolonged Trendelenburg potentially increases the potential risk of blindness in patients
suffering from moderate- or high-pressure glaucoma17 and

may particularly impact the older patient with respect to
ability to ventilate and cardiac output.
Despite these concerns, several series suggest that robotic surgery is safe and effective in the older patient, albeit
with a possible increased perioperative morbidity.1820 The
series from Guy et al, 18 for example, demonstrated increased rates of perioperative surgical and medical complications in the older age group with robotic surgery when
compared with younger patients with the same procedure.
Specifically, older patients had higher rates of perioperative surgical (8.3% vs. 5.2%; p , .001) and medical
(12.3% vs. 6.7%; p , .001) complications, longer length of
stay (2.00 vs. 1.67 days; p , .001), and lower rates of
discharge to home (88.8% vs. 96.8%; p , .001). As age
increased, perioperative surgical and medical complications also increased in a linear fashion. In contrast, in the
series of 41 women 70 and older from the series by Vaknin
et al,7 older and younger patients had the same rate of
postoperative complications. It is clear, however, that the
robotic approach is preferable to the open approach in the
older population. When compared directly to laparotomy,
the robotic approach was preferable for women age 70 or
older, with a decrease in complication rates, surgical blood
loss, and hospital stay.19
Lymphadenectomy
The role of lymphadenectomy in endometrial cancer continues to be one of some controversy, particularly in the
older patient. There is no doubt that adding lymph node
dissection will increase operating time as well as the potential for perioperative morbidity. In the series by de Marzi
et al,12 performance of lymphadenectomy significantly increased the rate of postoperative complications, but other
studies suggest that lymphadenectomy is safe in the older
patient.16,20 Although all patients in the study by Scribner
et al16 were staged, this practice led to a high conversion rate
to laparotomy and increased OR time. Other studies have
confirmed that staging procedures in elderly patients are
associated with a significantly longer operating time.21
Further studies are needed to weigh the benefits of lymphadenectomy on overall survival versus the potential
complications in this group of patients.
Surgery for the Older Patient With Cancer: Is There a

Bias?
Population database studies suggest that there is a selection
bias with respect to performing surgery for older patients,
both for hysterectomy and lymph node dissection. With
respect to hysterectomy specifically, two different SEER
analyses demonstrated that older patients with endometrial
cancer were less likely to have a hysterectomy.6,22 In the
study from Ahmed et al,22 a survival analysis was done within
each age group to compare the endometrial cancer-specific
survival rate of those who received cancer-directed surgery
(hysterectomy at a minimum) to those who did not. They
found a significant reduction in the hazard ratio (HR) for
cancer-specific death when cancer-directed surgery was

undertaken within each age category, adjusted for radiation
therapy, histology, and stage. Rauh-Hain et al23 performed a
similar analysis using the national cancer database, although
this group only considered women with advanced stage
(stage III or IV) disease and also demonstrated that elderly
women were less likely to be treated with surgery. Patients
younger than age 55 had surgery more frequently compared
with patients age 75 to 84 (97.2% compared with 95.8%;
p , .001) and age 85 or older (97.2% compared with 94.8%;
p , .001) and a higher rate of lymph node dissection (78.7%
compared with 70.5%, p , .001; and 78.7% compared with
59.5%, p , .001, respectively). In a multivariate analysis,
which took into account medical comorbidities as well as
adjuvant therapy, older women also had a higher risk of
death when compared with women younger than age 55
(women age 7584, HR 2.38; 95% CI, 2.142.65; women age
85 or older, HR 3.16; 95% CI, 2.763.61).
In contrast to the population-based studies above, most
single-institution studies from tertiary care centers, reported in large part by gynecologic oncologists, suggest that
all patients should be offered surgery, regardless of age.
These types of retrospective reviews represent a selection
bias of sorts and by study design do not account for those
women who were never offered surgery. They do suggest,
however, that gynecologic oncologists may be more likely to
offer surgery to older women with endometrial cancer than
other providers.
It is also important to note that database studies by their
very nature cannot provide us with data regarding medical
comorbidities affecting the elderly or the desires of the
patient and her family regarding goals of care. In the study by
de Marzi et al,12 for example, three women in their late 80s
declined surgery; this patient decision to decline an offered
surgery would not be captured in a population-based database study. Nevertheless, the studies point to an alarming
trend, confirming the national bias toward not offering
surgery to the older patient with endometrial cancer, regardless of fitness.
Both population-based studies and institutional studies
suggest an age bias regarding lymph node dissection. In the
SEER study from Wright et al,6 older women were less likely
to have lymph node dissection performed when compared
with women age 65 to 69 (43.5% for women age 6569,
37.5% for age 8084, and 24.8% for women 85 and older).
Lowery et al20 reported similar results in their SEER analysis,
with lymph node dissection performed in 41% of the age 80
or older population. Institution-based studies from Europe
reveal similar results. For example, in the study by de Marzi
et al,12 lymphadenectomy was performed more often in
younger patients (84.5% in women younger than 75 vs.
15.5% in women 75 and older). Bogani et al15 noted that we
usually omit execution of retroperitoneal staging in this
group. In contrast, single-institution studies from the
United States report higher node staging rates in older
patients. This aggressive approach to lymphadenectomy,
however, is not without morbidity.
169
In summary, surgery should be considered as a first step

for an older patient, if that patient is deemed medically fit. A
minimally invasive approach appears to be safe and offers
the best postoperative recovery despite longer operating
times. The role for complete lymphadenectomy is still
controversial, particularly in the population of older women,
and should be individualized; however, the plans for adjuvant therapy should also be considered when planning the
surgical procedure.
RADIOTHERAPY FOR ENDOMETRIAL CANCER IN

THE OLDER PATIENT
Radiotherapy remains an essential component of the
treatment of both early and locally advanced endometrial
cancer. Over the past decade, data from large randomized
controlled trials have established the place of radiation in
adjuvant treatment of uterine cancers. Risk groups based on
clinicopathological factors have been devised to identify
those patients who might benefit most from additional
treatment (Table 2). In the older patient, in whom age itself
is a risk factor, the use of pelvic radiotherapy needs to be
considered carefully, balancing the advantage of high local
control rates with potential toxicity and lack of survival
benefit.24,25
Although initially slow to integrate new three-dimensional
radiotherapy planning techniques into routine treatment,
the gynecology oncology community has now embraced the
technology, and intensity modulated radiotherapy (IMRT),
volumetric modulated arc therapy (VMAT), and imageguided radiotherapy are in widespread use. IMRT uses
multiple beams divided into smaller beamlets of varying
dose intensity, allowing more accurate shaping of the delivered dose to the target. This technique has been shown in
theoretical planning studies to reduce the dose to nearby
organs such as bowel and bladder by up to 60%26 and
translates to a reduction in both acute and late radiation
toxicity.27,28 The reduction in toxicity is likely to be especially
beneficial in an elderly population who have a higher incidence of underlying bowel and bladder problems with
urgency and incontinence often present.
One of the downsides of IMRT is the high number of
monitor units needed to deliver the plan compared with
old-fashioned conventional treatment (estimated to be
two to three times higher), which leads to an increase in
lower dose radiation to other parts of the body. The
long-term effect of IMRT is not known, but the potential
increased risk of second malignancy cannot be ignored

and is of particular significance when even older patients
may be expected to have a life expectancy exceeding
10 years.
A newer technique that provides similar highly conformal
radiation fields uses a continuous rotation of the radiation
source through 360 degrees. The advantage of this arc
technique is twofold. First, the speed of delivery is much
faster compared with IMRT. A typical pelvic field of two full
arcs might be delivered in less than 5 minutes. This is more
comfortable for patients, particularly the elderly, who
sometimes find it difficult to hold a full bladder. It may
thereby reduce the potential for inaccuracy as a result of
movement and also allow faster throughput in the working
day. Second, the number of monitor units required to deliver
arc treatment is approximately two-thirds that of IMRT,
leading to a lower total dose delivered to the patient.
Pelvic insufficiency fractures occur when the elastic resistance of bone is reduced and unable to withstand the
stress of weight bearing and are most frequent in older
women with postmenopausal osteoporosis. Pelvic radiation
is also a recognized cause of pelvic insufficiency fractures
and, although not always symptomatic, may give rise to
significant pain. VMAT and IMRT, with careful planning to
avoid areas such as the sacrum, may reduce the incidence of
fractures, which is estimated to occur in up to 50% of patients who undergo pelvic radiotherapy.29,30
The integration of three-dimensional imaging into radiotherapy and the use of intensity modulated radiation to
produce radiation treatment plans that maximize dose to
tumor while minimizing dose to normal tissues has been a
major step forward in radiotherapy. But it has also revealed
the uncertainties that exist in affirming that we are actually
treating what was intended. With shaping of fields and the
steep dose gradients around target, it has become more
important than ever to avoid a geographical miss of the
target either by not correctly identifying the target or
through internal organ movement. Delivering this highly
focused radiotherapy places the onus on the radiation oncologist to ensure accurate delineation of the radiation
target. Courses and online training aids have been developed, and atlases, comprising serial axial CT or MRIs with
anatomy clearly labeled, are now widely available to help the
oncologist.31,32 In particular, variation in bladder filling and
rectal dimensions, as a result of either gas or feces, may
have a significant impact on the position of the vagina and, if
present, the uterus and cervix. In elderly patients in whom
TABLE 2. Suggested Radiotherapy by Risk Group

Low Risk
Intermediate Risk
High Risk
IA G1, G2, 1B G1 without LVSI
1A G3, IBG1 or 2, with LVSI
1A G3 LVSI, IB G3, II, III
No treatment94
Vault brachytherapy95
Pelvic radiotherapy94,96
Risk of relapse is 10%
Brachytherapy will decease vaginal recurrence rate to 2%
Survival exceeds 95%

Abbreviation: LVSI, lymphovascular space invasion.
170
constipation and or bladder weakness is often an issue,

careful consideration is needed as to how best to ensure
consistency in rectal and bladder dimensions on a day-to-day
basis. Daily use of enemas, which is advocated by many, is
not always appropriate for older patients who may find it
physically challenging to administer the enema and manage
the result. For all patients undergoing pelvic radiotherapy,
whether old or young, choosing an appropriate margin
around the clinical target volume has become as critical to
success as actually delineating the original target and must
be sufficient to avoid missing the target but not so large as to
lose the advantage of reducing the volume of normal tissue
irradiated.
In the past, simple kilovoltage imaging was the mainstay of
treatment verification. Its major limitation is that it only
displays bony anatomy. Although this is helpful in providing a
rapid check that the patient is correctly aligned, it gives no
indication as to how the target is positioned relative to the
radiation field. More sophisticated on-board CT imaging
features almost all modern linear accelerators, allowing realtime visualization of not just bony landmarks but also other
internal structures and even tumors.
Many studies have been published looking at movement
of pelvic organs during a course of radiotherapy and
guidelines suggested on suitable margins to account for dayto-day change in bladder and rectal filling.33,34 These margins, which are based on population data, are necessarily
generous, as allowance must be made for the largest movement possible. So, although they are a practical way of ensuring target coverage, they potentially lead to larger treatment
volumes and in particular more normal tissue likely to be
irradiated.35
Daily individualized image-guided adaptive radiotherapy is
the best way to ensure both accuracy of delivery and reduction of total volume irradiated. This has been made
possible with the ready availability of cone beam CT scanning
on linear accelerators, allowing a CT scan of the radiation
field to be obtained in just a few minutes immediately before
radiation that can then be merged and compared with the
original planning scan. Daily variation in organ movement
can be partly accounted for by creating a so-called library of
plans using differing margins around the target. Each day a
plan can be selected that is the best fit. This has been used
with some success in bladder cancer radiotherapy.36,37
Taking a daily cone beam CT with real-time review and
selecting a suitable plan on every patient requires time and
highly trained individuals. Most busy radiotherapy departments are short of one or both and cannot offer this approach
for all.
Older age confers an additional risk factor in uterine
cancer, and careful consideration should be given to offering
adjuvant treatment to this group. Modern radiation techniques have given us treatment that is better tolerated with
improved delivery with significantly less severe acute and
late toxicity. These techniques combined with good medical,
nursing, and psychosocial support should mean that age
need not necessarily be a barrier to radiation.
THE UTILITY OF COMPREHENSIVE GERIATRIC

ASSESSMENT ACROSS ENDOMETRIAL CANCER
CONTINUUM
An approach toward whole care for the older patient with
cancer is critical for delivering best quality of care. Data on
older patients with endometrial cancer are just emerging;
however, data on older patients with various cancers in
different phases of their treatment prove the importance
of such an approach. Although health care providers rely
on standard methods of assessing functional status (e.g.,
Karnofsky performance status or American Society of Anesthesiologists classification), CGA has been shown to be
complementary to standard assessment.38 Shared decisionmaking among patients, families, and health care providers
may be impacted in up to 40% of cases as a result of new
information through performing CGA. Comprehensive geriatric assessment can predict surgical complications, length
of stay, and hospital disposition following surgery, hospital
readmission, and overall survival.3941 Older patients with
cancer with cognitive deficits could be especially at high risk
for delirium, which can lead to prolonged hospital stay and
possibly poorer overall survival.40,42,43 Despite these data,
only 6.4% of surgeons use CGA in their preoperative evaluation, and only one out of three surgeons collaborates with
geriatricians to optimize care of older patients with cancer
preoperatively.44
Various models of collaborative care between geriatricians
and/or geriatric-allied health care professional and surgical
services have been proposed.45 Although the effectiveness
and sustainability of these models must be further examined, the importance of performing CGA in the preoperative
evaluation and acting proactively based on findings by incorporating different supportive services (e.g., physical
therapy, nutrition) is clear. As a result of these findings, the
American College of Surgeons and the American Geriatrics
Society have issued a best practice guideline to incorporate
geriatric assessment in the preoperative assessment of older
patients undergoing surgery to fully assess older patients for
their fitness to undergo surgery.46
For older patients with recurrent or metastatic endometrial cancer in need of chemotherapy, CGA has significant
value. Two major chemotherapy toxicity calculators (Cancer
and Aging Research Group47 and the Chemotherapy Risk
Assessment for High-Age Patients48) have developed based
on the concept of CGA. Although these two calculators can
be useful for oncologists, patients, and their families to
understand the risk of chemotherapy toxicity in older patients with cancer, the exact intervention based on this
finding is not clear. One approach would be to modify the
chemotherapy regimen for a patient at high risk for developing chemotherapy toxicity; however, another approach could be to involve and empower the primary care
provider/geriatrician to monitor these patients more intensively for early and accurate detection of chemotherapy
toxicity and acting promptly and adequately to manage
these toxicities. Although these calculators are useful to give a
171
broad overview of patients risk for chemotherapy toxicities, it

is important to note any geriatrics deficit could be associated
with adverse outcomes and needs proper intervention.47,49
Moreover, CGA can help with selecting the proper chemotherapy agent. For example, selecting a neurotoxic
chemotherapy agent for treatment of metastatic endometrial cancer when the patient has fallen multiple times as a
result of neuropathy secondary to poorly controlled diabetes may not be in the best interest of an older patient
with metastatic endometrial cancer. The need for incorporating CGA in daily oncology practice becomes much
more important when one study showed that only 10% of
patients who reported falls during CGA (performed as a
research tool) had a documented history of falls in their
medical chart by medical oncologists.50
For older patients with endometrial cancer who require
oral hormone therapy, the issue of polypharmacy needs to
be addressed, as polypharmacy could be associated with
lower compliance rate and drug-drug interaction.51 Referral
to a pharmacist and/or primary care provider/geriatrician to

reassess the patients medication list and to discontinue
medications that are not absolutely necessarily could increase the compliance rate of taking hormonal therapies for
endometrial cancer. Patients with impaired cognitive status
with no reliable social support may not be the best candidates for oral treatments. Like preoperative care, life expectancy of older patients with recurrent or metastatic
endometrial cancer should be assessed, and the impact of
cancer treatment on both longevity and quality of life should
be discussed with the patient and his/her family.
In summary, for older patients with cancer in general and
in older patients with endometrial cancer in particular, a
whole-patient approach should be part of standard of care.
Comprehensive geriatric assessment is a useful tool for such
an approach, and deficits found through CGA can be managed
by involving various supportive services as well as empowering primary care providers/geriatricians in comanagement
of these patients.
References
1. National Cancer Institute. SEER Stat Fact Sheets: Endometrial Cancer. http://
seer.cancer.gov/statfacts/html/corp.html. Accessed January 22, 2016.
2. United Nations. World Populating Ageing 1950-2050. http://www.un.org/
esa/population/publications/worldageing19502050/pdf/90chapteriv.pdf.
3. Arnold M, Karim-Kos HE, Coebergh JW, et al. Recent trends in incidence
of five common cancers in 26 European countries since 1988: Analysis of
the European Cancer Observatory. Eur J Cancer. 2015;51:1164-1187.
4. Mariani A, Webb MJ, Keeney GL, et al. Low-risk corpus cancer: is
lymphadenectomy or radiotherapy necessary? Am J Obstet Gynecol.
2000;182:1506-1519.
5. Gayar OH, Robbins JR, Parikh K, et al. Hysterectomy for uterine adenocarcinoma in the elderly: tumor characteristics, and long-term
outcome. Gynecol Oncol. 2011;123:71-75.
6. Wright JD, Lewin SN, Barrena Medel NI, et al. Endometrial cancer in the
oldest old: Tumor characteristics, patterns of care, and outcome.
Gynecol Oncol. 2011;122:69-74.
7. Vaknin Z, Perri T, Lau S, et al. Outcome and quality of life in a
prospective cohort of the first 100 robotic surgeries for endometrial
cancer, with focus on elderly patients. Int J Gynecol Cancer. 2010;20:
1367-1373.
8. Laine C, Davidoff F. Patient-centered medicine. A professional evolution. JAMA. 1996;275:152-156.
9. Kogan AC, Wilber K, Mosqueda L. Person-centered care for older adults
with chronic conditions and functional impairment: a systematic literature review. J Am Geriatr Soc. 2016;64:e1-e7.
11. Walker JL, Piedmonte MR, Spirtos NM, et al. Laparoscopy compared
with laparotomy for comprehensive surgical staging of uterine cancer:
Gynecologic Oncology Group Study LAP2. J Clin Oncol. 2009;27:
5331-5336.
12. De Marzi P, Ottolina J, Mangili G, et al. Surgical treatment of elderly
patients with endometrial cancer ($ 65 years). J Geriatr Oncol. 2013;4:
368-373.
172
13. Kornblith AB, Huang HQ, Walker JL, et al. Quality of life of patients
with endometrial cancer undergoing laparoscopic international
federation of gynecology and obstetrics staging compared with
laparotomy: a Gynecologic Oncology Group study. J Clin Oncol.
2009;27:5337-5342.
14. Ciavattini A, Di Giuseppe J, Cecchi S, et al. Gynecologic laparoscopy
in patients aged 65 or more: feasibility and safety in the presence of
increased comorbidity. Eur J Obstet Gynecol Reprod Biol. 2014;175:
49-53.
15. Bogani G, Cromi A, Uccella S, et al. Perioperative and long-term outcomes of laparoscopic, open abdominal, and vaginal surgery for endometrial cancer in patients aged 80 years or older. Int J Gynecol Cancer.
2014;24:894-900.
16. Scribner DR Jr, Walker JL, Johnson GA, et al. Surgical management of
early-stage endometrial cancer in the elderly: is laparoscopy feasible?
Gynecol Oncol. 2001;83:563-568.
17. Rupp-Montpetit K, Moody ML. Visual loss as a complication of nonophthalmic surgery: a review of the literature. Insight. 2005;30:10-17.
18. Guy MS, Sheeder J, Behbakht K, et al. Comparative outcomes in older
and younger women undergoing laparotomy or robotic surgical staging
for endometrial cancer. Am J Obstet Gynecol. Epub 2015 Nov 17.
19. Lavoue V, Zeng X, Lau S, et al. Impact of robotics on the outcome of
elderly patients with endometrial cancer. Gynecol Oncol. 2014;133:
556-562.
20. Lowery WJ, Gehrig PA, Ko E, et al. Surgical staging for endometrial
cancer in the elderly - is there a role for lymphadenectomy? Gynecol
Oncol. 2012;126:12-15.
21. Giannice R, Susini T, Ferrandina G, et al. Systematic pelvic and aortic
lymphadenectomy in elderly gynecologic oncologic patients. Cancer.
2001;92:2562-2568.
22. Ahmed A, Zamba G, DeGeest K, et al. The impact of surgery on survival of
elderly women with endometrial cancer in the SEER program from
1992-2002. Gynecol Oncol. 2008;111:35-40.
23. Rauh-Hain JA, Pepin KJ, Meyer LA, et al. Management for elderly women
with advanced-stage, high-grade endometrial cancer. Obstet Gynecol.
2015;126:1198-1206.
24. Creutzberg CL, van Putten WL, Koper PC, et al. Surgery and postoperative radiotherapy versus surgery alone for patients with stage-1
endometrial carcinoma: multicentre randomised trial. PORTEC Study
Group. Post Operative Radiation Therapy in Endometrial Carcinoma.
Lancet. 2000;355:1404-1411.
25. Keys HM, Roberts JA, Brunetto VL, et al; Gynecologic Oncology
Group. A phase III trial of surgery with or without adjunctive external pelvic radiation therapy in intermediate risk endometrial
adenocarcinoma: a Gynecologic Oncology Group study. Gynecol
Oncol. 2004;92:744-751.
26. Taylor A, Powell ME. Conformal and intensity-modulated radiotherapy
for cervical cancer. Clin Oncol (R Coll Radiol). 2008;20:417-425.
27. Mundt AJ, Mell LK, Roeske JC. Preliminary analysis of chronic gastrointestinal toxicity in gynecology patients treated with intensity-modulated
whole pelvic radiation therapy. Int J Radiat Oncol Biol Phys. 2003;56:
1354-1360.
28. Mundt AJ, Lujan AE, Rotmensch J, et al. Intensity-modulated whole
pelvic radiotherapy in women with gynecologic malignancies. Int J
29. Shih KK, Folkert MR, Kollmeier MA, et al. Pelvic insufficiency fractures
in patients with cervical and endometrial cancer treated with postoperative pelvic radiation. Gynecol Oncol. 2013;128:540-543.
30. Kwon JW, Huh SJ, Yoon YC, et al. Pelvic bone complications after radiation therapy of uterine cervical cancer: evaluation with MRI. AJR Am J
Roentgenol. 2008;191:987-994.
31. Small W Jr, Mell LK, Anderson P, et al. Consensus guidelines for delineation of clinical target volume for intensity-modulated pelvic radiotherapy in postoperative treatment of endometrial and cervical
cancer. Int J Radiat Oncol Biol Phys. 2008;71:428-434.
32. Taylor A, Rockall AG, Powell ME. An atlas of the pelvic lymph node
regions to aid radiotherapy target volume definition. Clin Oncol (R Coll
Radiol). 2007;19:542-550.
33. Chan P, Dinniwell R, Haider MA, et al. Inter- and intrafractional tumor
and organ movement in patients with cervical cancer undergoing radiotherapy: a cinematic-MRI point-of-interest study. Int J Radiat Oncol
Biol Phys. 2008;70:1507-1515.
34. Taylor A, Powell ME. An assessment of interfractional uterine
and cervical motion: implications for radiotherapy target volume
definition in gynaecological cancer. Radiother Oncol. 2008;88:
250-257.
35. Bondar ML, Hoogeman MS, Mens JW, et al. Individualized nonadaptive and online-adaptive intensity-modulated radiotherapy treatment
strategies for cervical cancer patients based on pretreatment acquired
variable bladder filling computed tomography scans. Int J Radiat Oncol
Biol Phys. 2012;83:1617-1623.
36. Lutkenhaus LJ, Visser J, de Jong R, et al. Evaluation of delivered dose
for a clinical daily adaptive plan selection strategy for bladder cancer
radiotherapy. Radiother Oncol. 2015;116:51-56.
37. Foroudi F, Pham D, Rolfo A, et al. The outcome of a multi-centre
feasibility study of online adaptive radiotherapy for muscle-invasive
bladder cancer TROG 10.01 BOLART. Radiother Oncol. 2014;111:
316-320.
38. Repetto L, Fratino L, Audisio RA, et al. Comprehensive geriatric assessment adds information to Eastern Cooperative Oncology Group
performance status in elderly cancer patients: an Italian Group for
Geriatric Oncology Study. J Clin Oncol. 2002;20:494-502.
39. Schmidt M, Neuner B, Kindler A, et al. Prediction of long-term mortality
by preoperative health-related quality-of-life in elderly onco-surgical
patients. PLoS One. 2014;9:e85456.
40. Large MC, Reichard C, Williams JT, et al. Incidence, risk factors, and
complications of postoperative delirium in elderly patients undergoing
radical cystectomy. Urology. 2013;81:123-128.
41. Badgwell B, Stanley J, Chang GJ, et al. Comprehensive geriatric assessment of risk factors associated with adverse outcomes and resource
utilization in cancer patients undergoing abdominal surgery. J Surg
Oncol. 2013;108:182-186.
42. Robinson TN, Raeburn CD, Tran ZV, et al. Postoperative delirium in the
elderly: risk factors and outcomes. Ann Surg. 2009;249:173-178.
43. Korc-Grodzicki B, Sun SW, Zhou Q, et al. Geriatric assessment as a
predictor of delirium and other outcomes in elderly patients with
cancer. Ann Surg. 2015;261:1085-1090.
44. Ghignone F, van Leeuwen BL, Montroni I, et al. The assessment and
management of older cancer patients: A SIOG surgical task force survey
on surgeons attitudes. Eur J Surg Oncol. 2016;42:297-302. Epub 2015
Dec 17.
45. Dale W, Chow S, Sajid S. Socioeconomic considerations and sharedcare models of cancer care for older adults. Clin Geriatr Med. 2016;32:
35-44.
46. Chow WB, Rosenthal RA, Merkow RP, et al; American College of Surgeons National Surgical Quality Improvement Program; American
Geriatrics Society. Optimal preoperative assessment of the geriatric
surgical patient: a best practices guideline from the American College of
Surgeons National Surgical Quality Improvement Program and the
American Geriatrics Society. J Am Coll Surg. 2012;215:453-466.
47. Hurria A, Togawa K, Mohile SG, et al. Predicting chemotherapy toxicity
in older adults with cancer: a prospective multicenter study. J Clin
Oncol. 2011;29:3457-3465.
48. Extermann M, Boler I, Reich RR, et al. Predicting the risk of chemotherapy toxicity in older patients: the Chemotherapy Risk Assessment Scale for High-Age Patients (CRASH) score. Cancer. 2012;
118:3377-3386.
49. Soubeyran P, Fonck M, Blanc-Bisson C, et al. Predictors of early death
risk in older patients treated with first-line chemotherapy for cancer.
J Clin Oncol. 2012;30:1829-1834.
50. Guerard EJ, Deal AM, Williams GR, et al. Falls in older adults with cancer:
evaluation by oncology providers. J Oncol Pract. 2015;11:470-474. Epub
2015 Jul 14.
51. Balducci L, Goetz-Parten D, Steinman MA. Polypharmacy and
the management of the older cancer patient. Ann Oncol. 2013;
24:vii36-vii40.
52. Milla n-Calenti JC, Tubo J, Pita-Ferna ndez S, et al. Prevalence of
functional disability in activities of daily living (ADL), instrumental
activities of daily living (IADL) and associated factors, as predictors
of morbidity and mortality. Arch Gerontol Geriatrics. 2010;50:
306-310.
53. Gell NM, Wallace RB, LaCroix AZ, et al. Mobility device use in older
adults and incidence of falls and worry about falling: findings from the
2011-2012 national health and aging trends study. J Am Geriatr Soc.
2015;63:853-859.
54. Walling AD, Dickson GM. Hearing loss in older adults. Am Fam Physician.
2012;85:1150-1156.
55. Burke WJ, Roccaforte WH, Wengel SP. The short form of the Geriatric
Depression Scale: a comparison with the 30-Item Form. J Geriatr
Psychiatry Neurol. 1991;4:173-178.
56. Almeida OP, Almeida SA. Short versions of the geriatric depression
scale: a study of their validity for the diagnosis of a major depressive
episode according to ICD-10 and DSM-IV. Int J Geriatr Psychiatry. 1999;
14:858-865.
57. Weiss Wiesel TR, Nelson CJ, Tew WP, et al; Cancer Aging Research Group
(CARG). The relationship between age, anxiety, and depression in older
adults with cancer. Psychooncology. 2015;24:712-717.
58. Goodwin JS, Zhang DD, Ostir GV. Effect of depression on diagnosis,
treatment, and survival of older women with breast cancer. J Am Geriatr
Soc. 2004;52:106-111.
173
59. Prasad SM, Eggener SE, Lipsitz SR, et al. Effect of depression on diagnosis, treatment, and mortality of men with clinically localized
60. DiMatteo MR, Lepper HS, Croghan TW. Depression is a risk factor for
noncompliance with medical treatment: meta-analysis of the effects of
anxiety and depression on patient adherence. Arch Intern Med. 2000;
160:2101-2107.
61. Canoui-Poitrine F, Reinald N, Laurent M, et al; ELCAPA Study Group.
Geriatric assessment findings independently associated with clinical
depression in 1092 older patients with cancer: the ELCAPA Cohort
Study. Psychooncology. 2016;25:104-111.
62. Herrmann C. International experiences with the Hospital Anxiety and
Depression Scalea review of validation data and clinical results.
J Psychosom Res. 1997;42:17-41.
63. Kroenke K, Spitzer RL, Williams JB. The Patient Health Questionnaire-2:
validity of a two-item depression screener. Med Care. 2003;41:
1284-1292.
64. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16:606-613.
65. Zwahlen D, Hagenbuch N, Carley MI, et al. Screening cancer patients
families with the distress thermometer (DT): a validation study. Psychooncology. 2008;17:959-966.
66. Hurria A, Li D, Hansen K, et al. Distress in older patients with cancer.
J Clin Oncol. 2009;27:4346-4351.
67. Zachariae R, Pedersen CG, Jensen AB, et al. Association of perceived
physician communication style with patient satisfaction, distress,
cancer-related self-efficacy, and perceived control over the disease. Br J
Cancer. 2003;88:658-665.
68. Holland JC, Andersen B, Breitbart WS, et al. Distress management. J Natl
Compr Canc Netw. 2010;8:448-485.
69. Moser A, Stuck AE, Silliman RA, et al. The eight-item modified Medical
Outcomes Study Social Support Survey: psychometric evaluation
showed excellent performance. J Clin Epidemiol. 2012;65:1107-1116.
70. Gjesfjeld CD, Greeno CG, Kim KH. A confirmatory factor analysis of an
abbreviated social support instrument - The MOS-SSS. Res Soc Work
Pract. 2008;18:231-237.
71. Jyrkka J, Enlund H, Korhonen MJ, et al. Polypharmacy status as an
indicator of mortality in an elderly population. Drugs Aging. 2009;26:
1039-1048.
72. Hughes S, Jaremka LM, Alfano CM, et al. Social support predicts inflammation, pain, and depressive symptoms: Longitudinal relationships
among breast cancer survivors. Psychoneuroendocrinology. 2014;42:38-44.
73. Eom C-S, Shin DW, Kim SY, et al. Impact of perceived social support on
the mental health and health-related quality of life in cancer patients:
results from a nationwide, multicenter survey in South Korea. Psychooncology. 2013;22:1283-1290.
74. Kroenke CH, Quesenberry C, Kwan ML, et al. Social networks, social
support, and burden in relationships, and mortality after breast cancer
diagnosis in the Life After Breast Cancer Epidemiology (LACE) study.
75. Steptoe A, Shankar A, Demakakos P, et al. Social isolation, loneliness,
and all-cause mortality in older men and women. Proc Natl Acad Sci.
2013;110:5797-5801.
76. Zimet GD, Dahlem NW, Zimet SG, et al. The multidimensional scale of
perceived social support. J Pers Assess. 1988;52:30-41.
77. Cereda E. Mini nutritional assessment. Curr Opin Clin Nutr Metab Care.
2012;15:29-41.
78. Bellera CA, Rainfray M, Mathoulin-Pelissier S, et al. Screening older
cancer patients: first evaluation of the G-8 geriatric screening tool. Ann
Oncol. 2012;23:2166-2172.
174
79. Aaldriks AA, van der Geest LGM, Giltay EJ, et al. Frailty and malnutrition predictive of mortality risk in older patients with advanced
colorectal cancer receiving chemotherapy. J Geriatr Oncol. 2013;4:
218-226.
80. American Geriatrics Society 2015 Beers Criteria Update Expert Panel.
American Geriatrics Society 2015 Updated Beers Criteria for Potentially
Inappropriate Medication Use in Older Adults. J Am Geriatr Soc. 2015;
63:2227-2246.
81. Nightingale G, Hajjar E, Swartz K, et al. Evaluation of a pharmacistled medication assessment used to identify prevalence of and associations with polypharmacy and potentially inappropriate medication
use among ambulatory senior adults with cancer. J Clin Oncol. 2015;33:
1453-1459.
82. Park JW, Roh J-L, Lee SW, et al. Effect of polypharmacy and potentially
inappropriate medications on treatment and posttreatment courses in
elderly patients with head and neck cancer. J Cancer Res Clin Oncol.
Epub 2016 Jan 7.
83. Turner JP, Jamsen KM, Shakib S, et al. Polypharmacy cut-points in older
people with cancer: how many medications are too many? Support Care
Cancer. Epub 2015 Oct 9.
84. Borson S, Scanlan J, Brush M, et al. The mini-cog: a cognitive vital signs
measure for dementia screening in multi-lingual elderly. Int J Geriatr
Psychiatry. 2000;15:1021-1027.
85. Korc-Grodzicki B, Root JC, Alici Y. Prevention of post-operative delirium
in older patients with cancer undergoing surgery. J Geriatr Oncol. 2015;
6:60-69.
86. Folstein MF, Folstein SE, McHugh PR. Mini-mental state. A practical
method for grading the cognitive state of patients for the clinician.
J Psychiatr Res. 1975;12:189-198.
87. Nasreddine ZS, Phillips NA, Bedirian V, et al. The Montreal Cognitive
Assessment, MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc. 2005;53:695-699.
88. Katzman R, Brown T, Fuld P, et al. Validation of a short OrientationMemory-Concentration Test of cognitive impairment. Am J Psychiatry.
1983;140:734-739.
89. Charlson M, Szatrowski TP, Peterson J, et al. Validation of a combined
comorbidity index. J Clin Epidemiol. 1994;47:1245-1251.
90. Bje CR, Dalton SO, Grnborg TK, et al. The impact of comorbidity on
outcome in 12 623 Danish head and neck cancer patients: a population based study from the DAHANCA database. Acta Oncol. 2013;52:
285-293.
91. Sgaard M, Thomsen RW, Bossen KS, et al. The impact of comorbidity on
cancer survival: a review. Clin Epidemiol. 2013;5:3-29.
92. Currie CJ, Poole CD, Jenkins-Jones S, et al. Mortality after incident
cancer in people with and without type 2 diabetes: impact of metformin
on survival. Diabetes Care. 2012;35:299-304.
93. Bang D, Piccirillo J, Littenberg B, et al. The Adult Comorbidity Evaluation27 (ACE-27) test: a new comorbidity index for patients with cancer.
ASCO Annual Meeting Proceedings. 2000; abstr 1701.
94. Kong A, Johnson N, Kitchener HC, et al. Adjuvant radiotherapy for stage I
endometrial cancer: an updated Cochrane systematic review and metaanalysis. J Natl Cancer Inst. 2012;104:1625-1634.
95. Nout RA, Smit VT, Putter H, et al; PORTEC Study Group. Vaginal
brachytherapy versus pelvic external beam radiotherapy for patients with endometrial cancer of high-intermediate risk (PORTEC-2):
an open-label, non-inferiority, randomised trial. Lancet. 2010;375:
816-823.
96. Maggi R, Lissoni A, Spina F, et al. Adjuvant chemotherapy vs radiotherapy in high-risk endometrial carcinoma: results of a randomised
trial. Br J Cancer. 2006;95:266-271.
GYNECOLOGIC CANCER
Symptom Management for the

Patient with Gynecologic Cancer
CHAIR
Linda Van Le, MD
Chapel Hill, NC
SPEAKERS
Deborah Mayer, RN, PhD, AOCN, FAAN
University of North Carolina at Chapel Hill School of Medicine
Chapel Hill, NC
Mary McCormack, BSc, MSc, PhD, MBBS, FRCR
University College Hospital
VAN LE AND MCCORMACK
Enhancing Care of the Survivor of Gynecologic Cancer:

Managing the Menopause and Radiation Toxicity
Linda Van Le, MD, and Mary McCormack, BSc, MSc, PhD, MBBS, FRCR
OVERVIEW
It is expected that there will be 290,000 cases of gynecologic cancers in 2016. Of these cancers, 60,000 will be endometrial and
22,000 will be ovarianthe two most common gynecologic cancers. Endometrial and ovarian cancers occur in menopausal
women with mean ages of 60 and 63, respectively. The majority of endometrial cancers are early stage, and 5-year survival is
considered good at upwards of 75%. For ovarian cancer, while survival rates have improved, the 5-year survival rate for the
most common stage (stage III) is 40%. Thus, a substantial number of patients with gynecologic cancer are menopausal, and
a significant number of patients are survivors, particularly of endometrial cancers. It will be important for survivors of
gynecologic cancers to receive care tailored to their needs as women and to mitigate gender-specific side effects of their
cancer treatment.
ccording to the National Cancer Institute, an individual is

considered a cancer survivor from the time of diagnosis
through the balance of his or her life. The American Society
of Clinical Oncology (ASCO) endorses the importance of
survivorship care, broadly defined as care tailored to the
survivor.1 ASCO defines four components of survivorship
care: prevention and detection of new cancers and recurrent
cancer, surveillance for recurrence or new other primary
cancers, intervention for long-term and late effects from cancer
treatment, and coordination between specialists and primary
care providers to optimize survivor care.2 In the United
Kingdom, the National Cancer Survivorship Initiative, an initiative of the 2007 cancer reform strategy, was created to
support those living with and beyond cancer. In this article,
we address two important aspects of survivorship care for
women with gynecologic cancer: management of menopause
needs and care for the patient who has received radiation.
MANAGEMENT OF MENOPAUSE FOR PATIENTS

WITH CANCER
Menopause is the cessation of menstrual periods that occurs
at a median age of 52. Menopause occurs because of atresia
and dissolution of the original number of millions of oocytes
found at birth. For 5% of women, menopause may occur
after the age of 55 or between the ages of 40 and 45.
Menopause is officially declared after 12 months without
menses. The two most common symptom groups are vasomotor symptoms (VMS) and genitourinary symptoms of
menopause (GSM).3
Vasomotor symptoms (hot flashes) occur in up to 80% of

postmenopausal women. A hot flash is characterized by
development of heat in the torso and face that disseminates
to the rest of the body. These last 2 to 4 minutes and are
associated with sweating, followed by chills. Hot flashes can
continue for years after menopause (mean 7 years), with a
minority of women continuing to have hot flashes beyond
10 years.
A second major area of discomfort for menopausal women
is GSM, which includes vulvovaginal atrophy and urinary
tract dysfunction. In the absence of circulating estrogens,
the vulva and vagina epithelium lose elasticity, and the
epithelium thins. Symptoms associated with this include
burning, pain, dyspareunia, development of vulvar fissures,
and introital narrowing. Women may be chronically uncomfortable or uncomfortable at the time of intercourse.
Sexual activity and satisfaction may be diminished. Urinary
symptoms include dysuria and urinary frequency, and recurrent urinary tract infections may develop in patients
resulting from decreased integrity of the bladder epithelium.
Genitourinary symptoms of menopause may occur without
systemic vasomotor symptoms.
Other symptoms associated with menopause include
sleep disturbances (occurring independent of hot flashes),
depression, sexual dysfunction, joint pain, and cognitive
changes attributable to anxiety and depression. Physiologic
changes associated with menopause include bone loss and increased risk of cardiovascular disease. The data supporting development of dementia, arthritis, changes in body configuration,
From the Division of Gynecologic Oncology, University of North Carolina School of Medicine, Chapel Hill, NC; University College Hospital, London, United Kingdom.
Corresponding author: Linda Van Le, MD, The University of North Carolina at Chapel Hill, Campus Box 7572, P.O. Box B103, Chapel Hill, NC 27599-7572; email: lvl@med.unc.edu.
e270
SURVIVOR CARE FOR THE PATIENT WITH GYNECOLOGIC CANCER
skin changes, and balance have not been definitively shown

to be attributable to menopause. There are no indications
for hormone treatment of prevention or protection of other
health risks such as cardiovascular disease or colon cancer.
For patients with cancer, menopause can be accelerated
as a result of cancer treatment, including chemotherapy,
radiation therapy, and surgery. Various chemotherapy drugs
potentially induce ovarian failure, either temporarily or
permanently. A prime example is the alkylating drug cyclophosphamide. The breast cancer regimen cyclophosphamide, methotrexate, and fluorouracil is known to induce
amenorrhea in 70% of patients. Younger patients may recover ovarian function, whereas older patients are less likely
to do so. Pelvic radiation therapy is particularly toxic to the
ovaries. Although uncommon, young patients with ovarian
cancers and some patients with breast cancer will require
oophorectomy for successful treatment and will experience
menopause prematurely.
Treatment
Vasomotor symptoms can cause considerable discomfort
and affect quality of life. For mild symptoms, lifestyle
changes may suffice. Patients can wear light nightgowns, use
lighter bedding and cooling fans, drink cool water, and turn
down the thermostat at night. For patients with moderate to
severe VMS, estrogen replacement is the most effective
treatment. Hot flashes decrease by 75% compared
with 30%50% seen with administration of a placebo. For
patients without a uterus, estrogen treatment alone will
suffice. For patients with a uterus, addition of a progestin is
recommended. Currently, it is recommended that hormones
be offered to the immediately menopausal or younger
women requesting treatment of symptom relief. The Endocrine Society recommends treating symptomatic women
if they have experienced menopause within 10 years or are
younger than age 60, and are without health concerns such
as cardiovascular disease or stroke.4 For patients with
cancer, there will occasionally be younger patients in need of
treatment. The Womens Health Initiative did not include
KEY POINTS
Vasomotor symptoms and genitourinary symptoms of

menopause affect up to 80% of postmenopausal women
and diminish quality of life.
Patients with hormone-dependent cancers should be
offered nonhormonal treatment first.
Gastrointestinal symptoms are the most common side
effect of pelvic radiotherapy, with up to 50% of patients
experiencing some form of late toxicity.
Urologic adverse events are more modest; however,
they increase over time.
The most common side effects resulting from
pelvic radiation therapy are pelvic insufficiency
fractures.
study of this group, however, it is extrapolated that hormone

replacement is similarly safe for the younger patient. Risks of
combined estrogen and progestin replacement include a
slightly increased risk of breast cancer, coronary artery
disease, stroke, and venous thromboembolism (VTE). When
estrogen alone is administered, only an increased risk of
thromboembolic events is observed without risk of cardiovascular disease or breast cancer.5,6
There is a wide variety of choices for treatment of VMS.
Oral replacement is associated with higher levels of circulating estrogen metabolites than estrogens administered via
other routes. Transdermal and cutaneous (gel or spray)
formulations are also available and felt to be as effective.
There is a benefit to a transdermal approach, which delivers
estrogen directly to the systemic circulation and bypasses
liver processing (first-pass effect); the transdermal formulations do not result in a procoagulant affect and also do not
affect binding globulins, triglycerides, and C-reactive protein.
Patients with a uterus require addition of a progestin to decrease the risk of endometrial neoplasia. Progestins are also
available for transdermal, vaginal, or intrauterine delivery.
Duavee is a conjugated estrogen combined with bazedoxifene, a
selective estrogen receptor modulator (SERM). This drug prevents hot flashes and bone loss. Bazedoxifene counteracts the
stimulatory effect of estrogen on endometrium and is not
expected to stimulate endometrial hyperplasia. Its effect on
breast and uterine cancers is unknown, and, as such, use of
Duavee is contraindicated for these patients. The overall recommendation for estrogen or estrogen/progestin treatment is
to use the lowest dose possible, and there is evidence that this is
effective for treatment of VMS.
There are many hormone replacement regimens for the
patient with VMS, and these can be delivered in many
different ways. Some commonly prescribed low-dose regimens include: (1) 0.25 mg patch of Climara every week and
200 mg of Prometrium (micronized progesterone) orally
every day, (2) ClimaraPro patch applied weekly, and (3)
0.05 mg of estradiol taken orally every day and 2.5 mg of
medroxyprogesterone acetate taken orally every day (a generic
inexpensive combination of Estrace and Provera).
For some patients, a nonhormonal approach may be preferred. Use of selective serotonin reuptake inhibitors (SSRIs)
or serotonin and norepinephrine reuptake inhibitors is associated with a 25%70% decrease in VMS. Paroxetine mesylate
is the only U.S. Food and Drug Administrationapproved SSRI
for treatment of hot flashes. Gabapentin has been shown to
improve hot flashes that predominate at night. Pregabalin,
venlafaxine, and desvenlafaxine are also effective for treatment of hot flashes. The clonidine transdermal patch is another nonhormonal option for treatment of hot flashes.
Phytoestrogens such as the isoflavone genistein and daidezein
found in soy products are often suggested as an alternative
to treatment, as are herbal remedies; however, there is no
evidence to support their efficacy in treatment of VMS.
Treatment of GSM includes estrogen. Some of these
symptoms may improve among patients already taking
hormone replacement for VMS. For the patient with only
e271
GSM, a vaginal formulation should be prescribed. A continuum of treatment exists for treatment of vaginal symptoms ranging from use of nonhormonal vaginal moisturizers
and lubricants to vaginal estrogens. Moisturizers can be used
regularly to decrease vaginal atrophy symptoms. Lubricants
are used in addition to moisturizers during intercourse to
alleviate local symptoms of irritation and pain. Neither of
these improves urinary symptoms, however. If the patient
requires escalation of treatment, vaginal estrogens are
available. It is best to use the lowest effective dose to
minimize systemic absorption. Options for treatment of
vaginal symptoms include use of a low-dose vaginal ring,
which can be placed every 3 months. Another low-dose
choice is use of vaginal estradiol tablets (10 mg twice a week
vaginally). Both of these preparations have not been observed to increase systemic levels of circulating estrogen.
Estradiol creams and higher dose estradiol tablets increase
systemic levels of estrogen and require progestin supplementation if a uterus is in place. A new SERM, ospemifene,
was approved in 2013 for treatment of vulvovaginal atrophy.7 Patients with cancer were excluded from enrollment in
ospemifene studies, therefore, there is no safety data for its
use for patients with cancer. Side effects of ospemifene
include increase of VTE and increase in VSM. See the Sidebar
for hormonal and nonhormonal treatment options.
Nonhormone-Dependent Gynecologic Cancers

There is no contraindication for hormone treatment of menopausal symptoms for patients with cervical or vulvar cancers.
A recent randomized controlled trial of hormone replacement
among patients with ovarian cancer showed no increase in
recurrence; relapse-free survival and overall survival appeared
to be improved in the hormone-treated group. For younger
patients with germ cell tumors, hormone treatment is advocated. There are no data regarding safety of hormone treatment for patients with sarcomas or carcinosarcomas.
Other than breast cancers, it is uncommon for nongynecologic cancers to be hormone dependent. Unless there
are contraindications to estrogen treatment, these cancers
can be treated as in general menopausal patients.
Hormone-Dependent Cancers
Patients with breast cancer requiring menopausal symptom
management constitute a unique patient group. In 2003, the
HABITS trial and Stockholm trial, both randomized trials
studying menopausal estrogen replacement among patients
with breast cancer, were stopped prematurely because of
the observation of an increased risk of recurrent cancer in
the treatment arms.8 Long-term follow-up of both has not
observed a difference in survival, and the Stockholm trial did
not observe an increase in recurrence in the treatment arm.
Nonetheless, because there has been no rigorous trial clearly
demonstrating safety of hormone replacement in this group,
nonhormonal treatment of VMS is recommended for patients with breast cancer. For GSM, vaginal moisturizers and
lubricants are recommended. If symptoms do not resolve
e272
with nonhormonal treatment, low-dose vaginal estrogens (pill,

ring, or cream) that are not associated with increased systemic
estrogen levels may be considered. For patients taking tamoxifen, low-dose vaginal estrogens can be offered; any
minimal increase in systemic estrogen resulting from vaginal
treatment is blocked by tamoxifen. For patients taking aromatase inhibitors, estrogen administration is not recommended, and these patients should be offered nonhormonal
treatment. See the Sidebar for treatment options for GSM.
Patient with endometrial cancer also require thoughtful
management of menopause. The GOG 137 trial, published in
2006, studied 1,236 patients with stage I and II endometrial
cancer in a randomized control trial in which hormone replacement was offered to patients with endometrial cancer.9 This study was closed early because of enrollment
issues and was underpowered when it closed. It did conclude
that there was no notable difference in recurrence or death
between the treated and control patients. A 2014 metaanalysis of hormone replacement for patients with endometrial cancer concluded that there was no increased risk of
recurrence for these patients.10 However, there are no data
regarding hormone replacement for patients with advanced
disease. If estrogen treatment is chosen, the National
Comprehensive Cancer Network recommends waiting 6
12 months before starting estrogen treatment.
Granulosa cell tumors are uncommon and often hormonedependent. There are no data regarding the safety of hormone treatment in this group, and these patients should be
offered nonhormonal management.
If estrogens are indicated for patients with VTE, a
transdermal approach should be offered. Administration
of paroxetine for patients taking tamoxifen should be
prescribed with caution. Paroxetine inhibits the cytochrome
450-enzyme system and can decrease the efficacy of tamoxifen.
Female patients with cancer may experience premature
menopause because of necessary cancer therapeutics or
undergo spontaneous menopause by virtue of their age.
Quality of life has been shown to improve when symptomatic
patients are treated. There are many nonhormonal formulations, and it may be best to offer these first. However, some
patients may continue to be symptomatic from VSM, and
discussion of hormone treatment should be undertaken after
taking into consideration the patients cancer status. It is
widely acknowledged that definitive trials are lacking regarding treatment of hormone-dependent cancers. Given the
increase in the female survivor population, management of
menopause will require additional study in a timely fashion.
SURVIVING THE CURE: MANAGING THE LATE

EFFECTS OF RADIOTHERAPY
The late effects of radiotherapy (RT) are defined as those
that occur months to years after completion of a course of
RT. Such effects have become more important as the
number of long-term survivors increase. Pelvic RT is an
effective treatment of many gynecologic cancers but often
comes at a price. The extent of the problem is difficult to
gauge accurately, as these data have generally not been

collected prospectively in large randomized trials. Furthermore, patients may present to different specialists with
symptoms that are not immediately recognized as late effects of RT, therefore contributing to the general underreporting. The National Survivorship Initiative in the United
Kingdom11 identified four key needs of survivors of cancer:
A personalized survivorship care plan
Support to self-manage their condition where appropriate
Provision of information on long-term effects of living
with and beyond cancer
Access to specialist medical care for complications that
occur after cancer
There are wide variations (up to 23%) in the reported
prevalence of late toxicity following pelvic RT in women with
gynecologic cancers.12 The generally accepted prevalence
rate is approximately 10%.1315 It is likely that the classification of the late effects together with underreporting will
contribute to some of this variation.
Gastrointestinal Toxicity
Gastrointestinal (GI) symptoms are the most common side
effect of pelvic RT, with up to 50% of patients experiencing some form of late toxicity.16,17 A review of our institutional data showed that the mean time to presentation
of bowel symptoms was 8 and 10 months for women
following treatment of cervical and endometrial cancers,
respectively.18 Chronic GI symptoms include rectal bleeding,
FIGURE 1. Modified Algorithm for Approach to

Treating Patients With Gastrointestinal Symptoms
After Radiation Therapy
Abbreviations: QOL, quality of life; APC, argon plasma coagulation.

Adapted from Hauer-Jensen M, Denham JW, Andreyev HJ. Radiation
enteropathypathogenesis, treatment and prevention. Nat Rev Gastroenterol
Hepatol. 2014;11:470-479.
diarrhea, flatulence, frequency, urgency, incontinence, malabsorption, and pain. It is essential that the clinician take an
accurate and detailed history of the symptoms together with
establishing the precise details of all prior cancer therapies.
An initial assessment of the patient should include evaluation to exclude anemia and thyroid abnormalities, and
cross-sectional imaging should be ordered to exclude cancer
recurrence prior to referral to a gastroenterologist. An algorithm detailing an approach to the management of GI
radiation side effects, developed by colleagues at the Royal
Marsden Hospital in London, was published in a recent
Nature Reviews Gastroenterology & Hepatology review and
is presented in Fig. 1.19 Altered consistency of bowel habit,
frequency, urgency, and rectal bleeding are the most
commonly identified symptoms of late GI toxicity following
pelvic RT. Patients usually must undergo sigmoidoscopy/
colonoscopy for evaluation.20 The management of these
symptoms is complex and will depend upon the results of the
various investigations. Strategies include antibiotics to treat
small intestinal bacterial overgrowth, use of loperamide and
stoolbulking agents, and biofeedback to manage frequency
and urgency. Rectal bleeding from radiation-induced telangiectasia is common, and, in the majority of patients, it is
mild and requires no specific therapy. However, rectal
bleeding leading to anemia can have a considerable impact
on the patients quality of life. Options for management
often vary with resources and expertise but may include use
of argon plasma coagulation embolization or hyperbaric oxygen
therapy (HBO).11 A randomized controlled trial was undertaken
to assess the role of HBO in the treatment of chronic refractory
radiation proctitis. The trial, when analyzed according to intention to treat, demonstrated a notable improvement in the
Late Effects Normal Tissue Task Force-Subjective, Objective,
Management, Analytic score for patients treated with HBO. The
quality of life bowel bother subscale was also significantly
improved for patients who received HBO.21 Bowel obstruction,
perforation, and fistulas are uncommon, but more serious late
effects of RT are more frequently seen in survivors of cervical,
rather than endometrial, cancer.
Small intestinal bowel obstruction (SIBO) is estimated to
occur in 9% of patients after RT for gynecologic malignancies.22 However, the rates have been shown to be influenced
by both prior laparotomy and body weight.23 Management
of SIBO depends largely on the frequency and severity of
symptoms, with some patients experiencing only intermittent
episodes of SIBO that can be managed conservatively. In others,
the symptoms persist and progress to total obstruction over
weeks or months, and such patients require surgical
intervention.24 The type of procedure performed will vary
according to both patient factors and surgical expertise. In
expert hands, resection of the injured bowel with anastomosis is
the preferred procedure permitting resumption of enteral
nutrition postoperatively.25 Ogino et al26 reported late rectal
grade 3/4 complication rates of 6.8% and 8.1%, respectively,
among patients following radical pelvic RT for cervical cancer.
This was predominantly rectal bleeding and rectovaginal fistulae, although they did not differentiate further between the
e273
SIDEBAR. Nonhormonal and Hormonal

Treatment for Genitourinary Symptoms of
Menopause
Vaginal Moisturizers
Fresh Start
K-Y Silk-E
K-Y Liquibeads
Moist Again
Replens
Vaginal Lubricants
Astroglide
FemGlide
K-Y Jelly
Summers Eve
Vaginal Estrogen
Tablet: Estradiol (Vagifem) 10 mg daily first week, then two
times per week vaginally
Ring: Estradiol (Estring) 7.5 mg, insert every 90 days
Selective Estrogen Receptor Modulator: Ospemifeme
(Osphema), 60 mg orally daily
two. The fistula risk is less than 3% at 20 years, but pelvic surgery
before or after RT has been shown to almost double this risk.23
were urethral strictures and radiation cystitis. In contrast,

two large trials of postoperative RT in endometrial cancers28,29 reported no grade 3 to 4 urinary adverse effects
with a median follow-up of 5268 months. This finding may
be the result of the relatively short follow-up in these
studies. The management of ureteral strictures may involve
ureteral dilatation and intermittent self-catheterization,
whereas mild cystitis may be managed by dietary modifications. The management of severe urologic adverse events
is complex and beyond the scope of this article. In cases of
treatment for refractory chronic radiation injury, there may
be a role for HBO in symptom management.30
The impact of pelvic RT on bone health is increasingly
recognized and can manifest in a number of ways. The most
common events are pelvic insufficiency fractures. These may
be asymptomatic and identified only on MRI/CT/PET, or
patients may also present with pelvic pain and reduced
mobility. The median time to occurrence is 620 months after
radiation, and the majority is managed conservatively.31,32
Radiation treatment of gynecologic malignancies can be associated with considerable adverse effects, leading to a detrimental effect on quality of life. Therefore, the risk benefits should
always be carefully explored, and patients should be monitored
long term to facilitate early detection of serious adverse events
and subsequent prompt referral to an appropriate specialist.
Urologic Complications
Grade 1 and 2 urologic adverse events are relatively common; however, the risk of major urinary tract complications
is more modest. Unlike GI complications, which usually
present within 2 years, the risk of urologic complications
increases over time. A retrospective review of the complications following radical RT for early-stage cervical cancer
concluded that the actuarial risk of grade 3/4 urologic
complications was 11.1% at 10 years, 13% at 15 years, and
14.4% at 20 years.23,27 The most common adverse events
CONCLUSION
Cancer survival rates are increasing. This increase is the
result of improvements in screening, management of side
effects, and the availability of new treatments. The goal of
survivorship care is to improve overall health and quality of
life after cancer treatment. Women surviving gynecologic
cancer have unique needs. Understanding and addressing
these needs will help these patients celebrate and enjoy
their survivorship and improve their quality of life.
References
1. McCabe MS, Bhatia S, Oeffinger KC, et al. American Society of Clinical
Oncology statement: achieving high-quality cancer survivorship care.
J Clin Oncol. 2013;31:631-640.
2. American Society of Clinical Oncology. Key Components of Survivorship Care.
http://www.asco.org/practice-research/key-components-survivorship-care.
3. ACOG Practice Bulletin No. 141: management of menopausal symptoms. Obstet Gynecol. 2014;123:202-216.
4. Stuenkel CA, Davis SR, Gompel A, et al. Treatment of symptoms of the
menopause: an endocrine society clinical practice guideline. J Clin
Endocrinol Metab. 2015;100:3975-4011.
5. Rossouw JE, Anderson GL, Prentice RL, et al; Writing Group for the
Womens Health Initiative Investigators. Risks and benefits of estrogen
plus progestin in healthy postmenopausal women: principal results
from the Womens Health Initiative randomized controlled trial. JAMA.
2002;288:321-333.
6. Anderson GL, Limacher M, Assaf AR, et al; Womens Health Initiative Steering
Committee. Effects of conjugated equine estrogen in postmenopausal
e274
7.
8.
9.
10.
11.
women with hysterectomy: the Womens Health Initiative randomized

controlled trial. JAMA. 2004;291:1701-1712.
Simon JA, Kingsberg SA, Shumel B, et al. Efficacy and safety of flibanserin
in postmenopausal women with hypoactive sexual desire disorder:
results of the SNOWDROP trial. Menopause. 2014;21:633-640.
Fahlen M, Fornander T, Johansson H, et al. Hormone replacement
therapy after breast cancer: 10 year follow up of the Stockholm
randomised trial. Eur J Cancer. 2013;49:52-59.
Barakat RR, Bundy BN, Spirtos NM, et al; Gynecologic Oncology Group
Study. Randomized double-blind trial of estrogen replacement therapy
versus placebo in stage I or II endometrial cancer: a Gynecologic Oncology Group Study. J Clin Oncol. 2006;24:587-592.
Shim SH, Lee SJ, Kim SN. Effects of hormone replacement therapy on the
rate of recurrence in endometrial cancer survivors: a meta-analysis. Eur
J Cancer. 2014;50:1628-1637.
Andreyev HJ, Davidson SE, Gillespie C, et al; British Society of Gastroenterology; Association of Colo-Proctology of Great Britain and
Ireland; Association of Upper Gastrointestinal Surgeons; Faculty of
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
Clinical Oncology Section of the Royal College of Radiologists. Practice

guidance on the management of acute and chronic gastrointestinal
problems arising as a result of treatment for cancer. Gut. 2012;61:
179-192.
Kirwan JM, Symonds P, Green JA, et al. A systematic review of acute
and late toxicity of concomitant chemoradiation for cervical cancer.
Radiother Oncol. 2003;68:217-226.
Eifel PJ, Jhingran A, Bodurka DC, et al. Correlation of smoking history and
other patient characteristics with major complications of pelvic radiation therapy for cervical cancer. J Clin Oncol. 2002;20:3651-3657.
Vale CL, Tierney JF, Davidson SE, et al. Substantial improvement in UK
cervical cancer survival with chemoradiotherapy: results of a Royal
College of Radiologists audit. Clin Oncol (R Coll Radiol). 2010;22:
590-601.
Guth U, Hadwin RJ, Schotzau A, et al. Clinical outcomes and patterns of
severe late toxicity in the era of chemo-radiation for cervical cancer.
Arch Gynecol Obstet. 2012;285:1703-1711.
Theis VS, Sripadam R, Ramani V, et al. Chronic radiation enteritis. Clin
Oncol (R Coll Radiol). 2010;22:70-83.
Andreyev J. Gastrointestinal symptoms after pelvic radiotherapy: a new
understanding to improve management of symptomatic patients.
Lancet Oncol. 2007;8:1007-1017.
Kuku S, Fragkos C, McCormack M, et al. Radiation-induced bowel injury:
the impact of radiotherapy on survivorship after treatment for
gynaecological cancers. Br J Cancer. 2013;109:1504-1512.
Hauer-Jensen M, Denham JW, Andreyev HJ. Radiation enteropathy
pathogenesis, treatment and prevention. Nat Rev Gastroenterol Hepatol. 2014;11:470-479.
Gami B, Harrington K, Blake P, et al. How patients manage gastrointestinal symptoms after pelvic radiotherapy. Aliment Pharmacol Ther.
2003;18:987-994.
Clarke RE, Tenorio LM, Hussey JR, et al. Hyperbaric oxygen treatment of
chronic refractory radiation proctitis: a randomized and controlled
double-blind crossover trial with long-term follow-up. Int J Radiat Oncol
Biol Phys. 2008;72:134-143.
22. Stanic S, Mayadev JS. Tolerance of the small bowel to therapeutic irradiation: a focus on late toxicity in patients receiving para-aortic nodal
irradiation for gynecologic malignancies. Int J Gynecol Cancer. 2013;23:
592-597.
23. Eifel PJ, Levenback C, Wharton JT, et al. Time course and incidence of
late complications in patients treated with radiation therapy for FIGO
stage IB carcinoma of the uterine cervix. Int J Radiat Oncol Biol Phys.
1995;32:1289-1300.
24. Tsai MS, Liang JT. Surgery is justified in patients with bowel obstruction
due to radiation therapy. J Gastrointest Surg. 2006;10:575-582.
25. Onodera H, Nagayama S, Mori A, et al. Reappraisal of surgical treatment
for radiation enteritis. World J Surg. 2005;29:459-463.
26. Ogino I, Kitamura T, Okamoto N, et al. Late rectal complication following
high dose rate intracavitary brachytherapy in cancer of the cervix. Int J
27. McIntyre JF, Eifel PJ, Levenback C, et al. Ureteral stricture as a late
complication of radiotherapy for stage IB carcinoma of the uterine
cervix. Cancer. 1995;75:836-843.
28. Cruetzberg CL, van Putten WL, Koper PC, et al. Surgery and postoperative radiotherapy versus surgery alone for patients with stage-1
endometrial carcinoma; multicentre randomised trial. PORTEC Study
Group. Post Operative Radiation Therapy in Endometrial Carcinoma.
Lancet. 2000;355:1404-1411.
29. Keys HM, Roberts JA, Brunetto VL, et al; Gynecologic Oncology Group. A
phase III trial of surgery with or without adjunctive external pelvic
radiation therapy in intermediate risk endometrial adenocarcinoma:
a Gynecologic Oncology Group study. Gynecol Oncol. 2004;92:744-751.
30. Craighead P, Shea-Budgell MA, Nation J, et al. Hyperbaric oxygen
therapy for late radiation tissue injury in gynecologic malignancies. Curr
Oncol. 2011;18:220-227.
31. Schmeler KM, Jhingran A, Iyer RB, et al. Pelvic fractures after radiotherapy
for cervical cancer: implications for survivors. Cancer. 2010;116:625-630.
32. Park SH, Kim JC, Lee JE, et al. Pelvic insufficiency fracture after radiotherapy in patients with cervical cancer in the era of PET/CT. Radiat
Oncol J. 2011;29:269-276.
e275
Best of the Rest: Top Abstracts

on Head and Neck Cancer From
20152016 Oncology Meetings
CHAIR
Sue S. Yom, MD, PhD
San Francisco, CA
SPEAKERS
Andreas Dietz, MD
University of Leipzig
Leipzig, Germany
Apar K. Ganti, MD
University of Nebraska Medical Center
Omaha, NE
YOM, GANTI, AND DIETZ
Whats New in Head and Neck Cancer: Key Findings in

20152016 From ECCO/ESMO, ASTRO, and the
Multidisciplinary Head and Neck Cancer Symposium
Sue S. Yom, MD, PhD, Apar K. Ganti, MD, MS, FACP, and Andreas Dietz, MD
OVERVIEW
Scientific investigation is extremely active in the treatment, management, and optimization of therapies for patients with
head and neck cancer. These issues have undergone recent rapid evolution in response to a changing epidemiology based on
an increasing proportion of HPV-associated oropharyngeal cancer with advances in multimodality technologies to improve
outcomes and reduce toxicity. Choices of definitive treatment of various anatomic subsites are being refined, balancing the
relative indications and advantages of surgery or chemoradiation-based strategies. The major potential influence of HPVassociated etiology on therapy selection, prognostic factors, response to treatment, survival outcomes, and post-treatment
surveillance has created a robust and distinct field of scientific inquiry around this patient subset. Meanwhile, for patient subsets
where prognosis remains poor, therapeutic intensification is being explored, and for recurrent/metastatic disease, improved
selection for salvage and novel systemic therapies are under development. For all patients with head and neck cancer, upholding
principles of equity and access to the highly specialized care that results in optimal outcomes should be the goal.
his article summarizes selected findings in head and neck

cancer research that were presented at several major
meetings over the past year. The European CanCer Organization (ECCO) and the European Society for Medical Oncology
(ESMO) held a joint meeting, the European Cancer Congress,
from September 2529, 2015, in Vienna. The 57th Annual
Meeting of the American Society for Radiation Oncology
(ASTRO) took place in San Antonio, Texas, from October
1821, 2015. The 2016 Multidisciplinary Head and Neck
Cancer Symposium, a joint conference of the American Head
and Neck Society, American Society of Clinical Oncology, and
American Society of Radiation Oncology, took place in
Scottsdale, Arizona, from February 1820, 2016.
SURGERY AS A PRIMARY TREATMENT

MODALITY
At the 2015 European Cancer Congress, Cheng et al presented an analysis of data from Taiwan national health
insurance claims and the Taiwan cancer registry database,
including data from 1,698 patients with oropharyngeal and
1,619 patients with hypopharyngeal stage III/IVA cancer
diagnosed between 2004 and 2009.1 The researchers compared outcomes among those who had surgery and those who
did not; in both groups, patients may or may not have received
concurrent chemoradiation therapy. Radical surgery was

performed on 35.29% and 37.63% of patients with stage III
and stage IVA oropharyngeal cancer, respectively, and 54.52%
and 48.85% of patients with stage III and stage IVA hypopharyngeal cancer, respectively. Primary surgery was associated with better overall cancer survival in most subset
analyses, but the potential effect of selection for surgery in
this population-based database could not be discounted.
Surgical treatment of advanced head and neck cancer is a
well-recognized weapon that is very often followed by radiotherapy with or without chemotherapy, but in some cases,
surgery can be avoided and radiotherapy or chemoradiotherapy
is an appropriate standard of care. In many centers, the choice
of primary modality continues to be heavily influenced by local
factors such as experience or availability. These data further
underscore the need for prospective research to define the
relative roles of surgery and radiotherapy within curative
treatment paradigms for these subsites.
The importance of the prognostic features identified at
surgical pathology was highlighted by a presentation from
researchers in India. This group measured tumor thickness
using intraoral ultrasound (IOUS) and correlated it with the
risk of nodal metastasis in early N0 oral cancer. A prospective
study was carried out for patients with cT1-2N0 oral cavity
From the Department of Radiation Oncology, University of California, San Francisco, San Francisco, CA; Department of Internal Medicine, University of Nebraska Medical
Center, Omaha, NE; Department of Head Medicine and Oral Health, University of Leipzig, Leipzig, Germany.
Corresponding author: Sue S. Yom, MD, PhD, University of California, San Francisco, 1600 Divisadero St., San Francisco, CA 94115; email: yoms@radonc.ucsf.edu.
176
WHATS NEW IN HEAD AND NECK CANCER
cancer in a high-volume head and neck surgical center from

2013 to 2015. A cutoff level of greater than 8.5-mm tumor
thickness on IOUS was derived (area under the curve, 0.81;
95% CI, 0.660.97; p = .005) that predicted lymph node
metastasis with a sensitivity and specificity of 88% and 60%,
respectively. Tumor thickness measuring less than 8 mm
and greater than 9 mm on IOUS developed 3.2% and 25.9%
lymph node metastasis, respectively (p = .01). This cutoff
recommendation could be valuable in designing clinical trials
based on indications for elective neck dissection in early oral
cancer.2
Meanwhile, a group from Belgium took a different approach to this problem, testing the clinical feasibility of
sentinel lymph node identification in a prospective phase II
study. The goal of the study was to reduce the electively
radiated volumes of the neck. Among 21 patients with cN0
oral cavity, oropharynx, larynx, or hypopharynx squamous
cell carcinomas, 99mTc nanocolloid injection was followed
with single-photon emission CT (SPECT).3 An average of 2.7
sentinel nodes were detected per patient. Only the neck
levels containing the four hottest draining nodes were
electively radiated. At a median follow-up of 14 months,
there were two fatal local relapses but no regional relapses.
HPV-ASSOCIATED OROPHARYNGEAL
SQUAMOUS CELL CARCINOMA
HPV-positive oropharyngeal squamous cell carcinoma (OPSCC)
is a distinct subtype of head and neck squamous cell carcinoma
(HNSCC) that has a significantly better prognosis compared
with HPV-negative OPSCC.4 However, it appears that smoking
history may be used to stratify HPV-positive OPSCC into more
distinct prognostic groups. Zevallos et al presented their study
of the molecular profile of HPV-positive OPSCC, an attempt to
KEY POINTS
Surgery and chemoradiation-based strategies are

considered appropriate up-front modalities for
definitive treatment, based on anatomic considerations,
the balance of toxicities, and morbidity.
For HPVassociated oropharyngeal cancers,
de-intensification of radiation therapy is under active
exploration in an attempt to reduce the long-term
toxicities of treatment.
Post-treatment surveillance remains a relatively
undefined area, although physical examination, direct
visualization, and patient-reported symptoms are
recognized as key indices to follow.
In the salvage setting, improved selection of patients for
surgical salvage may produce better outcomes, and
early results of reirradiation with novel technologies
such as proton beam therapy are promising.
For recurrent/metastatic disease, a number of
combinations of novel targeted therapies and/or
immunomodulatory therapies will move forward in the
near future.
understand the underlying genetic basis for the differential

outcomes.5 They conducted targeted next-generation sequencing for 66 patients with HPV-positive OPSCC (40
had more than 10 pack years of smoking and 26 had less
than 10 pack years). The group with the lower smoking
history had better outcomes. TP53, CDKN2A, KRAS, and
NOTCH1 mutations were seen almost exclusively in the
high-smoking group, whereas HLA-A mutations were more
common in the low-smoking group.
In a more clinical vein, Lukens et al reviewed a series of 174
patients with HPV-positive OPSCC who had been surgically
treated with transoral robotic surgery and neck dissection
followed by adjuvant chemoradiotherapy. The focus was to
characterize the influence of lymph node characteristics
on the risk of distant metastasis. The follow-up was
38 months. The probability of distant metastasis increased
proportionately when patients had four or more involved
nodes. The risk of distant metastasis was 14% for four or
more nodes, 18% for five or more nodes, 22% for more than
six nodes, and 28% for more than seven nodes. On univariate
analysis, macroscopic/gross extracapsular extension (ECE) predicted distant metastasis, with distant metastasis developing in
20% of patients wtih ECE versus 4.2% of patients with microscopic ECE, but on multivariate analysis, macroscopic/gross ECE
was not significant (p = .84).6
BIOLOGIC AND RADIOLOGIC PROGNOSTIC

FEATURES
Although various clinical factors, especially smoking, are
considered prognostic, there is a continued search for other
predictors of outcome that may be more biologically specific.
At ASTRO 2015, Khwaja et al presented a well-conducted geneexpression profiling study of HPV-positive OPSCC.7 Patients
among whom distant metastasis developed had an E6 expression level that was twofold higher. In the validation
cohort, the authors were able to identify a cutoff level of E6
expression (7.3, by receiver operating curve analysis) that was
correlated to a fivefold higher risk of distant metastasis. The
only minor inconsistency in this study was that the patients
were treated with either chemoradiation or surgery followed
by postoperative radiation with or without chemotherapy.
From a different angle, radiographic predictors have been
an ongoing area of investigation that has yet to yield a clear
standard pretreatment measurement tool, perhaps because
assessment tools should be specific for the type of treatment
to be received. For example, a study from Australia analyzed
the possible predictive value of fluorodeoxyglucose (FDG)positron emission tomography (PETCT) after two cycles of
taxane/platinum/fluorouracil (5-FU) induction in correlation
with outcome. Although a high overall survival (OS) at 89.5%
was achieved, the FDG-PET/CTbased volumetric measures
were not predictive for locoregional control (impact of
changes of SUVmax not reported).8
More detailed data concerning off-target FDG-PET parameters for outcome prognosis after chemoradiotherapy
were reported from Germany. Early occurrence of mucositis
177
higher than grade 1 was associated with better locoregional

control (p = .012) and OS (p = .017). Occurrence of dysphagia
higher than level 2, but not mucositis, at the end of treatment
was associated with favorable locoregional control (p = .031)
and OS (p = .008). Besides error-prone and observer-biased
determination of mucositis by clinical examination, the
evaluation of mucosa by PET during treatment showed a
superior correlation with patients outcome.9
Lastly, MAASTRO investigators reported a new radionomics
tool that has prognostic and predictive implications after
chemoradiotherapy. Five radiomic features of primary tumor
regions were evaluated. Strong correlations with OS were
described by combining special features. This noninvasive
new morphometric tool deserves further evaluation.10
DE-INTENSIFIED CHEMORADIOTHERAPY FOR

HPV-POSITIVE OROPHARYNGEAL SQUAMOUS
CELL CARCINOMA
Given the excellent prognosis of HPV-positive OPSCC, efforts
are underway to attempt de-escalation of therapy without
affecting outcomes. In a phase II trial that was reported at
ASTRO and the Multidisciplinary Head and Neck Cancer
Symposium, Chera et al included 43 patients with T0-3, N0N2c, and M0 HPV-positive OPSCC with a minimal or remote
smoking history.11,12 The treatment regimen was 60 Gy of
intensity-modulated radiation therapy with weekly cisplatin
(30 mg/m2). All patients then underwent a biopsy of the
primary site and a neck dissection of pretreatment positive
lymph node regions, irrespective of response. The pathologic complete response rate was 86%. The patients without
pathologic complete response had only microscopic foci of
residual disease. All patients were alive without recurrence
after a median follow-up of 21.3 months. The incidence of
acute Common Terminology Criteria for Adverse Events grade
3/4 toxicity was as follows: mucositis 34%, pain 5%, nausea
18%, vomiting 5%, dysphagia 39%, and xerostomia 2%. A
feeding tube was required temporarily for 39% of patients.
Likewise, in another attempt to decrease late toxicity
presented at both ASTRO13 and the Multidisciplinary Head
and Neck Cancer Symposium,14 Melotek et al incorporated
a response-adapted volume de-escalation approach using
response-to-induction chemotherapy as a guide to decrease
the extent of radiation volume in a nonselected population of
patients with locally advanced HNSCC. The regimen was
comprised of two cycles of chemotherapy (cisplatin 75 mg/m2,
paclitaxel 175 mg/m2, both on day 1, and weekly cetuximab
with or without everolimus). Patients with good response,
defined as a 50% reduction in the sum of gross tumor diameters,
received concurrent chemoradiation (paclitaxel, 5-FU, hydroxyurea, and 1.5 Gy twice-daily radiotherapy every other week).
The radiation dose was as follows: 75 Gy with the planning
target volume (PTV1) encompassing exclusively gross disease.
Patients with less than 50% response (NR) were treated with
volumes encompassing PTV1 and the next nodal station at
risk (PTV2) to 45 Gy, followed by a sequential boost to PTV1 to
75 Gy. Of the 94 patients in this study, 63% were HPV-positive.
178
Thirty HPV-positive and seven HPV-negative patients had a

good response, as defined before. The 2-year progressionfree survival (PFS) and OS rates were 93.1% and 92.1% for
good response, HPV-positive OPSCC and 74.0% and 95.2% for
NR HPV-positive OPSCC, respectively.
It has been an ongoing active controversy as to whether
de-intensification with the use of cetuximab could risk
compromising outcomes. A new abstract on this subject was
presented at ASTRO 2015 but is unlikely to settle the debate.15
In this study, the investigators found that patients who had
cetuximab given concurrently with radiotherapy had higher
rates of distant metastases than those who had received
chemoradiotherapy with a platinum agent. Maturing results of
the major study addressing this question, RTOG 1016, which
tested cetuximab versus cisplatin in HPV-positive oropharyngeal squamous cell carcinoma, are eagerly awaited.
INTENSIFICATION FOR HPV-NEGATVE

Although the good prognosis of patients with HPV-positive
OPSCC has led to trials in therapeutic de-intensification, for
patients with HPV-negative HNSCC who had significantly
worse outcomes, systemic therapy intensification may be
warranted. In an abstract presented at the Multidisciplinary
Head and Neck Cancer Symposium, Melotek et al investigated
the addition of cetuximab to induction chemotherapy and
concurrent chemoradiation for patients with locoregionally
advanced HNSCC.16 Patients were randomly selected to receive
two cycles of weekly cetuximab, paclitaxel, carboplatin, and
either cetuximab, 5-FU, hydroxyurea, and 1.5 Gy twice-daily
radiotherapy every other week to 75 Gy (CFHX) or cetuximab,
cisplatin, and accelerated radiotherapy with delayed concomitant boost to 72 Gy in 42 fractions (CPX). Fifty-six patients with
HPV-negative HNSCC (32 receiving CFHX and 24 CPX) were
included in the trial. The 2-year PFS was 75.0% for CFHX and
70.8% for CPX. The 5-year PFS and OS for the HPV-negative
cohort were 65.9% and 72.5%, respectively, suggesting a
possible role for cetuximab for this group of patients.
SURVEILLANCE
The optimal frequency of follow-up imaging after definitive
curative-intent therapy has not been defined. The National
Comprehensive Cancer Network guidelines do not recommend routine imaging in the absence of symptoms, except
chest CT scans as recommended for lung cancer screening (in
high-risk patients who meet the criteria). At the Multidisciplinary
Head and Neck Cancer Symposium, Frakes et al retrospectively
reviewed the charts of 246 patients with HPV-positive OPSCC
who were treated with definitive radiotherapy or chemoradiotherapy.17 Local control was achieved for 239 patients,
with a 3-year local control rate of 97.8%. All local failures
were detected by direct visualization or flexible laryngoscopy. The 3-year regional control rate was 95.3%. Risk
factors for regional failure included patients who had involvement of five or more nodes or level 4 lymph nodes. The
majority of regional recurrences (89%) were detected either
by symptoms or 3-month post-treatment PET/CT. Twentyone patients had distant metastases. Factors predicting for
distant metastases included lymph nodes larger than 6 cm,
bilateral lymphadenopathy, five or more nodes, or level 4
lymph node involvement. Of these 21 patients, 71% were
identified either from symptoms or on the 3-month posttreatment imaging. These results suggest that if the 3-month
post-treatment PET/CT is negative, no further routine imaging
is necessary. The results also reiterate the importance of a good
history and physical examination, including direct visualization.
POST-TREATMENT NECK DISSECTION

The previously standard practice of planned neck dissection
after radiation-based definitive treatment has waned.
As presented at the European Cancer Congress, 147 patients
with locoregionally advanced stage IVA (52%) or IVB (46%)
HNSCC from Portugal were treated with induction taxane/
platinum/5-FU chemotherapy followed by chemoradiotherapy
from July 2008 to March 2014. Fifty-three patients (37%)
underwent neck dissection because of persistence of disease (4/53, 7.5%), suspicious imaging (40/53, 75.5%), or previous bulky disease (9/53, 17%). A majority of the positive neck
dissections were N2c, instead of N3. There was no meaningful
difference in OS between patients who did not have neck
dissection and those who had a negative neck dissection.
The authors conclude that prophylactic neck dissection did
not seem to improve OS.18
Adding to the increasing conservatism around postradiotherapy neck dissection, Mehanna et al presented a
late-breaking abstract examining differences in the quality of
life and functional outcomes of treatment between HPVpositive and HPV-negative patients receiving primary chemoradiotherapy, using patient data from the PET-NECK trial
previously presented at ASCO 2015.19 In this trial, the use
of PET surveillance reduced the rate of post-treatment neck
dissection to 20%, without affecting OS. The new presentation
reported that p16-positive subjects have major decreases in
quality-of-life scores during the acute phase of treatment and
may require additional support during that period, but that
they recover better than p16-negative subjects, who have
lower quality of life in the longer term and may need more longterm support.
ACUTE AND LATE TOXICITY OF

CHEMORADIATION
As seen in the PET-NECK trial, toxicity remains an acute
and long-term concern for patients receiving chemoradiation.
Oral mucositis is one of the major acute adverse events associated with head and neck radiotherapy, with or without
chemotherapy, occurring in more than 90% of patients. The
incidence of severe mucositis (grade 3-4) is 66%,20 yet there
are no currently approved agents to decrease the burden
associated with this toxicity.
In this vein, Anderson et al presented a study of 46
patients who received increasing doses of GC4419, a superoxide dismutase mimetic, and concurrent cisplatin-based
chemoradiotherapy.21 The duration, incidence, severity, and

onset of severe oral mucositis seemed to be markedly improved
compared with historical controls. The efficacy of GC4419 was
duration dependent with maximum benefit seen among
patients who received 6 to 7 weeks of treatment. The drug
was well tolerated, with the main side effects being grade 3
gastroenteritis and nausea/vomiting (one patient each). In
addition, peri-infusional facial tingling was attributable to
GC4419. The dose levels chosen for a future randomized
controlled trial were 30 and 90 mg/day.
Regarding the early toxicity of systemic therapy, the first
safety report of the Italian INTERCEPTOR trial (randomized,
multicenter phase III study comparing chemoradiotherapy
vs. induction chemoradiotherapy followed by radiotherapy
and cetuximab) was presented at ECCO/ESMO, with 228
patients accrued thus far. The first 170 were considered in
the present analysis (85 patients in each arm). Overall, the
only significant difference between the two arms was grade 3 to
4 neutropenia (p = .017), and the trial will continue. The excess of
neutropenia was caused entirely by the induction phase.22
Late toxicity is likewise a concern, especially if consideration is given to further intensification of standard chemoradiotherapy therapies. Twelve-year follow-up data from
Belgium at ECCO/ESMO compared two nonrandomized
chemoradiotherapy cohorts with a platinum-based induction
regimen followed by chemoradiotherapy. Induction was
associated with a significantly longer time to distant metastases. A little over half (51.6%) of the induction patients survived beyond 5 years, whereas 5-year survival in the CCRT
group was 29.6%. Second HNSCC primaries and other malignancies accounted for 47% of all deaths beyond 5 years. The
most important late local toxicities among surviving patients in
both groups were dysphagia and xerostomia.23
The radiation oncology community has debated for some
time whether proton beam therapy has the ability to improve patient-reported outcomes compared with current
state-of-the-art photon-based techniques. Ahn et al presented
a series of 95 patients, of whom 57 (60%) were treated with
photon-based volumetric arc therapy (VMAT) and 45 (45%)
were treated with proton pencil beam scanning (PBS). Fifty-two
(55%) of the patients were irradiated after transoral robotic
surgery and 43 (45%) were treated definitively, with or without
concurrent chemotherapy. At the 6-month endpoint, using the
Performance Scale Status instrument, there was no overall
difference in diet, speech, or public eating. There was no overall
difference in gastrostomy usage. Data from the the EORTC
questionnaires showed there was no difference in global health,
physical function, or sensory domains. However, patients who
underwent VMAT lost 5.5% of baseline weight compared with
3.4% for those who underwent PBS. Patients undergoing VMAT
also used more pain medication. In particular, among patients
receiving concurrent cisplatin, PBS was significantly associated
with a higher global health score, less pain, lower painkiller use,
and lower percutaneous endoscopic gastrostomy use.24 Given
the inherent confounding factors that may be in play, further
study of these complex questions is needed and obtaining
higher-level evidence should be a priority.
179
SINONASAL TUMORS
Induction chemotherapy remains an area of controversy and
active ongoing interest in the treatment of sinonasal cancers. Response to induction chemotherapy before surgical
treatment of locally advanced epithelial sinonasal cancer is
recognized as a prognostic factor for outcome. Bossi et al from
Italy identified 63 patients with different types of advanced
epithelial sinonasal cancer who received platinum-based induction chemotherapy, either associated with 5-FU and
lederfolin or 5-FU and docetaxel. For neuroendocrine
cancers, cisplatin and etoposide alternating with ifosfamide
and doxorubicin were used. Thirty-four patients had a
recurrence (28 at the locoregional level and six at distant
sites); 14 patients received salvage surgery (12 on primary
tumor and two on neck nodes), but only two of those
remained free of disease (one patient on T and one on N level).
With a median follow-up of 45 months, 5-year OS and diseasefree survival rates were 58% and 40%, respectively. Only
the response to induction chemotherapy retained prognostic value for OS (p = .0017).25 Because of the rarity and
heterogeneity of these tumors, the current treatment options
remain unsatisfactory, and enrollment into international trials
is needed to make progress.
LOCALLY RECURRENT DISEASE MANAGEMENT

It is generally accepted that surgery has a major role for the
successful treatment of patients with locally recurrent HNSCC. A
retrospective analysis from Belgium had the goal of improving
the selection of patients for salvage surgery as well as identifying
the major negative prognostic features in surgical pathology. On
multivariate analysis, oropharyngeal primary site (p , .001),
positive resection margin (p = .002), extracapsular spread
(p = .003), and major postoperative complications (p , .001)
were independent negative prognostic factors for relapsefree survival. These data underline the poor outcome of
recurrent oropharyngeal HNSCC and the need for highly
reflective indications for salvage surgery in this group.26
For patients who are not determined to be surgical candidates, reirradiation remains a generally unsatisfying enterprise. There is a wave of emerging data about the role of proton
beam radiation therapy for reirradiation of recurrent HNSCC. In
one retrospective analysis of an ongoing prospective data
registry presented at the Multidisciplinary Head and Neck
Cancer Symposium, Chang et al reported results for 90 patients who received reirridation with proton beam radiation
therapy between 2011 and 2014.27 The actuarial 6-month
locoregional control and OS were 86.5% and 84.1%, respectively. There was one death as a result of disease progression. Grade 3 acute toxicity included mucositis (10.0%),
dermatitis (3.3%), dysphagia (2.2%), and esophagitis (1.1%). In
addition, there were three (5.3%) skin toxicities grade 3 or
worse: a neck ulcer, a chronic neck wound managed with
hyperbaric oxygen treatments, and a pneumocephalus from a
skull base defect. There was one death caused by treatmentrelated bleeding. Short follow-up limits the conclusions that
can be drawn, although the low rates of toxicity are attractive.
180
More studies are needed to elucidate the role of proton beam

radiation therapy in HNSCC.
MANAGEMENT OF RECURRENT/METASTATIC
DISEASE
The LUX-Head and Neck-1 (LHN) trial demonstrated that
afatinib was associated with an improved PFS compared with
methotrexate (2.6 vs. 1.7 months; p = .03).28 In an attempt to
predict which subsets of patients would most benefit from
afatinib therapy, Cohen et al evaluated the association of
ERBB-related biomarkers and p16 on the antitumor activity of
afatinib and methotrexate.29 They noted that afatinib had a
higher response rate in EGFR-directed therapynaive patients
with p16-negative tumors (27.5% vs. 4.8%) and patients with
p16-negative and EGFR-amplified tumors (15.5% vs. 0%). In
contrast, no responses were observed among patients with
p16-positive disease, suggesting that EGFR targeting does not
have a role in HPV-positive HNSCC.
Adkins et al conducted a phase I study of cetuximab and a
suspension formulation of pazopanib among 22 patients with
incurable HNSCC.30 Both cetuximab-naive and cetuximabrefractory patients were eligible. The regimen was well tolerated and the maximally tolerated dose was not reached for
pazopanib at the maximum dose level. Six patients achieved a
partial response (27%), with an additional 11 (50%) patients
achieving stable disease. Three of the six responses were seen
among patients with cetuximab- and platinum-resistant disease, suggesting that this combination may be worthy of future
study.
After failure of platinum, anti-EGFR, and taxanes, patients
with recurrent/metastatic HNSCC are considered refractory,
and methotrexate is the general option with palliative care
intent. Final results from the phase II French trial UNICANCER
ORL03 showed that cabazitaxel met the primary endpoint,
producing a 27.6% 6-week disease control rate with acceptable tolerability among heavily pretreated patients
with refractory recurrent/metastatic HNSCC.31 Among patients with recurrent/metastatic HNSCC refractory to prior
surgery, radiotherapy, or chemotherapy, preliminary phase II
data of first-line paclitaxel combined with panitumumab
showed clinical benefit (median PFS: 7.5 months; 95% CI,
4.98.3; and median OS: 9.9 months; 95% CI, 7.916.3).
This was a comparable efficacy to the standard-of-care
EXTREME regimen, but with a worse (unacceptable) safety
profile (15% fatal adverse events).32
So what can be currently expected for disease that is
refractory to first-line cetuximab and platinum? Retrospective Swiss data in a cohort of 117 patients described the
outcome after second-line regimens (monotherapies, mostly
with methotrexate [55%] or other agents [43%]; only one
patient received platinum combination chemotherapy).
More than half (54%) of the patients did not reach the
point of second-line treatment. The response rate (partial
response, complete response) in 49 patients was 8%, with
33% showing stable disease. Median PFS was 81 days
(95% CI, 5292) and OS was 184 days (95% CI, 115220).
In conclusion, second-line options after first-line cetuximab containing regimens are strongly limited.33
IMMUNE MODULATION IN SQUAMOUS CELL

CARCINOMA OF THE HEAD AND NECK
The advent of immunotherapy has heralded a new era in the
treatment paradigm for many malignancies, including HNSCC.
Multiple trials are underway evaluating the role of immunotherapies including PD-1 inhibitors in metastatic HNSCC.
Preclinical data suggest that radiotherapy may lead to antitumor immune responses. In an attempt to identify immunologic changes seen during radiotherapy, Sridharan et al
prospectively obtained blood samples from 16 consecutive
patients with HNSCC undergoing curative-intent radiotherapy
(with or without platinum-based chemotherapy) at the
beginning and end of therapy.34 The investigators found an
increase in the circulating CD8-positive T-effector cells,
CD4/PD-1positive cells, CD8/LAG3-positive cells, and regulatory T cells (Treg). There was an increase in PD-L1 levels, which
mirrored increases in CD8-positive T cells over the course of
therapy. Levels of CXCL10 decreased during treatment,
whereas those of CXCL16 increased. These findings demonstrate the complex immunologic effects of chemoradiotherapy
and suggest mechanisms for potential synergy among chemotherapy, radiotherapy, and immunotherapy in HNSCC,
which may be exploited in future trials.
Evaluation of these specific immunologic responses may
enable future optimization of the use of this novel class of
agents. In a similar experiment, Ferris et al found that a
combination of the Toll-like receptor 8 agonist motolimod
and chemotherapy resulted in a decrease in the phenotypic markers of suppression in Treg cells (CTLA-4, CD73,
and membrane-bound transforming growth factor-b) in tumor
and peripheral blood.35 In contrast, in nonresponders to
chemotherapy alone, they noted an induction of Treg cells.
Further data relevant to subgroups from KEYNOTE 012
have provided greater insight into the efficacy of the antiPD-1
antibody, pembrolizumab. In the subgroup analyses, the
overall response rate appeared higher for patients who
received fewer than two prior therapies (31/97; 32.0%; two
complete reponses vs. 29 partial responses) compared with
those who had received more than two prior therapies (10/63;
16.0%; 10 partial responses). The overall response rate for
subjects who were previously exposed to both platinum
and cetuximab therapy was 19.1% (18/94; 18 partial responses). Pembrolizumab appeared to provide the most
benefit for patients with recurrent/metastatic HNSCC who
have had smaller, less heavily pretreated tumors.36
As explorations of immunotherapy continue, new indications may emerge. The prevalence of PD-L1 on tumor- and
infiltrating immune cells as well as other characteristics
of immune biology were investigated among 161 patients
with nasopharyngeal carcinoma.37 PD-L1 expression was
detected at greater than 1% and greater than 5% cutoff of
induction chemotherapy in 75% and 17% of tumors, respectively. Baseline clinical characteristics of stage, sex,
and age, as well as clinical outcome data such as time to

local failure, PFS, and OS, did not correlate with PD-L1
expression at either of the cutoffs. Nonetheless, the high
PD-L1 expression rate will likely stimulate further studies
of checkpoint inhibitors in this subsite.38 Interestingly, plasma
Epstein-Barr virus DNA may correlate with response to pembrolizumab treatment in metastatic nasopharyngeal carcinoma.
CARE DELIVERY AND ECONOMICS OF

SQUAMOUS CELL CARCINOMA OF THE HEAD
AND NECK
With the current emphasis on equitable care delivery to all
patients, it is important to understand disparities in the care
of patients with HNSCC. Churilla et al sought to evaluate
the association between health insurance status and clinical
features at presentation, treatment received, and outcomes
among patients with HNSCC.39 Using the SEER database,
they analyzed 53,848 patients with squamous cell carcinoma
of the oral cavity, oropharynx, and larynx. They found that
a larger proportion of Medicaid (72.9%) and uninsured
patients (75.1%) had stage III or IV disease, compared with
insured patients (60.1%; p , .001). After adjusting for
clinical and demographic factors, uninsured patients were
less likely to receive cancer-directed surgery (odds ratio
[OR] 0.86; 95% CI, 0.770.97), whereas Medicaid and uninsured patients were less likely to receive radiotherapy
(Medicaid: OR 0.77; 95% CI, 0.720.81; uninsured: OR
0.68; 95% CI, 0.620.75). On multivariate analysis, patients
with Medicaid (hazard ratio [HR] 1.55; 95% CI, 1.491.62)
or uninsured status (HR 1.48; 95% CI, 1.381.58) had
inferior OS compared with insured patients.
Disparities may exist across multiple facets and at multiple
levels across the cancer journey. It is recognized that optimal cancer-related decision making requires very strong
collaborative work by experienced multidisciplinary teams,
and HNSCC outcomes have been previously reported to be
better in centers treating a high number of these patients.
An abstract presented at ASTRO 2015 analyzed the impact of
radiation oncology provider volume on the clinical outcomes
of patients.40 They found that for each additional patient with
head and neck cancer treated per year, there was a significant
reduction in the risk of gastrostomy tube placement (0.8%), the
rate of aspiration pneumonia (0.8%), and the incidence of a
secondary head and neck surgery (1.0%). Among patients
who received intensity-modulated radiation therapy, each
additional patient with head and neck cancer treated per
year by the radiation oncologist resulted in a decreased risk
of cancer-related death of 0.8%. Addressing disparities in
health care may involve not just providing availability of
basic services but focusing on access to high-level expertise
for these complex conditions.
Finally, another controversial issue revolves around the
relative costs of up-front surgery versus nonsurgical therapy.
There have been several conflicting reports on this question.
Pinheiro and Kramer retrospectively reviewed the records of
72 patients with T1-T3 OPSCC, of whom 42 were treated
181
with a surgical approach (22 received adjuvant treatment),

and 30 were treated nonsurgically.41 In this small analysis,
cost was defined as the reimbursement for all charges in a
6-month episode of treatment after a positive biopsy. The
authors found that the cost was significantly lower for those
treated with surgery only compared with the nonsurgical
group ($38,462 vs. $83,222; p = .03). The cost of surgery

followed by adjuvant chemoradiation was similar to that
associated with primary chemoradiation ($84,598 and
$83,222, respectively; p = .95). In this analysis, the opportunity for cost savings was seen in patients who did not
require adjuvant therapy.
References
1. Cheng CT, Terng SD, Lin CY, et al. 2804 Primary surgery for advanced
oropharyngeal and hypopharyngeal cancers: a nationwide study in
Taiwan. Eur J Cancer. 2015;51:S559.
2. Vishnoi JR, Gupta S, Kumar V, et al. 2847 Importance of tumor thickness
using intraoral ultrasound in predicting nodal metastasis in early oral
cancer. Eur J Cancer. 2015;51:S573.
3. Longton E, Lawson G, Installe J, et al. Individualized super selective
nodal radiation therapy for clinically negative lymph nodes (cN0) in
head and neck squamous cell carcinoma patients based on sentinel
node(s) identification: a prospective phase 2 study of SPECT/CT-guided
volume de-escalation. Int J Radiat Oncol Biol Phys. 93:S71-S72.
4. Panwar A, Batra R, Lydiatt WM, et al. Human papilloma virus positive
oropharyngeal squamous cell carcinoma: a growing epidemic. Cancer
Treat Rev. 2014;40:215-219.
5. Zevallos JP, Yim E, Brennan P, et al. Molecular profile of human papillomavirus positive oropharyngeal squamous cell carcinoma stratified
by smoking status. Int J Radiat Oncol Biol Phys. 2016, 94:864.
6. Lukens JN, Tangsriwong K, Mitra N, et al. Pathological factors predicting
the risk of distant failure for human papillomaviruspositive oropharyngeal squamous cell carcinoma (OPSCC). Int J Radiat Oncol Biol
Phys. 2015; 93:S77.
7. Khwaja SS, Baker C, Haynes W, et al. High E6 gene expression predicts
for distant metastasis and cancer-specific survival in patients with HPVpositive oropharyngeal squamous cell carcinoma. Int J Radiat Oncol Biol
Phys. 2015; 93:S75-S76.
8. Schuler T, Nahar K, Chan D, et al. 2828 Towards predicting tumour
response in patients with locally advanced head and neck cancer
treated with neoadjuvant chemotherapy. Eur J Cancer. 2015;51:S567.
9. Zschaeck S, Lock S, Leger S, et al. 2836 Off-target FDG-PET parameters
have prognostic value in head and neck squamous cell carcinomas
undergoing primary radiochemotherapy and can be used to generate
radiobiological hypotheses. Eur J Cancer. 2015;51:S570.
10. Ou D, Levy A, Blanchard P, et al. 2876 Radiomics feature quantification
in patients with locally advanced head and neck carcinomas treated
with chemoradiotherapy. Eur J Cancer. 2015;51:S582.
11. Chera BS, Amdur RJ, Tepper JE, et al. A prospective phase II trial of deintensified chemoradiation therapy for low risk HPV associated oropharyngeal
squamous cell carcinoma. Int J Radiat Oncol Biol Phys. 2015; 93:S2.
12. Chera BS, Amdur RJ, Tepper JE, et al. Phase 2 trial of deintensified
chemoradiation therapy for low-risk human papillomavirus-associated
oropharyngeal squamous cell carcinoma. Int J Radiat Oncol Biol Phys.
2015; 93:976-985.
13. Melotek JM, Villaflor VM, Brisson RJ, et al. Response-adapted volume
Deescalation (RAVD) in locally advanced head and neck squamous
cell cancer (LA-HNSCC): toxicity and quality of life (QOL) analyses.
Int J Radiat Oncol Biol Phys. 2015; 93:E335.
14. Melotek JM, Villaflor VM, Karrison TG, et al. Response-adapted volume
de-escalation (RAVD) in locally advanced head and neck cancer: efficacy
and human papillomavirus-positive subgroup analysis. Int J Radiat
Oncol Biol Phys. 2016; 94:865.
182
15. Weller MA, Ward MC, Berriochoa CA, et al. Cetuximab-based bioradiation
therapy is associated with higher rates of distant metastases than platinumbased chemoradiation therapy in human papillomaviruspositive oropharyngeal cancer. Int J Radiat Oncol Biol Phys. 2015, 93:S76-S77.
16. Melotek JM, Haraf DJ, Blair EA, et al. Final results of a randomized phase
2 trial investigating the addition of cetuximab to induction chemotherapy and accelerated or hyperfractionated chemoradiation therapy
for locoregionally advanced head and neck cancer: HPV-negative subset
analysis. Int J Radiat Oncol Biol Phys. 2016; 95:868.
17. Frakes JM, Naghavi AO, Strom T, et al. Detection of recurrence in human
papillomavirus-associated oropharynx squamous cell carcinoma. Int J
Radiat Oncol Biol Phys. 2016; 94:866-867.
18. Pires da Silva I, Mouta J, Winckler P, et al. 2818 The role of neck dissection (ND) after neoadjuvant chemotherapy (CT) with docetaxel,
cisplatin and 5-fluorouracil (TPF) followed by concomitant chemoradiation (CR) with cisplatin for patients with locoregionally advanced
squamous cell carcinomas of the head and neck (SCCHN). Eur J Cancer.
2015;51:S563-S564.
19. Mehanna H, Wong WL, McConkey CC, et al. 11LBA Differences in the
quality of life (QoL) and functional outcomes of treatment between
HPV associated (HPV+) and HPV- patients receiving primary chemoradiotherapy in PET-NECK - a multi-centre randomized phase III controlled trial (RCT) comparing PETCT guided active surveillance with
planned neck dissection (ND) for locally advanced (N2/N3) nodal
metastases (LANM) in patients with head and neck squamous cell
cancer (HNC) treated with primary radical chemoradiotherapy (CRT).
Eur J Cancer. 2015;51:S715.
20. Elting LS, Cooksley CD, Chambers MS, et al. Risk, outcomes, and costs of
radiation-induced oral mucositis among patients with head-and-neck
malignancies. Int J Radiat Oncol Biol Phys. 2007;68:1110-1120.
21. Anderson CM, Allen BG, Sun W, et al. Phase 1b/2a trial of superoxide
(SO) dismutase (SOD) mimetic GC4419 to reduce chemoradiation
therapyeInduced oral mucositis (OM) in patients with oral cavity or
oropharyngeal carcinoma (OCC). Int J Radiat Oncol Biol Phys. 2016; 94:
869-870.
22. Merlano M, Vecchio S, Bacigalupo A, et al. 2821 The phase III study
INTERCEPTOR in locally advanced head and neck cancer (LA-HNC).
Preliminary safety report. Eur J Cancer. 2015;51:S560.
23. Dewaele E, Vermorken J, Verschueren C, et al. 2817 12-year follow-up
(FU) data and late local toxicity of two cohorts of patients with
locoregionally advanced squamous cell carcinoma of the head and neck
(LA-SCCHN) treated with concomitant chemoradiation (CCRT) with or
without induction chemotherapy (ICT). Eur J Cancer. 2015;51:S563.
24. Ahn P, Sharma S, Zhou O, et al. A comparative quality of life cohort of
oropharyngeal squamous cell (OPSCC) patients treated with volumetric
modulated radiation therapy (VMAT) versus proton pencil beam
scanning (PBS). Int J Radiat Oncol Biol Phys. 2015; 93:S71.
25. Bossi P, Pala L, Orlandi E, et al. 2827 Role of induction chemotherapy in
the multimodal management of locally advanced epithelial sinonasal
cancer. Eur J Cancer. 2015;51:S567.
26. Hamoir M, Schmitz S, Holvoet E, et al. 2848 Salvage surgery in recurrent head and neck squamous cell carcinoma. Oncologic outcome and
prognostic factors. Eur J Cancer. 2015;51:S573.
27. Chang JHC, Romesser PB, Scher ED, et al. Proton beam reirradiation for
recurrent head and neck cancer: multi-institutional report on feasibility
and early outcomes. Int J Radiat Oncol Biol Phys. 2016;94:865.
28. Machiels JP, Haddad RI, Fayette J, et al; LUX-H&N 1 investigators. Afatinib
versus methotrexate as second-line treatment in patients with recurrent or
metastatic squamous-cell carcinoma of the head and neck progressing on or
after platinum-based therapy (LUX-Head & Neck 1): an open-label, randomised phase 3 trial. Lancet Oncol. 2015;16:583-594.
29. Cohen E, Licitra L, Burtness B, et al. Neck squamous cell carcinoma (R/M
HNSCC) patients (Pts) treated with afatinib versus methotrexate (MTX):
LUX-Head & Neck 1 (LHN1) Int J Radiat Oncol Biol Phys. 2016;94:868869.
30. Adkins D, Ley J, Wildes T, et al. A phase 1 trial of pazopanib added to
cetuximab in patients with incurable head and neck squamous cell
carcinoma (HNSCC). Int J Radiat Oncol Biol Phys. 2016;94:868.
31. Fayette J, Guigay J, Letourneau C, et al. 2800 Cabazitaxel in patients with
refractory recurrent or metastatic (R/M) squamous cell carcinoma of
the head and neck (SCCHN): final results of phase II trial UNICANCER
ORL03. Eur J Cancer. 2015;51:S557.
32. Del Barco Morillo E, Mesia R, Adansa Klain JC, et al. 2802 Phase II study
of first-line paclitaxel (PTX) with panitumumab (P) in patients with
metastatic or recurrent head and neck cancer: TTCC-2009-03 study. Eur
J Cancer. 2015;51:S558.
33. Siano M, Resteghini C, Cau MC, et al. 2865 Outcome of systemic
treatments after first line platinum and cetuximab treatment in patients
34.
35.
36.
37.
38.
39.
40.
41.
with recurrent/metastatic (RM) head and neck squamous cell cancer

(HNSCC): a retrospective analysis. Eur J Cancer. 2015;51:S578.
Sridharan V, Margalit DN, Curreri SA, et al. Systemic immunologic effects of definitive radiation in head and neck cancer. Int J Radiat Oncol
Biol Phys. 2016;94:864.
Ferris RL, Kansy BA, Gibson SP, et al. A phase 1b study of neoadjuvant
immune biomarker modulation with cetuximab and motolimod in head
and neck cancer (HNC). Int J Radiat Oncol Biol Phys. 2016;94:867-868.
Chow L, Mehra R, Haddad R, et al. 2866 Antitumor activity of the antiPD-1 antibody pembrolizumab in subgroups of patients with recurrent/
metastatic head and neck squamous cell carcinoma (R/M HNSCC):
exploratory analyses from KEYNOTE-012. Eur J Cancer. 2015;51:S579.
Kao HF, Hsu C, Huang HC, et al. 2860 Correlation between plasma
Epstein-Barr virus DNA and clinical response to pembrolizumab in
patients with advanced or metastatic nasopharyngeal carcinoma. Eur J
Cancer. 2015;51:S576.
Chan OSH, Kowanetz M, Ng WT, et al. 2877 Characterization of PD-L1
expression and immune cell infiltration in nasopharyngeal cancer (NPC).
Eur J Cancer. 2015;51:S582.
Churilla TM, Egleston B, Dong Y, et al. The impact of health insurance
status on the presentation, local management, and outcomes of
patients with head and neck cancer in the United States. Int J Radiat
Oncol Biol Phys. 2016;94:870.
Boero IJ, Xu B, Mell LK, et al. Impact of radiation oncologist provider
volume on clinical outcomes in head and neck cancer. Int J Radiat Oncol
Biol Phys. 2015; 93:S127.
Pinheiro AD, Kramar RW. Determinants of cost in the treatment of T1-T3
oropharynx cancer. Int J Radiat Oncol Biol Phys. 2016;94:866.
183
Integrating Immune Checkpoint

Inhibitors and Targeted Agents
With Surgery and Radiotherapy
for Patients With Head and Neck
Cancer
CHAIR
David M. Brizel, MD
Durham, NC
SPEAKERS
Andrew G. Sikora, MD, PhD
Baylor College of Medicine
Houston, TX
Nooshin Hashemi Sadraei, MD
University of Cincinnati
Cincinnati, OH
IMMUNOTHERAPY FOR HEAD AND NECK CANCER
Immunotherapy and Checkpoint Inhibitors in Recurrent and

Metastatic Head and Neck Cancer
Nooshin Hashemi Sadraei, MD, Andrew G. Sikora, MD, PhD, and David M. Brizel, MD
OVERVIEW
Immune surveillance is well recognized as an important mechanism to prevent development or progression of head and
neck cancers. Head and neck cancer cells can escape the immune system through multiple mechanisms including development of tolerance in T cells and inhibition of T-cellrelated pathways, generally referred to as checkpoint inhibitors.
This article highlights advances in immuno-oncology treatment approaches in recurrent and metastatic head and neck
squamous cell carcinoma. Clinical trials are discussed in detail, with an emphasis on response dynamics, oncologic efficacy,
safety, and tolerability of checkpoint inhibitors. In addition, developing concepts and ongoing studies in this setting are also
reviewed.
umor cells release antigens, which are captured by antigen presenting cells and subsequently presented to
T cells. Activated T cells then produce a cytotoxic response
resulting in cancer cell death. Memory T cells that are
formed subsequent to these events can result in a durable
antitumor response. However, T-cellrelated pathways can
cause inhibition of these immune responses and result in
tumor cells evading the immune system.1 These pathways
are generally referred to as checkpoints, which are responsible for preventing a chronic autoimmune or inflammatory status. Two of the most commonly discussed
checkpoint inhibitory mechanisms are CTLA-4 and PD-1/PD-L1,
which act at earlier and later stages of the immune response to tumors.1,2 CTLA-4 is present on the surface of
CD4 and CD8 cells and, through binding to CD80 and CD86,
can transmit an inhibitory signal to T cells. CD28 has a
stimulatory effect on T cells and competes with CTLA-4 for
the same ligands (CD80 and CD86); however, CTLA-4 has
greater affinity and is able to overcome effects of stimulatory CD28.
PD-L1 can be found on tumor cells, as well as immune cells
and antigen presenting cells. PD-1 is primarily found on the
surface of T cells. PD-L1 is the main ligand for PD-1, which has
been associated with poor prognosis in many cancer types.
PD-1 is expressed on T cells upon activation and results in
negative regulatory effects on immune function.
PD-1 and PD-L1 both interact with other ligands, too. Most
importantly, PD-1/PD-L2 activation can have similar effects
as PD-1/PD-L1 engagement (although PD-L2 function is less
thoroughly understood). PD-L1 can also activate CD80,

which, as previously discussed, is a known ligand for CTLA-4.
These preclinical findings have led to a new generation of
clinical trials and novel treatments based on single and dual
inhibition of CTLA-4 and PD-1, and in combination with
cytotoxic and targeted therapies.
IMMUNE CHECKPOINT INHIBITORS IN

RECURRENT/METASTATIC HEAD AND NECK
The majority of immune therapy in head and neck cancer has
so far focused on removal of the negative regulatory immune pathways. Many ongoing studies are investigating
the role of immune checkpoint inhibitors in recurrent/
metastatic head and neck cancer, including single agent
and combination of checkpoint inhibitors as dual-inhibition,
combination with immunomodulators, vaccines, chemotherapy,
biologics, and targeted agents, as well as radiation. Most of
these studies have not yet reported their results.
PD-1 Inhibitor
PD-1 inhibition has been the most extensively studied immunotherapeutic strategy in head and neck squamous
cell carcinoma, particularly with the agent pembrolizumab
(formerly MK-3475). This agent has shown promising efficacy compared with historical data and is thought to be well
tolerated (KEYNOTE 012; NCT01848834).3 In a large phase IB
study of pembrolizumab for patients with PD-L1positive
From the Division of Medical Oncology, University of Cincinnati College of Medicine, Cincinnati, OH; Department of OtolaryngologyHead and Neck Surgery, Baylor College of Medicine,
Houston, TX; Department of Radiation Oncology, Duke University Cancer Center, Durham, NC.
Corresponding author: Nooshin Hashemi Sadraei, MD, The Vontz Center for Molecular Studies, University of Cincinnati, 3125 Eden Ave., Cincinnati, OH 45267; email: hashemnn@
ucmail.uc.edu.
e277
HASHEMI SADRAEI, SIKORA, AND BRIZEL
recurrent/metastatic head and neck squamous cell carcinoma, the overall response rate was approximately 20%,
which was similar among patients who were HPV-positive
or HPV-negative.4 PD-L1 positivity in this study was defined
as at least 1% or more stromal or tumor expression of PD-L1
by immunohistochemistry. Based on this definition, 78% of
screened patients were considered positive and eligible for
the study. Among patients with PD-L1 positivity, those with
higher PD-L1 expression demonstrated a much higher response compared with those with a lower PD-L1 expression
(50% vs. 11%, respectively).4
Median progression-free (PFS) and overall survival (OS)
were favorable compared with historical cohorts of patients with heavily pretreated head and neck squamous
cell carcinoma; the median PFS was 9 months (range,
8 to 20 months) and the median OS was 12.6 months.
Unlike response rates, which did not seem to be affected
by HPV status, patients who were HPV-positive, as predicted, showed longer PFS and OS compared with patients who were HPV-negative (median PFS, 17 vs. 8 months,
respectively; median OS, not reached vs. 9 months,
respectively).
Subsequently, in a combined update of two cohorts of
patients with recurrent/metastatic head and neck squamous
cell carcinoma that was heavily pretreated, pembrolizumab
resulted in a response rate of 24% and an additional 25%
with stable disease,5 which was slightly improved compared
with earlier reports.4 The median survival in the combined
KEY POINTS
e278
Immune surveillance is an important mechanism in

prevention of cancer development and inhibition of
growth of tumors and metastasis. Checkpoint pathways
prevent immune response through multiple methods
including CTLA-4 and PD-1/PD-L1 pathways.
A variety of immune checkpoint inhibitors (PD-1/PD-L1
and CTLA-4 inhibitors) have been studied in recurrent
and metastatic head and neck cancers with encouraging
efficacy and toxicity data.
Although many of these treatments are being used with
encouraging results in a variety of different
malignancies, there appears to be some differences in
patterns of response and toxicities, which may be
disease- or population-specific, to some extent.
Studies are ongoing to identify reliable biomarkers.
To date, despite common use of PD-L1 as a biomarker,
consensus in detection methods, definition of positivity,
and its use as a prognostic or predictive biomarker is
lacking, despite common use of PD-L1 as a biomarker.
In addition to checkpoint inhibition, a variety of other
immune therapy strategies are under investigation in
head and neck cancers, some of which include
combination strategies, positive costimulation, and
targeting HPV-specific antigens in treatment of
HPV-related disease.
cohort of patients was 9.6 months (6.6 months to not

reached) with a 1-year estimated survival of 47%.5 This is a
very promising improvement compared with best historical
survival outcome data among patients with recurrent/
metastatic disease (median survival, 10.1 months with
chemotherapy doublet-cetuximab triplet regimen and
7.4 months with chemotherapy doublet therapy).6
Overall drug-related adverse events were 61%, with fatigue (17%) being the most common, followed by endocrinopathies (hypothyroidism), low appetite, pyrexia, and a
variety of skin-related toxicities, each reported in 5% to 8%
of patients. Compared with the use of other checkpoint
inhibitors, mostly in melanoma, no colitis or pneumonitis of
any grades were observed, and high-grade toxicities (grade
3 and above) were seen in 12% of cases.4,5
Many randomized studies subsequently investigated the
role of PD-1 inhibitors in recurrent and metastatic disease.
A randomized study of pembrolizumab compared with
investigators choice chemotherapy (KEYNOTE-040) for
patients with platinum-refractory disease is nearing completion of accrual (NCT02252042),7 and another trial is a
three-arm randomized study in the first-line recurrent and
metastatic setting (KEYNOTE-048) that will evaluate standard platinum/5-fluorouracil/cetuximab compared with
pembrolizumab alone or combination of chemotherapy
plus pembrolizumab (NCT02358031).8
More recently, CheckMate 141, a study of nivolumab,
another PD-1 inhibitor, was stopped early due to a survival
advantage observed among patients who were treated with
nivolumab over investigators choice of therapy (cetuximab,
methotrexate, or docetaxel) in a second-line, recurrent/
metastatic setting. Details of the results have not yet
been released (NCT02105636). 9
PD-L1 Inhibitor
Durvalumab (formerly MEDI4736) is an antiPD-L1 monoclonal antibody that was evaluated in a large global, multicenter, multiarm phase I/II trial in multiple disease
sites, including head and neck squamous cell carcinoma
(NCT01693562/CD-ON-MEDI4736-1108).10 This large trial
included components of dose escalation and dose exploration to define optimum biologic dose. Dose cohorts
started at 0.1 mg/kg and escalated in five consecutive cohorts to 10 mg/kg, which was then taken to the expansion
phase as the selected dose. Although the overall rate of
drug-related adverse events was high (46%), 10 mg/kg of
durvalumab every 2 weeks resulted in grade 3 or 4 drugrelated adverse events (determined by investigators) in only
7% of patients with head and neck cancer, a rate that was
similar to other malignancies.11 Among all types of cancer,
fatigue, gastrointestinal symptoms, endocrinopathies,
dyspnea, peripheral neuropathy, and a variety skin rashes
were reported as more common drug-related adverse
events. Drug-related colitis did not occur, and drug-related
pneumonitis (grade 1 to 2 only) was only reported in 3%
of patients with head and neck cancer. There were no
drug-related discontinuations, and, overall, durvalumab was

well-tolerated among patients with head and neck cancer and
was similar to responses observed in other cancers.12 Among
eight tumor types, response was seen as early as 5 weeks and
was very durable (56 weeks) in some cases. Overall response
rate was 10%, which was generally higher among patients who were PD-L1positive (22% vs. 5%), and, once response was achieved, duration of response was comparable
among patients who were PD-L1positive and PD-L1negative.11
Patients demonstrated a similar pattern: the response rate
was slightly higher among patients who were PD-L1
positive versus PD-L1negative (18% vs. 8%, respectively)
with a respectable duration of response (16 to over 55 weeks).
The overall response rate among patients with head and neck
cancer was 11%, and the disease control rate (response plus
stable disease) at 6 months was 15% (18% for PD-L1positive
vs. 11% for PD-L1negative).12
These findings were very encouraging from a safety and
tolerability standpoint and showed promising improvement
in response and duration of disease control, particularly in
the PD-L1positive population.
PD-L1 Inhibitor in Combination With CTLA-4 Inhibitor

A new series of combination studies in head and neck
squamous cell carcinoma were initiated based on additive or
synergistic effects of drug combinations observed in preclinical studies or other diseases. The PD-1 (nivolumab) and
CTLA-4 inhibitor (ipilimumab) combination was shown to
be superior to CTLA inhibition alone among patients with
metastatic malignant melanoma and superior to either
CTLA-4 or PD-1 inhibition alone among patients who were
PD-L1negative, resulting in significantly prolonged PFS.13
Other emerging data support distinct mechanisms and a
synergistic effect of CTLA-4 and PD-L1 inhibition on immune
regulation, which results in improved tumor response in
mouse models and for patients.14
Multiple ongoing studies are currently evaluating the role
of dual checkpoint inhibition in head and neck squamous
cell carcinoma. Some of these studies are directed toward
PD-L1negative disease and others are stratified based on
PD-L1 status.
Patients with recurrent/metastatic head and neck squamous cell carcinoma whose disease progresses during
platinum therapy will be offered PD-L1 inhibitor alone (those
who are PD-L1positive; 10 mg/kg durvalumab every
2 weeks; HAWK; NCT02207530)15 or those who are PDL1negative will be randomly assigned to one of three arms
to receive combination PD-L1 and CTLA-4 inhibition
(durvalumab and tremelimumab), PD-L1 inhibitor alone
(durvalumab), or CTLA-4 inhibitor alone (tremelimumab;
CONDOR; NCT02319044).16 The primary endpoint in both
studies is overall response rate. A separate large phase III
study will randomly assign patients after PD-L1 expression
based stratification to receive PD-L1 inhibitor alone (durvalumab), dual PD-L1/CTLA-4 inhibitor therapy (durvalumab and tremelimumab), or standard-of-care single-agent
chemotherapy. This study is powered to evaluate the coprimay

endpoints of PFS and OS (EAGLE; NCT02369874).17
PD-L1 Inhibitor in Combination With PD-1 Inhibitor

Some ongoing studies in advanced solid tumors are evaluating
dual PD-1 (MEDI0680)/PD-L1 (durvalumab) inhibition.18 These
studies are currently in earlier phases of dose escalation and
expansion (NCT02118337).19
ADDITIONAL IMMUNE THERAPY STRATEGIES

Positive Costimulation
Another strategy to augment immune system recognition
of cancer cells has focused on enhancement of positive
stimulatory pathways through cytokines and monoclonal
antibodies.
Preclinical data suggest that Toll-like receptor 8 (TLR8)
activation subsequently activates monocytes, natural killer
cells, and myeloid-derived dendritic cells and can potentially
have synergistic effect with chemotherapy and increase the
effectiveness of monoclonal antibody treatment.20
A randomized study of platinum/5-fluorouracil/cetuximab
plus a TLR8 agonist (VTX-2337) or placebo recently completed accrual in first-line recurrent/metastatic head and
neck squamous cell carcinoma (NCT01836029).21 Many
other products are being investigated as single agent or in
combination, including the CD137 agonist, urelumab
(NCT02110082, NCT02253992),22,23 and Ox40 monoclonal
antibodies (in other malignancies).
EGFR Monoclonal Antibody

Cetuximab is a mouse-human chimeric monoclonal antibody
against EGFR, a well-recognized target in head and neck
squamous cell carcinoma.24 Cetuximab has shown efficacy
in recurrent and metastatic head and neck squamous cell
carcinoma and in combination with radiation in the curative
setting.25,26 Several studies have demonstrated immunomodulatory effects with EGFR monoclonal antibodies
through multiple mechanisms including tumor-antigen
specific T-cell activation27,28 as a result of T-cell activation
through interaction with antigen presenting cells. Ongoing
studies are investigating cetuximab combination with
different checkpoint inhibitors, for example nivolumab
(NCT02124850),29 ipilimumab (NCT01860430),30 and urelumab (NCT02110082).22
Therapeutic HPV Vaccines

Given the encouraging results with HPV-specific vaccine
prevention studies,31 many ongoing trials are focused on
treatment of already established HPV-related head and neck
cancer. Subcutaneous injection of HPV-16 peptide vaccine
for patients with recurrent/metastatic head and neck
squamous cell carcinoma was well tolerated and able to
stimulate cellular and humoral immune response.32 Other
studies have used bacterial or viral vaccine-base to deliver
tumor-specific antigen and to use the dual advantage of bacterial/
viral-induced immunogenicity as well as tumor-specific
e279
antigens like E6 and E7, which have a well-established role

in HPV-related malignancies.33
Many such studies are ongoing or in the early development phase and some include multiple HPV-related malignancies (NCT02526316, NCT00019110).34,35
PREDICTORS OF RESPONSE TO CHECKPOINT

INHIBITORS
A post hoc analysis of nearly 200 patients treated with
pembrolizumab showed that patients who were less heavily
retreated (two or less prior lines of therapy) had higher response rates compared with heavily pretreated patients (32%
vs. 16%, respectively). Patients who had smaller volumes
of disease had improved response compared with those
with larger volumes of disease (36% vs. 14%, respectively).5
Small-volume disease was defined in one study as being
less than the median of the sum of the size of the lesions,
where the median was 96 mm (range, 10 to 664 mm).5
Patients with higher PD-L1 expression seem to derive a
greater benefit as demonstrated by a much higher response
compared with those with a lower PD-L1 expression when
treated with pembrolizumab (50% vs. 11%, respectively)4 or
durvalumab (22% PD-L1positive vs. 5% PD-L1negative).12
Interestingly, unlike many prior chemotherapy studies,
patients with HPV-negative and HPV-positive disease did not
demonstrate different response rates when treated with
PD-1 inhibition (25% HPV-negative vs. 24% HPV-positive).5
RESPONSE EVALUATION
Many recent studies have acknowledged differences in time
to initiation of response, as well as patterns of response
between immune therapy and cytotoxic chemotherapy.
Although chemotherapy and targeted therapy directly exert
their effect on cancer cells and can result in a more rapid
development of response, radiographic responses to immune therapy develop more slowly because of the nature of
its indirect effect on cancer cells through activation or
disinhibition of the immune system. The original CTLA-4
inhibitor studies in melanoma (with ipilimumab) showed a
rather slow time to initiation of response (3 to 4 months)36,37;
however, the more recent PD-1 (pembrolizumab) and PD-L1
(durvalumab) inhibitors demonstrate a shorter time to response, as early as 5 weeks,12 with a median time to response
of 9 weeks (range, 7 to 18 weeks).5,38
One of the challenges in response assessment and determination of progression of disease with such treatments
has been a concern for tumor-flare, a phenomenon that has
been attributed to initial T-cell infiltration into the tumor,
causing increase in the size of lesions. This finding has been
mostly reported among patients with melanoma who were
treated with CTLA-4 inhibition.37,39 Interestingly, unlike
patients with melanoma undergoing treatment with ipilimumab, there seem to be very few patients with head and
neck cancer who experience initial radiographic progression
of disease during treatment with PD-1 or PD-L1 inhibitors
who later show response to these treatments (for example,
e280
in the Keynote 012 pembrolizumab study, none of the

132 reported patients had a RECIST-defined radiographic
progression of disease prior to response).38 These data
suggest that, in comparison with patients with melanoma
receiving a CTLA-4 inhibitor, clinicians must have a lower
threshold for recognizing increase in size of disease as
true progression, especially if accompanied by worsening
symptoms.38,40
PROGRESS AND CHALLENGES

Immune-modulating treatments appear to have a clear
clinical benefit for patients with head and neck squamous
cell carcinoma, as shown by multiple clinical studies in the
recurrent/metastatic setting. This benefit seems to be independent of previous treatment, even being observed
among patients who have received multiple lines of therapy
and who have been assumed to have exhausted all options.
Although patients with poor performance status and multiple comorbidities have been excluded from these early
trials, the safety and toxicity profiles of these drugs have
been very promising, suggesting applicability to a wider
population of patients compared with many cytotoxic
chemotherapeutic agents.
Many questions remain to be answered. As far as design of
future studies, perhaps most importantly, we must address
whether the primary benefit should be measured by response or
survival. Although, among patients who have been heavily
pretreated, response rates of 10% to 20% are considered respectable (compared with the 3% to 13% historical response rates
among previously treated recurrent/metastatic patients25,41-43),
it is the duration of response and the improvement in OS that
clearly differentiates these drugs from chemotherapy (median
time to progression, 2 to 4 months; median OS, 6 to 7 months
in chemotherapy studies of patients who were previously
treated25,41,43). The use of response as a primary endpoint for
some studies may therefore underestimate the actual benefit.40
It also appears that the Immune-Related Response Criteria
(irRECIST), which was originally developed based on response
evaluation for patients with melanoma who were undergoing
CTLA-4 inhibition, may not be the most accurate method of
measuring response (and hence benefit) across all checkpoint
inhibitors and all types of disease.39,40,44 The search for a reliable biomarker to predict clinic benefit continues. The
prognostic value of PD-1 and PD-L1 expression has varied
across studies, potentially because of the difficulty of accurately
quantifying expression levels and determining appropriate
cutoff levels for antigen positivity. An additional confounder
is that upregulation of PD-1 and PD-L1 can be induced by
interferon-gamma, one hallmark of a strong intratumoral T-cell
response.45-47 In fact, several studies across different cancer
types have shown positive association of PD-L1 expression with
survival in head and neck squamous cell carcinoma48,49 and
other cancers.50-52 This seemingly paradoxical association is
consistent with upregulation of PD-1/PD-L1 in response to host
immune pressure and suggests that these molecules may serve
as a potential biomarker of vigorous antitumor immunity.
As the field of biomarker research is evolving, an additional

candidate biomarker of response, consistent with PD-1/PDL1 axis upregulation induced by antitumor immune
response, is the presence of so-called T-cell inflamed
phenotype gene expression signatures,53 which has been
suggested to predict benefit from antiPD-1 in patients
with head and neck squamous cell carcinoma.54
CONCLUSION
In summary, even in the absence of a RECIST- or irRECISTdefined response, there appears to be a heretofore
unrecognized population of long-term survivors with these

treatments. We must improve our understanding of the role
of PD-L1 positivity and other clinical and biologic predictive
biomarkers in patient selection and whether a specific patient population benefits more from single-agent versus
combination therapy. Other important issues include finding
the best combination treatment (with other immuno-oncology
products, chemotherapy, targeted therapy, or radiation), sequencing approach, and duration of treatment and whether
these treatments should be introduced earlier in more curative
settings.
References
2. Shekarian T, Valsesia-Wittmann S, Caux C, et al. Paradigm shift in
oncology: targeting the immune system rather than cancer cells.
Mutagenesis. 2015;30:205-211.
3. NCT01848834. Study of Pembrolizumab (MK-3475) in Participants
With Advanced Solid Tumors (MK-3475-012/KEYNOTE-012). https://
4. Chow LQ, Burtness B, Weiss J, et al. A phase Ib study of pembrolizumab
(pembro; MK-3475) in patients (pts) with human papilloma virus (HPV)positive and negative head and neck cancer (HNC). Ann Oncol. 2014;25:
v1-v41 (suppl 5; abstr LBA31).
5. Chow L, Mehra R, Haddad R, et al. Antitumor activity of the anti-PD-1
antibody pembrolizumab in subgroups of patients with recurrent/
metastatic head and neck squamous cell carcinoma (R/M HNSCC):
exploratory analyses from KEYNOTE-012. Eur J Cancer. 2015;51:S579
6. Vermorken JB, Mesia R, Rivera F, et al. Platinum-based chemotherapy
plus cetuximab in head and neck cancer. N Engl J Med. 2008;359:
1116-1127.
7. NCT02252042. Pembrolizumab (MK-3475) Versus Standard Treatment
for Recurrent or Metastatic Head and Neck Cancer (MK-3475-040/
KEYNOTE-040). https://clinicaltrials.gov/ct2/show/NCT02252042.
8. NCT02358031. A Study of Pembrolizumab (MK-3475) for First Line
Treatment of Recurrent or Metastatic Squamous Cell Cancer of the
Head and Neck (MK-3475-048/KEYNOTE-048). https://clinicaltrials.
9. NCT02105636. Trial of Nivolumab vs Therapy of Investigators Choice in
Recurrent or Metastatic Head and Neck Carcinoma (CheckMate 141).
https://clinicaltrials.gov/ct2/show/NCT02105636. Accessed March 18,
2016.
10. NCT01693562. A Phase 1/2 Study to Evaluate MEDI4736. https://
11. Segal NH, Hamid O, Hwu W, et al. A phase I multi-arm dose-expansion
study of the anti-programmed cell death-ligand-1 (PD-L1) antibody
MEDI4736: preliminary data. Ann Oncol. 2014;25:iv365 (suppl 4; abstr
1058PD).
12. Segal NH, Ou SHI, Balmanoukian AS, et al. Safety and efficacy of
MEDI4736, an anti-PD-L1 antibody, in patients from a squamous cell
carcinoma of the head and neck (SCCHN) expansion cohort. J Clin Oncol.
2015;33: (suppl; abstr 3011).
13. Larkin J, Hodi FS, Wolchok JD. Combined nivolumab and ipilimumab
or monotherapy in untreated melanoma. N Engl J Med. 2015;373:
1270-1271.
cancer. Nature. 2015;520:373-377.
15. NCT02207530. Phase II Study of MEDI4736 Monotherapy in Treatment
of Recurrent or Metastatic Squamous Cell Carcinoma of the Head
and Neck. https://clinicaltrials.gov/ct2/show/NCT02207530. Accessed
March 18, 2016.
16. NCT02319044. Phase II Study of MEDI4736, Tremelimumab, and
MEDI4736 in Combination w/ Tremelimumab Squamous Cell Carcinoma of the Head and Neck. https://clinicaltrials.gov/ct2/show/
17. NCT02369874. Study of MEDI4736 Monotherapy and in Combination
With Tremelimumab Versus Standard of Care Therapy in Patients With
Head and Neck Cancer. https://clinicaltrials.gov/ct2/show/NCT02369874.
18. Hamid O, Chow LQM, Tavakkoli F, et al. Phase I, open-label study of
MEDI0680, an anti-programmed cell death-1 (PD-1) antibody, in
combination with MEDI4736, an anti-programmed cell death ligand-1
(PD-L1) antibody, in patients with advanced malignancies. J Clin Oncol.
2015;33 (suppl; abstr TPS3087).
19. NCT02118337. A Phase 1, Open-label Study to Evaluate the Safety and
Tolerability of MEDI0680 (AMP-514) in Combination With MEDI4736 in
Subjects With Advanced Malignancies. https://clinicaltrials.gov/ct2/
20. Lu H, Dietsch GN, Matthews MA, et al. VTX-2337 is a novel TLR8 agonist
that activates NK cells and augments ADCC. Clin Cancer Res. 2012;18:
499-509.
21. NCT01836029. Chemotherapy Plus Cetuximab in Combination With
VTX-2337 in Patients With Recurrent or Metastatic Squamous Cell
Carcinoma of the Head and Neck. https://clinicaltrials.gov/ct2/show/
22. NCT02110082. Combination Study of Urelumab and Cetuximab in
Patients With Advanced/Metastatic Colorectal Cancer or Advanced/
Metastatic Head and Neck Cancer. https://clinicaltrials.gov/ct2/show/
23. NCT02253992. Study of the Safety and Tolerability of Urelumab Administered in Combination With Nivolumab in Solid Tumors and B-cell
Non-Hodgkins Lymphoma. https://clinicaltrials.gov/ct2/show/NCT02253992.
24. Grandis JR, Zeng Q, Drenning SD, et al. Normalization of EGFR mRNA
levels following restoration of wild-type p53 in a head and neck
squamous cell carcinoma cell line. Int J Oncol. 1998;13:375-378.
25. Vermorken JB, Trigo J, Hitt R, et al. Open-label, uncontrolled, multicenter phase II study to evaluate the efficacy and toxicity of cetuximab
as a single agent in patients with recurrent and/or metastatic squamous
e281
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
cell carcinoma of the head and neck who failed to respond to platinumbased therapy. J Clin Oncol. 2007;25:2171-2177.
Bonner JA, Harari PM, Giralt J, et al. Radiotherapy plus cetuximab for
squamous-cell carcinoma of the head and neck. N Engl J Med. 2006;354:
567-578.
Srivastava RM, Lee SC, Andrade Filho PA, et al. Cetuximab-activated
natural killer and dendritic cells collaborate to trigger tumor antigenspecific T-cell immunity in head and neck cancer patients. Clin Cancer
Res. 2013;19:1858-1872.
Ferris RL. Immunology and immunotherapy of head and neck cancer.
J Clin Oncol. 2015;33:3293-3304.
NCT02124850. A Phase Ib Study of Neoadjuvant of Cetuximab Plus
Motolimod and Cetuximab Plus Motolimod Plus Nivolumab. https://
NCT01860430. A Phase Ib Trial of Concurrent Cetuximab (ERBITUX) and
Intensity Modulated Radiotherapy (IMRT) With Ipilimumab (YERVOY) in
Locally Advanced Head and Neck Cancer. https://clinicaltrials.gov/ct2/
Schiller JT, Castellsague X, Garland SM. A review of clinical trials of
Human papillomavirus prophylactic vaccines. Vaccine. 2012;30(Suppl
5):F123-F138.
Zandberg DP, Rollins S, Goloubeva O, et al. A phase I dose escalation trial
of MAGE-A3- and HPV16-specific peptide immunomodulatory vaccines in
patients with recurrent/metastatic (RM) squamous cell carcinoma of the
head and neck (SCCHN). Cancer Immunol Immunother. 2015;64:367-379.
Sewell DA, Pan ZK, Paterson Y. Listeria-based HPV-16 E7 vaccines limit
autochthonous tumor growth in a transgenic mouse model for HPV-16
transformed tumors. Vaccine. 2008;26:5315-5320.
NCT02526316. Cisplatin-based Chemotherapy Combined With P16_3763 Peptide Vaccination in Patients With HPV-positive Cancers (VICORYX-2).
https://clinicaltrials.gov/ct2/show/NCT02526316. Accessed March 18,
2016.
NCT00019110. Vaccine Therapy in Treating Patients With Advanced
or Recurrent Cancer. https://clinicaltrials.gov/ct2/show/NCT00019110.
Hodi FS, ODay SJ, McDermott DF, et al. Improved survival with ipilimumab in patients with metastatic melanoma. N Engl J Med. 2010;363:
711-723.
Weber JS, ODay S, Urba W, et al. Phase I/II study of ipilimumab
for patients with metastatic melanoma. J Clin Oncol. 2008;26:5950-5956.
Seiwert TY, Haddad RI, Gupta S, et al. Antitumor activity and safety of
pembrolizumab in patients (pts) with advanced squamous cell carcinoma of the head and neck (SCCHN): Preliminary results from KEYNOTE012 expansion cohort. J Clin Oncol. 2015;33 (suppl; abstr LBA6008).
Wolchok JD, Hoos A, ODay S, et al. Guidelines for the evaluation of
Seiwert T. Immunotherapy for head and neck cancer. Paper presented
at: 2016 Multidisciplinary Head and Neck Cancer Symposium; February
2016; Scottsdale, AZ.
e282
41. Argiris A, Ghebremichael M, Gilbert J, et al. Phase III randomized,

placebo-controlled trial of docetaxel with or without gefitinib in recurrent or metastatic head and neck cancer: an eastern cooperative
oncology group trial. J Clin Oncol. 2013;31:1405-1414.
42. Fury MG, Sherman E, Lisa D, et al. A randomized phase II study of
cetuximab every 2 weeks at either 500 or 750 mg/m2 for patients with
recurrent or metastatic head and neck squamous cell cancer. J Natl
Compr Canc Netw. 2012;10:1391-1398.
43. Stewart JS, Cohen EE, Licitra L, et al. Phase III study of gefitinib
compared with intravenous methotrexate for recurrent squamous
cell carcinoma of the head and neck [corrected]. J Clin Oncol. 2009;27:
1864-1871.
44. Hoos A, Wolchok JD, Humphrey RW, et al. CCR 20th Anniversary
commentary: immune-related response criteriacapturing clinical activity in immuno-oncology. Clin Cancer Res. 2015;21:4989-4991.
45. Concha-Benavente F, Srivastava RM, Trivediet S, et al. Identification of
the cell-intrinsic and -extrinsic pathways downstream of EGFR and
IFNgamma that induce PD-L1 expression in head and neck cancer.
Cancer Res. 2016;76:1031-1043.
46. Abiko K, Matsumura N, Hamanishi J, et al. IFN-gamma from lymphocytes induces PD-L1 expression and promotes progression of ovarian
cancer. Br J Cancer. 2015;112:1501-1509.
47. Fang W, Zhang J, Honget S, et al. EBV-driven LMP1 and IFN-gamma upregulate PD-L1 in nasopharyngeal carcinoma: implications for oncotargeted therapy. Oncotarget. 2014;5:12189-12202.
48. Vassilakopoulou M, Avgeris M, Velcheti V, et al. Evaluation of PD-L1
expression and associated tumor-infiltrating lymphocytes in laryngeal
squamous cell carcinoma. Clin Cancer Res. 2016;22:704-713.
49. Oliveira-Costa JP, de Carvalho AF, da Silveira da GG, et al. Gene expression patterns through oral squamous cell carcinoma development:
PD-L1 expression in primary tumor and circulating tumor cells. Oncotarget. 2015;6:20902-20920.
50. Droeser RA, Hirt C, Viehl CT, et al. Clinical impact of programmed cell
death ligand 1 expression in colorectal cancer. Eur J Cancer. 2013;49:
2233-2242.
51. Lipson EJ, Vincent JG, Loyo M, et al. PD-L1 expression in the Merkel cell
carcinoma microenvironment: association with inflammation, Merkel
cell polyomavirus and overall survival. Cancer Immunol Res. 2013;1:
54-63.
52. Taube JM, Anders RA, Young GD, et al. Colocalization of inflammatory
response with B7-h1 expression in human melanocytic lesions supports
an adaptive resistance mechanism of immune escape. Sci Transl Med.
2012;4:127ra37.
53. Gajewski TF, Louahed J, Brichard VG. Gene signature in melanoma
associated with clinical activity: a potential clue to unlock cancer immunotherapy. Cancer. 2010;16:399-403.
54. Seiwert TY, Burtness B, Weiss J, et al. Inflamed-phenotype gene expression signatures to predict benefit from the anti-PD-1 antibody
pembrolizumab in PD-L1+ head and neck cancer patients. J Clin Oncol.
2015;33:15s (suppl; abstr 6017).
Multimodality Management of
Locoregionally Recurrent or
Second Primary Head and Neck
Cancer
CHAIR
Stuart J. Wong, MD
Medical College of Wisconsin
Milwaukee, WI
SPEAKERS
Dwight E. Heron, MD
University of Pittsburgh Medical Center
Pittsburgh, PA
Kerstin Stenson, MD
Rush University Medical Center
Chicago, IL
WONG ET AL
Locoregional Recurrent or Second Primary Head and Neck

Cancer: Management Strategies and Challenges
Stuart J. Wong, MD, Dwight E. Heron, MD, MBA, FACRO, FACR, Kerstin Stenson, MD, FACS, Diane C. Ling,
MD, and John A. Vargo, MD
OVERVIEW
Treatment of patients with locoregional recurrent or second primary head and neck squamous cell cancer (HNSCC) has been
guided by well-reasoned principles and informed by carefully tested chemotherapy and radiation regimens. However,
clinical decision making for this population is complicated by many factors. Although surgery is generally considered the
treatment of choice for patients with HNSCC with recurrent disease or new second primary disease in a previously irradiated
field, operability of cases is not always straightforward. Postoperative treatment is frequently warranted but carries
significant risk. In addition, the rapid rise in the incidence of HPV-associated HNSCC raises the question of whether
established treatment paradigms should be re-examined in this population of patients with a much better prognosis than the
non-HPV population. Furthermore, new radiation techniques and new systemic agents show early promising results in
recent clinical studies, suggesting potential for practice-changing effects in the future management of this disease. This
article examines each of the treatment modalities used in the care of patients with HNSCC with recurrent or new second
primary disease and provides a perspective to aid clinicians in the management of this disease.
he majority of patients newly diagnosed with HNSCC

present with locally advanced cancer and undergo
combined modality therapy, including radiation therapy. For
this patient group who subsequently fail in a local or regional
site, the therapeutic options are complicatedmost commonly because of overlap of the previous radiation portal. In
addition, the management of locoregional recurrent or
second primary head and neck cancer has grown in complexity as a result of the increase of the relative incidence
of HPV-associated HNSCC compared with non-HPV HNSCC.
Practicing oncologists have more treatment options at their
disposal, which further complicates the treatment decisionmaking process. For instance, reirradiation was considered
a treatment option limited to high-volume centers with
abundant experience with this therapy. It is now far more
acceptable to use reirradiation outside of the clinical trial
setting. The wide use of this salvage treatment modality is
evident by numerous publications demonstrating safety and
efficacy of concurrent chemotherapy and reirradiation.
It is well recognized that diagnosis of HPV-associated
HNSCC carries a more favorable prognosis compared with
non-HPV HNSCC.1 However, the failure pattern of HPVassociated oropharynx cancer (OPC) and its sensitivity to
treatment in the setting of recurrent disease is less well
appreciated. Examination of the treatment for patients

with HPV-positive OPC in Radiation Therapy Oncology
Group (RTOG) studies provides a clear picture as to the
failure patterns of this patient population.2 This analysis
showed that following disease progression, 2-year overall
survival was nearly twice as high for patients with HPVpositive OPC compared with patients who were HPVnegative (55% vs. 28%, respectively). The majority of
these patients fail within 1 year after completing definitive therapy. Although ostensibly it is logical to apply
the same oncologic principles to each of these two patient
groups, it is important to note that we do not have
abundant clinical trial information with long-term followup data to guide these treatment decisions. The sequence
of salvage treatment options for these two categories of
patients given the differences in outcome remains the
subject of future investigation.
In the following discussion, we will review treatment
options for patients with recurrent and second primary
HNSCC in the context of the evolving epidemiologic patterns
of this disease. In particular, we will discuss surgical approaches to this disease, the use of new and conventional
radiation therapy techniques, and systemic therapeutic
options.
From the University of Pittsburgh Cancer Institute, Pittsburgh, PA; University of Pittsburgh School of Medicine, Pittsburgh, PA; Rush University Medical Center, Chicago, IL; Medical
College of Wisconsin, Milwaukee, WI.
Corresponding author: Stuart J. Wong, MD, Medical College of Wisconsin, 9200 West Wisconsin Ave., Milwaukee, WI 53226; email: swong@mcw.edu.
e284
RECURRENT HEAD AND NECK CANCER
SURGICAL MANAGEMENT OF PREVIOUSLY

IRRADIATED HEAD AND NECK CANCER: SALVAGE
AND TREATMENT MANAGEMENT STRATEGIES
BEFORE AND AFTER REIRRADIATION
The goals of multimodality strategies for patients with advanced head and neck cancer encompass organ preservation
through less radical surgery while improving outcomes. As
survival improved and multimodality organ preservation
strategies became more successful, the role of surgery was
redefined and has become increasingly more prominent for
patients with recurrent disease. Most surgeons hold a general
dogma that performing surgery on a patient who has previously undergone radiation or chemoradiation (CRT) can be
a challenging task. The salvage surgical approach may be not
only technically challenging but also associated with a higher
incidence of complications, such as wound-healing problems,
fistula, and infection. This discussion will address some of the
challenges that patients with recurrent HNSCC confront when
they are faced with salvage surgery. Important preoperative
considerations will be discussed, specific site considerations
will be presented, and the rationale for surgical palliation will
be addressed.
Preoperative Considerations
Patient factors. Adequate access to nutrition is critical not
only for perioperative health and wound healing but also for
efficient postoperative care of the patient. Extended postoperative nil per os status should be anticipated for patients
with previous radiation or CRT who undergo large resections. One should strongly consider placement of a
gastric tube. Previous studies have shown the benefit of a
gastric tube compared with a nasogastric tube in terms of
patient comfort, cosmesis, durability, and ease of tube
feeding.3,4 It is also essential that patients are euthyroid
before any planned surgery. Thyroid-stimulating hormone
levels are routinely evaluated not only before surgery but
also for the entire lifetime of the patient who has undergone
radiation or CRT. Hypothyroidism negatively affects wound
healing and occurs in at least 40% of patients treated with
radiation or CRT.5,6 For patients and physicians to gain

perspective and evaluate risk/benefit ratio of surgery, patients with advanced head and neck cancer who have
completed radiation or CRT need a full metastatic workup
before surgery. This includes CT scan of the head, neck, and
chest and/or PET scan.
Creating a safe wound. Experience and literature document the increased risk of fistula and other complications for
patients undergoing salvage surgery who have received
prior radiation or CRT. Use of nonirradiated free tissue has
been shown to provide reliable reconstruction that allows
for maximal resection, unhampered by concerns of adequate tissue availability.7-10 For patients who undergo
surgery after prior irradiation, free tissue transfer with
microvascular anastomoses allows for successful wound
rehabilitation.11 Paramount in caring for our patients who
will undergo resection in an irradiated bed is the surgeons
understanding of the type and volume of tissue needing
resection, the vulnerability of the surrounding structures,
and the amount of nonirradiated tissue needed to reconstruct the defect. In essence, a safe wound must be
created to limit the potential toxicity from wound breakdown after salvage surgery or tissue necrosis from reirradiation. A safe wound is one that will offer protection of
carotid and vertebral arteries (preventing major vessel
rupture) and cover potential exposed bone in a hostile
wound environment (fistula or tissue necrosis). Surgeons
must carefully evaluate preoperative imaging to determine
what anatomic areas will be vulnerable in these situations
(Figs. 1 and 2). Some patients are scheduled to undergo
(chemo-)reirradiation without planned prior surgery. The
team is responsible for determining if carotid arteries or
other areas will become exposed with the added high
FIGURE 1. Patient With Recurrent Palate Carcinoma

After Chemoradiation
KEY POINTS
HPV remains prognostic in recurrent HNSCC, challenging

existing paradigms.
Salvage surgery remains standard whenever feasible.
Modern reirradiation with intensity-modulated
radiation therapy or stereotactic body radiation therapy
widens the therapeutic ratio of reirradiation.
The addition of cetuximab to 5-fluorouracil and cisplatin
has redefined the standard for systemic therapy in
HNSCC not amendable to surgery or reirradiation, even
in HPV-positive HNSCC.
Novel systemic therapies, including immune checkpoint
inhibitors, show promising early results warranting
additional ongoing prospective study.
Arrow shows internal carotid artery and adjacent tumor.
e285
WONG ET AL
FIGURE 2. Same Patient After Resection and

Reconstruction With Radial Forearm Free Flap
surgery. The influence of these latter factors has driven oncologists and scientists to develop reirradiation treatment
strategies. Salvage surgery as a first step seems to positively
affect survival, as does concomitant chemo-reirradiation.22,23
Despite these improvements, acute and late treatment toxicities remain high.
Specific Sites and Surgical Approaches
Vulnerable area of carotid is covered with healthy tissue.
radiation doses. If the patient is at risk for wound breakdown, if feasible, the surgeon should plan to cover vulnerable vessels or bony areas with vascularized tissue prior
to reirradiation. Often, the team must weigh the risk of
major vessel rupture (if the wound is not safe) with the risk of
delaying the start of reirradiation.
Rationale for salvage surgery. The treatment of patients with
recurrent advanced disease remains a challenge and demands
the thoughtful input of multidisciplinary teams. Although historical series have shown overall poor prognosis for this group of
patients in general, surgical salvage is considered the standard
treatment and offers the best opportunity for locoregional
control and perhaps cure when compared with chemotherapy
or reirradiation alone. In arguably one of the most notable
studies of salvage surgery, Goodwin et al found that 2-year
survival was predicted by stage (73% for stage I and 22% for
stage IV; p = .0005) and site (laryngeal 76%, oral cavity 47%, and
pharyngeal sites 24%; p = .0645).12 Others have found that time
to recurrence and interval from previous radiation, use of
chemotherapy, comorbidity, performance status, pre-existing
organ dysfunction, and N3 neck disease are also negatively
correlated with survival.13-22 Surgical salvage, even with negative margins, is associated with high recurrence rates, with
approximately 67% overall failure rate. Multiple factors including tumor-acquired radiation and chemotherapy resistance,
pathologic patterns of submucosal multiple microscopic nests,
and/or perineural, perivascular, or perilymphatic invasion are
thought to contribute to the recurrence rates after salvage
e286
Neck management. Salvage neck dissections for isolated

neck recurrence or in combination with primary site recurrence are a heterogeneous group. Dissections may be
straightforward radical neck dissection with coverage of
vessels with a simple pectoralis myocutaneous flap. Alternatively, salvage neck dissections may involve resection/
reconstruction of the carotid artery, cranial nerves, deep
neck muscles, and/or vertebral arteries. Jones et al noted a
31% 5-year survival in the 69 (of 589) patients undergoing a
second neck dissection. Younger and more fit patients and
those with low T and N stage at presentation had better
survival than older patients and those with high T and
N stage.24 Mourad et al presented data for 51 patients (nine
who had undergone prior radiation) who underwent carotid
resection and in-line carotid artery bypass grafting. There
was an 82% 2-year survival and low neuromorbidity (2 of 51
had perioperative stroke).25 These findings and others support aggressive neck resection and carotid replacement
particularly for younger patients.
Primary sites. There have been great technical advances in
surgical approaches for salvage of primary site recurrences.
These developments mirror those occurring in surgical approaches for the untreated patient. For example, in recurrent nasopharyngeal cancers, image-guided endosurgery
is precise and effective, offers improved recovery, and
is safer than open approaches.26 Salvage endoscopic
nasopharyngectomy has also been found to be superior to
intensity-modulated radiation therapy (IMRT) in terms of
survival and quality of life parameters.27 Transoral robotic
surgery has been found to be an oncologically sound alternative to open surgical procedures and is superior in
terms of functional and recovery outcomes.28 In addition,
transoral laser microsurgery has been shown to have
comparable oncologic outcomes with superior function
and less morbidity compared with open partial or total
laryngectomy.29
Palliation
Survival statistics, surgical complication rates, and treatment
toxicities, as significant as they are, do not provide a full
picture of the positive and negative effects of surgical
treatment on patients with recurrent cancer. More importantly, surgery has a compelling role in palliation for patients
with recurrent head and neck cancers. All major U.S.
medical, ethical, and religious organizations recognize that it
is imperative to treat the distressing symptoms of patients
who are dying.30 Palliative surgical treatment is nuanced and
requires knowledge of a patients unique medical condition

at the time, as well as the patients personality and family
dynamics. A clear definition of palliative surgery has evolved
and remains consistent with those standardized principles of
palliative care in general: Surgical procedures used for the
primary intention of improving quality of life or relieving
symptoms caused by advanced disease.30
The indications for palliative surgery are not well defined, though many would support palliative resection when
confronted with potential suffocation, carotid rupture,
prevention of quadriplegia (in the instance of tumor erosion
into cervical spine), or control of foul, ulcerative wounds.
Patients typically do not live longer (though a small percentage are cured by the operation with or without postoperative adjuvant treatment), but there is great potential
for better living during the patients remaining life. The
pattern of disease progression can change, where instead of
dying with a painful, odiferous, ulcerative mass, the patient
is allowed to die of distant disease.31-34 Resource utilization
associated with palliative-intent surgeries have been shown
to be similar to that associated with curative intent surgeries.35 However, patients with palliative-intent surgery
have differing resource needs.
A quote by Gaisford perhaps best summarizes the goals in
palliative surgery: Palliative surgical procedures can be of
greater magnitude than curative operations, and the head
and neck cancer surgeon must be ready, willing, and
ablebut mostly willingto carry them out. There is in the
last analysis no absolute rulethe surgeon who accepts
patients with cancer for their initial care is obligated to care
for them throughout the course of their disease.36
The multidisciplinary teams ability to support our patients who undergo salvage surgery has improved over
the years through more robust reconstructions and advances in perioperative care. Nonetheless, salvage surgery is
nearly always radical, morbid, and technically challenging.
Awareness of a patients social support is as critical as
knowledge of their comorbidities and physical ability to
withstand an often-long operation with even longer recovery. What must be clear is the patients willingness to
accept that survival, not function, is the overriding goal in
salvage treatment. Some patients may want to maintain as
much quality of life (as defined by the individual patient) and
choose palliative chemotherapy or hospice care. Knowing
our patients well enough to help guide them in their choice
of treatment remains one of the great privileges of being a
doctor.
ROLE OF CONVENTIONAL AND ABLATIVE

RADIOTHERAPY IN THE MANAGEMENT OF
RECURRENT, SECOND PRIMARY, AND/OR
PREVIOUSLY IRRADIATED HEAD AND NECK
CANCER
Conventional Reirradiation
Historically, reirradiation to locoregional recurrences or
new primary cancers within a previously irradiated field has
been discouraged because of concern over excessive normal
tissue toxicity. An early retrospective study by the Institut

Gustave-Roussy reported outcomes on 169 patients with
unresectable disease treated with curative-intent full-dose
reirradiation with or without chemotherapy.37 At a median
follow-up of 70 months, the median overall survival was
10 months with an overall survival rate of 21% at 2 years and
9% at 5 years. This modest improvement in salvage outcome
came at the expense of high toxicity rates, with 46% of
patients experiencing grade 3 or greater mucositis. Others
have reported similar results using conventional radiation
therapy techniques.38-40
Two phase II RTOG trials using split-course hyperfractionated reirradiation and chemotherapy subsequently
followed. RTOG 9610 found that concurrent hydroxyurea
and 5-fluorouracil (5-FU) resulted in a median survival of
8.5 months and overall survival rates of 15.2% and 3.8% at
2 and 5 years, respectively.41 RTOG 9911 used concurrent
cisplatin and paclitaxel, achieving a slightly better median
survival of 12.1 months and a 2-year survival rate of 25.9%.42
However, high toxicity was reported in both studies, with a
63%78% rate of grade 3 or greater acute toxicity, 22%37%
rate of grade 3 or greater late toxicity, and overall 8%9%
treatment-related death.
Given the consistently high rates of toxicity reported,
many have questioned whether treatment toxicity outweighs the potential benefits of reirradiation, especially
given that the median survival only marginally exceeds the
6 to 9 months of chemotherapy alone.43-46 Efforts have been
made to identify patients who might benefit the most from
this highly aggressive approach. In RTOG 9610, an interval
of more than 1 year from prior radiation therapy was
identified as a significant predictor of improved survival
(median survival, 9.8 months vs. 5.8 months; p = .036).41 In
a long-term study at The University of Chicago, the authors
identified reirradiation dose, triple-agent chemotherapy
(cisplatin, paclitaxel, and gemcitabine), and surgery as independent prognostic factors for overall survival, progressionfree survival, and locoregional control.22 Nineteen (16.5%)
patients died of treatment toxicity in this study, attesting to
the morbidity of this approach. More recently, HPV status
has been shown to be strongly prognostic in the setting
of reirradiation and may help to select patients more likely
to benefit from aggressive salvage treatments such as
reirradiation.2,47
Intensity-Modulated Radiation Therapy

Recent technologic advances such as IMRT offer potentially
superior local control rates in the reirradiation setting
with fewer late effects compared with more conventional
techniques. For instance, Lee et al reported a 2-year
locoregional control rate of 52% compared with 20% among
patients treated with IMRT and those treated with conventional radiation techniques, respectively.48 In addition,
patients with locoregional progression-free disease had
better 2-year overall survival compared with those with
locoregional failure, further demonstrating the need to
maintain locoregional control to improve survival. High
e287
WONG ET AL
treatment morbidity occurred, although at a lower rate than

reported in prior studies utilizing conventional techniques.
Acute and late-grade 3-4 toxicities were reported in 23% and
15% of patients, respectively.48 Similar findings have been
published by other authors.49-51
Stereotactic Body Radiation Therapy

Stereotactic body radiation therapy (SBRT) offers further opportunities to deliver highly conformal treatments with comparable local control and overall survival, shorter treatment
times, and decreased toxicity compared with conventional
techniques.52 The most common SBRT regimen consists of five
fractions delivered on alternating days,52 which has been
shown to be safer than treatment delivered on consecutive
days.53,54 There has been increasing interest in SBRT as an
alternative technique for reirradiation, with several phase I and
phase II trials showing promising safety and efficacy results.55-58
A phase I dose-escalation trial from the University of Pittsburgh
Cancer Institute showed that reirradiation up to 44 Gy using
SBRT is well tolerated in the acute setting.58 A subsequent
dose-escalation study identified a clear dose-volume response
with local control that became more prominent for gross tumor
volumes greater than 25 cc with longer follow-up times.59
The use of novel EGFR-targeted agents may help to further
improve disease outcome while minimizing additional toxicity.
Because concurrent cetuximab with radiation therapy has been
shown to significantly improve local control and survival for
primary head and neck cancer without major increases in
mucosal toxicity,60 there has been interest in the application of
this agent as a radiosensitizer in the reirradiation setting. This
was explored in a single-institution, matched, case-control
study, which found that cetuximab conferred a locoregional
and overall survival advantage without a major increase in
toxicity.61 In a subsequent phase II trial on SBRT with concurrent cetuximab for previously irradiated HNSCC, the authors
reported a median overall survival of 10 months, with a 1-year
progression-free survival rate of 60%. Acute and late-grade 3
toxicity were each observed in 6% of patients, much lower than
reported in studies of other techniques.57 A similar French
phase II study reported a median survival of 11.8 months, with
32% of patients experiencing grade 3 toxicities.55 However,
although the risk of severe late toxicity increases significantly
beyond 5 years,62 follow-up for most studies have been limited
to 1 to 2 years, pointing to the need for long-term studies.
Local control and lower rates of toxicity with SBRT may
theoretically affect quality of life. In a prospective evaluation
of patient-reported quality of life outcomes following SBRT
with or without cetuximab, overall quality of life and multiple domains commonly affected by reirradiation progressively improved following an initial 1-month decline,
speaking to the positive palliative benefits of improved local
control afforded by conformal reirradiation with SBRT.63
Postoperative Reirradiation
Reirradiation in the postoperative setting has been studied in
GETTEC/GORTEC 9901, a phase III trial in which patients who
e288
had previously undergone radiation were randomly assigned to

reirradiation (60 Gy over 11 weeks in 2 Gy daily fractions) with
concurrent hydroxyurea and 5-FU or observation following
salvage surgery. Disease-free survival and locoregional control
were both significantly superior in the reirradiation group, but
there was no difference in overall survival. The reirradiation
arm experienced significantly higher rates of grade 3-4 toxicities compared with the observation arm (39% vs. 10%).64
These results support a high threshold for postoperative
reirradiation following salvage surgery. Primary considerations
include high complication rates from combined therapy, large
radiation volumes required to cover the postoperative bed,
high competing risks of outside-field recurrence (such as distant metastases and second primary cancer), and a lack of a
survival advantage. However, small-field postoperative reirradiation may be appropriate in select cases with high-risk
features such as extracapsular extension or positive margins.
The use of SBRT in this setting was recently explored in a
small retrospective study of 28 patients with high-risk features (predominately extracapsular extension or close/positive
margins). At a median follow-up of 14 months, the 1-year
locoregional control and overall survival were 51% and 64%,
respectively, while rates of acute and late-grade 3 and higher
toxicities were low at 0% and 8%, respectively.23 The role of
postoperative SBRT is currently being explored in a randomized
phase II trial at the University of Pittsburgh Cancer Institute.
Considerations to Reduce Toxicity

Patient selection with regard to performance status, medical
comorbidities, time interval since last radiation, prior dose
received, extent of existing late toxicity burden, and tumor
location and size should be carefully performed. With regard
to treatment parameters, it has been reported that 96% of
failures occur within the gross tumor volume 95% isodose
line,65 suggesting that target volume should be limited to
only high-risk areas to minimize toxicity. A clinical target
volume confined to the gross target volume plus a margin is
recommended.48,49 Induction or concurrent systemic therapy, including novel targeted therapy agents, may allow a
reduction in radiation dose. An ongoing phase II trial is being
conducted by the GORTEC to evaluate reirradiation using
a hyperfractionated scheme with concurrent cetuximab
(NCT01237483). Finally, recent technologic advancements
such as IMRT and SBRT offer potentially noninferior oncologic outcomes with lower toxicity rates compared with
conventional methods, although supporting phase III data
continues to evolve.
SYSTEMIC THERAPEUTIC OPTIONS FOR

RECURRENT/SECOND PRIMARY HEAD AND NECK
CANCERS
Combination Chemotherapy for Unresectable
Recurrent or Second Primary HNSCC
Systemic therapy, in general, is considered to be the
least effective of the three treatment modalities used in
the management of HNSCC. This is because, as a single
treatment modality, chemotherapy alone does not have the

capability of eliciting curative treatment effectsunlike
surgery or radiation therapy. In terms of achieving a cure for
head and neck cancer, the primary role of systemic therapy
is as a sensitizer for radiation therapy. As described in
the preceding section, reirradiation has the possibility of
achieving long-term tumor control and survival for patients
with unresectable, locoregional recurrent HNSCC. This raises
the difficult question that oncologists frequently struggle
with in the clinicwhether systemic chemotherapy or
concurrent chemo-reirradiation should be used as the firstline treatment for a patient with unresectable locoregional
recurrent HNSCC. To address this question, it will be useful to
first review clinical trials that have examined systemic
chemotherapy regimens.
The EXTREME study was a phase III trial comparing cisplatin, 5-FU, and cetuximab with cisplatin and 5-FU for
patients with HNSCC with recurrent or metastatic disease.66
A maximum of six cycles of chemotherapy was administered in both arms. In the experimental arm, weekly
maintenance cetuximab was prescribed following the
course of combination systemic therapy. The results of this
study demonstrated an overall survival advantage favoring
the experimental arm, with a median overall survival of
10.1 months and 7.4 months for the chemotherapycetuximab arm compared with the chemotherapy-alone
arm, respectively. A subsequent ad hoc analysis of this
study showed that the beneficial effects of adding cetuximab
to chemotherapy were independent of HPV status.67 The
results of EXTREME study were surprising in the sense that
numerous phase III studies in the preceding decades failed to
demonstrate superiority of newer, presumably more active,
drug combinations compared with older, presumably less
active, drug regimens. ECOG 1395 was a phase III trial that
compared cisplatin and paclitaxel with cisplatin and 5-FU.46
The experimental armcisplatin and paclitaxelfailed to
demonstrate an overall survival advantage; the median
overall survival was 8.1 months and 8.7 months, respectively, for the cisplatin-paclitaxel and the cisplatin/5-FU
arms. A preceding phase III study compared cisplatin plus
5-FU with either drug alone.68 Although the cisplatin/5-FU
arm achieved a higher overall response rate compared with
the single-agent arms, this did not translate into an improvement in overall survival. Hence, in the context of this
historical background it is clear why the results of the EXTREME trial were practice changing and why this regimen
has been solidified as the current standard of care.
Systemic Therapy Compared With Reirradiation

The results of the EXTREME study raise a logical and highly
clinically relevant question: will the EXTREME regimen
provide a superior outcome compared with concurrent
chemo-reirradiation for patients with unresectable locoregional recurrent HNSCC? To answer this question would
require a head-to-head study comparing these two approaches; to date no such data exist. However, it is
extremely unlikely that a study with this design will ever

come to fruition based on previous experience with this
disease.
RTOG 0421 (NCT00113399) was a phase III study launched
in 2005 that compared reirradiation and concurrent chemotherapy with chemotherapy alone. This trial used the
cisplatin and paclitaxel reirradiation regimen of RTOG 9911
(as described in the preceding section), which showed an
incremental improvement over the preceding study, RTOG
9610. The 2-year survival rate demonstrated in RTOG 9911
was 26%, which compared favorably with survival data from
contemporary chemotherapy-alone studies.69 It was within
this backdrop that RTOG 0421 was launched.
Simultaneously, the GORTEC group initiated a phase III
study in France, GORTEC 9803, which also compared chemoreirradiation with chemotherapy alone.70 Unfortunately,
neither of these studies were able to complete accrual,
leaving this important clinical question unanswered. Even
though no conclusion could be drawn from these studies,
some useful observations were made. First, the inability to
complete these studies strongly suggested a bias on behalf
of treating physicians favoring reirradiation that hampered
accrual. Second, there seemed to be a shift in the widespread use and acceptance of reirradiation. Since the early
days after reirradiation was first described,37,40 it had been
considered a novel and potential very dangerous therapy,
performed primarily at larger-volume academic centers with
experience in this technique. However, by the time RTOG
0421 was activated, a shift toward general acceptance of
reirradiation had occurred regarding the safety and a level of
comfort with the procedure. The confluence of these two
issues doomed these trials from achieving completion,
leaving the central question unanswered.
Both of RTOG 0421 and the GORTEC 9803 studies were
designed in an era prior to the availability of the results of the
EXTREME study. The positive results of the EXTREME trial
actually make the question of reirradiation or systemic
chemotherapy for unresectable recurrent HNSCC even more
relevant and pressing. So in the absence of high-level evidence comparing reirradiation with systemic therapy, how is
it possible to choose the most appropriate therapy for a
patient with unresectable recurrent HNSCC in a previously
irradiated field?
Although a simple solution is not possible, a number of
considerations make it permissible to approach the decision
in a logical, evidenced-based manner. The first issue to be
considered is toxicity. Considerable toxicity is associated
with the EXTREME regimen; approximately 80% of patients
experienced grade 3 or 4 adverse events in either arm of the
EXTREME trial. Likewise concurrent chemo-reirradiation is
associated with high toxicity, including grade 5 events that
have been reported in nearly all studies examining concurrent chemo-reirradiationtypically in the range of
3%.71,72 Most of the deaths associated with reirradiation are
related to bleeding events; carotid rupture accounts for the
majority of life-threatening bleeding events. Hence, it is
advisable to carefully consider patients with tumors that
e289
WONG ET AL
have significant carotid involvement, such as circumferential

tumor encasement of the carotid. Likewise, structural issues
preclude other patients from reirradiation, such as recurrent
laryngeal cancer where retreatment can cause cartilaginous
necrosis. Similarly, patients may not be eligible for reirradiation when it is not feasible to spare the spinal cord from a
lifetime cumulative conventionally fractionated (1.8-2.0 Gy
per fraction) equivalent dose less than 54 Gy. A nomogram
published by Tanvetyanon et al provides a tool by which
practitioners can identify poor prognostic factors that may
help identify favorable populations and avoid exposure of
dangerous toxicity to other patients.15

The emergence of data indicating activity of immune
checkpoint inhibitors in HNSCC has potential major implications for management of this disease, including treatment
for patients with HNSCC with unresectable locoregional
recurrent disease. An expansion cohort of patients with
HNSCC were enrolled in the KEYNOTE-012 phase IB clinical
trial of pembrolizumab for recurrent and metastatic disease
who had 1% or greater PD-L1positive staining by immunohistochemistry.73 Of the 56 evaluable patients, the best
overall response rate was 20%, with similar response rates
seen in patients with HPV-positive disease and those with
HPV-negative disease. An expansion cohort of patients with
PD-L1unselected HNSCC enrolled in the KEYNOTE-012
demonstrated a best overall response rate of 25% for all
patients, 21% for patients with HPV-positive disease, and
27% for patients with HPV-negative disease.74 Of note, both
of the reports included patients whose response was
ongoingsuggesting durable response to therapy.
In contrast to conventional chemotherapy regimens, such
as EXTREME, long-term durable response and survivorship
occurs in less than 5% of patients. Phase III studies of immune checkpoint inhibitors for recurrent metastatic HNSCC
are in progress and potentially could establish a new
standard of care in years to come. If long-term survivorship is
achievable with immune checkpoint inhibitor therapy, this
treatment option would put systemic therapy in a playing
field with reirradiation.
Emerging in the literature is a growing body of evidence

that suggests radiation therapy and immune checkpoint
have complimentary effects.75,76 These studies indicate that
immune checkpoint inhibitors can overcome the negative
effects of radiation on the tumor microenvironment. Conversely, these data also suggest that the antigenic response
to radiation therapy may induce a heightened effectiveness
from immune checkpoint inhibition. It is possible to now
speculate that by combining an immune checkpoint inhibitor
with reirradiation, the best of both worlds may be achievable
in a single treatment plan. In other words, it is conceivable
to concurrently potentiate the effects of reirradiation
(i.e., enhancing locoregional disease control) while simultaneously augmenting the systemic immunologic effects
(i.e., enhancing durable treatment response and long-term
survival). To test this hypothesis, a new study, the
KEYSTROKE trial (RTOG Foundation Study 3507), will compare reirradiation with SBRT plus pembrolizumab with
reirradiation with SBRT alone. It is conceivable that future
studies examining immune checkpoint inhibitors with reirradiation may demonstrate favorable results that could
reduce our reliance on dogmatic approaches to patients with
unresectable recurrent or new second primary HNSCC.
CONCLUSION
Locoregional recurrent or second primary HNSCC in a previously irradiated field represents a challenging clinical
scenario for which surgical, reirradiation, and systemic
treatment options continue to evolve. As outlined herein,
management in general would favor salvage surgical resection for patients with resectable disease, followed by
consideration for adjuvant reirradiation with chemotherapy.
For patients with unresectable disease, modern reirradiation
(IMRT or SBRT) with concurrent systemic therapy is favored.
Finally, in cases where surgery is not feasible and reirradiation may pose too high of a risk of severe toxicity either
because of prior radiation dose to normal structures or
extent/involvement of the recurrence, palliative systemic
therapy is favored. Patients should be encouraged to enroll
in clinical trials incorporating novel systemic agents.
References
1. Gillison ML, Harris J, Westra W, et al. Survival outcomes by tumor
human papillomavirus (HPV) status in stage III-IV oropharyngeal
cancer (OPC) in RTOG 0129. J Clin Oncol. 2009;27:15 (suppl; abstr
6003).
2. Fakhry C, Zhang Q, Nguyen-Tan PF, et al. Human papillomavirus and
overall survival after progression of oropharyngeal squamous cell
carcinoma. J Clin Oncol. 2014;32:3365-3373.
3. Canadian Agency for Drugs and Technologies in Health. Nasogastric Feeding
Tubes versus Percutaneous Endoscopic Gastrostomy for Patients with Head
or Neck Cancer: A Review of Clinical Effectiveness and Guidelines. Canadian Agency for Drugs and Technologies in Health Ottawa (ON). http://
www.ncbi.nlm.nih.gov/books/NBK253809/. Accessed February 19, 2016.
e290
4. Wang J, Liu M, Liu C, et al. Percutaneous endoscopic gastrostomy versus

nasogastric tube feeding for patients with head and neck cancer:
a systematic review. J Radiat Res (Tokyo). 2014;55:559-567.
5. Rivelli TG, Mak MP, Martins RE, et al. Cisplatin based chemoradiation
late toxicities in head and neck squamous cell carcinoma patients.
Discov Med. 2015;20:57-66.
6. Cannon CR. Hypothyroidism in head and neck cancer patients: Experimental and clinical observations. Laryngoscope. 1994;104:1-21.
7. Kim AJ, Suh JD, Sercarz JA, et al. Salvage surgery with free flap reconstruction: factors affecting outcome after treatment of recurrent
head and neck squamous carcinoma. Laryngoscope. 2007;117:
1019-1023.
8. Dirven R, Swinson BD, Gao K, et al. The assessment of pharyngocutaneous fistula rate in patients treated primarily with definitive
radiotherapy followed by salvage surgery of the larynx and hypopharynx. Laryngoscope. 2009;119:1691-1695.
9. Kostrzewa JP, Lancaster WP, Iseli TA, et al. Outcomes of salvage surgery
with free flap reconstruction for recurrent oral and oropharyngeal
cancer. Laryngoscope. 2010;120:267-272.
10. Suh JD, Kim BP, Abemayor E, et al. Reirradiation after salvage surgery
and microvascular free flap reconstruction for recurrent head and neck
carcinoma. Otolaryngol Head Neck Surg. 2008;139:781-786.
11. Cohn AB, Lang PO, Agarwal JP, et al. Free-flap reconstruction in the
doubly irradiated patient population. Plast Reconstr Surg. 2008;122:
125-132.
12. Goodwin WJ Jr. Salvage surgery for patients with recurrent squamous
cell carcinoma of the upper aerodigestive tract: when do the ends justify
the means? Laryngoscope. 2000;110:1-18.
13. Kim J, Kim S, Albergotti WG, et al. Selection of ideal candidates for
surgical salvage of head and neck squamous cell carcinoma: effect of the
Charlson-Age Comorbidity Index and oncologic characteristics on 1-year
survival and hospital course. JAMA Otolaryngol Head Neck Surg. 2015;
141:1059-1065.
14. Ho AS, Kraus DH, Ganly I, et al. Decision making in the management of
recurrent head and neck cancer. Head Neck. 2014;36:144-151.
15. Tanvetyanon T, Padhya T, McCaffrey J, et al. Prognostic factors for
survival after salvage reirradiation of head and neck cancer. J Clin Oncol.
2009;27:1983-1991.
16. Gleich LL, Ryzenman J, Gluckman JL, et al. Recurrent advanced (T3 or T4)
head and neck squamous cell carcinoma: is salvage possible? Arch
Otolaryngol Head Neck Surg. 2004;130:35-38.
17. Richey LM, Shores CG, George J, et al. The effectiveness of salvage
surgery after the failure of primary concomitant chemoradiation in
head and neck cancer. Otolaryngol Head Neck Surg. 2007;136:98-103.
18. Arnold DJ, Goodwin WJ, Weed DT, et al. Treatment of recurrent and
advanced stage squamous cell carcinoma of the head and neck. Semin
Radiat Oncol. 2004;14:190-195.
19. Mandapathil M, Roessler M, Werner JA, et al. Salvage surgery for head
and neck squamous cell carcinoma. Eur Arch Otorhinolaryngol. 2014;
271:1845-1850.
20. Zafereo M. Surgical salvage of recurrent cancer of the head and neck.
Curr Oncol Rep. 2014;16:386-391.
21. Omura G, Saito Y, Ando M, et al. Salvage surgery for local residual or
recurrent pharyngeal cancer after radiotherapy or chemoradiotherapy.
Laryngoscope. 2014;124:2075-2080.
22. Salama JK, Vokes EE, Chmura SJ, et al. Long-term outcome of concurrent
chemotherapy and reirradiation for recurrent and second primary
head-and-neck squamous cell carcinoma. Int J Radiat Oncol Biol Phys.
2006;64:382-391.
23. Vargo JA, Kubicek GJ, Ferris RL, et al. Adjuvant stereotactic body
radiotherapy6cetuximab following salvage surgery in previously
irradiated head and neck cancer. Laryngoscope. 2014;124:
1579-1584.
24. Jones AS, Tandon S, Helliwell TR, et al. Survival of patients with neck
recurrence following radical neck dissection: utility of a second neck
dissection? Head Neck. 2008;30:1514-1522.
25. Mourad M, Saman M, Stroman D, et al. Carotid artery sacrifice and
reconstruction in the setting of advanced head and neck cancer.
Otolaryngol Head Neck Surg. 2015;153:225-230.
26. Naara S, Amit M, Billan S, et al. Outcome of patients undergoing salvage
surgery for recurrent nasopharyngeal carcinoma: a meta-analysis. Ann
Surg Oncol. 2014;21:3056-3062.
27. You R, Zou X, Hua Y. Salvage endoscopic nasopharyngectomy is
superior to intensity-modulated radiation therapy for local recurrence
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
of selected T1-T3 nasopharyngeal carcinoma - A case-matched comparison. Radiother Oncol. 2015;115:399-406.

White H, Ford S, Bush B, et al. Salvage surgery for recurrent cancers
of the oropharynx: comparing TORS with standard open surgical
approaches. JAMA Otolaryngol Head Neck Surg. 2013;139:
773-778.
Canis M, Martin A, Ihler F, et al. Transoral laser microsurgery in
treatment of pT2 and pT3 glottic laryngeal squamous cell carcinoma results of 391 patients. Head Neck. 2014;36:859-866.
Palliative Surgery: Definition, Principles, Outcomes Assessment. In
Dunn GP, Martensen R, Weissman D (eds). Surgical Palliative Care: A
Residents Guide. Chicago, IL: American College of Surgeons, 2009;131139.
Jang DW, Teng MS, Ojo B, et al. Palliative surgery for head and neck
cancer with extensive skin involvement. Laryngoscope. 2013;123:
1173-1177.
Miglani A, Patel VM, Stern CS, et al. Palliative reconstruction for the
management of incurable head and neck cancer. J Reconstr Microsurg.
Epub 2015 Dec 4.
Copeland EM III. The art of medicine at the end of life: a surgeons point
of view. Bull Am Coll Surg. 2008;93:17-18.
Conley JJ. Palliative surgery in the head and neck. Ann Otol Rhinol
Laryngol. 1962;71:585-590.
Cullinane CA, Borneman T, Smith DD, et al. The surgical treatment of
cancer: a comparison of resource utilization following procedures
performed with a curative and palliative intent. Cancer. 2003;98:
2266-2273.
Gaisford JC. Palliative surgery. JAMA. 1972;221:83-84.
De Crevoisier R, Bourhis J, Domenge C, et al. Full-dose reirradiation for
unresectable head and neck carcinoma: experience at the GustaveRoussy Institute in a series of 169 patients. J Clin Oncol. 1998;16:
3556-3562.
Brockstein B, Haraf DJ, Stenson K, et al. A phase I-II study of concomitant
chemoradiotherapy with paclitaxel (one-hour infusion), 5-fluorouracil
and hydroxyurea with granulocyte colony stimulating factor support for
patients with poor prognosis head and neck cancer. Ann Oncol. 2000;11:
721-728.
Spencer SA, Wheeler RH, Peters GE, et al. Concomitant chemotherapy
and reirradiation as management for recurrent cancer of the head and
neck. Am J Clin Oncol. 1999;22:1-5.
Haraf DJ, Weichselbaum RR, Vokes EE. Reirradiation with concomitant
chemotherapy of unresectable recurrent head and neck cancer: a potentially curable disease. Ann Oncol. 1996;7:913-918.
Spencer SA, Harris J, Wheeler RH, et al. Final report of RTOG 9610, a
multi-institutional trial of reirradiation and chemotherapy for unresectable recurrent squamous cell carcinoma of the head and neck. Head
Neck. 2008;30:281-288.
Langer CJ, Harris J, Horwitz EM, et al. Phase II study of low-dose
paclitaxel and cisplatin in combination with split-course concomitant
twice-daily reirradiation in recurrent squamous cell carcinoma of the
head and neck: results of Radiation Therapy Oncology Group Protocol
9911. J Clin Oncol. 2007;25:4800-4805.
Argiris A, Ghebremichael M, Gilbert J, et al. Phase III randomized,
placebo-controlled trial of docetaxel with or without gefitinib in recurrent or metastatic head and neck cancer: an eastern cooperative
oncology group trial. J Clin Oncol. 2013;31:1405-1414.
Bourhis J, Rivera F, Mesia R, et al. Phase I/II study of cetuximab in
combination with cisplatin or carboplatin and fluorouracil in patients
with recurrent or metastatic squamous cell carcinoma of the head and
neck. J Clin Oncol. 2006;24:2866-2872.
Burtness B, Goldwasser MA, Flood W, et al. Phase III randomized trial
of cisplatin plus placebo compared with cisplatin plus cetuximab in
e291
WONG ET AL
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
metastatic/recurrent head and neck cancer: an Eastern Cooperative

Oncology Group study. J Clin Oncol. 2005;23:8646-8654.
Gibson MK, Li Y, Murphy B, et al. Randomized phase III evaluation of
cisplatin plus fluorouracil versus cisplatin plus paclitaxel in advanced
head and neck cancer (E1395): an intergroup trial of the Eastern Cooperative Oncology Group. J Clin Oncol. 2005;23:3562-3567.
Davis KS, Vargo JA, Ferris RL, et al. Stereotactic body radiotherapy for
recurrent oropharyngeal cancer - influence of HPV status and smoking
history. Oral Oncol. 2014;50:1104-1108.
Lee N, Chan K, Bekelman JE, et al. Salvage reirradiation for recurrent
head and neck cancer. Int J Radiat Oncol Biol Phys. 2007;68:731-740.
Sulman EP, Schwartz DL, Le TT, et al. IMRT reirradiation of head and neck
cancer-disease control and morbidity outcomes. Int J Radiat Oncol Biol
Phys. 2009;73:399-409.
Biagioli MC, Harvey M, Roman E, et al. Intensity-modulated radiotherapy with concurrent chemotherapy for previously irradiated, recurrent head and neck cancer. Int J Radiat Oncol Biol Phys. 2007;69:
1067-1073.
Duprez F, Madani I, Bonte K, et al. Intensity-modulated radiotherapy for
recurrent and second primary head and neck cancer in previously irradiated territory. Radiother Oncol. 2009;93:563-569.
Rwigema JC, Vargo JA, Clump DA, et al. Stereotactic body radiotherapy
in the management of head and neck malignancies. Curr Cancer Ther
Rev. 2014;10:314-322.
Kodani N, Yamazaki H, Tsubokura T, et al. Stereotactic body radiation
therapy for head and neck tumor: disease control and morbidity
outcomes. J Radiat Res (Tokyo). 2011;52:24-31.
Yazici G, Sanl TY, Cengiz M, et al. A simple strategy to decrease fatal
carotid blowout syndrome after stereotactic body reirradiaton for
recurrent head and neck cancers. Radiat Oncol. 2013;8:242.
Lartigau EF, Tresch E, Thariat J, et al. Multi institutional phase II study of
concomitant stereotactic reirradiation and cetuximab for recurrent
head and neck cancer. Radiother Oncol. 2013;109:281-285.
Comet B, Kramar A, Faivre-Pierret M, et al. Salvage stereotactic reirradiation with or without cetuximab for locally recurrent head-andneck cancer: a feasibility study. Int J Radiat Oncol Biol Phys. 2012;84:
203-209.
Vargo JA, Ferris RL, Ohr J, et al. A prospective phase II trial of reirradiation with stereotactic body radiotherapy plus cetuximab in patients with recurrent previously-irradiated squamous cell carcinoma of
the head and neck. Int J Radiat Oncol Biol Phys. 2015;91:480-488.
Heron DE, Ferris RL, Karamouzis M, et al. Stereotactic body radiotherapy
for recurrent squamous cell carcinoma of the head and neck: results of a
phase I dose-escalation trial. Int J Radiat Oncol Biol Phys. 2009;75:
1493-1500.
Rwigema JC, Heron DE, Ferris RL, et al. The impact of tumor volume and
radiotherapy dose on outcome in previously irradiated recurrent
squamous cell carcinoma of the head and neck treated with stereotactic
body radiation therapy. Am J Clin Oncol. 2011;34:372-379.
Bonner JA, Harari PM, Giralt J, et al. Radiotherapy plus cetuximab for
squamous-cell carcinoma of the head and neck. N Engl J Med. 2006;354:
567-578.
Heron DE, Rwigema JC, Gibson MK, et al. Concurrent cetuximab with
stereotactic body radiotherapy for recurrent squamous cell carcinoma
of the head and neck: a single institution matched case-control study.
Am J Clin Oncol. 2011;34:165-172.
e292
62. Owen D, Iqbal F, Pollock BE, et al. Long-term follow-up of stereotactic

radiosurgery for head and neck malignancies. Head Neck. 2015;37:
1557-1562.
63. Vargo JA, Heron DE, Ferris RL, et al. Prospective evaluation of patientreported quality-of-life outcomes following SBRT 6 cetuximab for
locally-recurrent, previously-irradiated head and neck cancer. Radiother Oncol. 2012;104:91-95.
64. Janot F, de Raucourt D, Benhamou E, et al. Randomized trial of postoperative reirradiation combined with chemotherapy after salvage
surgery compared with salvage surgery alone in head and neck carcinoma. J Clin Oncol. 2008;26:5518-5523.
65. Popovtzer A, Gluck I, Chepeha DB, et al. The pattern of failure after
reirradiation of recurrent squamous cell head and neck cancer: implications for defining the targets. Int J Radiat Oncol Biol Phys. 2009;74:
1342-1347.
66. Vermorken JB, Mesia R, Rivera F, et al. Platinum-based chemotherapy
plus cetuximab in head and neck cancer. N Engl J Med. 2008;359:
1116-1127.
67. Vermorken JB, Psyrri A, Mesa R, et al. Impact of tumor HPV status on
outcome in patients with recurrent and/or metastatic squamous cell
carcinoma of the head and neck receiving chemotherapy with or
without cetuximab: retrospective analysis of the phase III EXTREME
trial. Ann Oncol. 2014;25:801-807.
68. Jacobs C, Lyman G, Velez-Garca E, et al. A phase III randomized study
comparing cisplatin and fluorouracil as single agents and in combination
for advanced squamous cell carcinoma of the head and neck. J Clin
Oncol. 1992;10:257-263.
69. Wong SJ, Machtay M, Li Y. Locally recurrent, previously irradiated head
and neck cancer: concurrent reirradiation and chemotherapy, or
chemotherapy alone? J Clin Oncol. 2006;24:2653-2658.
70. Tortochaux J, Tao Y, Tournay E, et al. Randomized phase III trial (GORTEC
98-03) comparing reirradiation plus chemotherapy versus methotrexate in patients with recurrent or a second primary head and neck
squamous cell carcinoma, treated with a palliative intent. Radiother
Oncol. 2011;100:70-75.
71. Wong SJ, Bourhis J, Langer CJ. Retreatment of recurrent head and neck
cancer in a previously irradiated field. Semin Radiat Oncol. 2012;22:
214-219.
72. McDonald MW, Moore MG, Johnstone PA. Risk of carotid blowout after
reirradiation of the head and neck: a systematic review. Int J Radiat
Oncol Biol Phys. 2012;82:1083-1089.
73. Seiwert TY, Burtness B, Weiss J, et al. A phase Ib study of MK-3475 in
patients with human papillomavirus (HPV)-associated and non-HPV
associated head and neck (H/N) cancer. J Clin Oncol. 2014;32:5s (suppl;
abstr 6011).
74. Seiwert T, Haddad R, Gupta S. Antitumor activity and safety of
pembrolizumab in patients (pts) with advanced squamous cell carcinoma of the head and neck (SCCHN) (2015) preliminary results from
KEYNOTE-012 expansion cohort. J Clin Oncol. 2015;33s (suppl; abstr
LBA6008).
75. Deng L, Liang H, Burnette B, et al. Irradiation and anti-PD-L1 treatment
synergistically promote antitumor immunity in mice. J Clin Invest. 2014;
124:687-695.
cancer. Nature. 2015;520:373-377.
HEALTH SERVICES RESEARCH AND QUALITY

OF CARE
Defining and Measuring Quality

CHAIR
Patricia A. Ganz, MD, FASCO
University of California, Los Angeles
Los Angeles, CA
SPEAKERS
David C. Miller, MD, MPH
University of Michigan Health System
Ann Arbor, MI
Michael J. Hassett, MD, MPH
Boston, MA
GANZ, HASSETT, AND MILLER
Challenges and Opportunities in Delivering High-Quality

Cancer Care: A 2016 Update
Patricia A. Ganz, MD, FASCO, Michael J. Hassett, MD, MPH, and David C. Miller, MD, MPH
OVERVIEW
Herein, both the rationale and scope of current initiatives aimed at improving the quality of cancer care delivery in the United
States are described. First, we discuss a recent report from the Institute of Medicine that issued a strong call for both the
development of quality measures in oncology and implementation of a learning health care system in which data and
experience from clinical practice can inform continuous improvements in cancer care. Second, we review the multiple,
diverse initiatives that are underway to identify, test, and validate quality measures for the entire spectrum of cancer care.
Finally, we discuss regional quality improvement collaboratives as one successful approach to creating a cycle of quality
measurement, identification of best practices, and implementation of changes in practice patterns that ultimately yield
improved care and outcomes for patients with cancer.
n 2013, the Institute of Medicine (IOM) released a report

with comprehensive, specific recommendations for improving the quality of cancer care in the United States, with a
special focus on addressing the challenges associated with
the expanding number of patients with cancer and cancer
survivors.1 A major factor that will strain the cancer care
delivery system is the increasing numbers of persons older
than age 65 who will be at highest risk for the development
of cancer because of the known association of age with many
cancers (e.g., breast, colon, prostate, lung, bladder, kidney,
and lymphoma). In addition, the projected shortfall in oncology providers will exacerbate this situation.2 We also
anticipate tremendous growth in the number of survivors of
cancer who may remain on treatments for extended periods
of time and/or have long-term and late effects of treatments
that will require ongoing medical care and surveillance.3
The increasing cost of care complicates these demographic
changes and expanding demand for services. Costs are not
limited to pharmaceuticals, but also include widespread
adoption and use of many new and expensive technologies
in the diagnosis and treatment of cancer (e.g., nextgeneration sequencing) to refine understanding of tumor
characteristics for use of more specific and directed therapies. Furthermore, the extensive subtyping of cancers will
create many small subsets of patients with cancer, making
some types of cancer even rarer than before. Immunomodulatory treatments and cell therapies, although they
hold great promise, are substantially more expensive and

require individualized treatments that will be challenging to
deliver to large populations of patients. To some extent
these novel therapies remain experimental, but there is
tremendous demand by the public and clinicians to adopt
these therapies in the earliest stages of development. This
is the context for consideration of the challenges and
opportunities in the delivery of high-quality cancer care.
DIRECT RESPONSES AND CHAMPIONS FOR

SPECIFIC INSTITUTE OF MEDICINE REPORT
RECOMMENDATIONS
The IOM report specifically cited the central position of the
patient in the care delivery system and the importance of
effective communication of diagnosis, prognosis, and
treatment options at the point of care, with an emphasis on
patient preferences and shared decision making.1 Teambased and coordinated care also were proposed as important strategies to reduce the fragmentation of care and to
address the shortages of some clinicians, as well as to
maximize opportunities for receipt of psychosocial and
palliative care along with cancer-directed therapies. In addition, there was a strong call for the development of quality
metrics specific to oncology, elimination of wasteful and
low-value care, and for implementation of a learning health
care system in which information from clinical practice in the
From the Department of Health Policy and Management, UCLA Fielding School of Public Health, Center for Cancer Prevention and Control Research, Jonsson Comprehensive Cancer
Center, Los Angeles, CA; Department of Medicine, Harvard Medical School, Division of Population Sciences, Dana-Farber Cancer Institute, Boston, MA; Dow Division of Health Services
Research, Department of Urology, Institute for Healthcare Policy and Innovation, University of Michigan Medical School, Ann Arbor, MI.
Corresponding author: Patricia A. Ganz, MD, Center for Cancer Prevention and Control Research, Jonsson Comprehensive Cancer Center, 650 Charles Young Dr. South, Room A2-125
CHS, Los Angeles, CA 90049; email: pganz@mednet.ucla.edu.
e294
IMPROVING THE QUALITY OF CANCER CARE
real world could inform improvements in the delivery of

cancer care. Although the IOM report addresses many other
issues, the areas mentioned above have been the focus of
considerable activity in the last few years, and they will be
the major focus of this update.
Recommendations 1 and 2 from the IOM report focused
on the central role of engaged patients in delivering highquality cancer care. The goals of these recommendations are
to ensure that the cancer care team has provided patients
and their families with understandable information on
cancer prognosis, treatment benefits and harms, palliative
care, psychosocial support, and estimates of total and outof-pocket costs of cancer care. These recommendations
pertain to newly diagnosed patients at entry into the cancer
care setting, as well as patients with advanced and recurrent
cancer, where the goals also include providing patients with
end-of-life care consistent with their needs, values, and
preferences. The IOM report identified specific actors to
take the lead in accomplishing these goals, tasking a broad
array of stakeholders in developing better ways to communicate this complex information to patients and their
families, and calling out the need to develop a personalized
cancer care plan that would reflect the patients needs and
preferences.
This aspirational goal is very important, but it will take
some time to accomplish. However, we were heartened by
proposals from several organizations to modify cancer care
reimbursement to account for the complexity of initial diagnosis and treatment planning (American Society of Clinical
Oncology [ASCO] Plan), and an aggressive proposal for a
payment modification scheme from the Centers for Medicare & Medicaid Services (CMS) called the Oncology Care
Model (OCM),4 which identified the elements of patient-
KEY POINTS
The 2013 IOM report, Delivering High-Quality Cancer

Care: Charting a New Course for a System in Crisis,
identified the aging of the population, the exploding cost
of care, and fragmentation and complexity of cancer
treatment as major threats to the care delivery system.
Recommendations from the report are being
championed by various entities, and this article
highlights a number of areas of progress.
Measuring the quality of care is a major challenge, and
new strategies are being implemented to report on and
evaluate quality in cancer care delivery.
Among the experiments in place are physician learning
collaboratives, such as the Michigan Urological Surgery
Improvement Collaborative, to improve on the quality
of prostate cancer care delivery.
The development of new payment models that fund
episodes of care or provide reimbursement to groups of
physicians responsible for the entire continuum of
cancer care will likely hasten greater accountability and
coordination of care.
centered care in their episode of care payment model directly from the IOM report (reflecting recommendation 10
and other aspects of the report). The latter also encouraged
participation from private health insurers. As of this writing,
the participating clinician groups and insurance plans have
not been announced, but this is expected to be implemented
in early 2016. Providing more financial resources to support
the degree of patient engagement in treatment decision
making and care, as well as ancillary supportive services, will
be critical if we are to enhance the delivery of quality cancer
care.
Other efforts underway have begun to address the cost of
cancer care and the development of tools to assist clinicians
at the point of care in discussing preferences and toxicities,
along with cost (multiple IOM report recommendations).
Among these are the ASCO Value Framework5 and work
underway by the National Comprehensive Care Network in
its framework for resource stratification. There is increasing
dialogue regarding these issues in the lay press and in
professional oncology journals.6 It is heartening to see so
many stakeholders beginning to focus on the value of care
that is being delivered, including elimination of low-value
and wasteful care.7-9 Additional efforts in progress are
working to promote a learning health care system (IOM
report recommendation 7), with multiple experiments in
progress and development of oncology-specific quality
metrics. To provide further elaboration on several of these
issues, the following sections describe ongoing research on
quality metrics and work being done to address the need to
improve the quality of cancer care.
METRICS FOR ASSESSING QUALITY OF CANCER

CARE
A fundamental requirement of any effort to improve the
quality of health care is the ability to measure practice
performance reliably, validly, and efficiently. Without
quality measures, identifying gaps in performance and
assessing the extent to which improvement efforts ameliorate these gaps are impossible. In fact, quality measures
could help address many of the challenges highlighted by the
IOM and outlined above, including the need to reduce care
fragmentation, improve access to psychosocial and palliative
care, administer precision cancer treatments, provide endof-life care consistent with patients wishes, and eliminate
wasteful low-value care. In this section, we will identify
organizations that have taken the lead in developing cancer
quality measures; provide an overview of the quality
measures that have been developed; and identify important
topics for which quality measures are lacking. Finally, we will
review some of the biggest challenges to the development
and use of measures to assess and improve care quality.
In the 2 decades since the IOM issued a call to action
through a series of landmark publications, including To Err Is
Human: Building a Safer Health System10 and Crossing the
Quality Chasm: A New Health System for the 21st Century,11
professional societies have taken the lead in developing
e295
quality measures for cancer care. Substantial contributors to

this effort have included the American College of Surgeons/
Commission on Cancer, American Medical Association,
ASCO, American Society of Hematology, American Society
for Radiation Oncology, College of American Pathologists,
National Comprehensive Cancer Network, and National
Hospice and Palliative Care Organization. In addition, a
number of organizations have endeavored to collect and
endorse quality measures to facilitate their use, including
the Agency for Healthcare Research and Qualitys National
Quality Measures Clearing House12 and the National Quality
Forums (NQFs) endorsement process.13-15
The framework proposed by Donabedian16 defines three
types of quality measures: structure, process, and outcomes
(Fig. 1). Although all measure types are relevant to cancer
care, most available measures focus on processes because
they are easier to develop, implement, and interpret. In fact,
process-based quality measures are currently used to support accountability, pay-for-performance, contractual, and
regulatory programs such as the Physician Quality Reporting
System, Meaningful Use, alternative quality contracts, and
others.17-19 One notable exception is the Quality Oncology
Practice Initiative (QOPI) from ASCO, which was designed to
foster practice-based quality improvement specifically to
facilitate an oncology practices effort to monitor and improve
its performance.20-22 QOPI assesses performance relative to
more than 100 cancer-focused measures twice yearly for
hundreds of practices and thousands of patients.20-23
Although many valid and reliable process-based cancer
quality measures have been developed, existing measures tend
to cover a limited spectrum of cancer care services. Specifically,
they focus on common malignancies (i.e., breast cancer, colorectal cancer, lung cancer), assess the performance of physicians or hospitals, focus on widely accepted standards, and
address initial management strategies. This leaves many important aspects of cancer care underevaluated, including less
common cancers, nonphysician providers, long-term follow-up
care, and end-of-life issues. In a 2009 report, the NQF highlighted these hurdles and suggested that building a comprehensive measurement system would require new measures to
be developed to address gaps in episodes of care and focus on
patients rather than individual providers or distinct care settings.24 In 2012, the Measure Applications Partnership (MAP)
suggested that there was an immediate need to assess crosscutting aspects of care, meaning those that are relevant to all
patients with cancer throughout the trajectory of their illness.25
Both organizations noted that new measures should focus on
outcomes, care coordination, care transitions, quality of life,
patient safety, and experience of care.15,25
In December 2012, ASCO convened a Collaborative Cancer
Measure Summit to identify topics for which new quality
measures were needed.26 The summit, which was attended
by 12 specialty societies and patient advocacy organizations,
highlighted issues similar to those raised by NQF and the
MAP. Specifically, ASCO encouraged the development of
measures to address (1) the use of palliative care/hospice
services; (2) planning and counseling prior to therapy
e296
FIGURE 1. Schematic Example of Donabedian Model

Describing the Components of Quality Care
(e.g., treatment plan generation, fertility counseling);

(3) the provision of interdisciplinary and multidisciplinary
coordinated care; (4) comprehensive symptom assessment
and treatment; and (5) patient experience of care. Furthermore, it emphasized the importance of measures that
assess outcomes, address cross-cutting topics that pertain to
different cancer types and care settings (e.g., chemotherapyinduced nausea), and focus on issues of interest to patients
rather than the health care system.
Past efforts to develop and implement quality measures highlight a number of challenges that will have to be
addressed if we are to build the comprehensive measurement system advocated for by IOM and NQF. First, we must
determine what aspects of care should be measured.
Usually, decisions regarding what to measure are based on
consensus opinions about which performance gaps are
important and where improvement efforts could have the
greatest impact rather than on the use of actual practice
performance data to identify substantial gaps and define
benchmarks.27 Second, we must gather the data needed to
assess performance efficiently and accurately. With limited
resources, we can only assess performance relative to a limited
set of measures. And third, we must interpret the results of
measurement efforts appropriately, so that the findings can be
used to create real and meaningful improvements in quality.
Figuring out what to measure seems simple, but it can
present a number of pitfalls. Historically, selecting topics for
measure development has followed one of two paths. In the
first, interested parties, often professional societies, lead
rigorous measure development efforts on topics that are of
particular interest to them. This tends to produce measures
for which there is high-quality evidence and widespread
consensus. However, measures can take a long time to
develop, are hard to update, and often focus on areas where
quality problems are less pronounced. In the second, hospitals, clinics, and provider organizations develop measures
on their own to support local quality improvement efforts.
This provides more flexibility, but it can generate measures
based on lower-quality evidence or less consensus, and
testing for reliability and validity are less common. In both
paths, topic selection is usually driven by providers and
payers, rather than patients.
Once specifications have been developed, the next challenge is determining how to gather the data needed to
reliably and efficiently report on a measure. Three common
methods are used. Many measures rely on administrative
data collected for other purposes, such as claims for health
care services rendered. Administrative data are readily
available and are often based on a common standard (e.g.,
ICD-9, ICD-10, CPT), but they lack clinical granularity and
substantial data-quality issues can exist (i.e., missing data,
consistency, accuracy). Some measures depend on manual
data abstraction. This is often done by a cancer registrar (e.g.,
the National Cancer Databases Rapid Quality Reporting
System)28 or a quality improvement staff member (e.g.,
ASCOs QOPI).22 Although these data can be quite granular,
they are costly and time consuming to collect. With the recent
widespread adoption of electronic health records (EHRs),
there is growing interest in the use of EHR data for quality
measurement. EHR data offer great promise, but identifying
important concepts from these data can be challenging because they contain a lot of unstructured free text. EHRs
encourage users to enter information into structured data
fields to facilitate quality analyses, but this creates a burden
on EHR users who must spend time entering data that may
not be integral to the care they are providing.
After all the data have been collected, the final challenge is to
interpret the results. In many circumstances, there are justifiable reasons for not providing recommended treatments,
making 100% concordance an unattainable goal. That said,
determining what the actual target should be is not straightforward. Using high-performing organizations to define a
benchmark is one option, but even high-performing organizations can improve in some situations and adjusting for legitimate differences between organizations can be difficult.
Models that account for health-risk and case-mix exist, but
they are imperfect, can be challenging to implement, and only
pertain to some measures.29-32 Often, organizations conducting
quality improvement projects select internal benchmarks and
plan to demonstrate improvement relative to past performance. But, how much improvement can actually be attained?
Lastly, if suboptimal performance is identified, who should be
responsible for that deficiency? Attribution is a particular
challenge in oncology, in which care often requires a multidisciplinary approach spanning several different care settings.
These limitations notwithstanding, some organizations have
been able to apply quality measures in the real-world setting
and have used the results of these efforts to effect meaningful
improvements in cancer care quality. The third section of this
manuscript highlights a number of these successes.
USING PRACTICE VARIABILITY TO IDENTIFY

OPPORTUNITIES FOR INTERVENTIONS TO
IMPROVE QUALITY
The Role of Quality Improvement Collaboratives
As described previously, patients, providers, health care
payers, and policymakers are increasingly interested in
expanding efforts to measure and improve the quality of
cancer care. As one mechanism for achieving this goal, some
physicians and surgeons have established practice- or hospitalbased regional quality improvement collaboratives.33,34 The
general organizational principles of such collaboratives include
centralized and standardized collection of high-quality clinical
outcomes data; provision of performance feedback to individual clinicians and practices; identification of clinical care
processes (e.g., diagnostic tests, type of surgical procedure)
that correlate with optimal patient outcomes; implementation
of these best clinical practices; and dissemination of findings
from collaborative activities to benefit patients across a broad
geographic region.
The Michigan Urological Surgery Improvement Collaborative (MUSIC) is an example of such a collaborative.
Established in 2011 with funding and support from Blue
Cross Blue Shield of Michigan, MUSIC is a physician-led
quality improvement collaborative currently comprised of
42 urology practices and 235 urologists from across Michigan (roughly 85% of the urologists in the state).35 The
collaborative also includes four patient advocates who play a
central role in all of our quality improvement activities. The
mission of MUSIC is to make Michigan the best place in the
world for prostate cancer care.
The collaborative is managed by a coordinating center,
housed administratively in the University of Michigan Department of Urology, and governed by an executive committee (comprised of MUSIC urologists from across the
state) and a core set of operating principles. The participation of each individual practice has been reviewed and
deemed not regulated by a local institutional review board.
A centerpiece of the collaborative is a secure, web-based
clinical registry. In each practice, a trained data coordinator
screens case encounters to identify patients who are eligible
for inclusion in the MUSIC registry. Eligible cases include
patients undergoing prostate biopsy, as well as patients with
newly diagnosed prostate cancer that have not received
prior cancer-directed treatment. Data are collected on patient demographics, cancer severity (including pathologic
details from needle biopsies and radical prostatectomies),
utilization and outcomes for radiographic staging studies,
comorbidities, cancer-directed therapies, and perioperative
complications and patient-reported outcomes for men undergoing radical prostatectomy. MUSIC uses multiple quality
assurance techniques to ensure the integrity of the registry
data.35 The overall quality of the registry is reflected in its
inclusion as a Qualified Clinical Data Registry for the CMS
Physician Quality Reporting System.
Translating Data Into Quality Improvement

Although essential, a registry (and its associated reports and
metrics) is only the first link in a chain that leads to improved
quality and better patient outcomes. Data from the registry
must be analyzed and presented back to clinicians in an
understandable and actionable format. Most importantly,
there must be engagement among participating clinicians,
including a desire to understand and respond to the data in
a way that ultimately yields better patient outcomes. Examples of such engagement include efforts to identify,
e297
communicate, and disseminate specific processes of care or

techniques (e.g., postoperative pain control pathways) that
are used by higher performing surgeons or practices. Importantly, the collaborative model relies on learning from all
participants because no single hospital, practice, or surgeon
excels in all clinical domains or has a monopoly on the
knowledge and experience that leads to superior outcomes.35
This learning most commonly occurs through direct interactions and discussions at collaborative-wide meetings
held several times per year. The meetings are attended by
clinical champions from each participating hospital or practice. These individuals review data and engage in vigorous
discussion and learning with peers; they also maintain responsibility for sharing the data and lessons learned with
partners and clinical team members in their local practices
and hospitals. Ultimately, the quality improvement loop must
be closed by translating the knowledge gained in the collaborative into specific interventions that yield measurable
improvements in patient care. This cycle of data collection and
measurement, analysis, discussion, interventions and then remeasurement is repeated for topics identified and deemed
important by collaborative members (Fig. 2).35 The net effect
is creation of an engine for comparative effectiveness research that is linked to a statewide quality improvement
consortium that serves as the effector arm for translating data
into action and, ultimately, better patient outcomes.
FIGURE 2. Example of a Quality Improvement Cycle
lowest reported at approximately 0.7%.37 We also recently

disseminated a toolkit that discusses the components of a
biopsy bundle that may help to sustain and/or further reduce
the rate of infectious hospitalizations. In addition to the use of
tailored or augmented antibiotic prophylaxis, the bundle
includes the use of a needle disinfectant, a practice associated
with lower infection risk based on our own data in the MUSIC
registry and reports in the published literature.38
Specific MUSIC Initiatives

In terms of specific quality initiative priorities, MUSIC is
focused on reducing morbidity with prostate biopsy, optimizing radiographic staging for men with newly diagnosed
prostate cancer, improving patient outcomes after radical
prostatectomy, and leveraging clinical evidence and shared
decision making to enhance the appropriateness of local
therapy for men with early-stage prostate cancer. Several of
these initiatives are described in more detail below.
Reducing hospitalizations following prostate biopsy. Over
the last 2 years, MUSIC has implemented process changes
for antibiotic prophylaxis that have achieved and sustained a
large reduction in prostate biopsy-related infectious hospitalizations. After identifying a baseline hospitalization rate
of approximately 1.5% (with wide variability across practices
in the state), and fluoroquinolone resistance as the root
cause for most episodes of postbiopsy sepsis, MUSIC developed two pathways to address the rising prevalence of
these bacteria.36 These included tailored antibiotic prophylaxis based on results from a rectal swab culture and
augmented antibiotic prophylaxis (i.e., the addition of a
second agent to standard fluoroquinolone therapy) based
on each communitys local antibiotic resistance profiles. The
development of these pathways was based on published
literature in this area and the specific clinical experience of
MUSIC urologists with the lowest rates of hospitalization.
The net result of this intervention has been a nearly 50%
reduction in the frequency of these significant clinical events.
Our current statewide rate of hospitalization is among the
e298
Optimizing radiographic staging for men with early-stage

prostate cancer. As one of its first activities, MUSIC
addressed ASCOs participation in the Choosing Wisely
priority of avoiding the use of staging radionuclide bone
scans and CT scans among men with early-stage prostate
cancer at low risk for metastases.39 By providing urologists
individual and practice-level feedback on their own practice
patterns relative to their peers, as well as levels of adherence
with guideline recommendations, the collaborative achieved a
rapid statewide decrease in the use of these radiographic
staging evaluations for men with low-risk tumors.40
However, based on further analysis of both the existing
literature and our own registry data, it became clear that
these imaging studies could safely be avoided in a much
larger population of men with newly diagnosed prostate
cancer. Therefore, we developed and disseminated new
imaging appropriateness criteria that both minimize unnecessary imaging and are extremely unlikely to omit a
positive study. The published MUSIC imaging criteria
recommend a bone scan for men with a Gleason score of 8
or higher or prostate-specific antigen (PSA) level of 20 ng/mL
or higher, and a CT scan for men with a Gleason score of 8 or
higher or a PSA level of 20 or higher, or clinical stage greater
than or equal to T3.41,42 After baseline variation in imaging practice patterns was established in 2012 and 2013, a
multidimensional quality improvement intervention was
implemented in 2014, which included video presentations,
dissemination of written guidelines, and site visits to each
practice. Each component of the intervention aimed to
educate providers on the use and rationale for the MUSIC

imaging criteria.
Among more than 10,000 patients newly diagnosed with
prostate cancer that were entered into the MUSIC registry
from March 2012 through July 2015, 8,271 patients (80%)
met the criteria to avoid imaging. Among this group of
patients, we observed a large decrease in the use of both
bone scans and CT scans from before and after dissemination of the improvement interventions. We estimate that
application of the MUSIC criteria for imaging saved patients
in Michigan from undergoing 196 and 260 potentially unnecessary bone scans and CT scans, respectively, over an 18month period. By decreasing the use of both bone scans and
CT scans among men with prostate cancer at low risk for
metastases, this statewide intervention enhanced the value
of care for this patient population through targeted improvement interventions.
Patient-reported outcomes after radical prostatectomy.
Radical prostatectomy is a common treatment for men with
early-stage prostate cancer. Improving functional outcomes
after this procedure remains a priority for the urological
community. To address this quality improvement challenge,
MUSIC is also creating a statewide electronic infrastructure,
known as MUSIC-PRO, which will allow measurement of
urinary and sexual function outcomes at a population-level
across diverse academic and community practices.
MUSIC-PRO began with five practices in April 2014, and
expanded to 17 practices by December 2015. For men
undergoing radical prostatectomy in participating practices,
the coordinating center emails patients with a link to an
online questionnaire on urinary function (score range 0-21)
and erectile function (score range 0-30) before and at 3, 6,
12, and 24 months after radical prostatectomy. Patients
without email are provided with paper surveys. The results
of questionnaires are delivered to providers through the
registry, including patient-specific reports that can be
reviewed during clinic visits. Surgeons also receive summary
data on their own patients in comparison with patients in
their practice and at all sites.
Among 989 men participating in MUSIC-PRO through late
2015, the overall response rate was 85%, with 70% of the
surveys completed electronically. Across the collaborative,
52% and 71% of patients recovered urinary function (defined as a score greater than 17) at 3 and 6 months, respectively. However, recovery of urinary function varied
substantially across five pilot practices. Among patients with
good baseline erectile function (defined as a score greater
than 22), 15% and 21% recovered erectile function at 3 and
6 months.
We are now capitalizing on this variation in patient-reported
outcomes across practices and surgeons to identify top performers and areas needing improvement. Further, these data
can be used for preoperative counseling and managing postoperative expectations. Leveraging this infrastructure, MUSIC is
developing a menu of strategies designed to improve functional recovery for patients across the state.
In summary, the partnership with Blue Cross Blue Shield of

Michigan and patients with prostate cancer, urologists in
Michigan are collecting clinically credible data on prostate
cancer care and outcomes, comparing and capitalizing on
variation in performance among practices and individual
surgeons, sharing best practices, and implementing changes
in clinical behavior. The net result has been more efficient
utilization of health care resources, improved care delivery
in our own environments, and enhancements in the quality,
value, and outcomes of treatment provided to men in
Michigan with prostate cancer.
CONCLUSION
As can be seen by the preceding discussion, there are many
opportunities and challenges associated with improving the
quality of cancer care delivery. With so many different clinicians
involved in the care of patients with cancer (e.g., surgeons,
medical oncologists, radiation oncologists, diagnostic radiologists, pathologists, nurses, primary care providers, and internal
medicine specialists), it is challenging to identify which practitioners are accountable for specific aspects of care. In
addition, as cancer is a disease process with many clinical
manifestations depending on the organ site and disease
presentation, there are many processes and outcomes of
care that need our attention. In terms of measurement, one
does not know where to start.
In contrast, the example of the hands-on urology collaboratives described in Michigan provides a window on
what can be accomplished when a group of clinicians set out
to examine how they are delivering care in specific situations. This is especially valuable when there is consensus on
what types of tests or procedures provide little value or
cause substantial morbidity. It is also important that examination of clinical management practices occurs in a safe
and constructive setting where the primary goal is to understand how and why some unnecessary tests or procedures are performed, and then to address the situation
based on this understanding. Learning how to implement
these best practices can be shared and supported through
the safety of the collaborative environment.
The issue of accountability for correct or harmful processes
or outcomes will likely be resolved with the emergence of
physician groups being merged into organizational units that
will deliver cancer care from diagnosis through end of life in a
bundled or episode-based reimbursement scheme, such as
accountable care organizations. Under this circumstance,
coordination and quality of care will be the preeminent goal,
and elimination of wasteful practices will be a central dogma.
Improvements in health and well-being will be the primary
goal, with measurement of patient-reported outcome measures (e.g., psychosocial well-being, control of symptoms).
The prevention of hospitalizations or emergency department
visits that are costly and dreaded by patients, will become
even more important. Fortunately, there are many experiments in process across the country that will help bring these
important outcomes into mainstream cancer care.
e299
References
1. Levit L, Balogh E, Nass S, et al. Delivering High-Quality Cancer Care:
Charting a New Course for a System in Crisis. Washington, DC: The
National Academies Press; 2013.
Pract. 2007;3:79-86.
3. de Moor JS, Mariotto AB, Parry C, et al. Cancer survivors in the United
States: prevalence across the survivorship trajectory and implications
for care. Cancer Epidemiol Biomarkers Prev. 2013;22:561-570.
4. Kline RM, Bazell C, Smith E, et al. Centers for Medicare and Medicaid
Services: using an episode-based payment model to improve oncology
care. J Oncol Pract. 2015;11:114-116.
5. Schnipper LE, Davidson NE, Wollins DS, et al; American Society of
Clinical Oncology. American Society of Clinical Oncology statement:
a conceptual framework to assess the value of cancer treatment options. J Clin Oncol. 2015;33:2563-2577.
6. Saltz LB. The value of considering cost, and the cost of not considering
value. J Clin Oncol. 2015; JCO647867. Epub 2015 Dec 14.
7. Schnipper LE, Lyman GH, Blayney DW, et al. American Society of Clinical
Oncology 2013 top five list in oncology. J Clin Oncol. 2013;31:4362-4370.
8. Schnipper LE. Quality oncology practice initiative: an approach to
solving the value equation. J Oncol Pract. 2014;10:219-220.
9. Nekhlyudov L, Levit L, Hurria A, et al. Patient-centered, evidence-based, and
cost-conscious cancer care across the continuum: translating the Institute of
Medicine report into clinical practice. CA Cancer J Clin. 2014;64:408-421.
10. Kohn LT, Corrigan J, Donaldson MS. To Err Is Human: Building a Safer
Health System. Washington, DC: National Academy Press; 2000.
11. Institute of Medicine. Committee on Quality of Health Care in America:
Crossing the Quality Chasm: A New Health System for the 21st Century.
Washington, DC: National Academy Press; 2001.
12. NQMC. National Quality Measures Clearinghouse (NQMC). https://
www.qualitymeasures.ahrq.gov/. Accessed February 27, 2016.
13. National Quality Forum. Measure Evaluation Criteria and Guidance for
Evaluating Measures for Endorsement. Washington, DC: National
Quality Forum; 2015.
14. The National Quality Forum. Quality of Cancer Care Measures. Project
Steering Committee Meeting; 2005.
15. Hassett MJ, Bach PB. The Current State of Cancer Quality Measurement.
https://www.qualityforum.org/Publications/2008/09/White_Paper,
_The_Current_State_of_Cancer_Quality_Measurement.aspx. Accessed
February 27, 2016.
16. Donabedian A. The quality of care. How can it be assessed? JAMA. 1988;
260:1743-1748.
17. The Tax Relief and Health Care Act, Pub. L No. 109-432, 2006.
18. Centers for Medicaid & Medicare Services. Physician Quality Reporting
System, 2013.
19. The Patient Protection and Affordable Care Act, Pub. L No. 111-148, 2010.
20. Blayney DW. Defining quality: QOPI is a start. J Oncol Pract. 2006;2:203.
21. Neuss M, Gilmore TR, Kadlubek P. Tools for measuring and improving
the quality of oncology care: the Quality Oncology Practice Initiative
(QOPI) and the QOPI certification program. Oncology (Williston Park).
2011;25:880, 883, 886-887.
22. Jacobson JO, Neuss MN, McNiff KK, et al. Improvement in oncology
practice performance through voluntary participation in the Quality
Oncology Practice Initiative. J Clin Oncol. 2008;26:1893-1898.
23. McNiff KK, Neuss MN, Jacobson JO, et al. Measuring supportive care in
medical oncology practice: lessons learned from the quality oncology
practice initiative. J Clin Oncol. 2008;26:3832-3837.
e300
24. National Quality Forum. Measurement Framework: Evaluating Efficiency Across Patient-Focused Episodes of Care. Washington, DC: National Quality Forum; 2009.
25. Measure Applications Partnership, The National Quality Forum. Performance Measurement Coordination Strategy for PPS-Exempt Cancer
Hospitals. Washington, DC: National Quality Forum; 2012.
26. Hassett MJ, McNiff KK, Dicker AP, et al. High-priority topics for cancer
quality measure development: results of the 2012 American Society of
Clinical Oncology Collaborative Cancer Measure Summit. J Oncol Pract.
2014;10:e160-e166.
27. Hassett MJ, Hughes ME, Niland JC, et al. Selecting high priority quality
measures for breast cancer quality improvement. Med Care. 2008;46:
762-770.
28. Winchester DP, Stewart AK, Phillips JL, et al. The National Cancer Data
Base: past, present, and future. Ann Surg Oncol. 2010;17:4-7.
29. Ellis RP, Fernandez JG. Risk selection, risk adjustment and choice:
concepts and lessons from the Americas. Int J Environ Res Public Health.
2013;10:5299-5332.
30. Pouw ME, Peelen LM, Lingsma HF, et al. Hospital standardized mortality
ratio: consequences of adjusting hospital mortality with indirect
standardization. PLoS One. 2013;8:e59160.
31. Wennberg DE, Sharp SM, Bevan G, et al. A population health approach to reducing observational intensity bias in health risk adjustment: cross sectional analysis of insurance claims. BMJ. 2014;
348:g2392.
32. Fiscella K, Burstin HR, Nerenz DR. Quality measures and sociodemographic risk factors: to adjust or not to adjust. JAMA. 2014;312:
2615-2616.
33. Miller DC, Murtagh DS, Suh RS, et al. Establishment of a urological
surgery quality collaborative. J Urol. 2010;184:2485-2490.
34. Share DA, Campbell DA, Birkmeyer N, et al. How a regional collaborative
of hospitals and physicians in Michigan cut costs and improved the
quality of care. Health Aff (Millwood). 2011;30:636-645.
35. Montie JE, Linsell SM, Miller DC. Quality of care in urology and the
Michigan Urological Surgery Improvement Collaborative. Urol Pract.
2014;1:74-78.
36. Womble PR, Dixon MW, Linsell SM, et al; Michigan Urological Surgery
Improvement Collaborative. Infection related hospitalizations after
prostate biopsy in a statewide quality improvement collaborative.
J Urol. 2014;191:1787-1792.
37. Womble PR, Linsell SM, Gao Y, et al; Michigan Urological Surgery
Improvement Collaborative. A statewide intervention to reduce hospitalizations after prostate biopsy. J Urol. 2015;194:403-409.
38. Issa MM, Al-Qassab UA, Hall J, et al. Formalin disinfection of biopsy
needle minimizes the risk of sepsis following prostate biopsy. J Urol.
2013;190:1769-1775.
39. ABIM Foundation. Imaging Tests for Early Prostate Cancer. http://www.
choosingwisely.org/patient-resources/imaging-tests-for-early-prostate-cancer/. Accessed January 20, 2016.
40. Ross I, Womble P, Ye J, et al. MUSIC: patterns of care in the radiographic
staging of men with newly diagnosed low risk prostate cancer. J Urol.
2015;193:1159-1162.
41. Merdan S, Womble PR, Miller DC, et al. Toward better use of bone scans
among men with early-stage prostate cancer. Urology. 2014;84:
793-798.
42. Risko R, Merdan S, Womble PR, et al. Clinical predictors and recommendations for staging computed tomography scan among men with
prostate cancer. Urology. 2014;84:1329-1334.

OF CARE
Removing Barriers to Clinical Trial

Participation
CHAIR
Joseph M. Unger, PhD, MS
Fred Hutchinson Cancer Center
Seattle, WA
SPEAKERS
Archie Bleyer, MD
St. Charles Medical Center
Bend, OR
Elise Cook, MD, MS
Houston, TX
CLINICAL TRIAL PARTICIPATION
The Role of Clinical Trial Participation in Cancer Research:

Barriers, Evidence, and Strategies
Joseph M. Unger, PhD, Elise Cook, MD, Eric Tai, MD, and Archie Bleyer, MD
OVERVIEW
Fewer than one in 20 adult patients with cancer enroll in cancer clinical trials. Although barriers to trial participation have
been the subject of frequent study, the rate of trial participation has not changed substantially over time. Barriers to trial
participation are structural, clinical, and attitudinal, and they differ according to demographic and socioeconomic factors. In
this article, we characterize the nature of cancer clinical trial barriers, and we consider global and local strategies for reducing
barriers. We also consider the specific case of adolescents with cancer and show that the low rate of trial enrollment in this
age group strongly correlates with limited improvements in cancer population outcomes compared with other age groups.
Our analysis suggests that a clinical trial system that enrolls patients at a higher rate produces treatment advances at a faster
rate and corresponding improvements in cancer population outcomes. Viewed in this light, the issue of clinical trial enrollment is foundational, lying at the heart of the cancer clinical trial endeavor. Fewer barriers to trial participation would
enable trials to be completed more quickly and would improve the generalizability of trial results. Moreover, increased
accrual to trials is important for patients, because trials provide patients the opportunity to receive the newest treatments.
In an era of increasing emphasis on a treatment decision-making process that incorporates the patient perspective, the
opportunity for patients to choose trial participation for their care is vital.
he path from initial development of a new cancer drug to

diffusion of the new therapy into the cancer treatment
community relies on clinical trials, which represent the final
step in evaluating the efficacy of new therapeutic approaches. It has been consistently estimated that less than
5% of adult patients with cancer enroll in cancer clinical
trials.1,2 Conversely, the vast majority of adult patients with
cancer (greater than 95%) do not participate in clinical trials,
even though 70% of Americans are estimated to be inclined
or very willing to participate in clinical trials.3 Thus, a large gap
exists between trial participation rates and the willingness of
patients to participate, suggesting that barriers to trial participation are numerous and frequently insurmountable.
Barriers to trial participation have been the subject of frequent study, but the rate of trial participation has not changed
substantially over time. The infrastructure supporting the
conduct of clinical trials has been designed to anticipate a
low, albeit steady, trial participation rate. The National Cancer
Institutes (NCI) cooperative group clinical trial treatment
program caps enrollment for its funded groups at 17,000 total
patients per year, representing 1% of the estimated 1.7 million
new cancer diagnoses in the United States in 2015.4,5
To understand the effect of clinical trial participation on

cancer population mortality and survival, one might
imagine a counterfactual system in which the cancer clinical
trial participation rate was much higher. Fortunately, such a
system already exists. Enrollment of children (younger than
age 15) to clinical trials has historically been much higher
than for adult cancers (greater than 50%).2,6,7 At the same
time, mortality rates have for children have been decreasing
since the 1970s, whereas for adults they have been decreasing only since the 1990s.8 The average reduction in the
rate of mortality from 1975 to 1995 was 2.6% per year for
patients younger than age 20.9 Interestingly, the reduction
was weakest among older children (age 15 to 19; 2.0% per
year), reflecting other studies which have found both lower
trial enrollment for adolescents and young adults with
cancer and lower rates of mortality reduction.10,11
These data are consistent with the idea that a clinical trial
system that enrolls patients at a higher rate produces
treatment advances at a faster rate, and concurrent survival
increases and mortality reductions in the cancer population.
In this context, the issue of clinical trial enrollment is viewed
as foundational, lying at the heart of the cancer clinical trial
From the Fred Hutchinson Cancer Research Center, Seattle, WA; The University of Texas MD Anderson Cancer Center, Houston, TX; Centers for Disease Control and Prevention, Atlanta,
GA; St. Charles Health System, Quality Department, Bend, OR.
Corresponding author: Joseph M. Unger, PhD, Fred Hutchinson Cancer Research Center, 1100 Fairview Ave. N, M3-C102, P.O. Box 19024, Seattle, WA 98109;
email: junger@fredhutch.org.
185
UNGER ET AL
endeavor.12 Therefore, the identification of specific barriers

to trial enrollment and efforts to remove such barriers represent critical research objectives for cancer investigators.
In this article, we attempt to characterize the specific
barriers to cancer clinical trial participation. In addition, we
consider the distinction between clinical trial enrollment
between children and adolescents with cancer. As suggested
above, these age-proximal patient groups provide a natural
observational contrast illuminating the association between
clinical trial enrollment rates and corresponding improvements in outcomes in the cancer population. We present
original data to make this case. Finally, we examine global and
local strategies to improve cancer clinical trial participation.
FIGURE 1. Model Pathway of Trial Enrollment Process
UNDERSTANDING BARRIERS TO CLINICAL TRIAL

PARTICIPATION
A patients decision of which cancer treatment to receive is
complex and deeply personal; the prospect of incorporating
clinical trial treatment into a patients care adds another
level of complexity. In this multifactorial decision-making
environment, patients may face several barriers to trial participation. As a guide to understanding the trial decisionmaking process, we present a simplified flow diagram (Fig. 1)
illustrating a representative pathway through which a patient
may receive care. This model has been the basis for multiple
studies examining barriers to clinical trial participation.13-15
The model indicates that after cancer diagnosis and a clinic
visit, an assessment of trial availability is made to identify
whether a trial exists at the institution for the patients
histology and stage. If a trial is available, an evaluation of trial
eligibility is made, and, if eligible, a trial is discussed with the
patient. The trial may then be offered to the patient, at which
KEY POINTS
186
Although barriers to trial participation have been the

subject of frequent study, the rate of trial participation
has not changed substantially over time.
Barriers to trial participation are structural, clinical, and
attitudinal, and they differ according to demographic
and socioeconomic factors.
An analysis of the specific case of adolescents with
cancer illustrates how a clinical trial system that enrolls
patients at a higher rate produces treatment advances
at a faster rate and corresponding improvements in
cancer population outcomes.
Fewer barriers to trial participation would enable trials
to be completed more quickly and would improve the
generalizability of trial results. Crucially, increased
accrual to trials is important for patients because trials
provide patients the opportunity to receive the newest
treatments.
In an era of increasing emphasis on a treatment
decision-making process that incorporates the patient
perspective, the opportunity for patients to choose trial
participation for their care is vital.
Abbreviation: SES, socioeconomic status.
point the patient makes a decision about whether to participate. An important note is that under this model, patient
attitudes toward clinical trial participation only come into
play at the end of an otherwise long process.
We have categorized barriers to trial participation as
structural (especially the absence of an available clinical
trial), clinical (i.e., not meeting eligibility), attitudinal (with
respect to both patients and physicians), and demographic
and socioeconomic. It is recognized that there is greater
fluidity between these categories than the model allows,
but simplifications were made to facilitate discussion.
Structural Barriers
To participate in a clinical trial, patients must first have
access to a cancer clinic. Access to a clinic can be influenced
by many different structural factors such as transportation,
travel costs, access to insurance, and availability of child
care.16 Uninsured patients, in particular, present with later
stage of disease and have worse cancer outcomes.17,18 To
the extent that such patients present at their cancer diagnosis with a greater comorbid burden, their likelihood of
eventually participating in a clinical trial is lower.19
Once a patient has access to cancer care, a major structural
barrier pertains to the availability of a clinical trial for the
patients histology and stage. Multiple prospective studies of
the cancer care decision-making process have examined the
extent to which trials are unavailable for patients. Lara et al
prospectively tracked barriers to cancer clinical trials at the
University of California (UC) Davis Cancer Center from 1997
to 2000.20 Among patients considered for trial availability,
there was no trial available to 47% of patients (Table 1). Javid
et al registered patients to a prospective survey study prior
to their treatment decision regarding their cancer care at a
diverse set of eight institutions. No trial was available for
nearly half of the patients (46%).13 Together, these and
earlier studies consistently show that once a patient has
access to cancer care, the absence of an available clinical trial
precludes participation for about half of all patients.14,21,22
TABLE 1. Clinical Trial Enrollment Patterns for Multiple Studies in the Literature
No. Patients Examined*
Lara et al20
171
13
Trial Unavailable
47%
Ineligible
Trial Participation Rate
Did Not Participate
8%
23%
22%
9,09
46%
14%
16%
24%
Klabunde et al14
2,339
60%
16%
7%
17%
Begg et al21
3,534
33%
33%
16%
16%
44,156
60%
18%
8%
14%
49%
18%
14%
19%
Javid et al
22
Hunter et al
Average**
*Assessed for trial availability.

**Unweighted.
Clinical Barriers
Even if a trial is available, patients may not be eligible. Studies
have found that a common reason for patient ineligibility
to available protocols is narrow eligibility criteria.3,14,21-24
Trial eligibility attempt to satisfy two opposing criteria.
On the one hand, eligibility must be sufficiently narrow to
produce a treatment effect that is approximately consistent
across the cohort. On the other hand, eligibility should be
sufficiently inclusive that the trial targets a meaningful population of patients for whom a new treatment would apply.25
Eligibility criteria may also exclude patients due to safety
concerns. However, trials are often criticized for having
eligibility criteria that are too narrow, sacrificing generalizability.26 These exclusions also make trials less accessible
for patients.
In the previously described prospective studies of trial
barriers, ineligibility was identified as a reason for nonparticipation by 18% of all patients on average (Table 1).13,14,20-22,27
The dominant reason for ineligibility exclusions is likely exclusions due to comorbid conditions. One recent study comprehensively cataloged the trial eligibility criteria for a set
of 21 trials in diverse cancer settings.28 The authors found
that the average number of eligibility criteria per trial was
16, 60% of which were related to comorbidity or performance status. Although prespecified trial eligibility criteria
that protect patient safety are crucial, it is also possible that
certain kinds of exclusions are unnecessary. A recent report
indicated trial eligibility criteria have increased in recent years
for both academic group and pharmaceutical-sponsored
clinical trials.29 This trend not only renders trials less accessible for many patients, it may also limit the generalizability
of trial results.
Physician Attitudes
As the agent linking patients to their cancer care, physicians
play an obvious and vital role in clinical trial participation.
A survey of oncologists in community cancer clinics found
that most agreed that clinical trials provide high-quality
care (87%) and benefit enrolled patients (83%).30 However,
physicians face their own barriers to trial enrollment, so
even if quality cancer care is assumed, physicians may treat
otherwise eligible patients off-protocol with one arm of
a trial, without actually entering the patient on the trial.31
Multiple earlier studies found physician decision or preference was the primary reason for nonparticipation in half of
the patients for whom a protocol was available and the
patient was eligible.21,22
A number of factors have been found to deter physician
recommendation for trial participation. In their role of
guiding patient care, physicians may have a strong inclination toward a specific treatment for a given patient.31-33 The
prospective study by Javid and colleagues found that the
nature of the study regimen was cited as a reason for not
discussing a trial with eligible patients by 56% of physicians.13 Physicians are also frequently concerned that clinical
trial participation can interfere with the physician-patient
relationship.31,34,35 Random assignment into a phase III trial
in particular subjects the treatment choice to uncertainty,
and physicians may anticipate that the introduction of uncertainty will subvert patient confidence in the physicians
expertise, even if, as indicated by the existence of a randomized clinical trial, multiple treatments of potentially
similar efficacy are available.
Practical considerations may also influence physicians
willingness to participate in trials. Physicians often lack
appropriate incentives to participate in clinical research.36
The time spent attending to the details of clinical trial enrollment and explaining clinical trials to patients can often
be prohibitive for physicians.33 In addition, in a busy clinic
environment, physicians may be less likely to refer patients
to trials if they believe the process will be too time
consuming.30,31,37 Oncologists who consider trial paperwork time consuming or who otherwise believe trial effort would be extra work are less likely to refer a patient
to a clinical trial.37 Finally, some physicians find the requirement of obtaining informed consent to be problematic, even though nearly all agree that informed consent is
necessary.30,31
Patient Attitudes
Efforts to reduce structural, clinical, and physician barriers to
trial participation are critical. However, the ultimate decision
regarding trial participation rests with the patient. Inevitably, the decision about whether to participate in a trial
will reflect a patients personal preferences, which may also
be influenced by family and friends.38
187
UNGER ET AL
Some proportion of patients are influenced by altruistic

motivations.15 However, the majority of patients are (appropriately) concerned primarily with finding the best
possible treatment of their disease.15,39,40 In the absence of
other barriers, a patient who believes that the best possible
treatment option is to be found in a clinical trial is more likely
to participate in that clinical trial.33
Patients have frequently reported being uneasy or fearful
about the prospect of participating in a clinical trials.41 In
some cases this could be due to a residual mistrust of
medical science due to past abuses, such as the infamous
Tuskegee Syphilis Study or the history of human experimentation with radiation following World War II.42,43 Attention in the last several decades to the process of rigorous
consent may have reduced these fears, especially for
younger generations of patients. Attention must also be paid
to providing consent forms which are easy to read, because
more complicated consent forms can induce anxiety.44
More generally, a fear of experimentation may be
expressed through a dislike of randomization.14,15,22,45-47
There is perhaps no stronger indication that a patient is
about to participate in an experiment than the revelation
that the patient will be randomly allocated to one of two or
more treatments. Fear of randomization has been identified
as the most commonly cited reason by patients for declining
trial participation.15 Recognizing this, some physicians avoid
the word randomization, relying instead on analogy to
describe the randomization process, though this may lead to
situations where patients sometimes do not understand that
their treatment has been randomly assigned.48,49
Patients are sometimes uneasy as well about the potential
toxic effects of chemotherapy in trials, especially for the
experimental therapies.38 Patients may already have a
strong sense of the particular treatment they wish to receive
after discussion with their physicians.33 Because trials
sometimes require more frequent monitoring than nontrial
care, traveling to and from a cancer clinic has been indicated
by many patients to be a reason for nonparticipation.15,20
Concern about how to pay for trials has been cited as a
reason for nonparticipation among about a quarter of patients, despite the fact that the majority of states mandate
that insurers cover the routine care costs of trials, as does
Medicare.15 In a review by Ford et al, cost concerns were
identified as the second most frequently indicated reason
for nonparticipation in trials in the literature.41
Demographic and Socioeconomic Disparities

Demographic and socioeconomic disparities in trial enrollment can occur anywhere along the pathway from initial
clinic visit until the patient ultimately makes their treatment
decision. The most consistent and largest disparity pertains
to age.1,15,50-53 Hutchins et al found that patients in cooperative group trials were much less likely to be age 65 or
older than those in the U.S. cancer population.50 Some
evidence suggests that attitudinal barriers on the part of
physicians play a role.13,54-56 In addition, older patients are
188
likely to have more comorbid burdens, inducing clinical

exclusions.57,58 To the extent that trials seek to reflect the
population of patients for whom new trial-proven treatments will be administered, better representation of older
patients in trials is critical. Recognizing this, in 2000,
Medicare was directed to cover the routine care costs of
clinical trial participation for its patients.59 Unfortunately,
the proportion of older patients in trials remains well below
the expected rate.28,51
Evidence as to the association of race with trial participation is mixed. A study by Murthy et al found that black
patients were underrepresented in NCI-sponsored breast,
lung, colorectal, and prostate cancer clinical trials from 2000
to 2002.1,60 In contrast, in sequential studies within SWOG, a
national clinical trials consortium, black patients were enrolled to trials in a representative fashion over an extended
period of examination.28,50,53 This was confirmed in a sample
of older patients with breast cancer.61 Evidence has also
been mixed for Hispanic patients.1,53 Even if enrollment of
minorities is adequate in the treatment trial setting, enrollment of minority healthy volunteers for prevention trials
has been decidedly more difficult and has generated welldesigned outreach programs for large individual trials.62-65
Given the increasingly diverse nature of the U.S. population,
continued attention to this issue is required.
Females may be somewhat under-represented in the
nonsex-specific cancers, although the magnitude of the
disparity is likely small.28,50,53 In addition, some evidence
suggests that any sex disparity in clinical trial enrollment
could be age-related, with older women less likely to enroll
in trials than their younger counterparts.1 Such a pattern
might indicate generational differences in attitudes toward
clinical research.
The examination of socioeconomic factors as a barrier to
participation has historically been hindered by the lack of
collection of patient-level socioeconomic status (SES) data.
This is unfortunate given the frequency with which the direct
and indirect costs of trial participation have been cited as
meaningful barriers.41 Two recent articles overcame this
limitation. The first used a web-based survey to engage
patients in their decision-making process.15 Among the
numerous demographic and socioeconomic factors examined, patients with lower income (less than $50,000 per
year) were 29% less likely to participate in trials than higherincome patients. Utilizing data from a prospective barriers
study, this observation was confirmed.13,66 Thus, the evidence to date of income disparities in trial enrollment is
fairly consistent.61 Given that trial treatment costs are not
substantially different than nontrial treatment costs,67 this
suggests that marginal direct costs play a role. A prior study
of the effect of the 2000 Medicare policy change on trial
participation found that patients with Medicare plus private
insurance participated at a higher rate following the policy
change, whereas patients with Medicare alone participated
at rates similar to those prior to the policy change.53 This
finding points to the prohibitive influence of copays and
coinsurance for some patients.
EVIDENCE FOR THE BENEFIT OF CLINICAL TRIALS

ON CANCER POPULATION OUTCOMES
OBSERVED THROUGH THE RELATIVE LACK OF
PROGRESS IN ADOLESCENTS AND YOUNG
ADULTS
The potential barriers to trial enrollment that patients face
are numerous. But, just how important are clinical trials for
progress against cancer? The answer to this question is
crucial, because if trial participation is ultimately unrelated
to cancer population survival gains, the issue of barriers to
trial participation has little importance. To examine this, we
studied the relationship between adolescents and young
adults (AYAs) and cancer population outcomes over time.
This is an ideal group in which to examine the impact of
clinical trials given that since 1980, AYA patients have had a
slower rate of cancer population survival improvement than
younger and older age groups by 5% to 13% in absolute
differences (Fig. 2). Concomitantly, cancer has become the
most frequent cause of death due to disease in AYAs.68
During the past decade (2000 to 2009), deaths due to cancer
declined in all age groups except in young adults age 20 to
29; in 25- to 29-year-olds, deaths due to cancer actually
increased.69
The survival disparity between AYAs and other patients
may be due in part to early achievements in improving
survival for AYAs, after which resources were directed toward research in other age groups. Also, cancer in AYAs have
potentially complex biologic signatures that neither pediatric oncologists nor adult-treating medical and hematologist oncologists are accustomed to treating.70 AYAs are also
less likely to have health insurance,71 especially prior to the
FIGURE 2. Increase in Absolute Percentage of Annual

5-Year Cancer-Specific Survival Rates Since 1973 to
1975 by Calendar Year and Age
Baseline is 1973 to 1975 average. Kaposi sarcoma is excluded due to the HIV/
AIDS epidemic during the 1980s and early 1990s; thyroid cancer is excluded
because of overdiagnosis and increasing survival inflation. Regressions are 4
polynomials. Data source is SEER 9 regions.75
advent of the Affordable Care Act, which could be associated

with delays in diagnosis and compromises in optimum diagnostic and therapeutic interventions.72 At the same time,
AYAs have had the lowest participation in clinical trials in
absolute terms than any other major age group.10 The
central issue then is the extent to which lack of clinical trial
activity in AYA patients with cancer accounts for the relative
lack of progress in improving cancer population outcomes.
All Cancers
The Cancer Therapy Evaluation Program (CTEP) of NCIs
Division of Cancer Therapy and Diagnosis has patient accrual
data from phase I, II, and III cancer treatment trials conducted by the NCI cooperative groups and NCI-designated
cancer centers. For this analysis, 371,302 patient entries
during 1997 to 2009 were examined. In addition, we used
cancer population data derived from the Surveillance, Epidemiology, and End Results registry, U.S. Census data, and
joinpoint statistical software analyses to examine trends in
U.S. cancer population estimates.73-78
Figure 3 shows the relationship between the average
percent change (APC) in the 5-year cancer-specific survival
rate from 1985 to 1999 and the accrual rate to national
cancer treatment trials during 2001 to 2006. Although this
comparison is confounded by time, there was a nearly 1:1
correlation over the entire age range that was strongly
significant. Patients age 15 to 34 had the lowest APC in
5-year survival. A similar pattern was found with respect to
cancer mortality, which isolates patients age 0 to 40 (Fig. 4).
Patients age 20 to 24 had a particularly poor reduction in
cancer mortality, as well as the lowest absolute number of
clinical trial accruals. In both cases, the correlation between
trial enrollment and, respectively, APC in 5-year survival and
mortality is clearly evident and highly significant (p , .001).
Acute Lymphoblastic Leukemia

The greatest effort during the last decade to increase accruals in AYAs was within the population with acute lymphoblastic leukemia (ALL), the most common pediatric
cancer. New clinical trials in ALL specifically designed for
AYAs were launched,79-81 the National Comprehensive
Cancer Network released practice guidelines for ALL,82 and
an increasing number of presentations and publications on
the topic occurred at national meetings and appeared in the
peer-reviewed medical literature.83-89 The effort was effective, with AYAs having the greatest accrual increase of all
cancers in ALL, up to twice that of the cancer with the next
greatest increase, acute myeloid leukemia.74,90 Perhaps in
part for this reason, the only increase in either absolute
accruals or accrual as a proportion of cases during the first
decade of the 21st century occurred among patients age
10 to 20 (Fig. 5).
Hence, the population with ALL was examined more
closely for a relationship between survival improvement and
clinical trial participation. As shown in Fig. 6, clinical trial
accruals as a proportion of ALL cases in the United States
189
UNGER ET AL
FIGURE 3. Comparison of Average Percent Change in

the 5-Year Cancer Survival Rate and Treatment Trial
Accruals, by 5-Year Age Intervals
referrals, participation, and conduct. Fortunately, NCIdesignated cancer centers are evaluating their own AYA
referral patterns and clinical trial determinants,92,93 and
intergroup efforts are underway within the current organizational structure of the federal clinical trials enterprise,
including the NCIs National Clinical Trials Network (NCTN),
to create novel opportunities for collaborative AYA oncology
research among the pediatric and adult NCTN groups.94,95 As
is done in England, age-specific biology, pharmacology,
proteomics, genomics, and clinician and patient behavior
studies embedded within clinical trials are required to
further improve survival for AYAs.96
GLOBAL AND LOCAL STRATEGIES TO IMPROVE

The open columns represent trial accruals during 2001 to 2006 and the colored
bars the average percent change (APC) in 5-year relative survival rate of all
invasive cancer except Kaposi sarcoma during 1985 to 1999. The red bars indicate
the adolescent and young adult (AYA) age group. The inset compares the APC in
5-year survival rate with the treatment trial accruals. Accrual data from the
National Cancer Institute Cancer Therapy Evaluation Program (CTEP) were
provided by Steve Friedman, Michael Montello, Troy Budd, and Samantha
Finnegan via the Freedom of Information Act. Survival data were obtained from
SEER 9 Regions.75 Kaposi sarcoma is excluded from the survival statistic because the
HIV/AIDS epidemic occurred during the 1980s and early 1990s, which substantively
altered the overall cancer survival rate in AYAs during those years.
during 2000 to 2009 drops precipitously for patients age

15 to 20. Figure 6 also shows the 5-year leukemia-specific
survival rate for patients with ALL as a function of single year
of age. Joinpoint analysis identified two inflections, ages 17 and
20, during which the 5-year survival rate decreased 23%. This
AYA ALL cliff constituted 30% of the overall decline from
95% at age 5 to less than 20% at age 70. The cliff patterns for
both accrual and survival are virtually superimposable,
which strongly suggests they are related, although other
factors, such as a switch from pediatric to adult treatment
regimens, could also contribute.91
We have illustrated the nature of clinical trial enrollment

barriers and established the potential link between trial
enrollment and improvements in cancer population survival.
Efforts to improve trial enrollment of cancer patients are
clearly needed. In this context, we propose the following
global and local strategies to improve trial participation.
Overall Strategies
In 2010, the NCI and the American Society of Clinical Oncology sponsored a Cancer Trial Accrual Symposium to
provide recommendations for trial recruitment. Summary
recommendations centered on the patient, community,
physician/provider, and site.97 This symposium led to many
recommendations at each level consistent with the overarching view that one size does not fit all when it comes
to recruitment to clinical trials. The organizers noted in
FIGURE 4. Comparison of Average Percent Reduction

in the Annual National Cancer Mortality Rate and
Treatment Trial Accruals, by 5-Year Age Intervals, Age
Younger Than 40
Clinical Trials Effect Summary

These data enable three fundamental conclusions. First,
both survival prolongation and mortality reduction in patients with cancer are correlated with clinical trial activity.
Second, the dependency of survival prolongation on treatment trial accrual has been apparent for all ages. Third, AYAs
have had the least trial participation and the least survival
prolongation and mortality reduction, particularly among
patients age 20 to 29.
It has been previously observed that the age-dependent
rate in the reduction of deaths attributed to cancer in the
United States is correlated with the age-dependent accrual
of young adults to national cancer treatment trials during the
same era.69 After suicide, cancer is the second leading cause
of death due to disease among AYAs. More is needed to
overcome the national AYA cancer-related death problem,
beginning with increased clinical trial availability, access,
190
The open columns represent trial accruals during 2000 to 2006 and the colored
bars the average percent reduction in national cancer mortality rate during 1990
to 1998. The red bars indicate the adolescent and young adult (AYA) age group.
The inset compares the mortality rate reduction with the treatment trial accruals.
Accrual data from the National Cancer Institute Cancer Therapy Evaluation
Program (CTEP) were provided by Steve Friedman, Michael Montello, Troy Budd,
and Samantha Finnegan via the Freedom of Information Act. Mortality data were
obtained from the National Center for Health Statistics via the SEER program.78
FIGURE 5. Absolute and Relative Trial Accrual Rates to

NCI Treatment Trials: Comparison of 2001 to 2003
Versus 2007 to 2009 Estimates
Comparison of 2001 to 2003 versus 2007 to 2009 for annual accruals to

treatment trials sponsored by National Cancer Institute (NCI)-sponsored
cooperative groups and NCI-designated cancer centers (red curves) and accrual
proportion of all patients in the United States with invasive cancer into the trials
by 5-year age intervals (green curves), by single years of age. The heavy curves
represent 2007 to 2009 and the thin curves 2001 to 2003. Accrual data from the
NCI Cancer Therapy Evaluation Program (CTEP) were provided by Steve Friedman,
Michael Montello, Troy Budd, and Samantha Finnegan via the Freedom of
Information Act. Accrual proportion (%) was estimated from cancer incidence in
SEER 9 regions and population data from the U.S. Census Bureau.73,75
particular that although clinical trial enrollment barriers

have been extensively studied, Few rigorously conducted
studies have tested interventions to address challenges to
clinical trials accrual.97 In this broader context, the NCIs
AccrualNet website provides strategies, tools, and other
resources to support clinical trials.98
Global Strategies
At the beginning of this article, we delineated many of the
specific challenges to clinical trial enrollment. Given the
need to accrue large numbers of patients in a shortened
timeline and the increased complexity of U.S.-based clinical trials, academic and industry sponsors are increasingly
exploring regions outside of the United States to conduct
trials, including in less developed regions of the world.
In the view of Barrios et al, the globalization of clinical trial
research is unavoidable.99 In a wide ranging review, they
propose the following solutions to some of the challenges
of clinical trial globalization:
1. Harmonize and share standards and goals for product
safety, quality, and efficacy.
2. Use finite resources more efficiently, share knowledge, and optimize inspection resources.
3. Engage global partners to advance regulatory science
and public health solutions.
4. Implement risk-based monitoring and clinical inspection.
5. Incorporate error reduction strategies and consider
regional variations in the standards of care and their
effect on trial results.
6. Decentralize laboratories in favor of developing regional expertise to provide carry-over benefits after
trial completion.
7. Streamline and advance bio-bank regulatory issues.
8. Invest in research and evidence-based cancer care
relevant to each region, including cancer registries
and clinical trial infrastructure.
In the authors view, the globalization of clinical research
is vital to speed up availability of life-saving medicines
throughout the world.99 In the setting of a domestic clinical
trial system that has been described as being in a state of
crisis, this view has added weight.12
The importance of differing cultural, scientific, ethical,
government, and logistic issues in each region must be
considered.100 One key ethical issue is whether the study
drug will be available at the end of the clinical trial. Availability may be affected by local religious customs concerning
contraceptive studies and ethical guidelines limiting pediatric clinical trials. An additional ethical issue in low resource countries involves whether to develop local resources
for testing or to use international vendors for that purpose.
Although developing local resources provides an often needed
benefit, this could involve higher costs and may also be deterred
by local shipping laws and other logistic barriers.
An excellent example of global recruitment is the START
trial. 101 START is a multicenter, phase III, randomized,
double-blind, placebo-controlled trial of the cancer vaccine
tecemotide in patients with non-small cell lung cancer with
unresectable stage III disease. The trial is sponsored by
Merck KGaA and EMD Serono, Inc., in 275 study centers in
33 countries worldwide. A continuous series of strategies
was implemented for patient recruitment and retention
throughout the duration of the trial.102 These strategies
were referred to as the START global patient recruitment
and retention continuum, with the overarching purpose of
raising awareness of the START trial and keeping it in the
forefront in physician communities and in the local START
sites to increase patient enrollment and retention. The
strategies target physicians, research staff, local sites, patients, and the local community. Physicians were provided
START informational calls (sometimes physician-to-physician),
visits, and meetings, and START information at national oncology meetings. Research staff received START education,
and recruitment tools including motivational videos. The
local sites were offered additional site funding for accelerated recruitment and START educational teleconferencing.
Patients received holiday and thank you cards and patient
START educational materials. The local community was
reached through local media outreach, public awareness
advertisements, and engagement of local site liaisons. Enrollment to the trial occurred from 2007 through 2011 and
reached full accrual of 1,513 patients, indicating a highly
successful recruitment effort.
Domestic trials may also partner with international collaborators to augment trial enrollment. The SWOG SELECT
trial for prostate cancer prevention used similar recruitment
191
UNGER ET AL
FIGURE 6. 5-Year Leukemia-Specific Survival Rates in

Patients with Acute Lymphoblastic Leukemia
Diagnosed From 2000 to 2012, and Estimated ALL
Treatment Trial Accrual Proportion From 2000 to
2009, by Single Years of Age
Each year of age was averaged from two consecutive years. Joinpoint analysis of
survival data identified ages 17 and 2074; linear regressions for survival data are for
age ranges 5 to 17, 17 to 20, and 20 to 70. Survival data were obtained from SEER 18
regions.77 Accrual data from the NCI Cancer Therapy Evaluation Program (CTEP) were
provided by Steve Friedman, Michael Montello, Troy Budd, and Samantha Finnegan
via the Freedom of Information Act. Accrual proportion (%) was estimated from
cancer incidence in SEER 9 regions and population data from the U.S. Census
Bureau.73,75
strategies as the START trial. The SELECT trial enrolled over

35,000 men in the United States, Canada, and Puerto Rico,
and completed enrollment 2 years ahead of schedule.62,63
On a smaller accrual scale, the contributions of the International Breast Cancer Study Group, in collaboration with
the cooperative groups of the NCI, provided necessary accrual to a trial evaluating ovarian failure in premenopausal
women with breast cancer.103
Local Strategies
Social media. Increasingly, social media platforms provide
an opportunity to communicate about clinical trials with
potential trial researchers and participants.104 The Quorum
Review Institutional Review Board (IRB) offered the following considerations for a plan to use social media in
research105,106:
1. Provide a rationale for the application of social media
to the target population.
2. Address the privacy and confidentiality concerns of
the social media applications to be used.
3. Vet all communications for sensitivity and potential for
harm, even if the content does not require IRB approval.
4. Provide a summary statement regarding the intended
uses of the social media account.
5. Closely monitor user-generated content (if allowed),
which is essential to a robust online community, for
patient protection and study integrity.
192
Moreover, predetermined IRB-approved responses to

anticipated user-generated content should be available to
allow immediate responses when required, and links to
the study website should be included to allow for integration
of study platforms.
The U.S. Food and Drug Administration (FDA) has provided
no specific guidance on the use of social media in clinical
research.107 The Recruitment Information Sheet states that
in the case of direct advertising, the information and mode
of communication should be reviewed by the IRB for evidence of coercion or implication of benefits to participation.
The Office of the Inspector General has released guidance
regarding the use of clinical trial websites, indicating that IRB
approval of a clinical trial listing, if limited to selected basic
information, is not required.108
InVentiv Health, a contract research organization, has
provided specific plans to assist researchers to plan digital
recruitment campaigns (Table 2).109 Their plans are premised on the idea that recruitment for global clinical trials is
an area ripe for re-engineering. More broadly, Facebook has
often been listed as the social media of choice; of the 73% of
online U.S. adults using social media, 71% use Facebook.
Often researchers using Facebook attempt to recruit from
the initial audience prior to forming a relationship. A better
approach may be to grow and engage your audience first,
before patients are recruited.110
Strategies to Address Demographic and

Socioeconomic Barriers
Elderly recruitment. Patients with concomitant illnesses
are often excluded from trials to ensure safety and to isolate
the cancer as the primary source of morbidity in the patient.
Unfortunately, this has the effect of excluding many patients
from trials, especially older patients with a greater comorbid
burden.28,57,58 Further, trials typically exclude patients with
prior cancers, even as the population of cancer survivors in
the United States is growing and which currently numbers
around 15 million.111 In this context, one strategy to remove
barriers to trials would be to remove unnecessary eligibility
criteria. Such an approach would improve access to trials,
especially for older patients, andbecause histology and
stage explain the vast majority of variation in cancer outcomes, rather than comorbid conditionswould result in
only limited loss of power to test the efficacy of new
treatments.28 One study estimated that if protocol exclusions related to organ system abnormalities and functional
status were relaxed, participation of older patients in clinical
trials would approach 60%, in line with cancer population
rates.51
Researchers should also consider increasing the number of
trials targeted to older patients, with due consideration to
potential safety issues.112 Several trials found no more
toxicity in elderly patients in chemotherapy-containing trials
than in younger patients, when patients were appropriately
selected.113,114 However, when chemotherapy was given
to patients age 80 or older, high risk of hospitalization or
TABLE 2. Proposed Steps to Plan Digital Recruitment

Campaigns109
Steps for Strategy Development
Audit digital channels and social
media sites to determine
where the targeted patients/
caregivers can be found
Steps for Campaign

Implementation
Place ads with key messages and
a call to action in the selected
sites and social media platforms
Monitor existing online chats for Link banner or pop-up ads to

the area of research interest
search engine results for
to understand the issues and
targeted health-related
concern to patients/caregivers
searches
and listen for patient speak so
it can be replicated in your
messages
Map the recruitment messages
to the targeted patients/
caregivers and appropriate
channels
Partner with advocacy groups and

place trial ads on these groups
online sites
Identify key bloggers and others

who could eventually raise
awareness and support trial
participation
Engage key opinion leaders as

online ambassadors to raise
awareness of the trial
Develop the creative concepts

for advertising and test the
messaging and imagery for
effectiveness
treatment discontinuation as a result of toxicity (even with

frequent dose modifications) were observed.115 The International Society of Geriatric Oncology recommends the
use of comprehensive geriatric assessment (CGA) in cancer
patients older than age 70.116,117 The CGA is time consuming, often leading to physician abandonment. Fortunately, CGA time requirements have led to the development
of prescreening tools used to determine whether full
screening with CGA is required, though there are inconsistent results regarding the validity of these tools.118-120
More generally, it is important to develop prediction models
capable of estimating risk of chemotherapy for octogenarians and nonagenarians with regards to toxicity and hospitalization.115 A comprehensive approach to the evaluation
of the older patient with cancer considers the patients
residence and fitness and includes an interdisciplinary team
to provide individualized care.121
Socioeconomic barriers. If marginal direct costs are prohibitive for some patients, then measures to cover these
costs would remove a critical barrier to enrollment. One
approach would be to cover the excess costs of clinical trials
for all patients, because even in an insured population,
copays and coinsurance have been shown to deter clinical
trial participation.53 Another potential approach would be to
provide payments to patients. In the United States, the
practice of paying patients for trial participation is widespread, but also contentious, highly variable, and lacking in
general guidance.122 One concern is that a payment inducement might alter a subjects assessment of potential
risks or impair their judgment, although there is little
evidence that payment inducements do or do not affect

assessment.123,124 A careful calibration of the size of any
monetary incentive would be necessary to avoid undue
influence.125
Measures to address socioeconomic disparities in recruitment may have a preferentially beneficial impact on
minority patients. For the SELECT trial, several strategies
specifically addressed patients with low socioeconomic
status.61,62 SELECT provided funds to sites semiannually to
offset travel expenses and meals, in addition to providing
patient retention items. Larger supplemental site grants
were awarded to 15 SELECT sites with potential to increase
minority recruitment through a competitive award mechanism. These additional funds were most commonly used to
provide additional staff time for minority recruitment. Sites
also provided reimbursement for food and/or transportation costs expended to participate in the trial.62
DISCUSSION
Both patients and physicians have been found to regard
clinical trial participation as a positive approach to cancer
care.3 Despite this, the complexity of the enrollment process
and the potential barriers faced by patients have combined
to make a successful clinical trial enrollment a rare event.
Clinical trials are the key step in advancing new treatments
from the research setting to the cancer care clinic.
Therefore, a thorough understanding of the nature of trial
enrollment patterns and barriers to enrollment is of paramount importance. The literature indicates that structural
barriers preclude patient participation in trials for half of all
cancer patients. Among patients for whom a trial is available,
about half (or a quarter of all patients) are excluded due to
eligibility issues with trial exclusion criteria. The remaining
patients are sometimes not offered the chance to participate
because of physician concerns, or decline due to patient
concerns. Structural, clinical, and attitudinal barriers to trials
can differ according to some important factors, especially
age. As a result, only a small portion of adult patients with
cancer participate in trials, less than 5%, a rate that has
remained fairly constant over decades.
Increasing accrual to clinical trials is important for multiple
reasons. Faster accrual would enable trials to be conducted
more quickly. The predominant reason that trials fail to
complete is poor accrual.126 These failures represent a lost
investment on the part of funding agencies. Moreover, when
clinical trials close because of failure to accrue, nonfinancial
costs are also incurred. Patients have exposures to study
drugs with potential or realized adverse events. Patients and
research staff may experience psychologic effects such as
loss of trust and morale. Informed consent documents rarely
include the risk of closure because of lack of study participation, despite the fact that about one in four randomized,
phase III trials have such an outcome.126
The more rapid completion of trials would enable new
treatments to be developed more quickly. We have shown
data indicating a compelling relationship between the
193
UNGER ET AL
incidence of clinical trial enrollments and improvements in

cancer population survival outcomes. Our focus was a natural
observational contrast between AYAs and other age groups
with cancer. We found a slower rate of progress in AYAs
compared with younger and older patients, which underscores
the need to increase the number of clinical trials available to
AYAs with cancer and their participation in them. By extension,
this observation also points to the need to increase trial enrollment for patients of any age group, or any demographic
group, because this could have a beneficial impact on increasing survival and reducing mortality from cancer.
Another important reason to increase clinical trial accrual is
to improve the generalizability of clinical trial results. Figure 7
illustrates how the vast majority of patients with cancer of all
agesbut, especially patients older than about age 15do
not participate in clinical trials.127 However, we have made
the case that there is a strong correlation between trial
participation and cancer population survival improvements.
Thus, trial results must generalize to nontrial patients, at
least to some degree. But, they may not do so in an efficient
manner, and cancer population survival gains may be lost in
the process. Under this rubric, greater participation leads to
greater generalizability, which leads to better cancer population outcomes. Patients who participate in cancer trials
are usually younger, healthier, and perhaps wealthier than
the typical patient who is not a trial participant. To the extent
that trials are more inclusive with respect to comorbid or
other conditions, adequately represent the demographic
makeup of the United States, and are easier to pay for, the
generalizability of trials would likely improve.
Finally, increased accrual to trials is important for patients,
because trials provide opportunity to receive the newest
treatments. The principle of equipoise posits that a properly
designed treatment trial tests a new or modified form of
therapy that is not known to have that benefit (otherwise
the trial would not be justified). In contrast, in addition to
access to the newest treatments, patients who participate
in clinical trials have other potential advantages, such as
FIGURE 7. Estimated Treatment Trial Accrual

Proportion of Patients Diagnosed With Cancer From
2008 to 2010 by Single Year of Age and the History of
SEER Representation of the United States Population
Accrual data from the NCI Cancer Therapy Evaluation Program (CTEP) were
provided by Steve Friedman, Michael Montello, Troy Budd, and Samantha
Finnegan via the Freedom of Information Act. Accrual proportion (%) was
estimated from cancer incidence in SEER 9, SEER 13, and SEER 18 regions and
population data from the U.S. Census Bureau.73,75-77
Modified from Bleyer A.127
access to potentially less expensive therapies (if the agent

is provided at no or reduced cost to the patient), to teams
of professional dedicated to the patients care, and to care
that is strictly directed by a protocol. Moreover, in an era
of increasing emphasis on a treatment decision-making
process that incorporates the patient perspective, the
opportunity for patients to choose trial participation for
their care should not be hindered by unnecessary barriers. In the end, the potential benefits of trial participation will be shared by patients, researchers, and future
generations.
References
1.
2.
3.
4.
5.
6.
194
Murthy VH, Krumholz HM, Gross CP. Participation in cancer clinical trials:
race-, sex-, and age-based disparities. JAMA. 2004;291:2720-2726.
Tejeda HA, Green SB, Trimble EL, et al. Representation of AfricanAmericans, Hispanics, and whites in National Cancer Institute cancer
treatment trials. J Natl Cancer Inst. 1996;88:812-816.
Comis RL, Miller JD, Aldige CR, et al. Public attitudes toward participation in cancer clinical trials. J Clin Oncol. 2003;21:830-835.
National Cancer Institute. The NCTN Budget. http://www.cancer.gov/
research/areas/clinical-trials/nctn/budget#1. Accessed January 13,
2016.
Howlader N, Noone AM, Krapcho M, et al (eds). SEER Cancer Statistics
Review, 1975-2012, National Cancer Institute. April 2015. http://seer.
cancer.gov/csr/1975_2012/. Accessed January 13, 2016.
Bond MC, Pritchard S. Understanding clinical trials in childhood
cancer. Paediatr Child Health. 2006;11:148-150.
7.
8.
9.
10.
Hunger SP, Lu X, Devidas M, et al. Improved survival for children and

adolescents with acute lymphoblastic leukemia between 1990 and
2005: a report from the Childrens Oncology Group. J Clin Oncol. 2012;
30:1663-1669.
Kohler BA, Sherman RL, Howlader N, et al. Annual Report to the Nation
on the status of cancer, 1975-2011, featuring incidence of breast
cancer subtypes by race/ethnicity, poverty, and state. J Natl Cancer
Inst. 2015;107:djv048.
Ries LAG, Smith MA, Gurney JG, et al (eds). Cancer incidence and
survival among children and adolescents: United States SEER Program
1975-1995. National Cancer Institute, SEER Program. NIH Pub. No. 994649. Bethesda, MD: National Institutes of Health; 1999.
Bleyer A, Budd T, Montello M. Adolescents and young adults with
cancer: the scope of the problem and criticality of clinical trials.
Cancer. 2006;107:1645-1655.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
Ferrari A, Bleyer A. Participation of adolescents with cancer in clinical

trials. Cancer Treat Rev. 2007;33:603-608.
Institute of Medicine. A National Cancer Clinical Trials System for the
21st Century: Reinvigorating the NCI Cooperative Group Program.
Washington, DC: The National Academies Press; 2010.
Javid SH, Unger JM, Gralow JR, et al. A prospective analysis of the
influence of older age on physician and patient decision-making when
considering enrollment in breast cancer clinical trials (SWOG S0316).
Oncologist. 2012;17:1180-1190.
Klabunde CN, Springer BC, Butler B, et al. Factors influencing enrollment in clinical trials for cancer treatment. South Med J. 1999;92:
1189-1193.
Unger JM, Hershman DL, Albain KS, et al. Patient income level and
cancer clinical trial participation. J Clin Oncol. 2013;31:536-542.
Rivers D, August EM, Sehovic I, et al. A systematic review of the factors
influencing African Americans participation in cancer clinical trials.
Contemp Clin Trials. 2013;35:13-32.
Roetzheim RG, Pal N, Tennant C, et al. Effects of health insurance and
race on early detection of cancer. J Natl Cancer Inst. 1999;91:
1409-1415.
Ward E, Halpern M, Schrag N, et al. Association of insurance with
cancer care utilization and outcomes. CA Cancer J Clin. 2008;58:9-31.
Robbins AS, Pavluck AL, Fedewa SA, et al. Insurance status, comorbidity level, and survival among colorectal cancer patients age 18 to 64
years in the National Cancer Data Base from 2003 to 2005. J Clin Oncol.
2009;27:3627-3633.
Lara PN Jr, Higdon R, Lim N, et al. Prospective evaluation of cancer
Begg CB, Zelen M, Carbone PP, et al. Cooperative groups and community hospitals. Measurement of impact in the community hospitals.
Cancer. 1983;52:1760-1767.
Hunter CP, Frelick RW, Feldman AR, et al. Selection factors in clinical
trials: results from the Community Clinical Oncology Program Physicians Patient Log. Cancer Treat Rep. 1987;71:559-565.
McCusker J, Wax A, Bennett JM. Cancer patient accessions into clinical
trials: a pilot investigation into some patient and physician determinants of entry. Am J Clin Oncol. 1982;5:227-236.
Simon MS, Brown DR, Du W, et al. Accrual to breast cancer clinical
trials at a university-affiliated hospital in metropolitan Detroit. Am J
Clin Oncol. 1999;22:42-46.
Green S, Benedetti J, Crowley J. Clinical Trials in Oncology, 2nd Ed.
Boca Raton, FL: CRC Press; 2003.
Newhouse JP, McClellan M. Econometrics in outcomes research: the
use of instrumental variables. Annu Rev Public Health. 1998;19:17-34.
Langford AT, Resnicow K, Dimond EP, et al. Racial/ethnic differences in
clinical trial enrollment, refusal rates, ineligibility, and reasons for
decline among patients at sites in the National Cancer Institutes
Community Cancer Centers Program. Cancer. 2014;120:877-884.
Unger JM, Barlow WE, Martin DP, et al. Comparison of survival
outcomes among cancer patients treated in and out of clinical trials.
J Natl Cancer Inst. 2014;106:dju002.
Clisant S, Clermont A, Adenis A, et al. Inflation in the number of
eligibility criteria for industry-sponsored phase II cancer clinical
trial: illustration over a 20-year period. Contemp Clin Trials. 2012;
33:459.
Somkin CP, Altschuler A, Ackerson L, et al. Organizational barriers to
physician participation in cancer clinical trials. Am J Manag Care. 2005;
11:413-421.
Benson AB III, Pregler JP, Bean JA, et al. Oncologists reluctance to
accrue patients onto clinical trials: an Illinois Cancer Center study.
J Clin Oncol. 1991;9:2067-2075.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
Melisko ME, Hassin F, Metzroth L, et al. Patient and physician attitudes

toward breast cancer clinical trials: developing interventions based on
understanding barriers. Clin Breast Cancer. 2005;6:45-54.
Prescott RJ, Counsell CE, Gillespie WJ, et al. Factors that limit the
quality, number and progress of randomised controlled trials. Health
Technol Assess. 1999;3:1-143.
Ross S, Grant A, Counsell C, et al. Barriers to participation in randomised controlled trials: a systematic review. J Clin Epidemiol. 1999;
52:1143-1156.
Taylor KM, Margolese RG, Soskolne CL. Physicians reasons for not
entering eligible patients in a randomized clinical trial of surgery for
breast cancer. N Engl J Med. 1984;310:1363-1367.
Institute of Medicine. Transforming Clinical Research in the United
States: Challenges and Opportunities: Workshop Summary. Washington, DC: The National Academies Press; 2010.
Siminoff LA, Zhang A, Colabianchi N, et al. Factors that predict the
referral of breast cancer patients onto clinical trials by their surgeons
and medical oncologists. J Clin Oncol. 2000;18:1203-1211.
Ellis PM, Butow PN, Tattersall MH, et al. Randomized clinical trials in
oncology: understanding and attitudes predict willingness to participate. J Clin Oncol. 2001;19:3554-3561.
Cassileth BR, Lusk EJ, Miller DS, et al. Attitudes toward clinical trials
among patients and the public. JAMA. 1982;248:968-970.
Daugherty C, Ratain MJ, Grochowski E, et al. Perceptions of cancer
patients and their physicians involved in phase I trials. J Clin Oncol.
1995;13:1062-1072.
Ford JG, Howerton MW, Lai GY, et al. Barriers to recruiting underrepresented populations to cancer clinical trials: a systematic
review. Cancer. 2008;112:228-242.
Jones JH. Bad Blood: The Tuskegee Syphilis Experiment. New York: Free
Press; 1993.
Lederer S. Subjected to Science: Human Experimentation in America
after the Second World War. Baltimore, MD: Johns Hopkins University
Press; 1995.
Coyne CA, Xu R, Raich P, et al; Eastern Cooperative Oncology Group.
Randomized, controlled trial of an easy-to-read informed consent
statement for clinical trial participation: a study of the Eastern Cooperative Oncology Group. J Clin Oncol. 2003;21:836-842.
Jenkins V, Fallowfield L. Reasons for accepting or declining to participate in randomized clinical trials for cancer therapy. Br J Cancer.
2000;82:1783-1788.
Llewellyn-Thomas HA, McGreal MJ, Thiel EC, et al. Patients willingness to enter clinical trials: measuring the association with perceived
benefit and preference for decision participation. Soc Sci Med. 1991;
32:35-42.
McQuellon RP, Muss HB, Hoffman SL, et al. Patient preferences for
treatment of metastatic breast cancer: a study of women with earlystage breast cancer. J Clin Oncol. 1995;13:858-868.
Jenkins VA, Fallowfield LJ, Souhami A, et al. How do doctors explain
randomised clinical trials to their patients? Eur J Cancer. 1999;35:
1187-1193.
Hietanen P, Aro AR, Holli K, et al. Information and communication in
the context of a clinical trial. Eur J Cancer. 2000;36:2096-2104.
Hutchins LF, Unger JM, Crowley JJ, et al. Underrepresentation of
patients 65 years of age or older in cancer-treatment trials. N Engl J
Med. 1999;341:2061-2067.
Lewis JH, Kilgore ML, Goldman DP, et al. Participation of patients 65
years of age or older in cancer clinical trials. J Clin Oncol. 2003;21:
1383-1389.
Talarico L, Chen G, Pazdur R. Enrollment of elderly patients in clinical
trials for cancer drug registration: a 7-year experience by the US Food
and Drug Administration. J Clin Oncol. 2004;22:4626-4631.
195
UNGER ET AL
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
196
Unger JM, Coltman CA Jr, Crowley JJ, et al. Impact of the year 2000
Medicare policy change on older patient enrollment to cancer clinical
trials. J Clin Oncol. 2006;24:141-144.
Kemeny MM, Peterson BL, Kornblith AB, et al. Barriers to clinical trial
participation by older women with breast cancer. J Clin Oncol. 2003;
21:2268-2275.
Kornblith AB, Kemeny M, Peterson BL, et al; Cancer and Leukemia
Group B. Survey of oncologists perceptions of barriers to accrual of
older patients with breast carcinoma to clinical trials. Cancer. 2002;95:
989-996.
Townsley CA, Selby R, Siu LL. Systematic review of barriers to the
recruitment of older patients with cancer onto clinical trials. J Clin
Oncol. 2005;23:3112-3124.
Havlik RJ, Yancik R, Long S, et al. The National Institute on Aging and
the National Cancer Institute SEER collaborative study on comorbidity
and early diagnosis of cancer in the elderly. Cancer. 1994;74:
2101-2106.
Yancik R. Cancer burden in the aged: an epidemiologic and demographic overview. Cancer. 1997;80:1273-1283.
Centers for Medicare & Medicaid Services. Medicare Coverage. Clinical
Trials: Final National Coverage Decision. https://www.cms.gov/Medicare/
Coverage/ClinicalTrialPolicies/downloads/finalnationalcoverage.
Stewart JH, Bertoni AG, Staten JL, et al. Participation in surgical oncology clinical trials: gender-, race/ethnicity-, and age-based disparities. Ann Surg Oncol. 2007;14:3328-3334.
Gross CP, Filardo G, Mayne ST, et al. The impact of socioeconomic
status and race on trial participation for older women with breast
cancer. Cancer. 2005;103:483-491.
Cook ED, Moody-Thomas S, Anderson KB, et al. Minority recruitment
to the Selenium and Vitamin E Cancer Prevention Trial (SELECT). Clin
Trials. 2005;2:436-442.
Cook ED, Arnold KB, Hermos JA, et al. Impact of supplemental site
grants to increase African American accrual for the Selenium and
Vitamin E Cancer Prevention Trial. Clin Trials. 2010;7:90-99.
Fouad MN, Corbie-Smith G, Curb D, et al. Special populations recruitment for the Womens Health Initiative: successes and limitations. Control Clin Trials. 2004;25:335-352.
Moinpour CM, Atkinson JO, Thomas SM, et al. Minority recruitment in
the prostate cancer prevention trial. Ann Epidemiol. 2000;10:S85-S91.
Unger JM, Gralow JR, Albain KS, et al. Patient income level and cancer
clinical trial participation: a prospective survey study. JAMA Oncol.
2016;2:137-139.
Goldman DP, Berry SH, McCabe MS, et al. Incremental treatment costs
in national cancer institute-sponsored clinical trials. JAMA. 2003;289:
2970-2977.
Centers for Disease Control and Prevention National Center for Health
Statistics. http://www.cdc.gov/nchs. Accessed September 13, 2010.
Bleyer WA. Potential favorable impact of the affordable care act of
2010 on cancer in young adults in the United States. Cancer J. 2010;16:
563-571.
Bleyer A, Barr R, Hayes-Lattin B, et al; Biology and Clinical Trials
Subgroups of the US National Cancer Institute Progress Review Group
in Adolescent and Young Adult Oncology. The distinctive biology of
cancer in adolescents and young adults. Nat Rev Cancer. 2008;8:
288-298.
Bleyer WA, Barr R, Keohan ML. Adolescents and young adults with
cancer: what will it take to improve outcome. Am Soc Clin Oncol Educ
Book. 2001;21:125-140.
Martin S, Ulrich C, Munsell M, et al. Delays in cancer diagnosis in
underinsured young adults and older adolescents. Oncologist. 2007;
12:816-824.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
U.S. Census Bureau. National Intercensal Estimates. (2000-2010).

http://www.census.gov/popest/data/intercensal/national/nat2010.
html. Accessed January 28, 2016.
Joinpoint Regression Program. Version 4.1.1. August 2014; Statistical
Research and Applications Branch, National Cancer Institute. http://
surveillance.cancer.gov/joinpoint. Accessed January 15, 2016.
Surveillance Epidemiology, and End Results (SEER) Program (www.
seer.cancer.gov) SEER*Stat Database: Incidence - SEER 9 Regs Research Data, Nov 2014 Sub (1973-2012) ,Katrina/Rita Population
Adjustment. - Linked To County Attributes - Total U.S., 1969-2013
Counties, National Cancer Institute, DCCPS, Surveillance Research
Program, Surveillance Systems Branch, released April 2015, based on
the November 2014 submission.
seer.cancer.gov) SEER*Stat Database: Incidence - SEER 13 Regs Research Data, Nov 2014 Sub (1992-2012) ,Katrina/Rita Population
Adjustment. - Linked To County Attributes - Total U.S., 1969-2013
Counties, National Cancer Institute, DCCPS, Surveillance Research
Program, Surveillance Systems Branch, released April 2015, based on
the November 2014 submission.
seer.cancer.gov) SEER*Stat Database: Incidence - SEER 18 Regs Research Data + Hurricane Katrina Impacted Louisiana Cases, Nov 2014
Sub (1973-2012 varying) - Linked To County Attributes - Total U.S.,
1969-2013 Counties, National Cancer Institute, DCCPS, Surveillance
Research Program, Surveillance Systems Branch, released April 2015,
based on the November 2014 submission.
seer.cancer.gov) SEER*Stat Database: Mortality - All COD, Aggregated
With State, Total U.S. (1969-2012) ,Katrina/Rita Population
Adjustment., National Cancer Institute, DCCPS, Surveillance Research Program, Surveillance Systems Branch, released April 2015.
Underlying mortality data provided by NCHS (www.cdc.gov/nchs).
Breitenbach K, Stock W. Intergroup Trial C10403: a pediatric treatment approach to improve outcomes in adolescents and young adults
with acute lymphoblastic leukemia. J Adolesc Young Adult Oncol.
2011;1:107-108.
NCT02043587. Chemotherapy With Liposomal Cytarabine CNS Prophylaxis for Adult Acute Lymphoblastic Leukemia & Lymphoblastic Lymphoma. https://clinicaltrials.gov/ct2/show/NCT02043587?
term=acute+lymphoblastic+leukemia+wieduwilt&rank=1. Accessed
January 28, 2016.
DeAngelo DJ, Stevenson KE, Dahlberg SE, et al. Long-term outcome
of a pediatric-inspired regimen used for adults aged 18-50 years with
newly diagnosed acute lymphoblastic leukemia. Leukemia. 2015;29:
526-534.
Bleyer A. How NCCN guidelines can help young adults and older
adolescents with cancer and the professionals who care for them.
Dombret H, Cluzeau T, Huguet F, et al. Pediatric-like therapy for adults
with ALL. Curr Hematol Malig Rep. 2014;9:158-164.
Hallboo k H, Gustafsson G, Smedmyr B, et al; Swedish Adult Acute
Lymphocytic Leukemia Group; Swedish Childhood Leukemia Group.
Treatment outcome in young adults and children .10 years of age
with acute lymphoblastic leukemia in Sweden: a comparison
between a pediatric protocol and an adult protocol. Cancer. 2006;107:
1551-1561.
Hayakawa F, Sakura T, Yujiri T, et al; Japan Adult Leukemia Study Group
(JALSG). Markedly improved outcomes and acceptable toxicity in
adolescents and young adults with acute lymphoblastic leukemia
following treatment with a pediatric protocol: a phase II study by the
Japan Adult Leukemia Study Group. Blood Cancer J. 2014;4:e252.
86.
Hocking J, Schwarer AP, Gasiorowski R, et al. Excellent outcomes for

adolescents and adults with acute lymphoblastic leukemia and
lymphoma without allogeneic stem cell transplant: the FRALLE-93
pediatric protocol. Leuk Lymphoma. 2014;55:2801-2807.
87. Ram R, Wolach O, Vidal L, et al. Adolescents and young adults with
acute lymphoblastic leukemia have a better outcome when treated
with pediatric-inspired regimens: systematic review and metaanalysis. Am J Hematol. 2012;87:472-478.
88. Ribera JM, Oriol A, Sanz MA, et al. Comparison of the results of the
treatment of adolescents and young adults with standard-risk acute
~ol de Tratamiento en
lymphoblastic leukemia with the Programa Espan
Hematologa pediatric-based protocol ALL-96. J Clin Oncol. 2008;26:
1843-1849.
89. Usvasalo A, Raty R, Knuutila S, et al. Acute lymphoblastic leukemia in
adolescents and young adults in Finland. Haematologica. 2008;93:
1161-1168.
90. Stock W, La M, Sanford B, et al; Childrens Cancer Group; Cancer and
Leukemia Group B studies. What determines the outcomes for adolescents and young adults with acute lymphoblastic leukemia
treated on cooperative group protocols? A comparison of Childrens
Cancer Group and Cancer and Leukemia Group B studies. Blood. 2008;
112:1646-1654.
91. Douer D. Is asparaginase a critical component in the treatment of
acute lymphoblastic leukemia? Best Pract Res Clin Haematol. 2008;
214:647-658.
92. Collins CL, Malvar J, Hamilton AS, et al. Case-linked analysis of clinical
trial enrollment among adolescents and young adults at a National
Cancer Institute-Designated comprehensive cancer center. Cancer.
2015;121:4398-4406.
93. Sender L, Zabokrtsky KB. Adolescent and young adult patients with
cancer: a milieu of unique features. Nat Rev Clin Oncol. 2015;12:
465-480.
94. Freyer DR, Felgenhauer J, Perentesis J; COG Adolescent and Young
Adult Oncology Discipline Committee. Childrens Oncology Groups
2013 blueprint for research: adolescent and young adult oncology.
Pediatr Blood Cancer. 2013;60:1055-1058.
95. Weiss AR, Nichols CR, Freyer DR. Enhancing adolescent and young
adult oncology research within the National Clinical Trials Network:
Rationale, progress, and emerging strategies. Semin Oncol. 2015;42:
740-747.
96. Stark D, Bowen D, Dunwoodie E, et al. Survival patterns in teenagers
and young adults with cancer in the United Kingdom: Comparisons
with younger and older age groups. Eur J Cancer. 2015;51:2643-2654.
97. Denicoff AM, McCaskill-Stevens W, Grubbs SS, et al. The National
Cancer Institute-American Society of Clinical Oncology Cancer Trial
Accrual Symposium: summary and recommendations. J Oncol Pract.
2013;9:267-276.
98. AccrualNet. Strategies, tools, and resources to support accrual to
clinical trials. https://accrualnet.cancer.gov/. Accessed January 28,
2016.
99. Barrios CH, Werutsky G, Martinez-Mesa J. The global conduct of
cancer clinical trials: challenges and opportunities. Am Soc Clin Oncol
Educ Book. 2015:e132-e139.
100. Tierney G, Sieher S. Key strategies for effective globalization of clinical trials. Pharmaceutical Outsourcing. http://www.pharmoutsourcing.
com/Featured-Articles/150621-Key-Strategies-for-Effective-Globalizationof-Clinical-Trials/. Accessed September 25, 2013.
101. Butts CA, Socinski MA, Mitchell P, et al. START: a phase III study of
L-BLP25 cancer immunotherapy for unresectable stage III non-small
cell lung cancer. J Clin Oncol. 2013;31:15s (suppl; abstr 7500).
102. Gaumond B, Ferrande L, Houde J, et al. Global patient recruitment
continuum for a phase III non-small cell lung cancer clinical trial. Paper
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
120.
presented at: National Cancer Institute American Society of Clinical

Oncology Cancer Trial Accrual Symposium: Science and Solutions;
April 2010; Bethesda, MD.
Moore HC, Unger JM, Phillips KA, et al; POEMS/S0230 Investigators.
Goserelin for ovarian protection during breast-cancer adjuvant chemotherapy. N Engl J Med. 2015;372:923-932.
Frandsen M, Walters J, Ferguson SG. Exploring the viability of using
online social media advertising as a recruitment method for smoking
cessation clinical trials. Nicotine Tob Res. 2014;16:247-251.
Lamberti MJ, Kush R, Kubick W, et al. An examination of eClinical
Technology usage and CDISC standard adoption. Ther Innov Regul Sci.
2015:2168479015586003.
Top considerations for using social media in researchit all starts
with a plan. Seattle, WA; Quorum Review IRB. https://www.ucdmc.
ucdavis.edu/biorepositories/pdfs/misc-biobank/Top-considerationsfor-using-social-media-in-research.pdf. Accessed February 15, 2016.
U.S. Food and Drug Administration. Recruiting Study Subjects Information Sheet. www.fda.gov/regulatoryinformation/guidances/
ucm126428.htm. Accessed February 15, 2016.
Office of Inspector General. Clinical Trial Web Sites. A promising tool
to foster informed consent. http://oig.hhs.gov/oei/reports/oei-0197-00198.pdf. Accessed February 15, 2016.
InVentiv Health Clinical Division. E-Recruiting: using digital platforms, social, media, and mobile technologies to improve clinical trial
enrollment. http://www.inventivhealth.com/docs/e-Recruiting_
Using_Digital_Platforms_Social_Media_and_Mobile_Technologies_
to_Improve_Clinical_Trial_Enrollment.pdf. Accessed February 15,
2016.
Gossen R. Great Facebook pages for research sites. Engage and recruit
clinical research participants. Paper presented at: 23rd Society of
Clinical Research Associates Annual Meeting; September 2014;
Orlando, FL.
American Cancer Society. Cancer Treatment and Survivorship Facts &
Figures 2014-2015. Atlanta: American Cancer Society; 2014.
Denson AC, Mahipal A. Participation of the elderly population in
clinical trials: barriers and solutions. Cancer Contr. 2014;21:209-214.
Chen H, Cantor A, Meyer J, et al. Can older cancer patients tolerate
chemotherapy? A prospective pilot study. Cancer. 2003;97:
1107-1114.
Mariano C, Francl M, Pope J, et al. Comparison of toxicity experienced
by older versus younger patients enrolled in breast cancer clinical
trials. Clin Breast Cancer. 2015;15:73-79.
Sud S, Lai P, Zhang T, et al. Chemotherapy in the oldest old: the
feasibility of delivering cytotoxic therapy to patients 80 years old and
older. J Geriatr Oncol. 2015;6:395-400.
Extermann M, Aapro M, Bernabei R, et al; Task Force on CGA of the
International Society of Geriatric Oncology. Use of comprehensive
geriatric assessment in older cancer patients: recommendations from
the task force on CGA of the International Society of Geriatric Oncology (SIOG). Crit Rev Oncol Hematol. 2005;55:241-252.
Extermann M, Hurria A. Comprehensive geriatric assessment for older
patients with cancer. J Clin Oncol. 2007;25:1824-1831.
Hamaker ME, Jonker JM, de Rooij SE, et al. Frailty screening methods
for predicting outcome of a comprehensive geriatric assessment in
elderly patients with cancer: a systematic review. Lancet Oncol. 2012;
13:e437-e444.
Kellen E, Bulens P, Deckx L, et al. Identifying an accurate pre-screening
tool in geriatric oncology. Crit Rev Oncol Hematol. 2010;75:243-248.
Luciani A, Ascione G, Bertuzzi C, et al. Detecting disabilities in older
patients with cancer: comparison between comprehensive geriatric
assessment and vulnerable elders survey-13. J Clin Oncol. 2010;28:
2046-2050.
197
UNGER ET AL
121. Balducci L, Colloca G, Cesari M, et al. Assessment and treatment of

elderly patients with cancer. Surg Oncol. 2010;19:117-123.
122. Grady C. Payment of clinical research subjects. J Clin Invest. 2005;115:
1681-1687.
123. Bernstein M. Payment of research subjects involved in clinical trials is
unethical. J Neurooncol. 2003;63:223-224.
124. Grady C, Dickert N, Jawetz T, et al. An analysis of U.S. practices of
paying research participants. Contemp Clin Trials. 2005;26:365-375.
198
125. U.S. Department of Health and Human Services. Federal Policy for the
Protection of Human Subjects (Common Rule). http://www.hhs.gov/
ohrp/humansubjects/commonrule/. Accessed February 15, 2016.
126. Unger JM, Barlow WE, Ramsey SD, et al. The scientific impact of
positive and negative phase 3 cancer clinical trials. JAMA Oncol. Epub
2016 Mar 10.
127. Bleyer A. In and out, good and bad news, of generalizability of SWOG
treatment trial results. J Natl Cancer Inst. 2014;106:dju027.

OF CARE
Using Social Media, Wearables,

and Electronic Medical Records to
Improve Quality of Cancer Care
CHAIR
Michael J. Fisch, MD, MPH
Houston, TX
SPEAKERS
Arlene E. Chung, MD, MHA, MMCi
Durham, NC
Melissa K. Accordino, MD
New York-Presbyterian Hospital
New York, NY
FISCH, CHUNG, AND ACCORDINO
Using Technology to Improve Cancer Care: Social Media,

Wearables, and Electronic Health Records
Michael J. Fisch, MD, MPH, Arlene E. Chung, MD, MHA, MMCi, and Melissa K. Accordino, MD
OVERVIEW
Digital engagement has become pervasive in the delivery of cancer care. Internet- and cellular phonebased tools and
systems are allowing large groups of people to engage with each other and share information. Health systems and individual
health professionals are adapting to this revolution in consumer and patient behavior by developing ways to incorporate the
benefits of technology for the purpose of improving the quality of medical care. One example is the use of social media
platforms by oncologists to foster interaction with each other and to participate with the lay public in dialogue about science,
medicine, and cancer care. In addition, consumer devices and sensors (wearables) have provided a new, growing dimension
of digital engagement and another layer of patient-generated health data to foster better care and research. Finally,
electronic health records have become the new standard for oncology care delivery, bringing new opportunities to measure
quality in real time and follow practice patterns, as well as new challenges as providers and patients seek ways to integrate
this technology along with other forms of digital engagement to produce more satisfaction in the process of care along with
measurably better outcomes.
igh-quality medical care is expected to be efficacious,

safe, timely, patient centered, efficient, and equitable.1
None of these attributes of high-quality medical care are
possible in a vacuum. Physicians must be engaged with each
other to share ideas, collaborate, and establish best practice
norms and creative new solutions. Social media has rapidly
become one of the normative approaches to foster these
critical connections. Compared with traditional media, social
media is updated more frequently and it is more interactive,
while also being archivable and searchable.2 Three major
needs for connection that social media helps oncologists
meet are (1) finding and prioritizing medical literature and
medical meetings information; (2) finding the current location, interests, and qualifications of other physicians; and
(3) crowdsourcing or surveying each other about best
practices or approaches to clinical scenarios.
FINDING AND PRIORITIZING MEDICAL

LITERATURE
The medical literature is central to ongoing medical education. Early findings of new research as well as expert
summaries of existing literature are shared at medical
meetings, and this work is more formally disseminated in
medical journal articles. How oncologists and other health
professionals identify new information from meetings and
published literature and prioritize what to read continues

to evolve. Traditional techniques involved subscribing to
key journals, browsing other journals in shared collections
(libraries, offices), gathering in journal clubs, discussing
specific articles at morning reports or interdisciplinary
rounds, and attending local and national meetings where
new data are presented and reference literature is cited.
We occasionally visit online textbooks or information
resources and we also encounter lay press stories citing
medical articles. In todays world, as noted by Dr. Bryan
Vartebedian (@Doctor_V), weve reached a point where
social media is now part of the professional workflow.3
Oncology clinicians are finding both efficiency and enjoyment from using social media platforms such as Twitter
to keep up with the medical literature and share information from meetings. On Twitter, one can follow
many key journals (and can even follow collections of
journals on a list) and quickly scan through articles as they
come out and are tweeted by the journals. More importantly, when oncologists follow each other on Twitter
(and collections of colleagues on lists), there is a sort of
Bayesian quality to social media whereby the prior
probability of seeing an article or hearing about a new
finding at a meeting4 is enhanced by retweets that amplify
the most important articles.
From AIM Specialty Health, Chicago, IL; Outcomes Research Program, Lineberger Comprehensive Cancer Center, University of North Carolina School of Medicine, Chapel Hill, NC;
Department of Medicine, Herbert Irving Comprehensive Cancer Center, Columbia University Medical Center, New York, NY.
Corresponding author: Michael J. Fisch, MD, MPH, AIM Specialty Health, 8600 West Bryn Mawr Ave., Suite 800, Chicago, IL 60631; email: fischm@aimspecialtyhealth.com.
200
USE OF SOCIAL MEDIA IN ONCOLOGY PRACTICE
FINDING AND CONNECTING TO OTHER

PHYSICIANS
In addition to curating the medical literature, Twitter and
other social media platforms (such as Facebook) also provide a venue for connection. For example, if a physician posts
or tweets the link to an article, other colleagues can reply
with their own comments. Patient and layperson perspectives come forth frequently because others besides oncologists can see these tweets and engage in dialogue. This
element helps health professionals maintain a broader and
more patient-centered perspective on the literature. An example of an exchange between oncologists and a cancer
survivor is shown in Fig. 1.
Twitter does not lend itself to in-depth analysis or detailed
commentary, as the comments are limited to 140 characters.
Oncology professionals also use Twitter to connect to blogs
with extended commentary on the literature and other key
issues in oncology care. These blogs can be individually
managed (such as 33charts by @Doctor_V) or they can be
guest blogs distributed by a physicians institution or professional organization (such as the American Society of
Clinical Oncologys ASCO Connection) or through another
physicians popular blog (such as KevinMD). Another
professional-oriented site is ResearchGate. This site, which is
geared toward academic researchoriented professionals,
allows oncology professionals to connect directly with investigators to ask questions about their articles, to follow
their most recent publications, and to even request reprints
from individuals.
Doximity, a unique platform restricted to health professionals with a National Provider Identifier number, is a
more restricted setting for conversation and, this is more
comfortable for some clinicians. Articles can be found
KEY POINTS
Social media is increasingly used by oncology clinicians

to find and prioritize medical literature, connect with
other providers, and interact with the public about
cancer care.
Consumer devices and sensors (called wearables) are
growing in popularity and allow for patient monitoring
in their own environments outside of clinical
encounters.
Patient-generated health data from wearables provide
an opportunity to enhance comparative effectiveness
research in oncology.
Electronic health records are widespread in oncology
and are being used to measure adherence to quality
metrics and often to connect with patients through
secure portals.
Electronic health records have the potential to improve
the quality of cancer care by enhancing education and
awareness of evidence-based guidelines and by serving
as a resource for quality assessment and as a tool for
clinical and population-based research
directly through Doximity and commented on and/or

liked. The latter allows others to see what articles are
most endorsed. The most powerful feature of Doximity is
that it allows clinicians to find each other by searching the
Doximity search engine (even when the physician you are
searching for has not engaged with Doximity). Try it: search
for yourself on this site. You will find that Doximity identifies
not only your location but also your medical school, residency and fellowship (along with listing your colleagues who
trained with you), certifications, licenses, and insurance that
your practice accepts. Individuals can also list their publications and other aspects about themselves by expanding
the profile. Physicians could, for example, search for colleagues in their residency and find out where they now
practice and their current specialty. In addition, Medstro has
emerged as another popular social network for physicians.
Medstro features discussion groups with a tight connection
to the New England Journal of Medicine, as well as contests
and interactive challenges featuring engagement with academic subspecialty experts. Job searching is also integral to
both Doximity and Medstro.
CROWDSOURCING AND SURVEYING OTHER

PHYSICIANS
Social media provides a convenient mechanism for checking
in with other physicians, even when those physicians are
outside of the physicians local peer group or organization.
For example, if a clinician wonders Is there a valid instrument that can be used to measure fear of cancer recurrence?, the question could be crowdsourced on Twitter
by simply typing the question and pressing send. This
would be distributed to the clinicians followers. One could
also distribute the question outside of ones followers to
other groups of oncology professionals using hashtags. For
example, adding #pallonc at the end of the question would
distribute the tweet not only to an individuals followers but
also to anybody else worldwide who is tuned in to this
topic by searching or streaming tweets with this hashtag.
Dr. Matthew Katz (@subatomicdoc), a radiation oncologist,
has led the oncology community by developing Cancer Tag
Ontology and Oncology Tag Ontology (via symplur.com) to
codify this way of indexing tweets.5
Another sort of question might be a clinical case scenario,
such as, Is there a role for post-treatment PET/CT following
chemoRT for stage III NSCLC? This type of question could be
directed to Twitter also, and those replying are likely to give a
short answer and include a link to a relevant article or web
resource. However, there is a forum for oncologists available
for addressing clinical scenarios via theMednet.org. This new
forum is facilitated by Dr. Nadine Housri (@nadinehousri),
who is also a radiation oncologist. It is free and, like Doximity,
restricted to health professionals. It is not a place for discussion of specific patients, but it is useful for general case
scenarios, and the forum is well facilitated and allows for
social interactions between various disciplines and between community and academic practitioners. Another
201
FIGURE 1. Example of an Oncology-Related Exchange on Twitter
This example of an exchange on Twitter begins with a tweet by an oncology fellow in Boston sharing an ASCO Post article describing a peer-reviewed manuscript about the
association between symptoms and employment outcomes in cancer survivors. Hashtags and specific Twitter handles are used to help disseminate the tweet. A picture of the data is
also included by adding a tiny URL link to the picture file. A cancer survivor replies to the tweet, raising the issue of fear of recurrence. Another oncologist replies to the survivor,
referencing a specific article about key associations related to fear of recurrence.
kind of question that lends itself to social interaction

between physicians is How many oncologists are using the XYZ regimen in first-line treatment of multiple
myeloma? Dr. Simrit Parmar (@spa718), an academic
hematologist/oncologist, has developed a platform called
MDRing that works via the web or smartphone (as an
iPhone application) to bring cancer professionals together to answer survey-type questions like this. MDRing
has been used by clinical cooperative groups and other
202
organizations and by individual oncologists as an educational tool and a practical tool for clinical research
planning.
CONNECTING WITH PATIENTS

Physicians do not need social media to connect with their
own patients. Connecting with our patients occurs through
direct interactions and also through secure interactions
available through electronic health record (EHR) systems or
other platforms approved by our clinical organizations. It is

well recognized that use of social media can bring risks to
individual health professionals and their organizations.
Physicians must be aware of unintentional online disclosure
of patient information, because even subtle oversteps of
boundaries raise concerns for professional reputation and
responsibility.6 Nevertheless, there is nothing that prevents
our patients from engaging us through publicly available
social media sites such as Facebook and Twitter. Fear of this
sort of contact is one reason that some clinicians are reluctant to use social media. Dr. Don Dizon and a working
group of the ASCO Integrated Media and Technology
Committee noted that oncologists must not confuse the role
of provider of information and provider of care, must understand regulations related to state licensure and medical
records, must recognize the importance of transparency and
disclosure of potential conflicts of interest, and must avoid
disclosures of preliminary results or nonpublic information.7
Guidance is also commonly included in the social media
policies of health care organizations and clinical research
organizations.
With boundaries understood, it is appropriate for clinicians to interact with patients and the lay public in the
broader sense. That is, individual cancer survivors and
their family members can be followed on social media,
and they are a key component of the healthy exchange of
ideas and perspectives about illness, science, and health
care. Oncologists can use this opportunity to help educate
the public, to advocate for science and for clinical trial
participation, and to promote good health behaviors and
principles of sound medical care. Social media can also be
used by oncologists to demonstrate their humanity,
compassion, and deep dedication to the service of others.
Some oncologists use Facebook for this purpose. For
example, Dr. Anas Younes (@DrAnasYounes), an academic
lymphoma specialist, has shared information about
lymphoma, cancer, and other health care topics using this
platform and built a following represented by greater than
1,500 likes on his Facebook page. Moreover, organizations such as ASCO and specific health care organizations widely use public Facebook pages. The Mayo Clinic is
one example of a health care organization that has fully
embraced social media, adopting a patient-centered approach of reaching out to innovate in this evolving space
where patients are seeking information. The Mayo Clinic
Center for Social Media (#mccsm) has greater than 1.25
million followers on Twitter and greater than 750,000
likes on its public Facebook page. Of note, Facebook use
is most effective when the user is able to provide regular
input, whereas Twitter is easier to use sporadically. Other
social media platforms such as Instagram, Pinterest, and
even Snapchat are used less often for sharing information
about cancer, but there is certainly opportunity to harness these popular sites for the same purposes. As noted by
Dr. Michael Thompson (@mtmdphd), social media platforms
will always evolve, but it is the content, curation, and
connectivity that matter.8
WEARABLE TECHNOLOGY
In addition to connecting through words and media, another
dimension of digital engagement comes from the growing
availability and affordability of various consumer devices
and sensors that can be worn or placed on the body. Known
as wearables, these devices provide an unparalleled opportunity to monitor a persons health and experience in his
or her real-world environment. The current model of health
care for treatment or prevention is still mainly focused on
face-to-face encounters with providers in the clinical setting;
however, as we move away from this traditional model of
care, wearables could provide additional insights about the
patient experience and enhance the quality of cancer care.
Wearable technology refers collectively to electronics that
can be worn on or close to the body. Wearables encompass a
growing number of different types of items such as wristworn sensors that detect and measure physical activity and
intensity (such as a Fitbit device or Apple Watch), rings and
glasses with sensors, earbuds, and textiles or fabrics with
embedded sensors to monitor heart rate and temperature,
to more invasive versions such as an implantable microchip
or smart tattoo. It is estimated that there will be over 111
million worldwide shipments of wearables in 2016, with
most being in the form of watches and wristbands (92%).9
Users are almost equally split between males and females
and are mostly younger than age 55,9 although adoption
among older adults is expected to be the sector of largest
growth. It is also anticipated that the wearables market will
grow over 64% to $25 billion in 2019, with over 245 million
devices sold.10
HEALTH MEASUREMENT OPPORTUNITIES

THROUGH GROWING ADOPTION OF
WEARABLES
Patient-generated health data (PGHD) from wearables, such
as biometric data like heart rate and blood pressure readings
or physical activity data, are some of the more commonly
captured data from wearables. PGHD are defined as healthrelated dataincluding health history, symptoms, biometric
data, etc.created, recorded, gathered, or inferred by or
from patients/caregivers to help address a health concern.11 PGHD from wearables could provide a more holistic
view of the health and status of a patient.
As consumer adoption of wearables increases and there
are advances in the underlying technologies, the opportunity to collect data on a population level is more feasible and
could provide important advances in oncology comparative
effectiveness research. Recently, there have also been
several key funding initiatives that aim to incorporate
wearable device data. For example, the National Institutes of
Health Precision Medicine Initiative Cohort Program will
capture physiologic and environmental exposures. Moreover, several of the Patient-Centered Outcomes Research
Institutefunded Patient-Powered Research Networks collect PGHD from wearables in an effort to collect data
in a nationally representative sample at lower costs than
203
traditional methods of data collection. Dietary data and

physical activity data are examples of data that are traditionally difficult to capture on a continuous basis in research
studies or for clinical care. Capturing data such as these
could be used to better understand the influence of these
factors on cancer risk and progression among populations of
patients.
Physical activity, in particular, is an important factor in
cancer survivorship, both for prevention and quality of life
and symptom management, and may also improve a
patients ability to tolerate cancer treatment by improving
fatigue, quality of life, and physical functioning.12-14
Pedometry using commercially available fitness trackers
such as Fitbits has shown that large decreases in daily
steps may correlate with more severe symptoms, impaired physical health, and restricted daily activities.15
Thus, tracking physical activity using a wearable device
could potentially be used for monitoring symptoms and
severity in those who may not be able to self-report
symptoms because of various barriers such as literacy
or severity of illness. The use of a wearable physical activity tracker could also potentially monitor levels of
activity and provide a target for behavioral interventions
during treatment and survivorship using just-in-time
adaptive methodologies that would not have been possible without the real-time continuous activity tracking
afforded by wearables. One might imagine that a patient
with breast cancer undergoing treatment could wear a
physical activity tracker and that the PGHD captured
showed lower than usual levels of activity for this patient.
Decreased physical activity could trigger a cascade of
activities such as offering a patient-reported outcome
questionnaire to determine whether the patient has
worsening symptoms. If there was no detection of
worsening of symptoms, the patient could be encouraged
to be more active, which could be personalized based on
prior activity levels and other factors such as whether the
patient has anemia. The patient could receive a personalized and tailored text message to her wrist-worn
wearable about the benefits of activity during treatment and could be given specific goals for increasing
activity safely. This PGHD could also be shared with her
oncologist and primary care physicians, such that the care
team was aware of the patients health status and experiences, which could inform treatment decisions or
changes in management. Wearables could also be integrated with other smart home monitors and other
sensing systems, such as smart bottles that could detect
when a user has not opened a pill bottle and send alerts to
patients or caregivers, or these devices may be able to
detect a fall or impaired gait such that assistance could
be provided to prevent complications. Another scenario
in which wearables also have promise is for remote
monitoring for cardiovascular complications such as
arrhythmias during treatment with cardiotoxic chemotherapeutics and in survivorship care. As digital patient
engagement advances and matures, further applications
204
for wearables in cancer care are likely to emerge where

the wearable provides utility and value.
CHALLENGES IN THE USE OF WEARABLES

As is the case with other emerging technologies, the opportunity to monitor patients more continuously with
wearables comes with its own unique challenges. Because
the use of wearables in health care is still nascent, it is not
clear what types of data from wearables will be the most
effective to improve symptoms, the patient experience, and
short- and long-term health outcomes. Best practices for
integrating this PGHD into clinical workflows to ensure
appropriate review and action are not yet available, but as
practices and health care systems begin to have more experience with PGHD, there will be better guidance on how to
engage patients in a learning health ecosystem and what will
be acceptable within provider workflows. In addition, in the
next 2 years, the Office of the National Coordinator for
Health Information Technology will be developing a policy
framework for identifying best practices, gaps, and opportunities for the use of PGHD, which will provide further
guidance.
Among the largest vendors of EHRs, most allow patients to
submit wearables data in various formats for provider review. An example of a wearables workflow within clinical
care might include the following: (1) a patient receives a
message from his or her provider in the patients portal
account, which provides instructions on how to connect the
wearable device account to allow for transmission of these
data to the health care system via the patient portal; (2) the
patient uses the wearable device for a certain interval to
capture data; and (3) the PGHD are transmitted either in real
time or asynchronously and made available for providers to
review within EHR workflows. Currently, data visualization
and presentation is generally constrained by the EHR vendors specifications and is presented mostly in tabular format. These data are usually presented at the patient level
and not at a population level. Wearables could generate
large volumes of asynchronous or continuous streams of
data, which would be very challenging to manage and review
without intelligent filtering and summarization for both
patients and providers.16 Therefore, informatics solutions to
assist with transforming these data into actionable information will be imperative.
Current adoption of wearables is low and rarely sustained
beyond several months. However, this is felt to be attributable to cost, the lack of integration with other devices, and
the absence of interactive and dynamic feedback. Security
and privacy concerns are also a barrier to adoption. Overall,
the tangible benefits of wearables currently lag behind rapid
growth of the device market, so it is expected that adoption
will ramp up as software platforms provide better personalization and insights, which could also better sustain patient
engagement. In part, the lack of accurate, validated biometric data from the majority of consumer wearables is
also a barrier to adoption within health care. In the present
state, it is difficult to decipher which wearable devices and
what types of PGHD will provide the most value within

cancer care. Current studies in progress piloting the use of
PGHD will be important in determining what may be the
most effective types of wearables or PGHD that could improve the patient experience during cancer treatment and in
survivorship.
FUTURE DIRECTIONS WITH WEARABLES

As the wearables market matures, we will likely see a
movement away from primarily wrist-worn devices to bodyconforming sensors embedded in flexible patches or stickers
and toward medical-grade sensors that provide accurate and
valid data. In addition, we should see improvements in
physical comfort, devices being more unobtrusive, battery
life, and interoperability with other devices and software
applications. We will also see more of a connected ecosystem of appliances, home-based sensors, and wearables
such that richer information about an individuals activities
of daily living alongside biometric data could be captured.
These data could provide insights about various facets of a
patients health and environment, which, if collected on a
population level, could provide important insights about the
influence of behavioral, lifestyle, and environmental factors
on cancer risk and progression.
ELECTRONIC HEALTH RECORDS IN CANCER CARE

DELIVERY
Social media and wearables reflect interactions that occur
largely outside of the health care environment. In the very
fabric of health care delivery is the need to document health
status and elements of care. In recent years, the practice of
medical oncology has become increasingly complicated. As
reimbursement models transition from fee-for-service to
newer shared-risk models, a higher emphasis is placed on
the delivery of high-quality and high-value care. EHR use
reflects the new digital landscape that extends far beyond
mere documentation, as the EHR brings features that have
been shown to improve quality and efficiency of care.17 As
such, practices have been incentivized to use an EHR because it is seen as a generalizable intervention to expand the
use of technology to improve the quality of our care.18
Within oncology, EHR use can improve quality in real time
and can be used as a tool to measure adherence to quality
metrics within individual practices and help identify care
disparities.
An EHR, as defined by the Institute of Medicine, is an
electronic system with the ability to collect and store patient
data, supply information, permit providers to enter orders
(known as computerized provider-order entry), and provide
decision support. In addition, EHRs may be able to provide
electronic communication, administrative support, patient
support, and population health management reporting.19
Benefits of EHRs include improved quality of care by increased provider adherence to evidence guidelines, reduction in medical errors by improvement in accuracy and
clarity of data, and the potential for cost reduction.20-24 EHRs
also have the ability to share secured information with

patients directly through portals, which may improve patient satisfaction through increased data sharing.25,26 Limitations of EHRs include negative effects in provider-patient
communication, such as reduced eye contact, higher operating costs, unreliability of electronic systems, and workflow limitations.27,28
There are notable racial and age-related disparities in
the delivery of cancer care.29-31 An evaluation of administrative data showed a 66% increase in adoption of
basic EHRs in federally qualified health centers from 2010
to 2012 (from 29.8% to 49.6%). However, disparities
existed; centers with higher proportions of low-income
patients had lower EHR adoption rates compared with
centers with smaller proportions of low-income patients
(45.1% vs. 55.5%, p , .05).32 These results were similar
to another population-based study that showed EHR
adoption to be lowest in areas with high concentrations of
minority and low-income populations.33 In regard to EHR
portal access, disparities also exist. A review of four
university nephrology practices showed that older age
(odds ratio [OR] 0.29), black race (OR 0.50), and having
Medicaid insurance compared with private insurance
(OR 0.53) were associated with decreased portal enrollment.34 Another cross-sectional evaluation of portal utilization revealed that older and nonwhite patients were
less likely to enroll; after enrolling, younger patients and
men were less likely to solicit advice or medication refills
compared with older patients and women.35 Potential
barriers to portal enrollment include lack of resources,
patient comfort and proficiency, lack of provider enthusiasm and communication, health literacy, and distrust in
the health system.32,34,36-43 Interventions to reduce disparities in enrollment must address both access and attitudinal barriers.
THE EFFECT OF ELECTRONIC HEALTH RECORDS

ON QUALITY OF CARE
EHRs have the ability to improve the quality of cancer care
in a multitude of ways, including educational alerts to improve adherence to evidence-based guidelines, improved
patient communication through patient portals, as a resource for individual and practice-based quality selfassessment, and as a tool for cancer care delivery research.
Educational electronic alerts have been successful
at improving quality measures in a number of clinical
settings.44-48 Kucher et al found that an EHR alert that
identified patients at risk for deep-vein thrombosis (DVT)
increased use of appropriate DVT prophylaxis and led to
reduced rates of venous thromboembolism.46 Another study
alerted providers to a patients risk of contrast-induced
acute kidney injury (AKI) when imaging studies with contrast were ordered. After the alert was implemented, rates
of prophylaxis increased and incidence of contrast-induced
AKI decreased.47 Selvan et al investigated an electronic alert
triggered by erythropoietin-stimulating agent (ESA) order
entry if a patients hemoglobin was 12 g/dL or above
205
(considered inappropriate use) at a single institution.

Postalert rates of inappropriate ordering of ESAs decreased
by 80%.48 An EHR alert triggered by a patient-reported
pain score of 4 or greater led to increased documentation
of pain care plans for patients with cancer with moderate
to severe pain by 20%.49 Another EHR intervention, which
has the ability to provide real-time clinical decision support through a combination of alerts and links to clinical
guidelines, is ASCO CancerLinQ, a rapid learning system.50-52
A rapid learning system refers to a technology platform
that allows data to be collected from different EHR systems
to be collected, aggregated, analyzed, and translated into
information aimed to improve treatment and drive scientific
learning and discovery from every patient encounter.50,53,54
As health care delivery continues to shift toward systembased practice, knowledge of detailed evidence-based
guidelines does not solely fall on individual providers but
rather the system as a whole, and EHRs have the capability of
providing this type of decisional support to improve quality
of care.
Another important quality measure in oncology is ensuring
that patients receive appropriate treatment of their individual cancer.50 Because more cancer therapies are now
delivered orally, a focus on medication adherence is becoming increasingly important. Medication nonadherence is
common among patients with chronic conditions, in whom
adherence rates are estimated to be around 50%.55,56 In
women with early-stage hormone receptorpositive breast
cancer, noninitiation, early discontinuation, and nonadherence to hormonal therapy are common and are associated with poorer survival.57-63 Approximately one-third
of patients with chronic myelogenous leukemia are nonadherent to their tyrosine kinase inhibitors during treatment, and this is associated with poorer outcomes, including
treatment resistance.64,65 Barriers to nonadherence are
multifactorial but include poor provider-patient communication, side effects, costs of and access to medications, and
patient behaviors and education.56,66-68 Through patient
portals, EHRs have the ability to help foster secure communication and thus bridge the gap between patients and
providers and allow improved communication between all
members of the health care team, including staff who can
answer questions, provide symptom management, and
assist with prior authorizations to help offset high out-ofpocket costs. A recent randomized trial demonstrated that a
web-based platform of symptom self-reporting between
patients with cancer who are receiving chemotherapy and
providers was associated with improved quality of life, fewer
hospitalizations, and a longer duration of palliative chemotherapy.69 In addition, single-institution studies have
shown that access to medical records and the ability to
request medication refills (via patient portals) were both
associated with increased adherence to medications and
medical advice.70-72 Furthermore, patient portals may be a
valuable resource to provide patient education and even
individualized medication plans. Improved communication through the EHR may improve adherence to both
206
antineoplastic and supportive medications, and be associated with better quality of care.
To provide high-quality care, it is important to systematically assess our own practice to identify areas for improvement as well as disparities in care. The 1999 Institute of
Medicine report Ensuring Quality of Cancer Care called for a
system to measure and monitor quality of care and to reduce
disparities in cancer care.73 In response to this report, ASCO
developed the Quality Oncology Practice Initiative (QOPI) to
measure quality at the individual and practice levels by
medical record abstraction and facilitate continuous improvement.74 To globally address quality improvement
throughout the physician workforce, the Accreditation
Council for Graduate Medical Education has revised its
training requirements and implemented the Clinical
Learning Environmental Review (CLER) program designed to
improve education in patient safety and health care quality.75 As part of the CLER program, residents and fellows are
expected to receive specialty-specific data on their adherence to quality metrics and benchmarks within their patient
populations, to identify and reduce health care disparities
within their practice, and to foster habits of continuous
practice improvement.75 Through ASCOs CancerLinQ project, data from the EHR can be pulled to measure QOPI
compliance in real time.35 An EHR is a robust tool that can
pull individual and practice-specific data in real time to allow
providers the opportunity to measure their own performances and compare themselves with their peers, and thus
foster practice improvements on the individual, practice,
and, potentially, global levels.
In addition to continuous improvement, EHRs can affect
cancer care delivery on a larger scale as a tool for cancer care
delivery research. An upcoming pragmatic trial to be conducted through SWOG will use the EHR to assess the efficacy
of prophylactic colony-stimulating factors during the first
cycle of intermediate febrile neutropenia risk chemotherapy
(10%20%). Sites will be randomized to a standing-order
entry system for prophylactic colony-stimulating factors via
manipulation of the individual site EHR or usual care. In
addition to evaluating rates and resource utilization related
to febrile neutropenia, this study will evaluate whether a
standing-order entry system helps improve adherence to
clinical practice guidelines.76 Other areas for future cancer
care delivery research include improvement of content,
dissemination, and implementation of cancer survivorship
care plans; EHR triggers aimed at improving adherence to
medication at both the patient and physician levels; and EHR
educational alerts to help providers adhere to best practice
guidelines.77
Many opportunities exist to improve quality of care
among patients with cancer, especially in improving treatment adherence and reducing disparities. The EHR is a
valuable tool to aid in this regard, both in daily clinical
practice and in cancer care delivery research. However,
much like social media and wearables, interventions to reduce disparities of adoption and patient participation are
needed.
References
1. U.S. Institute of Medicine Committee on Quality of Health Care in
America. Crossing the Quality Chasm: A New Health System for the 21st
Century. Washington, DC: National Academy Press; 2001.
2. Thompson MA, Majhail NS, Wood WA, et al. Social media and the
practicing hematologist: Twitter 101 for the busy healthcare provider.
Curr Hematol Malig Rep. 2015;10:405-412.
3. Vartebedian B. Social Media Has Been Introduced to Physicians.
33charts, January 25, 2015. http://33charts.com/2015/01/socialmedia-introduced-to-physicians.html.
4. Fisch M. Putting Twitter to Use Among Oncologists: Shared Note
Taking at National Meetings and Other Stuff. ASCO Connection,
December 5, 2012. https://connection.asco.org/commentary/puttingtwitter-use-among-oncologists-shared-note-taking-national-meetingsand-other.
5. Katz MS, Utengen A, Anderson PF, et al. Disease-specific hashtags for
online communication about cancer care. J Clin Oncol. 2015;33 (suppl;
abstr 6520).
6. Mostaghimi A, Crotty BH. Professionalism in the digital age. Ann Intern
Med. 2011;154:560-562.
7. Dizon DS, Graham D, Thompson MA, et al. Practical guidance: the use of
social media in oncology practice. J Oncol Pract. 2012;8:e114-e124.
8. Thompson MA. Using social media to learn and communicate: it is not
about the tweet. Am Soc Clin Oncol Educ Book. 2015;35:206-211.
9. Plank W, Wyatt T. The Future of the Wearables Market. Wall Street
Journal, January 13, 2016. http://www.wsj.com/articles/the-future-ofthe-wearables-market-1452736738. Accessed January 18, 2016.
10. CCS Insight. Market Forecast: Wearables. Cupertino, CA: CCS Insight;
2014.
11. Shapiro M, Johnston D, Wald J, et al. Patient-Generated Health Data:
White Paper Prepared for the Office of the National Coordinator for
Health Information Technology, 2012. http://www.rti.org/pubs/
patientgeneratedhealthdata.pdf. Accessed December 1, 2014.
12. Hooke MC, Gilchrist L, Tanner L, et al. Use of a fitness tracker to promote
physical activity in children with acute lymphoblastic leukemia. Pediatr
Blood Cancer. Epub 2016 Jan 12.
13. Galv~ao DA, Taaffe DR, Spry N, et al. Combined resistance and aerobic
exercise program reverses muscle loss in men undergoing androgen
suppression therapy for prostate cancer without bone metastases:
a randomized controlled trial. J Clin Oncol. 2010;28:340-347.
14. Schmitz KH, Courneya KS, Matthews C, et al; American College of Sports
Medicine. American College of Sports Medicine roundtable on exercise
guidelines for cancer survivors. Med Sci Sports Exerc. 2010;42:
1409-1426.
15. Bennett AV, Reeve BB, Basch EM, et al. Evaluation of pedometry as a
patient-centered outcome in patients undergoing hematopoietic
cell transplant (HCT): a comparison of pedometry and patient reports of symptoms, health, and quality of life. Qual Life Res. Epub
2015 Nov 17.
16. Chung AE, Basch EM. Potential and challenges of patient-generated
health data for high-quality cancer care. J Oncol Pract. 2015;11:195-197.
17. Cheriff AD, Kapur AG, Qiu M, et al. Physician productivity and the
ambulatory EHR in a large academic multi-specialty physician group. Int
J Med Inform. 2010;79:492-500.
18. Centers for Medicare & Medicaid Services (CMS), HHS. Medicare and
Medicaid Programs; electronic health record incentive programstage 3
and modifications to meaningful use in 2015 through 2017. Final rules
with comment period. Fed Regist. 2015;80:62761-62955.
19. Institute of Medicine. Key Capabilities of an Electronic Health Record
System. Washington, DC: Institute of Medicine; 2003.
20. Chaudhry B, Wang J, Wu S, et al. Systematic review: impact of health

information technology on quality, efficiency, and costs of medical care.
Ann Intern Med. 2006;144:742-752.
21. Amarasingham R, Plantinga L, Diener-West M, et al. Clinical information
technologies and inpatient outcomes: a multiple hospital study. Arch
Intern Med. 2009;169:108-114.
22. Porterfield A, Engelbert K, Coustasse A. Electronic prescribing: improving the efficiency and accuracy of prescribing in the ambulatory
care setting. Perspect Health Inf Manag. 2014;11:1g.
23. Bubalo J, Warden BA, Wiegel JJ, et al. Does applying technology
throughout the medication use process improve patient safety with
antineoplastics? J Oncol Pharm Pract. 2014;20:445-460.
24. Meisenberg BR, Wright RR, Brady-Copertino CJ. Reduction in chemotherapy order errors with computerized physician order entry. J Oncol
Pract. 2014;10:e5-e9.
25. McNamara M, Arnold C, Sarma K, et al. Patient portal preferences:
perspectives on imaging information. J Assoc Inf Sci Technol. 2015;66:
1606-1615.
26. Beckjord EB, Rechis R, Nutt S, et al. What do people affected by cancer
think about electronic health information exchange? Results from the
2010 LIVESTRONG Electronic Health Information Exchange Survey and
the 2008 Health Information National Trends Survey. J Oncol Pract.
2011;7:237-241.
27. Kazmi Z. Effects of exam room EHR use on doctor-patient communication: a systematic literature review. Inform Prim Care. 2013;21:30-39.
28. Blumenthal D, Glaser JP. Information technology comes to medicine.
N Engl J Med. 2007;356:2527-2534.
29. Hershman DL, Wang X, McBride R, et al. Delay of adjuvant chemotherapy initiation following breast cancer surgery among elderly
women. Breast Cancer Res Treat. 2006;99:313-321.
30. Katz SJ, Lantz PM, Paredes Y, et al. Breast cancer treatment experiences
of Latinas in Los Angeles County. Am J Public Health. 2005;95:
2225-2230.
31. Maggard MA, OConnell JB, Lane KE, et al. Do young breast cancer
patients have worse outcomes? J Surg Res. 2003;113:109-113.
32. Jones EB, Furukawa MF. Adoption and use of electronic health records
among federally qualified health centers grew substantially during
2010-12. Health Aff (Millwood). 2014;33:1254-1261.
33. King J, Furukawa MF, Buntin MB. Geographic variation in ambulatory
electronic health record adoption: implications for underserved communities. Health Serv Res. 2013;48:2037-2059.
34. Jhamb M, Cavanaugh KL, Bian A, et al. Disparities in electronic health
record patient portal use in nephrology clinics. Clin J Am Soc Nephrol.
2015;10:2013-2022.
35. Goel MS, Brown TL, Williams A, et al. Disparities in enrollment and use
of an electronic patient portal. J Gen Intern Med. 2011;26:1112-1116.
36. Goel MS, Brown TL, Williams A, et al. Patient reported barriers to
enrolling in a patient portal. J Am Med Inform Assoc. 2011;18:i8-i12
(suppl 1).
37. Ferreira A, Correia A, Silva A, et al. Why facilitate patient access to
medical records. Stud Health Technol Inform. 2007;127:77-90.
38. Wright Nunes JA, Osborn CY, Ikizler TA, et al. Health numeracy: perspectives about using numbers in health management from African
American patients receiving dialysis. Hemodial Int. 2015;19:287-295.
39. Yamin CK, Emani S, Williams DH, et al. The digital divide in adoption and
use of a personal health record. Arch Intern Med. 2011;171:568-574.
40. Sarkar U, Karter AJ, Liu JY, et al. Social disparities in internet patient
portal use in diabetes: evidence that the digital divide extends beyond
access. J Am Med Inform Assoc. 2011;18:318-321.
207
41. Butler JM, Carter M, Hayden C, et al. Understanding adoption of a

personal health record in rural health care clinics: revealing barriers and
facilitators of adoption including attributions about potential patient
portal users and self-reported characteristics of early adopting users.
AMIA Annu Symp Proc. 2013;2013:152-161.
42. Wade-Vuturo AE, Mayberry LS, Osborn CY. Secure messaging and diabetes management: experiences and perspectives of patient portal
users. J Am Med Inform Assoc. 2013;20:519-525.
43. Ronda MC, Dijkhorst-Oei LT, Rutten GE. Reasons and barriers for using a
patient portal: survey among patients with diabetes mellitus. J Med
Internet Res. 2014;16:e263.
44. Klatt TE, Hopp E. Effect of a best-practice alert on the rate of influenza vaccination of pregnant women. Obstet Gynecol. 2012;119:
301-305.
45. Meyfroidt G, Wouters P, De Becker W, et al. Impact of a computergenerated alert system on the quality of tight glycemic control. Intensive Care Med. 2011;37:1151-1157.
46. Kucher N, Koo S, Quiroz R, et al. Electronic alerts to prevent venous
thromboembolism among hospitalized patients. N Engl J Med. 2005;
352:969-977.
47. Cho A, Lee JE, Yoon JY, et al. Effect of an electronic alert on risk of
contrast-induced acute kidney injury in hospitalized patients undergoing computed tomography. Am J Kidney Dis. 2012;60:74-81.
48. Selvan MS, Sittig DF, Thomas EJ, et al. Improving erythropoietinstimulating agent administration in a multihospital system through
quality improvement initiatives: a pre-post comparison study. J Patient
Saf. 2011;7:127-132.
49. Ranpura V, Agrawal S, Chokshi P, et al. Improving documentation of pain
management at MedStar Washington Cancer Institute. J Oncol Pract.
2015;11:155-157.
50. Jacobson JO, Neuss MN, Hauser R. Measuring and improving value of
care in oncology practices: ASCO programs from Quality Oncology
Practice Initiative to the Rapid Learning System. Am Soc Clin Oncol Educ
Book. 2012;32:e70-e76.
51. American Society of Clinical Oncology. ASCO CancerLinQ. http://
cancerlinq.org/. Accessed January 4, 2016.
52. Sledge GW Jr, Miller RS, Hauser R. CancerLinQ and the future of cancer
care. Am Soc Clin Oncol Educ Book. 2013;33:430-434.
53. Abernethy AP, Etheredge LM, Ganz PA, et al. Rapid-learning system for
cancer care. J Clin Oncol. 2010;28:4268-4274.
54. Etheredge LM. A rapid-learning health system. Health Aff (Millwood).
2007;26:w107-w118.
55. Osterberg L, Blaschke T. Adherence to medication. N Engl J Med. 2005;
353:487-497.
56. McDonald HP, Garg AX, Haynes RB. Interventions to enhance patient
adherence to medication prescriptions: scientific review. JAMA. 2002;
288:2868-2879.
57. Livaudais JC, Hershman DL, Habel L, et al. Racial/ethnic differences in
initiation of adjuvant hormonal therapy among women with hormone
receptor-positive breast cancer. Breast Cancer Res Treat. 2012;131:
607-617.
58. Kimmick G, Anderson R, Camacho F, et al. Adjuvant hormonal therapy
use among insured, low-income women with breast cancer. J Clin Oncol.
2009;27:3445-3451.
59. Neugut AI, Hillyer GC, Kushi LH, et al. Non-initiation of adjuvant hormonal therapy in women with hormone receptor-positive breast
cancer: the Breast Cancer Quality of Care Study (BQUAL). Breast Cancer
Res Treat. 2012;134:419-428.
208
60. Partridge AH, Wang PS, Winer EP, et al. Nonadherence to adjuvant
tamoxifen therapy in women with primary breast cancer. J Clin Oncol.
2003;21:602-606.
61. Hershman DL, Kushi LH, Shao T, et al. Early discontinuation and nonadherence to adjuvant hormonal therapy in a cohort of 8,769 earlystage breast cancer patients. J Clin Oncol. 2010;28:4120-4128.
62. Hershman DL, Shao T, Kushi LH, et al. Early discontinuation and nonadherence to adjuvant hormonal therapy are associated with increased
mortality in women with breast cancer. Breast Cancer Res Treat. 2011;
126:529-537.
63. Yood MU, Owusu C, Buist DS, et al. Mortality impact of less-thanstandard therapy in older breast cancer patients. J Am Coll Surg. 2008;
206:66-75.
64. Noens L, van Lierde MA, De Bock R, et al. Prevalence, determinants, and
outcomes of nonadherence to imatinib therapy in patients with chronic
myeloid leukemia: the ADAGIO study. Blood. 2009;113:5401-5411.
65. Ganesan P, Sagar TG, Dubashi B, et al. Nonadherence to imatinib adversely affects event free survival in chronic phase chronic myeloid
leukemia. Am J Hematol. 2011;86:471-474.
66. Sonpavde G, Periman PO, Bernold D, et al. Sunitinib malate for metastatic castration-resistant prostate cancer following docetaxel-based
chemotherapy. Ann Oncol. 2010;21:319-324.
67. Neugut AI, Subar M, Wilde ET, et al. Association between prescription
co-payment amount and compliance with adjuvant hormonal therapy
in women with early-stage breast cancer. J Clin Oncol. 2011;29:
2534-2542.
68. Accordino MK, Hershman DL. Disparities and challenges in adherence to
oral antineoplastic agents. Am Soc Clin Oncol Educ Book. 2013;33:
271-276.
69. Basch E, Deal AM, Kris MG, et al. Symptom monitoring with patientreported outcomes during routine cancer treatment: a randomized
controlled trial. J Clin Oncol. 2016;34:557-565.
70. Wright E, Darer J, Tang X, et al. Sharing physician notes through an
electronic portal is associated with improved medication adherence:
quasi-experimental study. J Med Internet Res. 2015;17:e226.
71. Lyles CR, Sarkar U, Schillinger D, et al. Refilling medications through an
online patient portal: consistent improvements in adherence across
racial/ethnic groups. J Am Med Inform Assoc. Epub 2015 Sep 2.
72. Ross SE, Moore LA, Earnest MA, et al. Providing a web-based online
medical record with electronic communication capabilities to patients
with congestive heart failure: randomized trial. J Med Internet Res.
2004;6:e12.
73. Hewitt ME, Simone JV (eds); National Cancer Policy Board, Institute of
Medicine and National Research Council. Ensuring Quality Cancer Care.
Washington, DC: National Academy Press; 1999.
74. Neuss MN, Desch CE, McNiff KK, et al. A process for measuring the
quality of cancer care: the Quality Oncology Practice Initiative. J Clin
Oncol. 2005;23:6233-6239.
75. Accreditation Council for Graduate Medical Education. CLER Pathways
to Excellence. https://www.acgme.org/acgmeweb/Portals/0/PDFs/
CLER/CLER_Brochure.pdf. Accessed December 28, 2015.
76. Ramsey SD. A Pragmatic Trial to Improve Colony Stimulating Factor
Use in Cancer. http://www.pcori.org/research-results/2015/pragmatictrial-improve-colony-stimulating-factor-use-cancer. Accessed December
28, 2015.
77. Mayer DK, Birken SA, Check DK, et al. Summing it up: an integrative
review of studies of cancer survivorship care plans (2006-2013). Cancer.
2015;121:978-996.
MYELODYSPLASTIC SYNDROMES, AND
ALLOTRANSPLANT
Evolving Therapies in Acute

Myeloid Leukemia: Progress
at Last?
CHAIR
Daniel J. DeAngelo, MD, PhD
Boston, MA
SPEAKERS
Eytan M. Stein, MD
New York, NY
Farhad Ravandi, MD
Houston, TX
DEANGELO ET AL
Evolving Therapies in Acute Myeloid Leukemia: Progress at

Last?
Daniel J. DeAngelo, MD, PhD, Eytan M. Stein, MD, and Farhad Ravandi, MD
OVERVIEW
Acute myeloid leukemia (AML) is an acquired disease characterized by chromosomal translocations and somatic mutations
that lead to leukemogenesis. Systemic combination chemotherapy with an anthracycline and cytarabine remains the
standard induction regimen for fit adults. Patients who achieve complete remission generally receive postinduction
therapy with cytarabine-based chemotherapy or an allogeneic bone marrow transplant. Those unfit for induction chemotherapy are treated with hypomethylating agents (HMAs), low-dose cytarabine, or they are offered supportive care alone
with transfusions and prophylactic antimicrobials. The revolution in understanding the genetics of AML, facilitated by nextgeneration sequencing, has led to many new drugs against driver mutations. Better methods of identification of leukemic
blasts have provided us with better means to detect the disease left behind after cytotoxic chemotherapy regimens. This
measurable residual disease has been correlated with poorer relapse-free survival, demonstrating the need for novel
strategies to eradicate it to improve the outcome of patients with acute leukemias. In this article, we discuss adapting and
improving AML therapy by age and comorbidities, emerging targeted therapies in AML, and minimal residual disease (MRD)
assessment in AML.
cute myeloid leukemia is most commonly diagnosed in

older adults, with a median age of diagnosis of 66 years.1
Given the aging population in the United States, the incidence of AML will grow dramatically in the coming
decades.2 Older patients with AML define a distinct subgroup
with a worse prognosis and different biological characteristics as compared to their younger counterparts.3 In addition, many older patients have significant comorbidities
and functional impairment that further complicates therapeutic options.3,4 As a result of the biologic complexity, high
rates of chemotherapy drug resistance, and poor host factors, the 5-year disease-specific survival of patients older
than age 65 remains less than 5%.
In spite of the many advances made in other hematologic
diseases, induction chemotherapy with cytarabine and an
anthracycline remains the standard initial approach for older
patients with AML. Although associated with high rates of
death from toxicity, it still offers the best chance of longerterm survival.5 Less-intensive treatment and palliative approaches are currently being explored for patients who
cannot tolerate standard anthracycline induction therapy.6
Previous studies have identified age, performance status,
comorbidities, and cytogenetic risk groups as important
factors in stratifying older patients with AML.7 However,
these factors do not sufficiently address the heterogeneity

present in this older population. New studies suggest
that geriatric evaluation including psychological, cognitive,
physical, and comorbidities can identify deficits affecting
morbidity and mortality among patients with cancer and
should be incorporated into the initial evaluation for older
adults with AML.8-11 In addition, molecular profiling can help
to further identify a biologically resistant group of patients
who may benefit from targeted therapeutic approaches.
Regardless of whether standard induction chemotherapy,
less-intensive therapy, or targeted approaches are used, the
definition of complete remission is being redefined by our
ability to detect small residual subclones or MRD.
ADAPTING AND IMPROVING THERAPY

ACCORDING TO AGE AND COMORBIDITIES
Acute myeloid leukemia is a clonal hematopoietic stem cell
disorder, for which the treatment, and in fact cure, hinges
upon the successful eradication of the leukemic clone. To
this end, induction chemotherapy is given to reduce the level
of the leukemic burden below the level of detection and to
restore normal hematopoiesis. Postremission consolidation,
alone or followed by an autologous or allogeneic stem cell
transplant, is then used to eradicate the remaining leukemic
From the Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA; Memorial Sloan Kettering Cancer Center, Weill Cornell Medical College, New York, NY;
The University of Texas MD Anderson Cancer Center, Houston, TX.
Corresponding author: Daniel J. DeAngelo, MD, PhD, Dana-Farber Cancer Institute, 450 Brookline Ave., Boston, MA 02215; email: daniel_deangelo@dfci.harvard.edu.
e302
EVOLVING THERAPIES IN ACUTE MYELOID LEUKEMIA
burden in hopes of cure. The backbone of remission induction therapy for younger patients (age , 60) and a subset of older patients deemed fit enough to undergo such
treatment has consisted of high-dose cytotoxic chemotherapy often consisting of 7 days of continuous infusion
cytarabine combined with 3 days of an anthracycline,
otherwise known as the 7 plus 3 regimen. Remission rates
with induction therapy vary and depend on many factors
including patient age, cytogenetics, molecular profiling,
antecedent hematologic disorder, prior chemotherapy, and
type of leukemia. Over the past several decades, induction
therapy has been intensified with the aim of improving
remission rates, extending disease-free survival and, ultimately, overall survival. For elderly patients (age . 60)
induction strategies have concentrated on less-toxic induction strategies using HMAs, monoclonal antibodies,
nucleoside analogs, and other novel compounds.
Health Care Utilization

Older patients with AML spend a large portion of their life
after diagnosis in the hospital or clinic. Given the short life
expectancy of the vast majority of older patients with AML
and the low likelihood of cure, it is important to consider the
impact of cancer therapy on their quality of life, which includes the time spent in the hospital and receiving medical
care throughout their illness. One randomized study compared intensive induction versus supportive care of older
patients with AML, with hydroxyurea or low-dose cytarabine
used as cytoreductive agent when needed.12 In this study,
patients treated with intensive chemotherapy had a
10-week survival advantage compared with those treated
with supportive care. However, there were no major differences in patients time spent in the hospital. Another
study looked retrospectively at 330 consecutive older patients diagnosed with AML to examine their health care
utilization.13 The median number of hospitalizations for the
entire cohort was 4.2 (range, 1 to 18). Patients who died
KEY POINTS
The treatment of AML, especially in elderly patients,

remains a difficult endeavor with high therapy-related
toxicity and poor long-term survival.
Efficacy of therapy in AML has been assessed by
morphologic assessment, with complete remission
correlating with improved outcomes.
Several retrospective studies have clearly demonstrated
the adverse prognostic value of persistent MRD after
induction and consolidation therapy.
Standardization and universal availability of these
assays are key to their potential future use to direct
risk-adapted therapy.
New target-specific agents with novel mechanisms of
action are likely to be more effective in eradicating MRD,
thereby improving outcomes.
spent a mean of 28.3% of their life from diagnosis in the

hospital and 13.8% of their life attending outpatient appointments, and only 23% used hospice services. Within
30 days before death, 84.5% of patients were hospitalized,
and 61.0% died in the hospital. Although intensive induction
offers a minority of patients a potentially curative therapy,
the patients who died spent more than 50% of their life from
diagnosis in the hospital or clinic. However, with the introduction of HMAs, as well as improvement in supportive
care measures, it is unclear whether the findings of this study
would be replicable in the modern era.
Pretreatment Evaluation
Induction chemotherapy is a complicated and rigorous process that is required to obtain clinical remission.14 Before
the initiation of induction chemotherapy, it is important to
establish the presence of comorbid conditions that could
potentially complicate the treatment of the patient. It is
clear that all elements of the patients history and physical examination are important. However, particular attention should be paid to a prior history of a hematologic
disorderfor example, a myelodysplastic or myeloproliferative disorder, prior therapy for malignancies including
chemotherapy or radiation therapy, and a family history of
disease. Patients with secondary or treatment-related AML
have a significantly worse prognosis, which is predominantly
influenced by the higher incidence of unfavorable cytogenetic abnormalities. A history of cardiac disease such as
congestive heart failure would mandate careful monitoring
and potential alteration of induction chemotherapy because
anthracycline-based regimens are standard. Patients typically receive large amounts of intravenous fluids that accompany initial chemotherapy as well as antibiotics and
blood and platelet transfusions, and therefore patients with
cardiac abnormalities at the time of diagnosis need rigorous
and careful monitoring.
A study of 101 patients with AML age 65 or older at the
Dana-Farber Cancer Institute underwent geriatric assessment.15 With a median age of 72, only 38% of patients had a
hematopoietic stem cell transplant comorbidity index of
greater than 1, and only three patients required help with
activities of daily living. The majority of patients (78%) received treatment and, of those, 45% underwent induction
chemotherapy and 44% received either decitabine or azacitidine. The 1-year disease-specific survival was 39%. On
multivariate analysis, age at diagnosis, cytogenetic risk
group, secondary AML versus de novo, comorbidity score
(hematopoietic stem cell transplant comorbidity index) of
at least 1, difficulty with strenuous activity, and pain were
independent prognostic factors for survival. The most
amazing fact is that even when the analysis was limited to
patients with ECOG performance status of 1 or lower, difficulty with strenuous activities, pain, and comorbidity
remained independent predictors. Thus, geriatric assessment
variables are independent prognostic factors for survival,
even among patients with the best performance status.
e303
DEANGELO ET AL
Acute Myeloid Leukemia Ontogeny

Acute myeloid leukemia is a biologically heterogeneous
disease that is typically classified into three clinically
distinct categories. Secondary AML (s-AML) represents
transformation from a prior diagnosis of myelodysplastic
syndrome or myeloproliferative neoplasm (MPN); therapyrelated AML develops as a complication of exposure to
leukemogenic therapies; and de novo AML arises in the
absence of an identified exposure or prior stem cell disorder. A retrospective, targeted, mutational analysis of
194 patients with s-AML or therapy-related AML was
compared with 100 unselected older patients with AML.16
The presence of a mutation in SRSF2, SF3B1, U2AF1, ZRSR2,
ASXL1, EZH2, BCOR, or STAG2 was greater than 95% specific
for the diagnosis of s-AML. The presence of one of these
secondary-type mutations defines a subtype of elderly
de novo AML that share clinical and pathologic similarities with confirmed cases of s-AML. Patients with
secondary-type mutations have a higher likelihood of intrinsic therapy resistance, as reflected by a lower rate
of complete remission with standard intensive induction
chemotherapy.3,17,18 In contrast, older patients with de
novo/pan-AML mutations achieved a higher rate of complete remission and rarely required reinduction. Thus, genetic ontogeny may highlight a subset of older patients with
secondary-type mutations who, on the basis of age, genetic
characteristics, and induction outcomes, may have had an
unrecognized period of antecedent myelodysplasia before
AML diagnosis. In elderly AML, genetic ontogeny may account for the differences in chemosensitivity and clinical
outcomes.
Less Intensive Strategies

Low-dose cytarabine (LDAC) has been used for the treatment of elderly, unfit AML patients for several decades.
Many studies have highlighted the fact that even very small
doses of cytarabine (20 mg/m2 administered once or twice
daily for 1014 days per month) can induce complete remission in 8% to 18% of patients with AML and can prolong
survival compared with best supportive care.19 However,
patients with adverse cytogenetics seldom achieve remission with LDAC. Despite this, LDAC may still be a
therapeutic option for geriatric patients without poor-risk
karyotype AML who prefer to self-administer drugs at home
without the need for daily clinic visits or hospitalization.
Because of the tolerability of LDAC, there are many clinical
trials exploring the combination of LDAC with novel investigational agents in an attempt to improve outcomes.
Inhibitors of the DNA methyltransferase (DNMT) azacitidine (Vidaza, Celgene) and decitabine (Dacogen, Bloomington, MN) are approved for the treatment of patients
with myelodysplastic syndrome.20-22 The current World
Health Organization definition of AML now includes patients
with a blast count of between 20% and 29% because these
patients were previously considered as having myelodysplasia.23,24 The study by Fenaux et al demonstrated an
e304
improved median overall survival among patients treated

with acacitadine compared with conventional care approaches, which included supportive care, LDAC, and 7 plus 3.25
Although most studies using HMA therapy in AML patients demonstrate lower complete remission rates
(10%20%) compared with conventional intensive chemotherapy, approximately 30% of patients have evidence of
disease response or stabilization and, importantly, have
overall survival rates equivalent or superior to other conventional treatments. In addition, HMA therapy is well
tolerated over long periods of time, with fewer hospital days
and transfusions.
At The University of Texas MD Anderson Cancer Center,
the outcomes of 671 older patients with AML treated
with up-front intensive chemotherapy or HMA therapy
were retrospectively analyzed.26 Although the complete
remission rates were significantly lower with HMA therapy
versus intensive chemotherapy (28% vs. 42%), there was no
difference in 8-week early death, 2-year relapse-free survival, or median overall survival (6.7 vs. 6.5 months). The
only group of patients who benefited from intensive induction were those with NPM-1mutated AML, coming back
to the importance of AML ontogeny. The group from The
Ohio State University demonstrated a very high overall
response rate of 75% using a 10-day decitabine regimen,
even in those patients with a complex karyotype.27 However, the 10-day decitabine regimen is associated with
treatment mortality rates similar to those of intensive
chemotherapy in elderly individuals. Thus, the vast majority
of older individuals desiring largely outpatient therapy
should be considered for subcutaneous azacitidine, 5-day
decitabine, or LDAC therapy.
The novel nucleoside analog clofarabine (Genzyme,
Cambridge, MA) has been studied extensively in older patients with de novo AML. Clofarabine inhibits ribonucleotide
reductase and DNA polymerase. It is resistant to cleavage by
purine nucleoside phosphorylase and to deamination by
adenosine deaminase.28-30 A complete remission rate of
32% was seen in a phase II trial of patients with relapsed or
refractory hematologic malignancies.31,32 Clofarabine has
been combined in an effort to modulate cytarabine triphosphate accumulation in subsequent phase I and II trials.33,34
Clofarabine as a single agent was studied in 112 patients
older than age 60 who were deemed to be unlikely candidates to benefit from standard conventional chemotherapy. A complete remission rate was 38%, and the
overall response rate was 46% with a median disease-free
survival of approximately 33 weeks.35 Clofarabine has now
been compared in multiple randomized phase III studies,
including the MRC study of older adults who are not fit for
chemotherapy, comparing low dose ara-c versus clofarabine. Although the clofarabine arm had a higher remission
rate, there was no difference in overall survival.36 In addition, the recent ECOG 2906 study was presented at the
American Society of Hematology (ASH) Annual Meeting
comparing 3 plus 7 versus clofarabine, and the standard
arm was superior.37
Conclusion
The treatment of AML, especially for elderly patients,
remains a difficult endeavor with high therapy-related
toxicity and poor long-term survival. The impact on health
care utilization should not be underestimated, even in those
patients choosing less-intensive strategies. It is hoped that
the development of targeted therapies either as single
agents of in combination with current modalities will result
in improved outcomes.
EMERGING TARGETED THERAPIES IN ACUTE

MYELOID LEUKEMIA
Acute myeloid leukemia is an acquired disease characterized by chromosomal translocations and somatic mutations
that lead to leukemogenesis. Systemic combination chemotherapy with an anthracycline and cytarabine remains
the standard induction regimen for fit adults. Patients
who achieve a complete remission generally receive postinduction therapy with cytarabine-based chemotherapy or
an allogeneic bone marrow transplant.34 Those unfit for
induction chemotherapy are treated with hypomethylating
agents or low-dose cytarabine, or they are offered supportive care alone with transfusions and prophylactic
antimicrobials.35,38
The revolution in understanding the genetics of AML,
facilitated by next-generation sequencing, has led to many
new drugs against driver mutations. These novel therapies have emerged in parallel with antibodies against cell
surface proteins expressed on leukemic myeloblasts. The
past 10 years have seen the development of small-molecule
inhibitors of commonly mutated proteins and antibody-based
therapies with early clinical efficacy unseen in prior studies.
Small-Molecule Inhibitors
A variety of mutated and overexpressed proteins contribute alone or in combination to leukemogenesis. This everenlarging number of molecular targets has led to small
molecules that inhibit mutant or overexpressed proteins.
The list of agents in development is long, but a few inhibitors
stand out as showing exceptional promise in recent clinical
studies (Table 1).
The most targeted mutated protein is FLT3 because of

its role as the most common recurrent genetic alteration
among patients with de novo AML. FLT3 internal tandem
duplications (FLT3-ITD) occurs in 30% of AML patients and
portend a poor prognosis, particularly among patients with a
high allelic ratio.39,40
Midostaurin. Midostaurin (PKC-412) is a moderately potent
inhibitor of FLT3-ITD and FLT3 tyrosine kinase domain (TKD)
mutations and inhibits other kinases such as c-KIT, PDGFR-b,
VEGFR-2, and protein kinase C. A phase II study of midostaurin monotherapy for patients with relapsed and refractory FLT3-positive AML showed minimal clinical activity;
only 1 patient achieved a partial remission (PR).41 A followup phase IB study combined midostaurin with induction
chemotherapy (7 plus 3) without regard to FLT3 status and
demonstrated that the combination was safe and well tolerated; complete remissions were seen in 92% of patients
with FLT3-ITD compared with 74% of patients with FLT3 wildtype.42 On the basis of these data, the RATIFY study, an
international randomized phase III study of midostaurin or
placebo in combination with induction and consolidation
chemotherapy, was designed. The outcomes, reported at
the 2015 American Society of Hematology (ASH) Annual
Meeting, showed improved 5-year overall survival in the
midostaurin arm (51.4% vs. 44.2%), regardless of whether
patients were censored at the time of stem cell transplant,
despite no difference in the rates of complete remission at
60 days.43 The superiority of midostaurin/chemotherapy
over placebo/chemotherapy was consistent regardless of
allelic burden (high vs. low), FLT3-ITD, or FLT3-TKD. Patients receiving midostaurin had an increased frequency of
grade 3-4 desquamating rash. The overall survival benefit
in combination with the favorable toxicity profile makes
midostaurin in combination with induction and consolidation chemotherapy the new standard of care for patients
with FLT3-mutated AML.
Quizartinib. More potent inhibitors of FLT3-ITD and FLT3-TKD
are in development. Quizartinib is a selective inhibitor of
FLT3-ITD but lacks activity against FLT3-TKD. The composite
complete remission rate in phase II studies of quizartinib
TABLE 1. Response Rate in Clinical Trials of Targeted Agents for the Treatment of Acute Myeloid Leukemia*
Agent
Mechanism of Action
Suggested Patient Population
Quizartinib
FLT3 inhibitor
FLT3 + AML
Notes
Single-agent activity against FLT3-ITD
Resistance emerges in most patients
ASP-2215
FLT3 inhibitor with activity

against TKD resistance mutation
FLT3-ITD or FLT3-TKD
Single-agent activity against FLT3-ITD and FLT3-TKD

with CRc rate of 46% in relapsed/refractory AML
Single-agent activity (37% ORR) in R/R AML
AG-221
IDH2 inhibitor
IDH2 mutated
AG-120
IDH1 inhibitor
IDH1 mutated
Single-agent activity (35% ORR) in R/R AML
ABT-199
BCL-2 inhibitor
Ongoing investigation
CR/CRi rate of 71% when combined with HMAs

in untreated AML in older patients
Abbreviations: AML, acute myeloid leukemia; CRc, cytogenic complete remission; ORR, overall response rate; R/R, relapsed/refractory; CRi, incomplete peripheral blood count
recovery; HMA, hypomethalyting agent.
*Response criteria between trials do not necessarily use the international working group criteria and, therefore, between-trial comparisons are difficult to make.
e305
DEANGELO ET AL
as a single agent in relapsed and refractory FLT3-mutated

AML ranges between 44% and 54%.44-47 The median duration of response is between 11.3 and 12.7 weeks. Many
patients who receive quizartinib acquire mutations in
the TKD of the FLT3 gene (D835 and F691) and resistance
to ongoing treatment. Because of this, quizartinib may be
best used as monotherapy bridge to a potentially curative allogeneic bone marrow transplant. A randomized
open-label phase III study of quizartinib versus chemotherapy (NCT02039726) with a primary endpoint of overall
survival is currently accruing patients, and a randomized
double-blind study of quizartinib or placebo in combination
with induction and consolidation chemotherapy is in development (NCT02668653).
Gilteritinib. Gilteritinib (ASP-2215), also a potent inhibitor
of FLT3, differs from quizartinib in its ability to inhibit both
FLT3-ITD and FLT3-TKD mutations. Results of a phase I/II
study among 165 patients receiving 80 mg or greater of
gilteritinib were reported at the 2015 ASH Annual Meeting.48
Among patients with FLT3-ITD, the composite complete
remission rate was 46%, and it was 9% for those with a
FLT3-TKD. Responses were similar regardless of whether
patients had received prior FLT3directed therapy. The
median duration of response was 15.9 weeks, similar
to what is seen in clinical studies with quizartinib. A phase
I study of ASP2215 in combination with induction and
consolidation chemotherapy is ongoing (NCT02236013),
and a randomized phase III study of ASP2215 versus salvage
chemotherapy is accruing patients (NCT02421939).
IDH1 and IDH2

IDH, the enzyme that converts isocitrate to alpha-ketoglutarate
in the mitochondria (IDH2) or the cytoplasm (IDH1) as part
of the citric acid cycle, is mutated in 15% and 10% of patients with de novo AML, respectively.49 The prevalence of
IDH mutations increases with age.50,51 Mutant IDH enzymes
acquire neomorphic activity and catalyze the conversion of
alpha-ketoglutarate into beta-hydroxygutarate (2-HG). Increased intracellular 2-HG causes inhibition of TET enzymes
and subsequent arrest in myeloblast maturation.52-54 Inhibitors of mutant IDH1 and mutant IDH2 are currently
in phase I clinical trials (NCT02381886, NCT01915498, and
NCT02074839).
Interim results of a phase I/II study of the IDH2 inhibitor
AG-221 (Agios/Celgene), presented at the 2015 ASH Annual
Meeting, demonstrated an overall response rate of 37%
among 159 patients with relapsed/refractory AML with
a composite complete remission of 27%. Duration of
response was 6.9 months.55 Similarly, a phase I study of
the IDH1 inhibitor AG-120 demonstrated an overall response rate of 35%, with a composite complete remission
rate of 33%.56 Accrual has started on a phase I study
exploring the safety of combining AG-120 and AG-221
with both induction and consolidation chemotherapy
and with the HMA, 5-azacitidine (NCT02632708 and
NCT0267792).
e306
TABLE 2. Antibody-Based Therapies for Acute Myeloid

Leukemia (Excluding Immunotherapy) in Clinical
Development
Agent
Gemtuzumab
Ozagomicin
Mechanism of
Action
Notes
Anti-CD33 antibody-drug Withdrawn from U.S.

conjugate
market; efforts to
reintroduce on the
basis of ongoing
clinical studies
Vadastuximab Anti-CD33 antibody-drug CR/CRi of 65% when

Tariline
conjugate
given in combination
(SGN-CD33A)
with HMAs in older
patients with
untreated AML in
ongoing study
IMGN779
Anti-CD33 antibody-drug Payload of alkylating

conjugate
agent (DGN-462);
studies beginning
in early 2016
LintuzumabAC225
Anti-CD33 antibodyradiopharmaceutical
Ongoing phase I/II study

in combination with
low-dose cytarabine
JNJ-56022473
Anti-CD123
Antibody against CD123;

trials beginning
KHK2823
Anti-CD123
Antibody against CD123;

trials beginning
ADCT-301
Anti-CD25
Antibody-drug conjugate
against CD25; trials
beginning
AGS67E
(Anti-CD37)
Anti-CD37
Antibody-drug conjugate
against CD37; trials
beginning
Abbreviations: CR, complete remission; CRi, complete response with incomplete blood
count recovery; HMA, hypomethylating agent; AML, acute myeloid leukemia.
BCL-2 Inhibitors
BCL-2 overexpression has been implicated in the maintenance and survival of AML cells in vitro and is associated with
resistance to chemotherapy and poor survival among patients with AML. The BCL-2 inhibitor venetoclax (ABT-199)
showed a complete remission/complete response with
incomplete blood count recovery in five of 32 patients
in a phase II clinical study.57 Preclinical models suggest that
the combination of venetoclax and HMAs are synergistic. On
the basis of preclinical data, a phase I study of the combination of venetoclax and decitabine or 5-azacitidine was
initiated. As of September 2015, the overall response rate
(34 patients) is 76%, with 71% of patients having a complete
remission or complete response with incomplete blood
count recovery, without differences in response between
patients who received decitabine or patients who received
5-azacitidine.58 The study is proceeding to an expansion
stage (NCT02203773).
Cell Surface Targets: Antibody-Based Therapy

Antibody-based therapies against cell surface proteins, either alone or conjugated to chemotherapeutic or radiopharmaceutical agents, continue to be developed. To date,
only one antibody-drug conjugate against CD33, gemtuzumab ozagomicin, was granted accelerated approval by the
U.S. Food and Drug Administration but was subsequently
withdrawn from the market because of the inability to
confirm clinical benefit in a phase III study. Despite this, the
effort to target CD33 and other cell surface proteins
continues (Table 2).
Perhaps the antibody-drug conjugate with the most encouraging clinical data is SGN-CD33A, a novel antibody-drug
conjugate with highly stable dipeptide linkers that enable
uniform drug loading of a pyrrolobenzodiazapene dimer,
which cross-links DNA, leading to cell death. An interim
analysis of a phase I dose-escalation study of SGN-CD33A
among patients with relapsed CD33-positive AML or patients
who declined intensive therapy, the composite complete
remission rate was 27%.59 More impressive is an ongoing
study of SGN-CD33A in combination with HMAs (23 patients), where the complete remission/complete response
with incomplete blood count recovery rate is 65%.60 The
median overall survival has not yet been reached, and at a
median follow-up of 7.7 months, 72% of patients remained
alive and in the study. This impressive early clinical activity
has led to the development of the CASCADE study, a randomized phase III study of HMAs in combination with
SGN-CD33A versus HMAs alone for patients with newly
diagnosed AML who are not candidates for intensive induction chemotherapy.
MINIMAL RESIDUAL DISEASE ASSESSMENT IN

ACUTE MYELOID LEUKEMIA: DOES IT MATTER?
Why Is Detection of Minimal Residual Disease
Relevant?
Predicting outcome of therapy has been and continues to be
an integral component of the treatment of patients with
AML.61 This is based largely on the realization that certain
subtypes of AML are less sensitive to the effects of cytotoxic chemotherapy and, as such, less likely to be cured
with the established and currently available combination chemotherapy regimens. This prediction, be it highly
imperfect, has been largely used to determine the
best postremission strategy (including allogeneic stem cell
transplantation or continued chemotherapy) and has been
based mainly on pretreatment patient- and disease-related
variables such as age, performance status, cytogenetic and
molecular aberrations, and presence or absence of antecedent hematologic disorders or prior DNA-damaging
therapy for other disorders or malignancies.61-64 As such,
this determination is based largely on collective historical
information and does not rely on the response of the individual patient to the specific therapeutic intervention he
or she received for induction of their initial remission.
Clearly, the sensitivity of an individual patients leukemic
cells to the initial therapy is likely to be important in determining the degree of reduction of disease burden and the
probability that it would be associated with relapse-free
survival and cure.
Lack of achieving a response to the initial therapy, per se,

identifies the individual as having a less favorable disease,
thereby diminishing their likelihood of survival.65 Traditionally, response has been determined using morphologic
assessment declaring a patient as being in complete remission when microscopic examination of the bone marrow and peripheral blood fails to detect abnormal immature
cells, and there is evidence of resumption of normal bone
marrow function with normalization of peripheral blood
counts. This assessment is highly insensitive, typically relying
on the examination of only 500 cells in the marrow specimen
and is very much dependent on the skills of the examiner and
the sampling technique. It has also been clear that such
morphologic remission, although associated with a higher
likelihood of achieving a cure, is not equivalent with complete eradication of the leukemia cells and, without further
postremission therapy, will likely be associated with relapse
in most, if not all, patients.66 This is a result of the persistence
of morphologically undetectable leukemia, commonly referred to a MRD. Detection of this residual leukemia is highly
dependent on the sensitivity and specificity of the technique
used, leading to its description by some as measurable
residual disease.67
Therefore, the traditional definition of complete remission is an insensitive and imperfect assessment of the
susceptibility of leukemic cells to the treatment administered. More sensitive assays such as multiparameter flow
cytometry (MFC), polymerase chain reaction (PCR) analysis
of aberrant and leukemia-specific genes, and, more recently,
next-generation sequencing have been used to try to detect
submicroscopic leukemia cells persisting in complete remission, thereby providing an improved assessment of the
effectiveness of therapy.68 Several published reports among
patients with AML have examined the utility of such residual
disease assessment and have demonstrated the prognostic
value of detecting MRD at the time of morphologic remission
in AML. This section reviews the available literature and
discusses the potential ways that this can influence future
therapeutic decision making for patients with AML.
Available Assays for Detecting Minimal Residual

Disease
Currently available techniques for detecting residual disease in complete remission include MFC, which relies on
detection of leukemia-associated immunophenotypes, PCR,
and, more recently, next-generation sequencingbased
assays to detect aberrant genetic events in the leukemic
cells.69-71 Although karyotypic analysis to detect persistent
cytogenetically abnormal clones at the time of remission is
an old and less sensitive strategy, its value in predicting
outcome has been reported by several investigators.72,73
Leukemia-associated immunophenotypes are aberrant
patterns of antigen expression that can be identified in the
leukemic blasts of most patients at diagnosis.74,75 They include aberrant expression of lymphoid antigens, aberrant
expression levels of normal antigens, and co-expression or
e307
DEANGELO ET AL
asynchronous expression of early and late antigens, leading

to temporal aberrancy. The advantages of this assay include
its wide availability, its applicability to more than 95% cases
of AML, and its relative rapidity of assessment, which offers
its potential utility for rapid therapeutic decision making.
The disadvantages often include challenging interpretations
requiring expert analysis (multiple leukemia-associated
immunophenotypes in different blast subsets for the same
patients, potential for phenotypic shifts between diagnosis,
and relapse blasts) and lower sensitivity depending on the
antibody panel used.
Molecular markers such as gene fusion transcripts, gene
mutations, and aberrantly overexpressed genes have also
been used effectively as markers for detecting residual
leukemia burden.67,71 Quantitative reverse-transcription
PCR assays are now well established for detecting recurrent fusion transcripts such as PML-RARA, CBFB-MYH11,
and RUNX1-RUNX1T1 and have the advantage of being
highly sensitive and specific in these subsets of AML that
account for approximately 20% of cases overall.76 Their
value in predicting outcome is now universally accepted.
This is particularly true for patients with acute promyelocytic
leukemia in whom achievement of a complete molecular
remission with a sensitive assay is now considered as a
requisite for long-term leukemia-free survival.77
To increase the applicability of molecular testing to a larger
population of AML patients (particularly those with a normal
karyotype), a number of recurrent aberrant gene mutations such as FLT3, NPM1, and DNMT3A have been evaluated
as markers for MRD assessment.78-80 However, despite
this, a large proportion of patients with AML lack a leukemiaspecific aberrant molecular target for MRD monitoring. This
has led some investigators to examine normal genes such as
WT1 that are overexpressed in leukemia cells as potential
markers for MRD assessment, establishing thresholds that
predict higher likelihood of disease relapse.81 Novel and
sensitive next-generation sequencing assays that can detect
all coexisting mutations in leukemic clones at the time of
diagnosis and remission have also begun to be used for MRD
monitoring, although there are few retrospective studies
reported to date.82 The advantage of molecularly-based
MRD monitoring is its high sensitivity, particularly using
the more modern assays.
Comparative studies evaluating morphology, flow, and
PCR assays in the same patients have been limited. In a study
from St. Jude Childrens Hospital, follow-up bone marrow
samples from 203 children and adolescents with newly diagnosed AML were evaluated by morphologic assessment,
flow cytometry, and PCR for fusion transcripts.83 Among
samples with more than 5% blasts by morphology, 8.2%
were MRD-positive ($ 0.1%) by flow cytometry, whereas
57.5% of samples with more than 5% blasts by morphology
were negative by flow. Virtually all (99%) of samples negative
by PCR were also negative by flow, but only 9.6% of samples positive by PCR were also flow-positive, with analysis of RUNX1-RUNX1T1 and CBFB-MYH11 accounting for
most discrepancies. Flow MRD after induction was an
e308
independent prognostic factor for relapse on multivariate

analysis, and the prediction was not improved by PCR or
morphologic data.83 Other investigators have reported that
MRD by flow and type of response (complete remission,
complete remission with incomplete platelet recovery, or
complete remission with incomplete blood count recovery)
are important prognostic factors in AML and should be used
independently to predict outcome.84
Reported Studies of Minimal Residual Disease

Monitoring in Acute Myeloid Leukemia
The potential value of detection of aberrant mutant genes in
predicting outcome in AML is best apparent among patients
with normal karyotype, because this is the group in which
there is most uncertainty regarding the role of allogeneic
stem cell transplant in first remission. Mutations in several
genes including FLT3, NPM1, CEBPA, DNMT3A, TET2, and
IDH1/2 have been described and have been correlated with
outcome in large cohorts of patients with AML.85-87 Several
of these have been studied as potential markers for MRD
monitoring. The potential utility of mutant FLT3 as a marker
of MRD has been controversial, with some investigators
suggesting that it can be reliably used in complete remission
to predict relapse, and others arguing that it may not be ideal
because of the probability of its gain or loss at the time of
relapse.78,88 In contrast, the utility of NPM1 mutations as reliable markers of persistent disease is becoming more established, with several studies demonstrating that, using real-time
quantitative PCR for NPM1, achieving an MRD-negative status
after induction and after consolidation is associated with a
considerable reduction in the risk of relapse and improvement
in survival.79,89,90 Most recently, a large prospective trial
conducted by the United Kingdom National Cancer Research
Institute AML Working Group (UK NCRI) further demonstrated
that MRD monitoring for mutated NPM1 was the only major
prognostic factor for relapse or death, and, on sequential
monitoring, relapse was reliably predicted by a rising level of
NPM1-mutated transcripts.91
The value of MRD monitoring for acute promyelocytic
leukemia and for core-binding factor (CBF) leukemias has
been more established. In a study by the U.K. group of
patients with acute promyelocytic leukemia treated with alltrans retinoic acid and chemotherapy, persistent MRD after
consolidation and MRD recurrence were the most important
predictors of relapse in multivariate analysis.77 Similarly,
several studies have reported the value of establishing
thresholds for RUNX1-RUNX1T1 and CBFB-MYH11 after induction and consolidation in predicting the likelihood of
relapse and survival.92-94 Furthermore, recent reports have
suggested a value for MRD-directed risk stratification to
improve outcome for patients with CBF AML.95
Although MFC has been extensively used to detect MRD
in pediatric studies (and particularly for patients with acute
lymphoblastic leukemia), data in the adult population are
more limited.96 Early reports using more limited assays
demonstrated the feasibility of this strategy in predicting
relapse and survival.97,98 Several reports in pediatric AML

further established the prognostic value of MRD assessed
by flow and demonstrated its potential use in risk-adapted
therapy.99,100 More recently, investigators from the HOVAN/
SAKK AML study group and the UK NCRI have reported the
prognostic value of flow-based MRD assessment in predicting the outcome in younger and older AML cohorts,
respectively.101,102 Furthermore, data from the Fred
Hutchison Cancer Center have clearly demonstrated that
patients transplanted in CR1 or CR2 have a substantially
lower likelihood of relapse post-transplant and improved
survival if they are determined to be MRD-negative using a
flow cytometrybased assay.103-105
Future Prospects for Minimal Residual Disease

Eradication Using Target-Specific Agents
Perhaps the most important value of persistent MRD is that
it indicates that continuation of current treatment strategy
has a decreased likelihood of success, and a change in

treatment, if available, is necessary. Clearly, this is dependent on the availability of agents with alternative
mechanisms of action that are able to eradicate MRD. Although allogeneic stem cell transplant has been traditionally
considered to be the alternate strategy in this setting, the
available data referred to before suggest that it may not be
the most effective strategy to eradicate MRD.103,104 Novel
agents such as molecularly targeted drugs (FLT3 or IDH
inhibitors) or monoclonal antibodybased agents including
antibody-drug conjugates and bispecific antibodies, and,
potentially, checkpoint inhibitors and chimeric antigen receptor T cells, may provide us with such alternate strategies
to deal with persistent MRD remaining after cytotoxic
regimens. Whether these concepts for maintenance therapy
can lead to improved outcomes with or without subsequent
transplantation will require complex design of future clinical
trials.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Howlader N, Krapcho M, Neyman N, et al (eds). SEER Cancer Statistics

Review, 1975-2012, National Cancer Institute. http://seer.cancer.gov/
csr/1975_2012/. Accessed April 6, 2016.
U.S. Census Bureau. U.S. Interim Projections by Age, Sex, Race
and Hispanic Origin. https://www.census.gov/population/projections/
data/national/usinterimproj.html. Accessed April 7, 2016.
Appelbaum FR, Gundacker H, Head DR, et al. Age and acute myeloid
leukemia. Blood. 2006;107:3481-3485.
Wedding U, Rohrig B, Klippstein A, et al. Impairment in functional
status and survival in patients with acute myeloid leukaemia. J Cancer
Res Clin Oncol. 2006;132:665-671.
Behringer B, Pitako JA, Kunzmann R, et al. Prognosis of older patients
with acute myeloid leukemia receiving either induction or noncurative
treatment: a single-center retrospective study. Ann Hematol. 2003;
827:381-389.
ODonnell MR, Abboud CN, Altman J, et al. Acute myeloid leukemia.
J Natl Compr Canc Net. 2011;9:280-317.
Sekeres MA, et al. Decision-making and quality of life in older adults
with acute myeloid leukemia or advanced myelodysplastic syndrome.
Leukemia. 2004;18:809-816.
Hurria A, Browner IS, Cohen HJ, et al. Senior adult oncology. J Natl
Compr Canc Netw. 2012;10:162-209.
Hurria A, Gupta S, Zauderer M, et al. Developing a cancer-specific
geriatric assessment: a feasibility study. Cancer. 2005;104:
1998-2005.
Klepin HD, Geiger AM, Tooze JA, et al. The feasibility of inpatient
geriatric assessment for older adults receiving induction chemotherapy for acute myelogenous leukemia. J Am Geriatr Soc. 2011;59:
1837-1846.
Sorror ML, Maris MB, Storb R, et al. Hematopoietic cell transplantation (HCT)-specific comorbidity index: a new tool for risk assessment before allogeneic HCT. Blood. 2005;106:2912-2919.
Lowenberg B, Zittoun R, Kerkhofs H, et al. On the value of intensive
remission-induction chemotherapy in elderly patients of 65+ years
with acute myeloid leukemia: a randomized phase III study of the
European Organization for Research and Treatment of Cancer Leukemia Group. J Clin Oncol. 1989;7:1268-1274.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
El-Jawahri AR, Abel GA, Steensma DP, et al. Health care utilization and
end-of-life care for older patients with acute myeloid leukemia.
Cancer. 2015;121:2840-2848.
Lowenberg B, Downing JR, Burnett A. Acute myeloid leukemia. N Engl J
Med. 1999;341:1051-1062.
Sherman AE, Motyckova G, Fega KR, et al. Geriatric assessment in
older patients with acute myeloid leukemia: a retrospective study of
associated treatment and outcomes. Leuk Res. 2013;37:998-1003.
Lindsley RC, Mar BG, Mazzola E, et al. Acute myeloid leukemia
ontogeny is defined by distinct somatic mutations. Blood. 2015;125:
1367-1376.
Anderson JE, Kopecky KJ, Willman CL, et al. Outcome after induction
chemotherapy for older patients with acute myeloid leukemia is not
improved with mitoxantrone and etoposide compared to cytarabine
and daunorubicin: a Southwest Oncology Group study. Blood. 2002;
100:3869-3876.
Goldstone AH, Burnett AK, Wheatley K, et al; Medical Research
Council Adult Leukemia Working Party. Attempts to improve treatment outcomes in acute myeloid leukemia (AML) in older patients: the
results of the United Kingdom Medical Research Council AML11 trial.
Blood. 2001;98:1302-1311.
Burnett AK, Milligan D, Prentice AG, et al. A comparison of low-dose
cytarabine and hydroxyurea with or without all-trans retinoic acid for
acute myeloid leukemia and high-risk myelodysplastic syndrome in
patients not considered fit for intensive treatment. Cancer. 2007;109:
1114-1124.
Kantarjian H, Issa JP, Rosenfeld CS, et al. Decitabine improves patient
outcomes in myelodysplastic syndromes: results of a phase III randomized study. Cancer. 2006;106:1794-1803.
Silverman LR, Demakos EP, Peterson BL, et al. Randomized controlled
trial of azacitidine in patients with the myelodysplastic syndrome:
a study of the cancer and leukemia group B. J Clin Oncol. 2002;20:
2429-2440.
Kantarjian H, Oki Y, Garcia-Manero G, et al. Results of a randomized
study of 3 schedules of low-dose decitabine in higher-risk myelodysplastic syndrome and chronic myelomonocytic leukemia. Blood.
2007;109:52-57.
e309
DEANGELO ET AL
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
e310
Harris NL, Jaffe ES, Diebold J, et al. The World Health Organization
classification of neoplastic diseases of the hematopoietic and lymphoid tissues. Report of the Clinical Advisory Committee meeting, Airlie House, Virginia, November, 1997. Ann Oncol. 1999;10:
1419-1432.
Vardiman JW, Harris NL, Brunning RD. The World Health Organization
(WHO) classification of the myeloid neoplasms. Blood. 2002;100:
2292-2302.
Fenaux P, Mufti GJ, Hellstrom-Lindberg E, et al; International Vidaza
High-Risk MDS Survival Study Group. Efficacy of azacitidine compared
with that of conventional care regimens in the treatment of higherrisk myelodysplastic syndromes: a randomised, open-label, phase III
Quintas-Cardama A, Ravandi F, Liu-Dumlao T, et al. Epigenetic therapy
is associated with similar survival compared with intensive chemotherapy in older patients with newly diagnosed acute myeloid leukemia. Blood. 2012;120:4840-4845.
Blum W, Garzon R, Klisovic RB, et al. Clinical response and miR-29b
predictive significance in older AML patients treated with a 10-day
schedule of decitabine. Proc Natl Acad Sci USA. 2010;107:7473-7478.
Jeha S, Razzouk B, Rytting M, et al. Phase II study of clofarabine in
pediatric patients with refractory or relapsed acute myeloid leukemia.
J Clin Oncol. 2009;27:4392-4397.
Jeha S, Gaynon PS, Razzouk BI, et al. Phase II study of clofarabine in
pediatric patients with refractory or relapsed acute lymphoblastic
leukemia. J Clin Oncol. 2006;24:1917-1923.
Jeha S, Gandhi V, Chan KW, et al. Clofarabine, a novel nucleoside
analog, is active in pediatric patients with advanced leukemia. Blood.
2004;103:784-789.
Kantarjian H, Gandhi V, Cortes J, et al. Phase 2 clinical and pharmacologic study of clofarabine in patients with refractory or relapsed
acute leukemia. Blood. 2003;102:2379-2386.
Kantarjian HM, Gandhi V, Kozuch P, et al. Phase I clinical and pharmacology study of clofarabine in patients with solid and hematologic
cancers. J Clin Oncol. 2003;21:1167-1173.
Kantarjian HM, Erba HP, Claxton D, et al. Phase II study of clofarabine
monotherapy in previously untreated older adults with acute myeloid
leukemia and unfavorable prognostic factors. J Clin Oncol. 2010;28:
549-555.
Bloomfield CD, Lawrence D, Byrd JC, et al. Frequency of prolonged
remission duration after high-dose cytarabine intensification in acute
myeloid leukemia varies by cytogenetic subtype. Cancer Res. 1998;58:
4173-4179.
Kantarjian HM, Thomas XG, Dmoszynska A, et al. Multicenter, randomized, open-label, phase III trial of decitabine versus patient choice,
with physician advice, of either supportive care or low-dose cytarabine for the treatment of older patients with newly diagnosed acute
myeloid leukemia. J Clin Oncol. 2012;30:2670-2677.
Burnett AK, Russell NH, Hunter AE, et al. Clofarabine doubles the
response rate in older patients with acute myeloid leukemia but does
not improve survival. Blood. 2013;122:1384-1394.
Foran JM, Zhuoxin S, Claxton DF, et al. North American Leukemia,
Intergroup Phase III Randomized Trial of Single Agent Clofarabine As
Induction and Post-Remission Therapy, and Decitabine As Maintenance Therapy in Newly-Diagnosed Acute Myeloid Leukemia in Older
Adults (Age $60 Years): A Trial of the ECOG-ACRIN Cancer Research
Group (E2906). Presented at: 57th ASH Annual Meeting & Exposition;
December 2015; Orlando, FL.
Dombret H, Seymour JF, Butrym A, et al. International phase 3 study of
azacitidine vs conventional care regimens in older patients with newly
diagnosed AML with .30% blasts. Blood. 2015;126:291-299.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
Kottaridis PD, Gale RE, Frew ME, et al. The presence of a FLT3 internal
tandem duplication in patients with acute myeloid leukemia (AML)
adds important prognostic information to cytogenetic risk group and
response to the first cycle of chemotherapy: analysis of 854 patients
from the United Kingdom Medical Research Council AML 10 and 12
trials. Blood. 2001;98:1752-1759.
Frohling S, Schlenk RF, Breitruck J, et al; AML Study Group Ulm. Acute
myeloid leukemia. Prognostic significance of activating FLT3 mutations in younger adults (16 to 60 years) with acute myeloid leukemia
and normal cytogenetics: a study of the AML Study Group Ulm. Blood.
2002;100:4372-4380.
Stone RM, DeAngelo DJ, Klimek V, et al. Patients with acute myeloid
leukemia and an activating mutation in FLT3 respond to a smallmolecule FLT3 tyrosine kinase inhibitor, PKC412. Blood. 2005;105:
54-60.
Stone RM, Fischer T, Paquette R, et al. Phase IB study of the FLT3
kinase inhibitor midostaurin with chemotherapy in younger newly
diagnosed adult patients with acute myeloid leukemia. Leukemia.
2012;26:2061-2068.
Stone RM, Mandrekar S, Sanford BL. The multi-kinase inhibitor
midostaurin (M) prolongs survival compared with placebo (P) in
combination with daunorubicin (D)/cytarabine (C) induction (ind),
high-dose C consolidation (consol), and as maintenance (maint)
therapy in newly diagnosed acute myeloid leukemia (AML) patients (pts) age 18-60 with FLT3 mutations (muts): an international
prospective randomized (rand) P-controlled double-blind trial (CALGB
10603/RATIFY [Alliance]). Blood. 2015;126:6.
Levis MJ, Perl AE, Dombret H, et al. Final results of a phase 2 openlabel, monotherapy efficacy and safety study of quizartinib (AC220) in
patients with FLT3-ITD positive or negative relapsed/refractory acute
myeloid leukemia after second-line chemotherapy or hematopoietic
stem cell transplantation. Blood. 2012;120:673.
Cortes JE, Kantarjian H, Foran JM, et al. Phase I study of quizartinib
administered daily to patients with relapsed or refractory acute
myeloid leukemia irrespective of FMS-like tyrosine kinase 3-internal
tandem duplication status. J Clin Oncol. 2013;31:3681-3687.
Cortes JE, Perl AE, Dombret H, et al. Final results of a phase 2 openlabel, monotherapy efficacy and safety study of quizartinib (AC220) in
patients .= 60 years of age with FLT3 ITD positive or negative
relapsed/refractory acute myeloid leukemia. Blood. 2012;120:48.
Schiller GJ, TM, Goldberg SL, et. al. Final results of a randomized phase
2 study showing the clinical benefit of quizartinib (AC220) in patients
with FLT3-ITD positive relapsed or refractory acute myeloid leukemia.
Altman JK, Perl AE, Cortes JE. Antileukemic activity and tolerability of
ASP2215 80mg and greater in FLT3 mutation-positive subjects with
relapsed or refractory acute myeloid leukemia: results from a phase
1/2, open-label, dose-escalation/dose-response study. Blood. 2015;
126:321.
Green CL, Evans CM, Zhao L, et al. The prognostic significance of IDH2
mutations in AML depends on the location of the mutation. Blood.
2011;118:409-412.
Paschka P, Schlenk RF, Gaidzik VI, et al. IDH1 and IDH2 mutations are
frequent genetic alterations in acute myeloid leukemia and confer
adverse prognosis in cytogenetically normal acute myeloid leukemia
with NPM1 mutation without FLT3 internal tandem duplication. J Clin
Oncol. 2010;28:3636-3643.
Marcucci G, Maharry K, Wu YZ, et al. IDH1 and IDH2 gene mutations
identify novel molecular subsets within de novo cytogenetically
normal acute myeloid leukemia: a Cancer and Leukemia Group B
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
Ward PS, Patel J, Wise DR, et al. The common feature of leukemiaassociated IDH1 and IDH2 mutations is a neomorphic enzyme activity
converting alpha-ketoglutarate to 2-hydroxyglutarate. Cancer Cell.
2010;17:225-234.
Figueroa ME, Abdel-Wahab O, Lu C, et al. Leukemic IDH1 and IDH2
mutations result in a hypermethylation phenotype, disrupt TET2
function, and impair hematopoietic differentiation. Cancer Cell. 2010;
18:553-567.
Lu C, Ward PS, Kapoor GS, et al. IDH mutation impairs histone demethylation and results in a block to cell differentiation. Nature. 2012;
483:474-478.
Stein EM, DiNardo C, Altman JK. Safety and efficacy of AG-221, a
potent inhibitor of mutant IDH2 that promotes differentiation of
myeloid cells in patients with advanced hematologic malignancies:
results of a phase 1/2 trial. Blood. 2015;126:323.
DiNardo C, de Botton S, Pollyea DA. Molecular profiling and relationship with clinical response in patients with IDH1 mutationpositive hematologic malignancies receiving AG-120, a first-in-class
potent inhibitor of mutant IDH1, in addition to data from the completed dose escalation portion of the phase 1 study. Blood. 2015;126:
1306.
Konopleva M, Pollyea DA, Potluri J, et al. A phase 2 study of ABT-199
(GDC-0199) in patients with acute myelogenous leukemia. Blood.
2014;124:118.
DiNardo C, Pollyea D, Pratz K. Phase 1b study of venetoclax (ABT199/GDC-0199) in combination with decitabine or azacitidine in
treatment-naive patients with acute myelogenous leukemia who
are $ to 65 years and not eligible for standard induction therapy.
Blood. 2015;126:327.
Stein AS, Walter RB, Erba HP, et al. A phase 1 trial of SGN-CD33A as
monotherapy in patients with CD33-positive acute myeloid leukemia
(AML). Blood. 2015;126:324.
Fathi AT, Erba HP, Lancet JE, et al. A novel well-tolerated regimen with
high remission rate in frontline unfit AML. Blood. 2015;126:454.
Dohner H, Weisdorf DJ, Bloomfield CD. Acute myeloid leukemia.
N Engl J Med. 2015;373:1136-1152.
Dohner H, Estey EH, Amadori S, et al; European LeukemiaNet. Diagnosis and management of acute myeloid leukemia in adults: recommendations from an international expert panel, on behalf of the
European LeukemiaNet. Blood. 2010;115:453-474.
Grimwade D, Hills RK, Moorman AV, et al; National Cancer Research
Institute Adult Leukaemia Working Group. Refinement of cytogenetic
classification in acute myeloid leukemia: determination of prognostic
significance of rare recurring chromosomal abnormalities among 5876
younger adult patients treated in the United Kingdom Medical Research Council trials. Blood. 2010;116:354-365.
Grimwade D, Ivey A, Huntly BJ. Molecular landscape of acute myeloid
leukemia in younger adults and its clinical relevance. Blood. 2016;127:
29-41.
Ravandi F. Primary refractory acute myeloid leukaemia - in search of
better definitions and therapies. Br J Haematol. 2011;155:413-419.
Stone RM. Is it time to revisit standard post-remission therapy? Best
Pract Res Clin Haematol. 2012;25:437-441.
Hourigan CS, Karp JE. Minimal residual disease in acute myeloid
leukaemia. Nat Rev Clin Oncol. 2013;10:460-471.
Ravandi F, Jorgensen JL. Monitoring minimal residual disease in acute
myeloid leukemia: ready for prime time? J Natl Compr Canc Netw.
2012;10:1029-1036.
Klco JM, Miller CA, Griffith M, et al. Association between mutation
clearance after induction therapy and outcomes in acute myeloid
leukemia. JAMA. 2015;314:811-822.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
Kern W, Voskova D, Schoch C, et al. Determination of relapse risk

based on assessment of minimal residual disease during complete
remission by multiparameter flow cytometry in unselected patients
with acute myeloid leukemia. Blood. 2004;104:3078-3085.
Grimwade D, Freeman SD. Defining minimal residual disease in acute
myeloid leukemia: which platforms are ready for prime time? Blood.
2014;124:3345-3455.
Chen Y, Cortes J, Estrov Z, et al. Persistence of cytogenetic abnormalities at complete remission after induction in patients with acute
myeloid leukemia: prognostic significance and the potential role of
allogeneic stem-cell transplantation. J Clin Oncol. 2011;29:2507-2513.
Marcucci G, Mrozek
K, Ruppert AS, et al. Abnormal cytogenetics at

date of morphologic complete remission predicts short overall and
disease-free survival, and higher relapse rate in adult acute myeloid
leukemia: results from cancer and leukemia group B study 8461. J Clin
Oncol. 2004;22:2410-2418.
San-Miguel JF, Vidriales MB, Orf~ao A. Immunological evaluation of
minimal residual disease (MRD) in acute myeloid leukaemia (AML).
Best Pract Res Clin Haematol. 2002;15:105-118.
Kern W, Haferlach C, Haferlach T, et al. Monitoring of minimal residual
disease in acute myeloid leukemia. Cancer. 2008;112:4-16.
Schnittger S, Weisser M, Schoch C, et al. New score predicting for
prognosis in PML-RARA+, AML1-ETO+, or CBFBMYH11+ acute myeloid
leukemia based on quantification of fusion transcripts. Blood. 2003;
102:2746-2755.
Grimwade D, Jovanovic JV, Hills RK, et al. Prospective minimal residual
disease monitoring to predict relapse of acute promyelocytic leukemia and to direct pre-emptive arsenic trioxide therapy. J Clin Oncol.
2009;27:3650-3658.
Schnittger S, Schoch C, Dugas M, et al. Analysis of FLT3 length mutations in 1003 patients with acute myeloid leukemia: correlation to
cytogenetics, FAB subtype, and prognosis in the AMLCG study and
usefulness as a marker for the detection of minimal residual disease.
Blood. 2002;100:59-66.
Kronke J, Schlenk RF, Jensen KO, et al. Monitoring of minimal residual
disease in NPM1-mutated acute myeloid leukemia: a study from the
German-Austrian acute myeloid leukemia study group. J Clin Oncol.
2011;29:2709-2716.
Debarri H, Lebon D, Roumier C, et al. IDH1/2 but not DNMT3A mutations are suitable targets for minimal residual disease monitoring in
acute myeloid leukemia patients: a study by the Acute Leukemia
French Association. Oncotarget. 2015;6:42345-42353.
Cilloni D, Renneville A, Hermitte F, et al. Real-time quantitative polymerase chain reaction detection of minimal residual disease by
standardized WT1 assay to enhance risk stratification in acute myeloid
leukemia: a European LeukemiaNet study. J Clin Oncol. 2009;27:
5195-5201.
Pastore F, Levine RL. Next-generation sequencing and detection of
minimal residual disease in acute myeloid leukemia: ready for clinical
practice? JAMA. 2015;314:778-780.
Inaba H, Coustan-Smith E, Cao X, et al. Comparative analysis of different approaches to measure treatment response in acute myeloid
leukemia. J Clin Oncol. 2012;30:3625-3632.
Chen X, Xie H, Wood BL, et al. Relation of clinical response and minimal
residual disease and their prognostic impact on outcome in acute
myeloid leukemia. J Clin Oncol. 2015;33:1258-1264.
Mrozek
K, Marcucci G, Paschka P, et al. Clinical relevance of mutations

and gene-expression changes in adult acute myeloid leukemia with
normal cytogenetics: are we ready for a prognostically prioritized
molecular classification? Blood. 2007;109:431-448.
e311
DEANGELO ET AL
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
e312
Patel JP, Gonen M, Figueroa ME, et al. Prognostic relevance of integrated genetic profiling in acute myeloid leukemia. N Engl J Med.
2012;366:1079-1089.
Schlenk RF, Dohner K, Krauter J, et al; German-Austrian Acute Myeloid
Leukemia Study Group. Mutations and treatment outcome in cytogenetically normal acute myeloid leukemia. N Engl J Med. 2008;358:
1909-1918.
Nazha A, Cortes J, Faderl S, et al. Activating internal tandem duplication mutations of the fms-like tyrosine kinase-3 (FLT3-ITD) at
complete response and relapse in patients with acute myeloid leukemia. Haematologica. 2012;97:1242-1245.
Schnittger S, Kern W, Tschulik C, et al. Minimal residual disease levels
assessed by NPM1 mutation-specific RQ-PCR provide important
prognostic information in AML. Blood. 2009;114:2220-2231.
Shayegi N, Kramer M, Bornhauser M, et al; Study Alliance Leukemia
(SAL). The level of residual disease based on mutant NPM1 is an
independent prognostic factor for relapse and survival in AML. Blood.
2013;122:83-92.
Ivey A, Hills RK, Simpson MA, et al; UK National Cancer Research
Institute AML Working Group. Assessment of minimal residual disease
in standard-risk AML. N Engl J Med. 2016;374:422-433.
Corbacioglu A, Scholl C, Schlenk RF, et al. Prognostic impact of minimal
residual disease in CBFB-MYH11-positive acute myeloid leukemia.
J Clin Oncol. 2010;28:3724-3729.
Krauter J, Gorlich K, Ottmann O, et al. Prognostic value of minimal
residual disease quantification by real-time reverse transcriptase
polymerase chain reaction in patients with core binding factor leukemias. J Clin Oncol. 2003;21:4413-4422.
Yin JA, OBrien MA, Hills RK, et al. Minimal residual disease monitoring
by quantitative RT-PCR in core binding factor AML allows risk stratification and predicts relapse: results of the United Kingdom MRC
AML-15 trial. Blood. 2012;120:2826-2835.
Zhu HH, Zhang XH, Qin YZ, et al. MRD-directed risk stratification
treatment may improve outcomes of t(8;21) AML in the first complete
remission: results from the AML05 multicenter trial. Blood. 2013;121:
4056-4062.
96.
97.
98.
99.
100.
101.
102.
103.
104.
105.
Paietta E. Minimal residual disease in acute myeloid leukemia: coming

of age. Hematology Am Soc Hematol Educ Program. 2012;2012:35-42.
San Miguel JF, Martnez A, Macedo A, et al. Immunophenotyping investigation of minimal residual disease is a useful approach for predicting relapse in acute myeloid leukemia patients. Blood. 1997;90:
2465-2470.
Venditti A, Buccisano F, Del Poeta G, et al. Level of minimal residual
disease after consolidation therapy predicts outcome in acute myeloid
leukemia. Blood. 2000;96:3948-3952.
Rubnitz JE, Inaba H, Dahl G, et al. Minimal residual disease-directed
therapy for childhood acute myeloid leukaemia: results of the AML02
multicentre trial. Lancet Oncol. 2010;11:543-552.
Langebrake C, Creutzig U, Dworzak M, et al; MRD-AML-BFM Study
Group. Residual disease monitoring in childhood acute myeloid
leukemia by multiparameter flow cytometry: the MRD-AML-BFM
Study Group. J Clin Oncol. 2006;24:3686-3692.
Terwijn M, van Putten WL, Kelder A, et al. High prognostic impact of
flow cytometric minimal residual disease detection in acute myeloid
leukemia: data from the HOVON/SAKK AML 42A study. J Clin Oncol.
2013;31:3889-3897.
Freeman SD, Virgo P, Couzens S, et al. Prognostic relevance of
treatment response measured by flow cytometric residual disease
detection in older patients with acute myeloid leukemia. J Clin Oncol.
2013;31:4123-4131.
Walter RB, Gooley TA, Wood BL, et al. Impact of pretransplantation
minimal residual disease, as detected by multiparametric flow
cytometry, on outcome of myeloablative hematopoietic cell transplantation for acute myeloid leukemia. J Clin Oncol. 2011;29:
1190-1197.
Walter RB, Buckley SA, Pagel JM, et al. Significance of minimal residual
disease before myeloablative allogeneic hematopoietic cell transplantation for AML in first and second complete remission. Blood.
2013;122:1813-1821.
Araki D, Wood BL, Othus M, et al. Allogeneic hematopoietic cell
transplantation for acute myeloid leukemia: time to move toward a
minimal residual disease-based definition of complete remission?
J Clin Oncol. 2016;34:329-336.
ALLOTRANSPLANT
Molecularly and Phenotypically

Defined Subtypes of Acute
Lymphoblastic Leukemia:
Implications for Management
CHAIR
Charles G. Mullighan, MD, MBBS, MSc
Memphis, TN
SPEAKERS
Sabina Chiaretti, MD, PhD
Sapienza University of Rome
Rome, Italy
Susan M. OBrien, MD
University of California, Irvine
Irvine, CA
CHIARETTI ET AL
Advances in the Genetics and Therapy of Acute Lymphoblastic

Leukemia
Sabina Chiaretti, MD, PhD, Valentina Gianfelici, MD, PhD, Susan M. OBrien, MD, and
Charles G. Mullighan, MBBS, MD
OVERVIEW
Acute lymphoblastic leukemia (ALL) remains an important cause of morbidity in children and adults. In this article, we
highlight advances in the genetics and therapy of three key subtypes of ALL: T-cell ALL, BCR-ABL1 (Philadelphia [Ph]
chromosonepositive), and Ph-like ALL. T-ALL is an aggressive disease that accounts for about 15% and 25% of ALL among
pediatric and adult cohorts, respectively, and exhibits a multistep nature of cancer initiation and progression. The integration
of cytogenetics, molecular biology, and immunophenotype analyses has led to the identification of defined T-ALL subgroups,
such as early T-cell precursor ALL and novel lesions with a prognostic role, for which specific inhibitors are being developed.
Phpositive ALL was historically regarded as a subtype of ALL with a poor prognosis, and allogeneic stem cell transplant was
recommended for all patients who could undergo this procedure. The deep complete responses seen with combination
tyrosine kinase inhibitors (TKIs) and chemotherapy in Ph-positive ALL, and the reports of long-term survival among some
patients not undergoing allogeneic stem cell transplant, has raised the question of whether there is a subset of patients who
could be cured without this intervention. Ph-like ALL is a subtype of B-progenitor ALL common among older children and
adults and associated with a diverse range of genetic alterations that activate kinase signaling. Ph-like ALL is also associated
with poor outcome, for which precision medicine trials identifying kinase alterations and testing TKI therapy are being
developed.
-cell ALL (T-ALL) is a genetically heterogeneous disease

that is caused by the accumulation of lesions acting in a
multistep pathogenic process involving cell growth, proliferation, survival, and differentiation during thymocyte
development.1 It accounts for about 15% and 25% of ALL
among pediatric and adult cohorts, respectively, and is
slightly more frequent in males than females, particularly
in older children and adolescents.2 As for other ALL subtypes, prognosis is far superior for children than adults;
however, it must be emphasized that the outcome of
these patients is now similar to that of patients with
B-cell ALL (B-ALL) because of the use of more intensive
treatments.
Until recently, there was a limited understanding of the
genetic basis of T-ALL. The improvement of cytogenetic
assays and integration with techniques including gene expression profiling, single nucleotide polymorphism microarray analysis, and next-generation sequencing (NGS) have
enabled detailed characterization of the genomic complexity
of T-ALL. The most frequent lesions can be categorized as
chromosomal translocations, duplications and deletions of

DNA, deregulated gene expression, and mutations.
The most frequent translocations involve the 14q11 (T-cell
receptor alpha and delta, TRA and TRD) and 7q34 (TRB) regions, juxtaposing the T-cell receptor (TCR) genes to pivotal
transcription factor genes, such as TAL1, TAL2, LYL1, OLIG2
(BHLHB1), LMO1, LMO2, TLX1, TLX3, NKX2-1, NKX2-2, NKX2-5,
HOXA genes, MYC, and MYB. Further improvement of cytogenetic analysis and the advent of NGS will likely result in
identification of additional partners of rearrangement to TCR
genes. Chromosomal rearrangements can also lead to fusion
genes, among which the most important from a clinical
standpoint are represented by those involving ABL1, mostly
represented by NUP214-ABL1, EML1-ABL1, and ETV6-ABL1
rearrangement, for which the addition of TKIs may be of
benefit, as well as those involving KMT2A (MLL; MLLrearranged). Finally, DNA copy number alterations are identified including deletion of CDKN2A/B and duplication of MYB.3
Gene expression profiling studies have refined the characterization of molecular groups of T-ALL, which are reflective
From the Department of Cellular Biotechnologies and Hematology, Sapienza University of Rome, Rome, Italy; Chao Family Comprehensive Cancer Center, School of Medicine, University
of California, Irvine, CA; Department of Pathology, St. Jude Childrens Research Hospital, Memphis, TN.
Corresponding author: Charles G. Mullighan, MBBS, MD, St. Jude Childrens Research Hospital, 252 Danny Thomas Place, MS 342, Memphis, TN 38105; email: charles.mullighan@
stjude.org.
e314
GENETICS AND THERAPY OF ACUTE LYMPHOLASTIC LEUKEMIA
of a maturational arrest and are characterized by the overexpression of known transcription factors (TLX1, TLX3, TAL1,
LMO1, HOXA genes, NKX genes, LYL1, and MEF2C) and, more
importantly, to identify novel subgroups (early T-cell precursor ALL, see below).4-7
The most frequent mutations are those in NOTCH1 and
FBXW7, in genes involved in the JAK/STAT and Ras/PI3K/AKT
pathways, in epigenetic regulators (PHF6, SUZ12, EZH2,
TET2, H3F3A, and KDM6A), in transcription factors or regulators of transcription (i.e., LEF1, WT1, BCL11B) and genes
involved in mRNA maturation and ribosome activity
(i.e., CNOT3, RPL5, and RPL10). Furthermore, at relapse,
mutations of NT5C2 known to confer resistance to chemotherapy with 6-mercaptopurine and 6-thioguanine are
common.8,9
NOTCH1 signaling is critical for T-cell differentiation, and
mutations in this gene are present in up to 60% of T-ALL
cases. In the same pathway, mutations in FBXW7, which
increase NOTCH1 stability, can be detected in about 20% of
cases. Although a favorable outcome has been overall reported for NOTCH1 mutations, the scenario is more complex
if concomitant lesions are identified. Indeed, a comprehensive prognostic model has been recently proposed by the
GRAAL group10 that defined low-risk patients as those harboring NOTCH1 and FBXW7 mutations and high-risk patients
as those without these mutations or harboring lesions involving Ras/PTEN.
Similarly, the PI3K/PTEN/AKT/mTOR pathway is critical for
cell metabolism, proliferation, and survival. Several mechanisms can lead to its deregulation: mutations or deletions in
PTEN have been reported in around 12% of adults and 13%
to 22% of children, whereas PI3K and AKT mutations have
KEY POINTS
T-cell acute lymphoblastic leukemia is an aggressive

disease that accounts for about 15% and 25% of ALL
among pediatric and adult cohorts, respectively.
Recent scientific advances have allowed identification of
T-ALL subgroups such as early T-cell precursor ALL and
novel lesions with a prognostic role.
Ph-positive ALL was historically regarded as a subtype of
ALL with a poor prognosis, and allogeneic stem cell
transplant was recommended for all patients who could
undergo this procedure.
The deep complete responses seen with combination
tyrosine kinase inhibitors and chemotherapy in Phpositive ALL, and the reports of long-term survival in
some patients not undergoing allogeneic stem cell
transplant, has raised the question of whether there is a
subset of patients who could be cured without this
intervention.
Ph-like ALL, common among older children and adults, is
associated with poor outcome and is characterized by
a range of genetic alterations activating kinase signaling
that are sensitive to TKI therapy.
been described in rare cases. Moreover, NRAS/KRAS mutations are found in around 10% of adults and 4% to 10% of
children.10-14 Finally, interleukin (IL)-4 and IL-7 may upregulate PI3K/AKT/mTOR signaling in T-ALL.15 The prognostic
role of PTEN and NRAS/KRAS alterations is still under investigation and may differ between children and adults.
Among children, the presence of inactivating PTEN lesions,
as well as NRAS/KRAS mutations, do not affect outcome,11,14
whereas among adults their presence is associated with
poorer prognosis.10,12
The JAK/STAT pathway is also often aberrantly activated in
T-ALL. Activating mutations are found mainly in IL7R, JAK1,
JAK3, and STAT5B; moreover, inactivating mutations and
deletions have been described in PTPRC and PTPN2, which
encode phosphatases that regulate the activation of JAK1.16,17
The prognostic role of these lesions is still debated; JAK1
mutations have been associated with chemotherapy refractoriness but with contrasting results. Similarly, the prognostic role of JAK3 mutations needs to be more extensively
assessed.18-21 Finally, STAT5B mutations have been correlated
with an increased risk of recurrence in pediatric cases.22,23
Given this complex genomic scenario, several targeted
approaches have been developed and might be used in the
context of a personalized medicine.
NOTCH1 inhibitors represent an attractive therapeutic
target, in light of the high incidence of mutations. g-secretase
inhibitors (GSI), in use for patients with Alzheimer disease, were the first compounds developed for T-ALL.
However, their side effects, particularly at the gastrointestinal level, made their use not feasible24; these effects
might be overcome by the concomitant administration of
dexamethasone.25
Similarly, PI3K/PTEN/AKT/mTOR inhibitors are appealing
agents. Among them, rapamycin has been extensively investigated. Although its use as a single agent can be limited
by the prolonged inhibition of TORC1 and activation of alternative pathways, a synergistic effect between rapamycin
(or its derivatives) and various chemotherapeutic agents
(i.e., idarubicin, doxorubicin, and dexamethasone) have
been described.26 Therefore, several clinical trials are currently evaluating the effectiveness of the combination of
rapamycin derivatives with chemotherapy among pediatric
patients with relapsed ALL. In particular, in the phase I
RAD001 study, everolimus is combined with reinduction
chemotherapy for pediatric patients with relapsed ALL
(NCT01523977). Furthermore, a second phase I study
designed to assess the effectiveness of CCI-779 (temsirolimus)
among children whose disease has relapsed has recently
concluded, with results not published (NCT01403415).
The novel dual PI3K/mTOR inhibitor NVP-BEZ235 is an
orally bioavailable imidazoquinoline derivative, which exerts antiproliferative and proapoptotic effects in several
T-ALL cell lines.27 Similarly, NVP-BKM120 is a proapoptotic
pan-PI3K inhibitor. Currently, two ongoing clinical trials are
evaluating the effectiveness of BEZ235 and NVP-BKM120
among patients with relapsed or refractory acute leukemia
(NCT01756118 and NCT01396499, respectively).
e315
CHIARETTI ET AL
A variety of JAK inhibitors have been developed. The

JAK1/JAK2-selective inhibitor ruxolitinib is already approved
by the U.S. Food and Drug Administration (FDA) for the
treatment of patients with myelofibrosis, and the JAK3selective inhibitor tofacitinib received FDA approval for
the treatment of patients with rheumatoid arthritis. These
data demonstrate that JAK kinase inhibitors can be administered safely and open new possibilities for the treatment of T-ALL with IL7R, JAK1, or JAK3 mutations. In vitro
studies have shown the efficacy of ruxolitinib in reducing the
proliferation of cells that express mutations in JAK1, IL7R,
and PTPRC.16 Furthermore, Degryse et al showed that
treatment with the JAK3-selective inhibitor tofacitinib reduced the white blood cell count and caused leukemic cell
apoptosis in leukemic mice that expressed JAK3 mutant
variants.28
Finally, BCL-2 antagonists, and in particular ABT-199,
which have proven effective in chronic lymphocytic leukemia, might also represent a therapeutic option in T-ALL;
a specific effect has been demonstrated in TLX3- or HOXApositive primary T-ALL and in more immature T-ALL,29 as well
as among patients who harbor STAT5B mutations.22
Lastly, the concomitance of different alterations suggests
that the combination of specific inhibitors that target different pathways might prove useful for these patients and
minimize the risk of resistance to treatment with single
agents.
EARLY T-CELL PRECURSOR ACUTE

LYMPHOBLASTIC LEUKEMIA
A distinct subgroup of T-ALL is represented by early T-cell
precursor ALL.6 This subset, initially identified by gene expression profiling, can be easily recognized by flow
cytometry because it is characterized by distinct cell surface
features: absence of CD1a and CD8, weak CD5 expression,
and the expression of one or more myeloid or stem cell
associated markers. Several genomic lesions have been
identified, including a low frequency of NOTCH1 mutations,
and mutations in DNMT3A, FLT3, IDH1, IDH2, and ETV6 have
been reported. Interestingly, FLT3 mutations can be detected in up to 35% of cases, thus implying the possibility of
novel therapeutic strategies.30 Furthermore, mutations
occurring in genes that regulate cytokine receptor and Ras
signaling (67%), inactivating lesions that disrupt hematopoietic development (58%), and histone-modifying genes
(48%) have been reported, suggesting that early T-cell
precursor ALL shares a similar genomic background with
acute myeloid leukemia.
From a clinical standpoint, early T-cell precursor ALL was
initially associated with a dismal outcome because of its poor
response to chemotherapy and high rate of resistance or
early relapse.6 Notably, recent studies have reported poor
outcomes in cases with genomic rearrangements associated
with HOXA deregulation. Bond et al analyzed patients enrolled in the GRAAL-2003 and -2005 studies and showed that
patients with HOXA-positive disease had higher rates of an
e316
early T-cell precursorlike immunophenotype. Strikingly, the

presence of an early T-cell precursorlike immunophenotype
conferred marked differences in outcome within the HOXApositive group, which were mirrored by corresponding increases in cumulative incidence of relapse. In contrast, these
survival differences were not seen among patients with
HOXA-negative disease, where early T-cell precursor and nonearly T-cell precursor cases had similar 5-year event-free
survival (EFS) and cumulative incidence of relapse.31 Similarly, Matlawska-Wasowska analyzed a cohort of pediatric
and young adults and found overlap between MLL-rearranged
and the early T-cell precursor ALL cases, and confirmed that
both are risk features for induction failure.32
Overall, these findings lead to some important considerations. The prompt recognition of early T-cell precursor
cases is improving outcome because patients are rapidly
directed toward intensive treatments, including allogeneic
transplant during the first complete response (CR). In contrast, refined molecular characterization is permitting further stratification of these cases and recognition of very
high-risk cases. Finally, the presence of acute myeloid
leukemiarelated features prompts investigation of the
use of myeloid-directed therapies.
In conclusion, extensive research has refined the genomic
profile of T-ALL. In fact, improvement in cytogenetic and molecular analysis and the integration with immunophenotype
have led to the identification of different subgroups and
novel lesions that result in deregulated pathways. This effort
paves the way toward a personalized approach that can
include targeted therapies, depending on the underlying
molecular lesion(s), and possibly should include a combined
approach with conventional therapy and personalized targeted approaches.
PHILADELPHIA CHROMOSOMEPOSITIVE ALL:

TRANSPLANT OR NO TRANSPLANT
Philadelphia chromosomepositive ALL was historically
considered one of the most high risk forms of ALL. Although
CR rates with induction chemotherapy were reasonable,
relapse was almost inevitable, and a key strategy was to get
patients to undergo allogeneic stem cell transplant (SCT),
which was considered the only potentially curative option.
Given the age of patients with ALL, in particular Ph-positive
ALL (which increases with each decade of life), many patients
were unable to be undergo this potentially curative therapy.33
Approximately 30% of patients were able to receive a SCT in
the past; in addition to older age, other contraindications to
SCT included lack of remission, early relapse, or poor performance status from complications due to chemotherapy.34
Thus, long-term survival rates were quite poor.
The advent of TKI therapy has dramatically improved
outcomes for patients with Ph-positive ALL. Complete responses are substantially deeper than those achieved with
chemotherapy alone, as reflected in the high rate of major
molecular response (MMR) or complete molecular response
(CMR) after a combination of TKI and chemotherapy.35 The
fact that CMR can occur has raised the question of whether
some patients can be cured without undergoing SCT. Small
cohorts of patients who were not candidates for SCT and
were treated with chemotherapy and TKIs have provided
some information on long-term outcomes without SCT.
No randomized trials have addressed this issue, and
all of the published clinical trials regarding outcomes in
Ph-positive ALL include variable transplant approaches
(myeloablative versus reduced-intensity conditioning) and
choice, dose, and duration of TKI. Post-transplant maintenance is an additional variable. Most data involve a combination of imatinib and chemotherapy because imatinib
was the first available TKI. There are also some data that
autologous SCT may be beneficial in this setting. Autologous
SCT has not been shown to be an improvement over chemotherapy in the overall treatment of ALL, but one of the
presumed reasons is the lack of true minimal residual disease (MRD) among patients prior to undergoing autologous
SCT. With a TKI and chemotherapy combination, this low
level or absence of disease is possible.
Bassan describes the northern Italy experience with 94 patients with Ph-positive ALL who received 600 mg of imatinib
daily with chemotherapy; 31 patients did not receive imatinib and constituted a control cohort.36 Patients were to
undergo myeloablative allogeneic SCT (if related or unrelated donor was available) or autologous SCT followed by
imatinib maintenance. Nine patients, including four who had
not received imatinib, underwent autologous SCT and further imatinib therapy, and their outcomes were similar to
patients undergoing allogeneic SCT. A multivariate model
suggested that imatinib and SCT favorably affected overall
survival (OS), but SCT included patients undergoing autologous SCT. A confounding element to the data is that
imatinib and/or dasatinib was administered to patients with
persistent or increasing partial CR after SCT.
The PETHEMA and GETH groups37 reported results from
30 patients who received 400 mg of imatinib daily and chemotherapy. Sixteen patients underwent allogeneic (primarily
myeloablative) or autologous SCT. Interestingly, of the five
patients who underwent autologous SCT, four were MRD
negative. The only patient who was MRD positive experienced
relapse; none of the other four patients relapsed. Of the
patients who received imatinib maintenance, two were alive
at 40 and 60 months into CR. In a Saudi Arabian analysis of Phpositive pediatric ALL, SCT was limited to only children with
related donors.38 Twelve patients underwent allogeneic SCT
and 10 were treated with chemotherapy and imatinib. After a
median follow-up of 42.2 months, there was no statistical
difference in EFS or OS. Interestingly, two children who relapsed after SCT responded to TKI and were still in CR.
Long-term follow-up of the GRAAPH-2003 trial was recently published and involved 45 patients with Ph-positive
ALL.39 Notably, postremission therapy was partly decided by
MRD results. Imatinib doses of 600 to 800 mg per day were
given with consolidation chemotherapy and improved OS to
52% compared with 20% prior to imatinib. Overall survival
was 50% after myeloablative allogeneic SCT (24 patients),
33% for patients without a SCT (nine patients), and 80% after
autologous SCT (10 patients). The favorable outcomes with
autologous SCT may also have been related to the requirement that patients have a marrow MRD ratio of less
than 10e-4. Seven patients had a CMR. The three patients
who did not achieve a CMR experienced disease relapse
after SCT. Imatinib was given to four patients after SCT; three
were alive in first CR at 37, 41, and 46 months. Treatmentrelated mortality was particularly high for patients who
underwent a matched unrelated donor SCT, and the authors
noted that whether autologous SCT with a low or negative
MRD level should be performed instead of allogeneic SCT
using an HLA-matched unrelated donor in patients with
Ph-positive ALL is a critical issue that should further be
addressed through a prospective controlled study.
Another trial that utilized MRD status to guide choice of
therapy was recently published. The GRAALL group randomly assigned 268 patients with Ph-positive ALL to 800 mg
of imatinib therapy with either reduced-intensity chemotherapy (arm A) or hyperfractionated cyclophosphamide plus
vincristine, doxorubicin, and dexamethasone (hyperCVAD)
therapy (arm B).40 Patients then underwent allogeneic SCT
(myeloablative, later amended to allow reduced-intensity
conditioning) if they had a related or 9/10 or 10/10 HLAcompatible unrelated donor. Other patients who experienced MMR after two cycles (MMR2) were to undergo
autologous SCT. Thirty-nine patients were eligible for autologous SCT and 28 of those patients underwent SCT during
first remission. Among patients who achieved an MMR2, the
outcome was similar for autologous and allogeneic SCT.
Allogeneic SCT was as effective for patients who did not
experience early MMR as for those who did. Of note, patients who underwent autologous SCT received maintenance with a TKI, which was not planned for the allogeneic
SCT cohort. This again speaks to the question of the benefit
of autologous SCT compared with just continued TKI therapy. The authors concluded that the data strongly suggests that favorable patients with low white blood cell
count and/or those with good early MRD response could be
treated with nonallogeneic postremission therapies.
The UKALLXII/ECOG 299 is the largest prospective trial that
assessed the role of imatinib in the treatment of Ph-positive
ALL.41 The analysis compared outcomes of a large group of
patients treated between 1993 and 2003 (pre-imatinib) with
patients who received early or late imatinib (480 to 600 mg).
Both the CR and OS rates were better in the imatinib cohort.
The improvement in OS was derived partially from the use of
imatinib that enabled a higher percentage of patients to
undergo allogeneic SCT. There was also a modest benefit to
imatinib in terms of EFS. Again, a small number of five
patients underwent autologous SCT. There was one early
death and one relapse. The other three patients were alive in
CR at 3.5, 8, and 8 years. The conclusion was that the data did
not support the omission of allogeneic SCT from the
treatment of Ph-positive ALL, but that limited data on the
use of reduced-intensity conditioning suggested that it could
be beneficial. No attempt was made to allocate patients
e317
CHIARETTI ET AL
depending on their MRD status, and there was no analysis of

the relationship of MRD status to outcome.
The issue of the need for allogeneic SCT is also complicated
by the fact that most trials did not use TKI maintenance after
SCT. Recent data suggest that TKI maintenance after allogenic SCT can markedly improve outcomes. Pfeifer et al
performed a randomized trial of prophylactic or MRDtriggered imatinib after allogeneic SCT.42 Five-year survival was high in both groups and substantially higher than in
their previous study in which the interval between SCT and
imatinib was 5 months compared with less than 2.5 months.
Early occurrence of MRD positivity or positivity at higher
levels identified a subset of patients who did not benefit
from imatinib; whether second-generation TKIs would
produce a better outcome in this group is not known.
In addition to using MRD as a potential decision-making
tool for the use of SCT, another question is: are there
prognostic factors beyond the presence of Ph positivity that
further define subgroups? Mullighan et al published data
that showed a high percentage of children with Ph-positive
ALL had a deletion in the Ikaros (IKZF1) gene.43 Kim et al
investigated its relevance among 118 adult patients with
Ph-positive ALL who received imatinib-based chemotherapy
and underwent allogeneic SCT (myeloablative or reducedintensity conditioning depending on age).44 IKZF1 deletions
were detected among 78.8% of patients and such patients
had a trend for a higher rate of relapse. However, in multivariate analysis, the MRD kinetics were most closely related to outcomes, whereas neither IKZF1 deletions nor their
functional subtypes were independently important.
Dose intensity of imatinib and the effect on outcomes has
not been well-explored in the setting of Ph-positive ALL. Lim
et al intended to administer 600 mg per day of imatinib to all
patients starting from day 8 of induction chemotherapy
through remission induction and consolidation chemotherapy.45 For patients not proceeding to SCT, a further
2 years of imatinib was planned. Lim et al showed that dose
intensity during the initial 7 to 8 weeks strongly correlated
with median CR duration as well as OS. The most common
reason for less dose intensity was gastrointestinal toxicity.
This suggests the choice of chemotherapy in conjunction
with imatinib might also be important.
In summary, the advent of TKIs has dramatically improved
the prognosis of patients with Ph-positive ALL. The goal of
treatment is still to deliver allogeneic SCT when possible.
However, long-term data from cohorts of patients who could
not undergo SCT suggest that a cure faction is possible even
without this intervention. The key question is, how can we
identify patients who may enjoy long-term survival without
the risks of SCT? Examining data across trials is complicated
by the fact that although most trials have used imatinib, the
doses have varied substantially and very little data using
second-generation TKIs exist. In addition, autologous
transplant has been used in a minority of patients, some of
whom have had long-term survival but most of whom also
received imatinib maintenance, raising the question of the
contribution of the autologous SCT versus continued TKI
e318
therapy. The general thinking is that long-term survival is not

possible without achieving MRD negativity. Thus, most
physicians would still recommend SCT for patients who did
not achieve this level of MRD. However, the question of
whether MRD status can be used to select patients for lessintensive therapy has not yet been answered in a randomized clinical trial. Of note, the upcoming Intergroup trial
of patients with Ph-positive ALL will randomly assign patients to SCT or no SCT (for patients who are MRD-negative).
This trial will provide important information about selecting
patients who may experience long-term survival without
undergoing SCT.
PH-LIKE ACUTE LYMPHOBLASTIC LEUKEMIA:

GENETICS, DIAGNOSIS, AND MANAGEMENT
Ph-like ALL is a recently described subtype of B-ALL
characterized by a gene expression profile similar to that
of BCR-ABL1 positive (Ph-positive) ALL, a diverse range of
genetic alterations activating tyrosine kinase signaling,
mutation of lymphoid transcription factor genes such as
IKZF1 (Ikaros), and poor outcome. The high frequency of
kinase activating alterations in this form of ALL and the
potential for targeting these lesions with currently available TKIs have resulted in intense interest in implementing
diagnosis and testing of Ph-like ALL and underlying genetic
alterations in prospective clinical trials.
In 2009, two groups identified childhood ALL cases in which
the gene expression profile of leukemic cells lacking a known
chromosomal rearrangement was similar to that of Ph-positive
ALL.46,47 The Childrens Oncology Group identified a group of
high-risk BCR-ABL1negative B-ALL cases with a high frequency of IKZF1 alteration (which is also a hallmark of
Ph-positive ALL)43 and poor outcome that exhibited similarity of the transcriptomic signature of these cases to that
of Ph-positive ALL using gene set enrichment analysis. The
Dutch used hierarchical clustering of gene expression
profiling data to show clustering of Ph-positive with
Ph-negative cases.47 As there is no single founding genomic
alteration (such as BCR-ABL1) common to all cases, and as
identification rests on demonstrating similarity with the
expression profile of Ph-like ALL, the frequency of Ph-like ALL
varies between studies and the gene expression classifier
used.48 However, multiple studies have shown that Ph-like
ALL is common, particularly with increasing age, and it
is associated with poor outcome and commonly harbors
kinase-activating lesions.
A study of over 1,700 childhood, adolescent (age 16 to 21),
and young adult (age 22 to 39) ALL cases performed by the
Childrens Oncology Group, the St. JudeWashington University Pediatric Cancer Genome Project, and institutions
studying adults with ALL defined the prevalence and genetic
basis of ALL.49 Using gene expression microarrays to define
the signature of Ph-positive ALL and statistical prediction,
Ph-like ALL was shown to be present in approximately 10% of
standard-risk childhood ALL cases, with prevalance increasing by age to over 25% of cases among young adults.49
There are fewer data describing the prevalence of Ph-like ALL

in older adults. Data from the German Multicenter Study
Group for Adult ALL (GMALL) suggested that the frequency
of Ph-like cases peaked among adolescents with a subsequent decline.50 Recent analysis of almost 700 adult ALL
cases from the United States, United Kingdom, and Europe
identified a frequency of 26% among young adults and 20%
among adult patients over the age of 40.51 A similar prevalence among adults was reported in 127 patients enrolled in
Dutch-Belgian HOVON protocols.52 The outcome of Ph-like
ALL is poor in the majority of studies,46,47,49,52-57 particularly
for adolescents and adults for whom EFS and OS rates are
equal or are inferior to other high-risk B-ALL subtypes such as
Ph-positive and MLL-rearranged ALL.52 An exception is the
St. Jude Total Therapy Study XV, in which children with
Ph-like ALL did not have an inferior outcome, although many
were reclassified as high risk due to suboptimal treatment
response (as measured by levels of MRD) and subjected to
more intensive therapy including allogeneic bone marrow
transplantation.58 Importantly, several children undergoing bone marrow transplantation commonly had kinaseactivating lesions targetable with ABL1 inhibitors such as
imatinib and dasatinib.
Multiple chromosomal rearrangements, sequence mutations, and structural genetic alterations activate cytokine
receptor and tyrosine kinase signaling pathways in Ph-like
ALL. Approximately half of cases have rearrangement of
CRLF2 (encoding cytokine receptorlike factor 2), either as a
rearrangement with the immunoglobulin heavy chain locus
(IGH-CRLF2) or a focal deletion upstream of CRLF2 resulting
in expression of a P2RY8-CRLF2 fusion, both of which result
in deregulated expression of intact CRLF2, which may be
detected by flow cytometry.59-61 Less commonly, a point
mutation of CRLF2 is present (p.Phe232Cys) that also
results in receptor activation.62 Approximately half of CRLF2rearranged cases have activating mutations of Janus kinases,
particularly in the pseudokinase domain of JAK2 but also in
JAK1 and JAK3. CRLF2, which forms a heterodimeric receptor
with IL-7 receptor alpha (IL7RA) for the cytokine thymic
stromal lymphopoietin (TSLP), is a scaffold for mutant JAK2,
and the two alterations are cotransforming. CRLF2 cases
lacking JAK2 mutations commonly have other mutations
deregulating JAK-STAT signaling, including mutations of
IL7R15,49 and deletions of SH2B3 (also known as LNK), which
is a negative regulator of JAK-STAT signaling.63,64
Initial sequencing studies of tyrosine kinase genes failed to
identify sequence mutations in Ph-like ALL apart from those
in Janus kinases.54 Subsequent transcriptome sequencing
(RNA-seq) of a small number of Ph-like ALL cases identified a
diverse range of chimeric fusions involving tyrosine cases in
cases lacking CRLF2 rearrangement.65 Transcriptome sequencing of over 150 cases, with whole-genome sequencing
of a subset, provided a detailed understanding of the genetic
alterations driving kinase signaling in Ph-like ALL.58 Cases
lacking CRLF2 rearrangements have alterations deregulating
multiple signaling pathways, including ABL1-class chromosomal rearrangements, JAK2 rearrangements, EPOR
rearrangements, mutations or deletions activating JAK-STAT

or FLT3 signaling, Ras mutations, uncommon kinase alterations (e.g., NTRK3, FGFR1), and, in a proportion of cases, no
known kinase alteration. ABL1-class rearrangements include
those involving ABL1, ABL2 (also known as ABL-related
gene, or ARG), CSF1R (the macrophage CSF receptor), and
PDGFRB.66 In each case, the kinase is the 39 (C-terminal) fusion
partner and rearranged to multiple different 59 (N-terminal)
fusion partners, and the fusion preserves the tyrosine kinase
domain, resulting in activation of signaling. This group of
rearrangements is termed ABL1-class as the transforming
effects of each are abrogated by ABL1 inhibitors such as
imatinib and dasatinib. Cases with EBF1-PDGFRB rearrangement exhibit a particularly high rate of suboptimal treatment
response, induction failure, and relapse.67 Similarly, JAK2
rearrangements involve a highly diverse range of fusion
partners (at least 20), preserve the JAK2 kinase domain,
remove part of or the entire inhibitory pseudokinase domain, and are particularly common with increasing age.
Rearrangements of EPOR (encoding the erythropoietin receptor) involve at least three different immunoglobulin and
immunoglobulin-like genes that result in deregulation of
receptor expression and truncation of the cytoplasmic tail of
the receptor, which results in stabilization of receptor expression and heightened JAK-STAT signaling following stimulation with EPO.68
Several observations indicate that these alterations activate kinase signaling and are amenable to TKI therapy.
ABL1-class and JAK fusions expressed in cell lines result in
proliferation and activation of downstream signaling pathways, both of which are inhibited by TKIs. Human leukemic
cells also exhibit signaling activation sensitive to TKIs; TKIs
synergize with chemotherapeutic agents to inhibit proliferation of human Ph-like ALL cells in vivo in xenograft
models. Ph-like cells also exhibit activation of additional
signaling pathways such as PI3K/mTOR signaling,69 and
combinatorial treatment with PI3K inhibitors and BCL-2
inhibitors with a TKI is efficacious.70-72 Anecdotal reports
also indicate that patients with Ph-like ALL are responsive to
the addition of TKI therapy.49,73-75 Multiple patients with
poorly responsive or refractory B-ALL have been found to
have Ph-like ALL by gene expression profiling and/or testing
for driving kinase rearrangements, and they have frequently
shown dramatic, complete, and durable responses with the
addition of a TKI, particularly for ABL1-class fusions. Importantly, there are few data regarding the use of JAK inhibitors in Ph-like cases with JAK-STAT activating lesions.
Existing preclinical data suggest that Ph-like cases with JAK2
fusions, EPOR rearrangements, and IL7R/SH2B3 alterations
are sensitive to type I JAK inhibitors such as ruxolitinib,
particularly in combination with chemotherapy,49,65,76 but
CRLF2-rearranged cells are less responsive yet sensitive to
type II JAK inhibitors that inhibit JAK2 in its inactive
conformation.77
There is consequently great interest in identifying patients
with Ph-like ALL and defining the underlying genetic lesions
to implement logical TKI therapy. In view of the diverse range
e319
CHIARETTI ET AL
of kinase-activating alterations, which continues to grow,

identification of all such lesions requires NGS-based technology. The most appropriate diagnostic approach used is
dependent on the clinical goal. For example, the majority of
cases with ABL1-class rearrangements can be readily detected with fluorescence in situ hybridization for the kinase
rearrangements, and CRLF2 overexpression arising from
rearrangement can be identified by flow cytometry. However,
identification of all Ph-like cases requires gene expression
profiling using microarray, RNA-seq, or Taqman low-density
array (TLDA) approaches,78 which are informative for research
studies but less suited for real-time diagnostic testing. Several
commercial solutions for nucleic acid capture of kinase loci
and NGS are available (e.g., Archer, Foundation Medicine79),

and it is likely that prospective trials will use these or agnostic
genome sequencing at diagnosis.
In view of the success of ABL1 inhibitors in the management of Ph-positive ALL,80 the use of these agents in ABL1class Ph-like ALL is logical but requires testing in prospective
trials. There is limited experience regarding the safety of
combining JAK inhibition with chemotherapy in ALL, and this
will require evaluation prior to widespread clinical use. Finally, there are less common kinase alterations (TYK2,
NTRK3) for which drugs have not yet been tested in ALL, and
these are an active area of research and drug discovery in
Ph-like ALL.
References
1. Van Vlierberghe P, Ferrando A. The molecular basis of T cell acute
lymphoblastic leukemia. J Clin Invest. 2012;122:3398-3406.
2. Chiaretti S, Vitale A, Cazzaniga G, et al. Clinico-biological features of
5202 patients with acute lymphoblastic leukemia enrolled in the
Italian AIEOP and GIMEMA protocols and stratified in age cohorts.
Haematologica. 2013;98:1702-1710.
3. Durinck K, Goossens S, Peirs S, et al. Novel biological insights in T-cell
acute lymphoblastic leukemia. Exp Hematol. 2015;43:625-639.
4. Ferrando AA, Neuberg DS, Staunton J, et al. Gene expression signatures
define novel oncogenic pathways in T cell acute lymphoblastic leukemia.
Cancer Cell. 2002;1:75-87.
5. Homminga I, Pieters R, Langerak AW, et al. Integrated transcript and
genome analyses reveal NKX2-1 and MEF2C as potential oncogenes in
T cell acute lymphoblastic leukemia. Cancer Cell. 2011;19:484-497.
6. Coustan-Smith E, Mullighan CG, Onciu M, et al. Early T-cell precursor
leukaemia: a subtype of very high-risk acute lymphoblastic leukaemia.
Lancet Oncol. 2009;10:147-156.
7. Chiaretti S, Messina M, Tavolaro S, et al. Gene expression profiling identifies a subset of adult T-cell acute lymphoblastic leukemia with myeloidlike gene features and over-expression of miR-223. Haematologica. 2010;
95:1114-1121.
8. Tzoneva G, Perez-Garcia A, Carpenter Z, et al. Activating mutations in
the NT5C2 nucleotidase gene drive chemotherapy resistance in relapsed ALL. Nat Med. 2013;19:368-371.
9. Vicente C, Cools J. The origin of relapse in pediatric T-cell acute lymphoblastic leukemia. Haematologica. 2015;100:1373-1375.
10. Trinquand A, Tanguy-Schmidt A, Ben Abdelali R, et al. Toward a
NOTCH1/FBXW7/RAS/PTEN-based oncogenetic risk classification of
adult T-cell acute lymphoblastic leukemia: a Group for Research in Adult
Acute Lymphoblastic Leukemia study. J Clin Oncol. 2013;31:4333-4342.
11. Jenkinson S, Koo K, Mansour MR, et al. Impact of NOTCH1/FBXW7
mutations on outcome in pediatric T-cell acute lymphoblastic leukemia
patients treated on the MRC UKALL 2003 trial. Leukemia. 2013;27:
41-47.
12. Beldjord K, Chevret S, Asnafi V, et al; Group for Research on Adult Acute
Lymphoblastic Leukemia (GRAALL). Oncogenetics and minimal residual
disease are independent outcome predictors in adult patients with
acute lymphoblastic leukemia. Blood. 2014;123:3739-3749.
13. Gutierrez A, Sanda T, Grebliunaite R, et al. High frequency of PTEN, PI3K,
and AKT abnormalities in T-cell acute lymphoblastic leukemia. Blood.
2009;114:647-650.
14. Bandapalli OR, Zimmermann M, Kox C, et al. NOTCH1 activation
clinically antagonizes the unfavorable effect of PTEN inactivation in
e320
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
BFM-treated children with precursor T-cell acute lymphoblastic leukemia. Haematologica. 2013;98:928-936.
Zenatti PP, Ribeiro D, Li W, et al. Oncogenic IL7R gain-of-function mutations in childhood T-cell acute lymphoblastic leukemia. Nat Genet.
2011;43:932-939.
Porcu M, Kleppe M, Gianfelici V, et al. Mutation of the receptor tyrosine
phosphatase PTPRC (CD45) in T-cell acute lymphoblastic leukemia.
Blood. 2012;119:4476-4479.
Kleppe M, Lahortiga I, El Chaar T, et al. Deletion of the protein tyrosine
phosphatase gene PTPN2 in T-cell acute lymphoblastic leukemia. Nat
Genet. 2010;42:530-535.
Flex E, Petrangeli V, Stella L, et al. Somatically acquired JAK1 mutations
in adult acute lymphoblastic leukemia. J Exp Med. 2008;205:751-758.
Jeong EG, Kim MS, Nam HK, et al. Somatic mutations of JAK1 and JAK3 in
acute leukemias and solid cancers. Clin Cancer Res. 2008;14:3716-3721.
Asnafi V, Le Noir S, Lhermitte L, et al. JAK1 mutations are not frequent
events in adult T-ALL: a GRAALL study. Br J Haematol. 2010;148:
178-179.
Vicente C, Schwab C, Broux M, et al. Targeted sequencing identifies
associations between IL7R-JAK mutations and epigenetic modulators
in T-cell acute lymphoblastic leukemia. Haematologica. 2015;100:
1301-1310.
Kontro M, Kuusanmaki H, Eldfors S, et al. Novel activating STAT5B
mutations as putative drivers of T-cell acute lymphoblastic leukemia.
Leukemia. 2014;28:1738-1742.
Bandapalli OR, Schuessele S, Kunz JB, et al. The activating STAT5B N642H
mutation is a common abnormality in pediatric T-cell acute lymphoblastic
leukemia and confers a higher risk of relapse. Haematologica. 2014;99:
e188-e192.
DeAngelo DJ, Stone RM, Silverman LB, et al. A phase I clinical trial of the
notch inhibitor MK-0752 in patients with T-cell acute lymphoblastic
leukemia/lymphoma (T-ALL) and other leukemias. J Clin Oncol. 2006;24
(suppl; abstr6585).
Real PJ, Tosello V, Palomero T, et al. Gamma-secretase inhibitors reverse glucocorticoid resistance in T cell acute lymphoblastic leukemia.
Nat Med. 2009;15:50-58.
Avellino R, Romano S, Parasole R, et al. Rapamycin stimulates apoptosis
of childhood acute lymphoblastic leukemia cells. Blood. 2005;106:
1400-1406.
Chiarini F, Grimaldi C, Ricci F, et al. Activity of the novel dual phosphatidylinositol 3-kinase/mammalian target of rapamycin inhibitor
NVP-BEZ235 against T-cell acute lymphoblastic leukemia. Cancer Res.
2010;70:8097-8107.
28. Degryse S, de Bock CE, Cox L, et al. JAK3 mutants transform hematopoietic cells through JAK1 activation, causing T-cell acute lymphoblastic
leukemia in a mouse model. Blood. 2014;124:3092-3100.
29. Peirs S, Matthijssens F, Goossens S, et al. ABT-199 mediated inhibition
of BCL-2 as a novel therapeutic strategy in T-cell acute lymphoblastic
leukemia. Blood. 2014;124:3738-3747.
30. Zhang J, Ding L, Holmfeldt L, et al. The genetic basis of early T-cell
precursor acute lymphoblastic leukaemia. Nature. 2012;481:157-163.
31. Bond J, Marchand T, Touzart A, et al. An early thymic precursor phenotype predicts outcome exclusively in HOXA-overexpressing adult
T-ALL: a GRAALL study. Blood. 2015;126:808.
32. Matlawska-Wasowska K, Kang H, Devidas M, et al. Mixed lineage
leukemia rearrangements (MLL-R) are determinants of high risk disease
in homeobox A (HOXA)-deregulated T-lineage acute lymphoblastic
leukemia: a Childrens Oncology Group study. Blood. 2015;126:694.
33. Chiaretti S, Foà R. Management of adult Ph-positive acute lymphoblastic leukemia. Hematology Am Soc Hematol Educ Program. 2015;
2015:406-413.
34. Fielding AK. Treatment of Philadelphia chromosome-positive acute
lymphoblastic leukemia in adults: a broader range of options, improved
outcomes, and more therapeutic dilemmas. Am Soc Clin Oncol Educ
Book. 2015;35:e352-e359.
35. Daver N, Thomas D, Ravandi F, et al. Final report of a phase II study of
imatinib mesylate with hyper-CVAD for the front-line treatment of adult
patients with Philadelphia chromosome-positive acute lymphoblastic
leukemia. Haematologica. 2015;100:653-661.
36. Bassan R, Rossi G, Pogliani EM, et al. Chemotherapy-phased imatinib
pulses improve long-term outcome of adult patients with Philadelphia
chromosome-positive acute lymphoblastic leukemia: Northern Italy
Leukemia Group protocol 09/00. J Clin Oncol. 2010;28:3644-3652.
37. Ribera JM, Oriol A, Gonzalez M, et al; Programa Espan~ol de Tratamiento
~ol de Trasplante Hemopoyetico Groups.
en Hematologa; Grupo Espan
Concurrent intensive chemotherapy and imatinib before and after stem
cell transplantation in newly diagnosed Philadelphia chromosomepositive acute lymphoblastic leukemia. Final results of the CSTIBES02
trial. Haematologica. 2010;95:87-95.
38. Salami K, Alkayed K, Halalsheh H, et al. Hematopoietic stem cell
transplant versus chemotherapy plus tyrosine kinase inhibitor in the
treatment of pediatric Philadelphia chromosome-positive acute lymphoblastic leukemia (ALL). Hematol Oncol Stem Cell Ther. 2013;6:34-41.
39. Tanguy-Schmidt A, Rousselot P, Chalandon Y, et al. Long-term follow-up
of the imatinib GRAAPH-2003 study in newly diagnosed patients with de
novo Philadelphia chromosome-positive acute lymphoblastic leukemia: a GRAALL study. Biol Blood Marrow Transplant. 2013;19:150-155.
40. Chalandon Y, Thomas X, Hayette S, et al; Group for Research on
Adult Acute Lymphoblastic Leukemia (GRAALL). Randomized study of
reduced-intensity chemotherapy combined with imatinib in adults with
Ph-positive acute lymphoblastic leukemia. Blood. 2015;125:3711-3719.
41. Fielding AK, Rowe JM, Buck G, et al. UKALLXII/ECOG2993: addition of
imatinib to a standard treatment regimen enhances long-term outcomes in Philadelphia positive acute lymphoblastic leukemia. Blood.
2014;123:843-850.
42. Pfeifer H, Wassmann B, Bethge W, et al; GMALL Study Group.
Randomized comparison of prophylactic and minimal residual
disease-triggered imatinib after allogeneic stem cell transplantation
for BCR-ABL1-positive acute lymphoblastic leukemia. Leukemia.
2013;27:1254-1262.
43. Mullighan CG, Miller CB, Radtke I, et al. BCR-ABL1 lymphoblastic leukaemia is characterized by the deletion of Ikaros. Nature. 2008;453:
110-114.
44. Kim M, Park J, Kim DW, et al. Impact of IKZF1 deletions on long-term
outcomes of allo-SCT following imatinib-based chemotherapy in adult
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
Philadelphia chromosome-positive ALL. Bone Marrow Transplant.

2015;50:354-362.
Lim SN, Joo YD, Lee KH, et al. Long-term follow-up of imatinib plus
combination chemotherapy in patients with newly diagnosed Philadelphia
chromosome-positive acute lymphoblastic leukemia. Am J Hematol. 2015;
90:1013-1020.
Mullighan CG, Su X, Zhang J, et al; Childrens Oncology Group. Deletion
of IKZF1 and prognosis in acute lymphoblastic leukemia. N Engl J Med.
2009;360:470-480.
Den Boer ML, van Slegtenhorst M, De Menezes RX, et al. A subtype of
childhood acute lymphoblastic leukaemia with poor treatment outcome:
a genome-wide classification study. Lancet Oncol. 2009;10:125-134.
Boer JM, Marchante JR, Evans WE, et al. BCR-ABL1-like cases in pediatric
acute lymphoblastic leukemia: a comparison between DCOG/Erasmus
MC and COG/St. Jude signatures. Haematologica. 2015;100:e354-e357.
Roberts KG, Li Y, Payne-Turner D, et al. Targetable kinase-activating
lesions in Ph-like acute lymphoblastic leukemia. N Engl J Med. 2014;371:
1005-1015.
Herold T, Baldus CD, Gokbuget N. Ph-like acute lymphoblastic leukemia
in older adults. N Engl J Med. 2014;371:2235.
Roberts KG, Payne-Turner D, McCastlain K, et al. High frequency and
poor outcome of Ph-like acute lymphoblastic leukemia in adults. Blood.
2015;126:2618.
Boer JM, Koenders JE, van der Holt B, et al. Expression profiling of adult
acute lymphoblastic leukemia identifies a BCR-ABL1-like subgroup
characterized by high non-response and relapse rates. Haematologica.
2015;100:e261-e264.
van der Veer A, Waanders E, Pieters R, et al. Independent prognostic
value of BCR-ABL1-like signature and IKZF1 deletion, but not high CRLF2
expression, in children with B-cell precursor ALL. Blood. 2013;122:
2622-2629.
Loh ML, Zhang J, Harvey RC, et al. Tyrosine kinome sequencing of
pediatric acute lymphoblastic leukemia: a report from the Childrens
Oncology Group TARGET Project. Blood. 2013;121:485-488.
Kiyokawa N, Iijima K, Yoshihara H, et al. An analysis of Ph-like ALL in
Japanese patients. Blood. 2013;122:352 (suppl; abstr 352).
Te Kronnie G, Silvestri D, Vendramini E, et al. Philadelphia-like signature
in childhood acute lymphoblastic leukemia: the AEIOP experience.
Blood. 2013;122:353 (suppl; abstr 353).
Stock W, Luger SM, Advani AS, et al. Favorable outcomes for older
adolescents and young adults (AYA) with acute lymphoblastic leukemia
(ALL): early results of U.S. Intergroup Trial C10403. Blood. 2014;124:
796.
Roberts KG, Pei D, Campana D, et al. Outcomes of children with BCRABL1like acute lymphoblastic leukemia treated with risk-directed
therapy based on the levels of minimal residual disease. J Clin Oncol.
2014;32:3012-3020.
Russell LJ, Capasso M, Vater I, et al. Deregulated expression of cytokine
receptor gene, CRLF2, is involved in lymphoid transformation in B-cell
precursor acute lymphoblastic leukemia. Blood. 2009;114:2688-2698.
Mullighan CG, Collins-Underwood JR, Phillips LA, et al. Rearrangement
of CRLF2 in B-progenitor- and Down syndrome-associated acute lymphoblastic leukemia. Nat Genet. 2009;41:1243-1246.
Harvey RC, Mullighan CG, Chen IM, et al. Rearrangement of CRLF2 is
associated with mutation of JAK kinases, alteration of IKZF1, Hispanic/
Latino ethnicity, and a poor outcome in pediatric B-progenitor acute
lymphoblastic leukemia. Blood. 2010;115:5312-5321.
Yoda A, Yoda Y, Chiaretti S, et al. Functional screening identifies CRLF2 in
precursor B-cell acute lymphoblastic leukemia. Proc Natl Acad Sci USA.
2010;107:252-257.
Bersenev A, Wu C, Balcerek J, et al. Lnk constrains myeloproliferative
diseases in mice. J Clin Invest. 2010;120:2058-2069.
e321
CHIARETTI ET AL
64. Bersenev A, Wu C, Balcerek J, et al. Lnk controls mouse hematopoietic

stem cell self-renewal and quiescence through direct interactions with
JAK2. J Clin Invest. 2008;118:2832-2844.
65. Roberts KG, Morin RD, Zhang J, et al. Genetic alterations activating
kinase and cytokine receptor signaling in high-risk acute lymphoblastic
leukemia. Cancer Cell. 2012;22:153-166.
66. Kobayashi K, Mitsui K, Ichikawa H, et al. ATF7IP as a novel PDGFRB fusion
partner in acute lymphoblastic leukaemia in children. Br J Haematol.
2014;165:836-841.
67. Schwab C, Ryan SL, Chilton L, et al. EBF1-PDGFRB fusion in paediatric
B-cell precursor acute lymphoblastic leukaemia (BCP-ALL): genetic
profile and clinical implications. Blood. Epub 2016 Feb 12.
68. Iacobucci I, Li Y, Roberts KG, et al. Truncating erythropoietin receptor
rearrangements in acute lymphoblastic leukemia. Cancer Cell. 2016;29:
186-200.
69. Tasian SK, Doral MY, Borowitz MJ, et al. Aberrant STAT5 and PI3K/mTOR
pathway signaling occurs in human CRLF2-rearranged B-precursor
acute lymphoblastic leukemia. Blood. 2012;120:833-842.
70. Shi C, Han L, Zhang Q, et al. Combined targeting of JAK2 with a type II
JAK2 inhibitor and mTOR with a TOR kinase inhibitor constitutes
synthetic activity in JAK2-driven Ph-like acute lymphoblastic leukemia.
Blood. 2015;126:2529.
71. Waibel M, Solomon VS, Knight DA, et al. Combined targeting of JAK2 and
Bcl-2/Bcl-xL to cure mutant JAK2-driven malignancies and overcome
acquired resistance to JAK2 inhibitors. Cell Reports. 2013;5:1047-1059.
72. Maude SL, Tasian SK, Vincent T, et al. Targeting JAK1/2 and mTOR in
murine xenograft models of Ph-like acute lymphoblastic leukemia.
Blood. 2012;120:3510-3518.
e322
73. Weston BW, Hayden MA, Roberts KG, et al. Tyrosine kinase inhibitor
therapy induces remission in a patient with refractory EBF1-PDGFRBpositive acute lymphoblastic leukemia. J Clin Oncol. 2013;31:e413-e416.
74. Lengline E, Beldjord K, Dombret H, et al. Successful tyrosine kinase
inhibitor therapy in a refractory B-cell precursor acute lymphoblastic
leukemia with EBF1-PDGFRB fusion. Haematologica. 2013;98:
e146-e148.
75. Kobayashi K, Miyagawa N, Mitsui K, et al. TKI dasatinib monotherapy
for a patient with Ph-like ALL bearing ATF7IP/PDGFRB translocation.
76. Maude S, Tasian S, Vincent T, et al. Targeting Jak2 and Mtor in xenograft
models of Crlf2-overexpressing acute lymphoblastic leukemia (All).
Pediatr Blood Cancer. 2012;58:1014.
77. Wu SC, Li LS, Kopp N, et al. Activity of the type II JAK2 inhibitor CHZ868 in
b cell acute lymphoblastic leukemia. Cancer Cell. 2015;28:29-41.
78. Harvey RC, Kang HN, Roberts KG, et al. Development and validation of a
highly sensitive and specific gene expression classifier to prospectively
screen and identify B-precursor acute lymphoblastic leukemia (ALL) patients with a Philadelphia chromosome-like (Ph-like or BCR-ABL1-like)
signature for therapeutic targeting and clinical intervention. Blood. 2013;
122:826.
79. He J, Abdel-Wahab O, Nahas MK, et al. Integrated genomic DNA/RNA
profiling of hematologic malignancies in the clinical setting. Blood.
2016. In press.
80. Schultz KR, Carroll A, Heerema NA, et al; Childrens Oncology Group.
Long-term follow-up of imatinib in pediatric Philadelphia chromosomepositive acute lymphoblastic leukemia: Childrens Oncology Group
study AALL0031. Leukemia. 2014;28:1467-1471.
ALLOTRANSPLANT
Progress in Myeloproliferative
Neoplasms: Are We Ready?
CHAIR
Ruben A. Mesa, MD, FACP
Mayo Clinic
Scottsdale, AZ
SPEAKERS
Francesco Passamonti, MD
Fondazione IRCCS Policlinico San Matteo
Pavia, Italy
Jeffrey Tyner, PhD
Oregon Health & Science University
Portland, OR
MESA AND PASSAMONTI
Individualizing Care for Patients With Myeloproliferative

Neoplasms: Integrating Genetics, Evolving Therapies, and
Patient-Specific Disease Burden
Ruben A. Mesa, MD, and Francesco Passamonti, MD
OVERVIEW
Individualized medicine is important for patients with myeloproliferative neoplasms (MPNs), including essential thrombocythemia, polycythemia vera, and myelofibrosis, which are heterogeneous in terms of genetic mutation profile, prognosis,
disease burden, and symptoms. Status of MPN driver mutations in JAK2, CALR, and MPL (or lack of one of these mutations)
and other myeloid mutations (ASXL1, SRSF2, CBL, and IDH1/2, among others) affects diagnosis and prognosis. Management
begins with estimating the prognosis, disease burden including MPN symptoms, and prevention of vascular events. Allogeneic stem cell transplantation is the definitive therapy in a subset of patients with myelofibrosis, the majority of whom
receive JAK inhibition with ruxolitinib to relieve splenomegaly and symptoms and to prolong survival. Ruxolitinib is now a
second-line therapy in polycythemia vera, with pegylated interferon being evaluated as a potential front-line therapy
compared with hydroxyurea. The therapeutic landscape is evolving to include new JAK inhibitors, which may affect
cytopenias (pacritinib and momelotinib), combination therapies including ruxolitinib, and novel targets such as pentraxin
and telomerase. Assessing the therapeutic efficacy (including symptom impact) and toxicity of these new approaches is
necessary to determine longitudinal management of MPNs in clinical practice and is a key component of individualizing
care for patients with MPNs.
yeloproliferative neoplasms account for three main

entities named polycythemia vera, essential thrombocythemia, and primary myelofibrosis.1,2 Polycythemia
vera is characterized by erythrocytosis with some degree of
leukocytosis and thrombocytosis and an overall increase in
bone marrow cellularity with well-defined symptoms.3 The
clinical picture of patients with essential thrombocythemia is
dominated by isolated thrombocytosis.4,5 Polycythemia vera
and essential thrombocythemia share several clinical features, most prominently a high risk of thrombosis and a
predisposition to evolve into acute myeloid leukemia or
secondary myelofibrosis.6-8 Among MPNs, primary myelofibrosis has the most heterogeneous clinical presentation,
including anemia, splenomegaly, leukocytosis or leukopenia,
thrombocytosis or thrombocytopenia, a variably high
symptom burden, and a variability of clinical responses to
therapy.9,10 A study among 1,581 patients with MPNs was
recently conducted at the Mayo Clinic in the United States
and at the University of Florence and Papa Giovanni XXIII
Hospital in Italy; median survival was 19.8 years, 13.5 years,
and 5.9 years for individuals with essential thrombocythemia,
polycythemia vera, and primary myelofibrosis, respectively.11
Classification of MPNs is performed according to the 2008

World Health Organization (WHO) criteria, which are based
on clinical, histopathologic, and molecular findings.12 An
updated WHO classification is expected in 2016, which will
likely include MPN driver mutations (JAK2, CALR, and MPL),
different cell count cutoffs, and an extended indication for
bone marrow biopsy.1
CLINICAL BURDEN OF MPNs: THE EFFECTS OF

MPN SYMPTOMS ON HEALTH-RELATED QUALITY
OF LIFE
Given their variable nature, MPNs have various clinical effects such as variable cytopenias, variable degrees of
splenomegaly, and a constellation of MPN-related symptoms.13 A patients clinical course is also affected by
comorbidities, therapy-related toxicity, and morbidity
owing to prior vascular events. Health-related quality of
life for patients with MPNs is a broader construct that
includes the above-listed features and also considers the
burden of medical care, disease-related financial burden
(e.g., decreased ability to work or medical disability),
and stress or effects on mood as a result of uncertainty
From the Mayo Clinic Cancer Center, Phoenix, AZ; Division of Hematology, Department of Clinical and Experimental Medicine, University of Insubria, Varese, Italy.
Corresponding author: Ruben A. Mesa, MD, Mayo Clinic Cancer Center, 5777 East Mayo Blvd., Phoenix, AZ 85259; email: mesa.ruben@mayo.edu.
e324
INDIVIDUALIZED THERAPY FOR MYELOPROLIFERATIVE NEOPLASMS
about the future (Sidebar). As one quantifies MPNrelated symptom burden and assesses the effects of
therapy and disease course, it is important to consider
that MPN symptom burden and health-related quality
of life are interrelated but distinct concepts. For example, one could theoretically improve a specific MPNassociated symptom such as pruritus; however, if the
agent used to achieve that response led to problematic
toxicity (e.g., gastrointestinal), then the net effect on
health-related quality of life could be adverse despite
symptom-specific improvement.
MPN SYMPTOM BURDEN PLAYS AN IMPORTANT

ROLE IN MORBIDITY AND MORTALITY
Individualized management of MPNs recognizes that MPNs
can have a long disease course. Although MPNs are clonal
hematopoietic malignancies, their course can frequently
exceed a decade, even among some patients with secondary
myelofibrosis. As a result, the cumulative effects of symptom
burden, decreased health-related quality of life, comorbidities, and medication-related toxicity all affect morbidity
and mortality.14 Vascular events early in the course of MPNs
can affect disease morbidity, ranging from very serious
events such as intra-abdominal thrombosis (which can lead
to portal hypertension, esophageal varices, and risks of
hemorrhage) to the obvious implications of stroke and
myocardial infarction or the more subtle effect of MPNs on
cognition, chronic headaches, and erythromelalgia. The
burden of specific symptoms can be debilitating, ranging
from unrelenting pruritus (e.g., lack of relief has led to
KEY POINTS
Individualizing care for patients with MPNs begins with

an accurate assessment of prognosis with current
clinical scoring systems and an estimation of disease
burden.
Genetic mutation testing (driver mutations such as JAK2
V617F, CALR, and MPL) in MPNs is helpful in making the
initial diagnosis, and more extensive mutation testing
(ASXL1, IDH1/2, etc.) is helpful in refining the prognosis
in select clinical scenarios, such as equivocal candidates
for stem cell transplantation.
MPNs have a range of disease-associated symptoms
that arise from the clonal disease, are associated with
pathologic cytokines and inflammation, and directly
contribute to disease morbidity and mortality.
Alleviation of MPNs symptoms, now measurable
through standardized instruments, is an integral part of
assessing therapeutic effects in both clinical trials and
clinical practice.
JAK inhibition is the cornerstone of medical therapy for
problematic myelofibrosis and polycythemia vera.
Expanding investigations with interferons, new JAK
inhibitors, JAK inhibitor combinations, and new targets
are on the horizon.
suicide) to overwhelming fatigue (which can lead to

medical disability and exacerbate mood disturbances).15
The advancing symptom burden seen with secondary
myelofibrosis is not only a problematic feature of the
disease but also likely directly contributes to the mortality
of the disorder. Analysis of causes of death in secondary
myelofibrosis shows that approximately one-third of
patients will succumb to acute myeloid leukemia and
likely mortal complications of cytopenias; however, the
remainder of patients succumb to a variety of causes
exacerbated by the debilitation and hypercatabolic state
caused by the disease (e.g., thrombosis, infection, or
exacerbation of comorbidities such as congestive heart
failure). 14
The identification of symptoms prevalent among patients
with MPNs is not new; symptom burden has long been
recognized as a key clinical feature of MPNs. However, the
degree of symptom burden was first quantified in 2007 in
a web-based survey of more than 1,000 patients, conducted
by the CMPD Education Foundation and Mayo Clinic.13
Identified symptoms included fatigue, night sweats, pruritus, and bone pain, among others.13
There is likely a pathogenetic link between the clonal
myeloid malignancy of an MPN and the patients direct
symptom burden. Part of this link may potentially be related
to inflammatory cytokines. Previous studies demonstrated
that interleukin (IL)-1, IL-2, IL-6, tumor necrosis factor-alpha,
and interferons may have a direct link between these two
areas (particularly in constitutional symptoms such as fevers,
night sweats, and weight loss), as well as between symptoms
and prognostic significance.14 Indeed, increases in inflammatory cytokines were observed in early studies with
ruxolitinib and improvement in these cytokines was seen
with response to therapy.16
ASSESSMENT OF THE MUTATION PROFILE IN

MPNs
The same driver mutations of JAK2, MPL, and CALR genes are
present in polycythemia vera, essential thrombocythemia,
SIDEBAR. Factors Impacting Health-Related

Quality of Life for Patients With
Myeloproliferative Neoplasms
MPN-derived constitutional symptoms (i.e., pruritus, night

sweats)
MPN-derived spleen-oriented symptoms (i.e., discomfort
or pain)
MPN vascular symptoms arising from elevated blood
counts
Toxicity from MPN-directed therapies
Complications from earlier MPN-related events such as
thrombosis
Burden of medical care related to MPN
Financial burden related to illness
Stress related to uncertainty in health
e325
MESA AND PASSAMONTI
and primary myelofibrosis, both in sporadic17-20 and in familial cases.21,22 These patients (10%15%, overall) are referred to as having triple-negative status. Among patients
with triple-negative status, new somatic JAK2 and MPL
variants have been discovered.23
In polycythemia vera, JAK2 mutations cover the whole
mutation profile, with V617F present among 95%97%
of patients4 and exon 12 mutations present among the
remainder.24 Very few patients present with CBL or LNK
mutations. MPL mutations have been identified in approximately 5% of patients with essential thrombocythemia
and in 5%10% of patients with secondary myelofibrosis.25
Mutations in the CALR gene have been mainly identified among
patients with essential thrombocythemia and/or primary
myelofibrosis without JAK2 or MPL mutations,18,26 covering
70% of the mutation profile for patients without JAK2/MPL
mutations.
Detection of JAK2 mutations is critical in the diagnostic
process of erythrocytosis and is a major criterion for diagnosis
of polycythemia vera.6 Regarding JAK2 allele burden, higher
burdens correlate unequivocally with enhanced myelopoiesis
of the bone marrow, leukocytosis, and increasing spleen size,
whereas they inversely correlate with platelet count.7,27 Although allele burden quantification is of interest, it is not
mandatory to diagnose polycythemia vera; a qualitative test
on whole blood leukocytes is a correct approach outside of
the research setting.
Concerning CALR mutations, all of the described mutations
are found mainly in a heterozygous state and are insertions
or deletions (type 1, 52-bp deletion, p.L367fs*46; and type 2,
5-bp TTGTC insertion, p.K385fs*47) in the last exon encoding
the C-terminal amino acids of the CALR protein.18,26
Information on CALR mutation in essential thrombocythemia is summarized as follows.19,28-30 First, 40%50% and
30%40% of patients display type 1 and type 2 variants,
respectively. Compared with mutant JAK2, both variants
were associated with higher platelet and lower hemoglobin
and leukocyte counts. Patients with a CALR mutation have a
lower risk of thrombosis than patients with a JAK2 mutation.
Evolution from essential thrombocythemia to polycythemia
vera seems unusual (even impossible). Patients with a CALR
mutation were classified into the lower-risk groups with
regard to both the standard risk stratification for thrombosis
(older than age 60 and prior thrombosis) and the International Prognostic System in Essential Thrombocythemia (IPSET) score31 for survival.
In the largest collaborative trial including 617 patients with
primary myelofibrosis, individuals with a CALR mutation
had a lower risk of developing anemia, thrombocytopenia,
and marked leukocytosis compared with other subtypes.32
Patients with a CALR mutation also had a lower risk of
thrombosis compared with patients with a JAK2 mutation. In
other studies, CALR mutations had a favorable effect on
survival that was independent of both risk (as determined by
the Dynamic International Prognostic Scoring System
[DIPSS] Plus model) and ASXL1 mutation status.19 A subsequent analysis identified that combined CALR and ASXL1
e326
mutation status was associated with different survival

(CALR+ASXL12 and CALRASXL1+ for best and worse survival, respectively).33 A recent analysis showed that the
favorable prognostic effect of CALR mutations on primary
myelofibrosis might be restricted to only those patients with
type 1 CALR mutations.34 In a recent study of 1,118 patients
with primary myelofibrosis, driver mutations were classified
prognostically in two distinct categories: favorable (type
1/type 1like CALR) and unfavorable (all other mutation
categories).35
On the basis of this information, molecular profiling of
MPN at diagnosis must include analysis of JAK2 (looking for
the V617F point mutation first),7 screening of exon 12
mutations of JAK2 (only in V617F-negative PV),24 and
screening of CALR and MPL mutations (in V617F-negative ET
and primary myelofibrosis).19
IS IT MANDATORY TO STUDY ADDITIONAL

MUTATIONS IN MPNs AT DIAGNOSIS?
Additional mutations of CBL, TET2, SF3B1, SRSF2, DNMT3A,
IDH1/2, EZH2, and ASXL1 genes are variably reported in
MPNs, although mainly among patients with secondary
myelofibrosis.36
A total of 879 patients with secondary myelofibrosis were
studied to determine the individual and combined prognostic relevance of somatic mutations in ASXL1, SRSF2, EZH2,
TET2, DNMT3A, CBL, IDH1, IDH2, MPL, and JAK2.37 Of these,
ASXL1, SRSF2, and EZH2 mutations interindependently
predicted shortened survival. However, only ASXL1 mutations remained significant in the context of the International
Prognostic Scoring System (IPSS) prognostic model.14
Leukemia-free survival was negatively affected by IDH1/2,
SRSF2, and ASXL1 mutations, whereas it was affected by
IDH1 and SRSF2 mutations in a Mayo Clinic cohort.37
Next-generation sequencing may simplify the identification of new additional mutations. By applying this approach,
mutations other than JAK2, CALR, or MPL occur among more
than 80% of patients with primary myelofibrosis, including
those with triple-negative status. A prior study revealed that
the absence of such mutations is independently favorable
for survival, whereas the prognostic effect of their presence
is influenced by ASXL1, CBL, RUNX1, and SRSF2 mutations.38
These mutations are not routinely assessed at the time of
diagnosis in daily clinical practice, although they may be
useful in some clinical situations as discussed above.
WHICH PROGNOSTIC SCORE SHOULD BE USED

TO PREDICT SURVIVAL IN MYELOFIBROSIS?
IPSS and DIPSS Models Based on Clinical Parameters
The IPSS was defined in 2009 on the basis of 1,054 patients
with primary myelofibrosis, excluding post-polycythemia
vera and post-essential thrombocythemia secondary myelofibrosis and prefibrotic primary myelofibrosis.14 Median
survival was 69 months. Parameters affecting the survival
and scoring system are reported in Table 1. For the low,
intermediate-1, intermediate-2, and high-risk categories,
median survival was 135 months, 95 months, 48 months, and

27 months, respectively.
The progressive nature of primary myelofibrosis generated interest in defining new so-called dynamic models, such
as the DIPSS and the DIPSS Plus model. In a nontimedependent analysis (models at diagnosis), patients are
assigned to a risk group on the basis of the assessment of
risk factors at diagnosis and they are followed in the same
category irrespective of the acquisition of other risk factors
during the disease course. According to a dynamic model,
patients contribute to the estimate of survival in a category
only as long as they do not acquire further risk factors, then
they shift to a higher category according to their new
score.
The DIPSS was developed based on a study of 525 patients
with primary myelofibrosis who were regularly followed,
using the same IPSS risk factors but a different scoring
system (Table 1).39 Median survival referring to patients
maintaining the same risk category over time was not
reached in the low-risk category, whereas median survival
was 14.2 years, 4 years, and 1.5 years in the intermediate-1,
intermediate-2, and high-risk categories, respectively.
The DIPSS Plus model was produced as a result of an
analysis of 793 patients with primary myelofibrosis; 428
patients were referred within their first year of diagnosis,
whereas 365 were referred after their first year of diagnosis.40 Unfavorable cytogenetics, red blood cell
transfusion need, and low platelet counts were added to
the DIPSS categories. For the low-, intermediate-1,
intermediate-2, and high-risk categories, median survival was 185 months, 78 months, 35 months, and
16 months, respectively.
TABLE 1. Clinically Driven Prognostic Model in Primary

Myelofibrosis
IPSS
Age > 65
DIPSS
1
Hemoglobin < 10 g/dL
Leukocyte Count
> 25 3 109/L
Blast Cells 1%
Constitutional Symptoms
Unfavorable Cytogenetics*
Red Blood Cell Need
Platelets
< 100 3 109/L
DIPSS Plus
To assess DIPSS status**
Molecular-Based Prognostic Models

On the basis of the IPSS and DIPSS prognostic models and the
available data on mutation platforms, it seemed obvious to
combine different clinical/molecular parameters with the
aim to distinguish patients at higher risk from those at lower
risk. To reinforce the role of mutations in disease prediction,
the additional prognostic value of the number of mutated
genes was taken into account.41 The presence of two or
more mutations predicted the worst outcomes. Median
survival was 2.6 years, 7.0 years, and 12.3 years for patients
with two, one, or no mutations, respectively. The presence
of two or more mutations was also associated with shortened leukemia-free survival.
By combining CALR, JAK2, MPL, triple-negative status, and
each single variable included in the IPSS, a molecular-based
risk model was developed.32 Five risk categories with significantly different survival were identified. Although this
model was based on the three driver mutations and some
IPSS clinical parameters, the Mutation-Enhanced International Prognostic Scoring System (MIPSS)42 also combined additional mutations (ASXL1, SRSF2, EZH2, and
IDH1/2) previously identified by next-generation sequencing.37 Factors included in the MIPSS model are reported in
Table 2. Four risk groups were defined.
The Genetics-Based Prognostic Scoring System prognostic
model was developed to further integrate cytogenetic and
molecular profiles (Table 2). The four risk categories were
associated with different survival, ranging from 2.2 years to
more than 17 years.43 Figure 1 reports how to use this
clinicalmutational-based prognostic model in the clinical
practice.
TABLE 2. Genetically Driven Prognostic Model in

Primary Myelofibrosis
MIPSS
GIPSS
Age > 6
1.5
Constitutional Symptoms
0.5
No
Hemoglobin < 10 g/dL
0.5
No
Platelets < 200 3 10 /L
1.0
No
Triple Negative
1.5
JAK2 or MPL Mutation
0.5
ASXL1 Mutation
0.5
No
No
SRSF2 Mutation
0.5
No
No
CALR Type 2, Type 2Like
No
No
No
Unfavorable Cytogenetics*
No
3 for very high risk;

2 for high risk
Data are presented as points values unless otherwise indicated. IPSS categories are as
follows: low (score 0), intermediate-1 (score 1), intermediate-2 (score 2), and high
(score $ 3). DIPSS categories are as follows: low (score 0), intermediate-1 (score 1 to 2),
intermediate-2 (score 3 to 4), and high (score 5 to 6). DIPSS Plus categories are as
follows: low (score 0), intermediate-1 (score 1), intermediate-2 (score 2 to 3), and high
(score 4 to 6).
*Complex karyotype or a single abnormality or two abnormalities, including 18, -7/7q-,
i(17q), -5/5q-, 12p-, inv(3), or 11q23 rearrangement.
**DIPSS status of intermediate-1 (1 point); intermediate-2 (2 points); and high
(3 points).
Abbreviations: IPSS, International Prognostic Scoring System; DIPSS, Dynamic
International Prognostic Scoring System.
Data are presented as points values unless otherwise indicated. MIPSS categories: low
(score 0 to 0.5), intermediate-1 (score 1 to 1.5), intermediate-2 (score 2 to 3.5), and
high (score $ 4).
*Very high risk indicates a monosomal karyotype, inv(3), i(17q), -7/7q-, 11q, or 12p
abnormalities; high risk indicates complex nonmonosomal karyotype, 2 abnormalities
not included in the very high-risk category, 5q-,18, other autosomal trisomies
except 19, and other sole abnormalities not included in other risk categories.
Intermediate risk indicates sole abnormalities of 20q-, 1q1, or any other sole
translocation, and Y or other sex chromosome abnormality. Low risk indicates normal
cytogenetics or sole abnormalities of 13q- or 19.
Abbreviations: MIPSS, Mutation-Enhanced International Prognostic Scoring System;
GPSS, Genetics-Based Prognostic Scoring System.
e327
MESA AND PASSAMONTI
PROGNOSTICATION IN POLYCYTHEMIA VERA

AND ESSENTIAL THROMBOCYTHEMIA
There is a global consensus that being age 60 or older at
diagnosis and presence of vascular events in the patients
history are the two most relevant prognostic factors in
predicting thrombosis in polycythemia vera and essential
thrombocythemia.44,45 Therefore, patients with at least one
of these risk factors at diagnosis are considered at high risk,
whereas those with none are considered at low risk.45 This
risk classification affects the therapeutic approach for patients with polycythemia vera and essential thrombocythemia. However, leukocyte counts, cardiovascular risk, JAK2/
CALR mutation status, and bone marrow features are reported
to affect thrombosis, progression, and survival.28,29,46,47
In polycythemia vera, the ECLAP study48 showed that
patients younger than age 65 without prior thrombosis have
an incidence of thrombosis of 2.5 per 100 persons per year,
those older than age 65 or with prior thrombosis have an
incidence of 5.0 per 100 persons per year, and patients older
than age 65 with prior thrombosis have an incidence of 10.9
per 100 persons per year.
In essential thrombocythemia, physicians can predict survival
according to the IPSET model,49 which was tested among 867
patients with WHO-defined essential thrombocythemia. The
model included the following parameters: older than age 60 (2

points), leukocyte count greater than 11 3 109/L (1 point), and
prior history of thrombosis (1 point). Three risk categories with
significantly different survival were identified: low (0 points,
median survival not reached), intermediate (1 to 2 points,
median survival 24.5 years), and high (3 to 4 points, median
survival 13.8 years).
Concerning thrombosis prediction in WHO-diagnosed
essential thrombocythemia, the IPSET thrombosis
model50 was generated for the same IPSET cohort but
considers risk factors including being older than 60 (1 point),
history of thrombosis (2 points), presence of cardiovascular
risk factors (1 point), and JAK2 (V617F) mutation (2 points).
On the basis of these factors, the IPSET thrombosis model
specifies the following risk categories: low (less than 2
points; thrombosis risk of 1.03 per 100 person-years), intermediate (2 points; thrombosis risk of 2.35 per 100
person-years), and high (more than 2 points; thrombosis
risk of 3.56 per 100 person-years). Although CALR mutation
status was introduced into the statistical analysis, the IPSET
model maintained its prognostic power.51
In another study, survival was assessed in 1,545 patients
with WHO-defined polycythemia vera.52 Parameters included in this model were older age, leukocytosis, and
FIGURE 1. Stepwise Approach to Estimating Prognosis in Myeloproliferative Neoplasms Incorporating Clinical

Prognostic Scoring System and Then Refining Prognosis With Mutation Status
Abbreviations: DIPSS, Dynamic International Prognostic Scoring System; ET, essential thrombocythemia; f-up, follow-up; HR, high risk; Int-1, intermediate-1 risk; Int-2,
intermediate-2 risk; IPSET, International Prognostic Score Essential Thrombocythemia; IPSS, International Prognostic Scoring System; LR, low risk; Med OS, median overall
survival; PMF, primary myelofibrosis; PV, polycythemia vera; WHO, World Health Organization.
e328
TABLE 3. Myeloproliferative Neoplasm Symptom Assessment Tools and Associated Instruments

Questionnaire
Goals
Number of Items
Reference
Myelofibrosis Symptom Assessment Form

(MF-SAF)
Myelofibrosis-specific symptom assessment
20
Mesa et al53
Mesa et al54
Myelofibrosis Symptom Assessment Form

2.0 (MF-SAF 2.0)
Includes the Brief Fatigue Inventory

Myelofibrosis symptom serial assessment on
clinical trials
Validated in the COMFORT-1 trial
Myeloproliferative Neoplasm Symptom Assessment Form (MPN-SAF)
Expanded to measure MPN symptoms across

ET, PV, and MF
27
Scherber et al15
Myeloproliferative Neoplasm Symptom Assessment Form Total Symptom Score

(MPN 10 or MPN-SAF TSS)
Derived as a subset of the most representative

symptoms from MPN-SAF
10
Emanuel et al55
Abbreviations: MPN, myeloproliferative neoplasm; ET, essential thrombocythemia; PV, polycythemia vera; MF, myelofibrosis.
venous thrombosis, which generated three risk categories

with different survival ranging from 10.9 to 27.8 years.
Figure 1 shows how to use prognostication in MPNs.
CREATION OF VALIDATION OF MPN SYMPTOM

ASSESSMENT TOOLS
The spectrum of symptom-related difficulties encountered
by patients with MPNs is varied and includes several areas of
symptom burden such as those that are constitutional in
origin and those that are those related to vascular flow or
enlargement of the spleen. Given that there were no validated instruments that incorporated this spectrum of difficulties, MPN-specific tools were necessary and were thus
developed utilizing the above-described 2007 web survey
as a foundation (Table 3).13 The Myelofibrosis Symptom
Assessment Form (MF-SAF) was the first instrument developed. The MF-SAF is a 20-item survey validated against
other outcome tools reported for patients with cancer and is
effective in capturing the presence and intensity of myelofibrosis disease symptoms.53 Subsequently, this instrument was further refined to a seven-item instrument
(MF-SAF 2.0) for use specifically in myelofibrosis clinical
trials.54 In an effort to have a broader instrument that could
represent the difficulties seen in essential thrombocythemia
and polycythemia vera, the MF-SAF was expanded to 27
items and was named the MPN Symptom Assessment
Form.15 In addition to the initial MF-SAF questions related to
abdominal discomfort, fatigue, coughing, night sweats, bone
pain, weight loss, inactivity, abdominal pain, pruritus, fever,
and quality of life, there was an additional focus on microvascular symptoms including insomnia, sexual dysfunction, vertigo, lightheadedness, dizziness, numbness, tingling,
headaches, and concentration. Each individual symptom
had a potential score ranging from 0 (absent) to 10 (worst
imaginable/as bad as it could be). Further refinement of this
instrument for frequent serial use resulted in a 10-item total
symptom score (the MPN-SAF-TSS or MPN 10).55 These 10 core
items included worst fatigue, early satiety, abdominal discomfort, concentration problems, inactivity, night sweats,
itching, bone pain, fever, and weight loss. We recommend
using the comprehensive MPN-SAF at diagnosis and at the time
of clinical changes, whereas the MPN 10 is helpful for routine

clinical monitoring visits or when used daily in a diary with
clinical trials. The tools are meant to be a springboard for
discussion between the patient and his or her provider,
and improvement in symptoms is always weighed against
medication-related toxicities.
WHAT LESSONS HAVE WE LEARNED FROM MPN

SYMPTOMS AND DISEASE FEATURES?
The MPN International Quality of Life Study Group has analyzed
the aggregate data gathered on symptom burden among patients with MPN to make several observations regarding these
diseases. Observations made across the spectrum of MPNs include that cytokine increases are likely linked to symptom burden.16 It has been observed that certain symptom profile clusters
can exist among patients with MPNs, and these symptom clusters
may potentially correlate with disease phenotypes with clinical
features (and possibly biologic features as well).56 Clinically relevant observations for problematic polycythemia vera (specifically that hydroxyurea treatment will fail for individual subgroups
or these individuals will have either palpable splenomegaly or a
persisting need for phlebotomies) show that patients with these
features all have worse symptom burden compared with patients
without these features. Furthermore, individuals who have more
than one of these problematic clinical features also have
worsening symptom burden.57 These latter observations may
have relevance regarding current choice of medical therapy.
DETERMINING A TREATMENT PLAN FOR

PATIENTS WITH MPNs
When determining the treatment plan for an individual
patient, it is necessary to consider the nuanced assessment
of prognosis, the importance of avoidance of vascular
events, the effect of symptom burden, the contribution of
splenomegaly to morbidity, and the likelihood of progression to acute myeloid leukemia.
Which Patients Are Candidates for Allogeneic Stem

Cell Transplantation and When?
Allogeneic stem cell transplantation can be curative for
patients with MPN; however, because of the associated
e329
MESA AND PASSAMONTI
TABLE 4. JAK Inhibitors and New Agents and Their Use, or Possible Use, in Myeloproliferative Neoplasm
Treatment
Agent
Disease
Trial
Comments
Reference
Ruxolitinib (Approved)
MF
COMFORT-1 and
COMFORT-2
Improved splenomegaly
Verstovsek et al63 and Harrison et al64
Improved MF-associated symptoms

Improved survival
PV
(second
line)
RESPONSE
MF
PERSIST-1 and
PERSIST-2
(ongoing)
Improved complete response rate
Vannucchi et al78
Decreased PV-associated symptoms
Pacritinib
Vannucchi et al70
Improved MF-associated symptoms

Able to be given in full dose
regardless of thrombocytopenia
Improvement in anemia
Momelotinib
MF
SIMPLIFY-1 and
SIMPLIFY-2
Ongoing phase III trials
NCT0196838 and NCT02101268
SIMPLIFY-2
Combination Trials All Including a
Ruxolitinib Base
Plus Azacitidine (Hypomethylating)
MF
Phase II
Ongoing trial
NCT01787487, Daver et al73
Plus Danazol (Androgen)
MF
Phase II
Ongoing trial
NCT01732445, Gowin et al72
Plus Pegylated Interferon Alfa-2a

(Interferon)
MF
Phase II
Ongoing trial
Mikkelsen et al76
Plus Panobinostat (HDAC Inhibitor)
MF
Phase II
Ongoing trial
NCT01433445, Harrison et al75
Plus Pomalidomide (IMID)
MF
Phase II
Ongoing trial
NCT01644110, Stegelmann et al74
Plus BKM-120 (PI3K Inhibitor)
MF
Phase II
Ongoing study
Durrant et al83
Plus LDE-225 (Hedgehog Inhibitor)
MF
Phase II
Ongoing trial
NCT01787552, Gupta et al84
PV
Phase III
Ongoing trial
NCT01949805
PRM-151 (Antifibrosis)
MF
Phase II
Ongoing trial
NCT01981850
Imetelstat (Telomerase Inhibitor)
MF
Phase II
Ongoing trial
NCT02426086
New Agents for MPNs Testing as

Single Agents
P1101/AOP2014
Proline Pegylated Interferon Alfa-2b
Abbreviations: MF, myelofibrosis; PV, polycythemia vera; HDAC, histone deacetylase; ET, essential thrombocythemia; IMID, immunomodulatory drug; PI3K, phosphoinositide
3-kinase.
expense and risk of transplant-related morbidity and mortality, HSCT is currently limited to a subset of patients with
MF. The European Group for Blood and Marrow Transplantation and the European LeukemiaNet international
working group recently provided indications for and management of allogeneic stem cell transplantation.58 Although
readers are referred to the guidelines by the European
Group for Blood and Marrow Transplantation and the European LeukemiaNet for further information, some points
are summarized as follows. Patients with intermediate-2
or high-risk disease and who are younger than age 70
should be considered candidates for allogeneic stem cell
transplantation.59 Patients with intermediate-1 risk of disease
and who are younger than age 65 should be considered
candidates if they present with adverse cytogenetics, a percentage of blasts in peripheral blood greater than 2%, or refractory, transfusion-dependent anemia. Decisions regarding a
pretransplant splenectomy should be made on a case-by-case
e330
basis. Patients with intermediate-2 or high-risk disease lacking

an HLA-matched sibling or unrelated donor should be enrolled
in a protocol using HLA nonidentical donors; the optimal intensity of the conditioning regimen must be defined.
What Options Are Available to Control Splenomegaly,

Cytoses, and Symptoms?
The European LeukemiaNet guidelines recommend control
of hematocrit and use of low-dose aspirin (unless contraindicated) for the prevention of thrombosis, as well as the
selective use hydroxyurea as the drug of choice when an
antimyeloproliferative effect is needed for patients with
MPNs. 45 However, data available on hydroxyurea are
scant, with the most complete study on hydroxyurea in
myelofibrosis retrospectively evaluating 40 patients.60
In the last few years, several medicines with anti-JAK
properties (named JAK inhibitors) have been studied
(Table 4).61 Among these, ruxolitinib is the only approved
JAK inhibitor for MPNs and available for clinical practice.

Other compounds entered phase III investigation (fedratinib,62 momelotinib, and pacritinib), and others are being
tested in phase I to phase II studies.
Ruxolitinib
Two prospective randomized trials with ruxolitinib have
been published: COMFORT-1 (155 patients treated with
ruxolitinib vs. 151 treated with placebo)63 and COMFORT-2
(146 patients treated with ruxolitinib vs. 73 treated with best
available therapy [BAT]).64 In the COMFORT-1 trial, the
primary endpoint (reduction of spleen volume of 35% or
more assessed by MRI) at week 24 was reached in 42% of
patients in the ruxolitinib arm and in 1% of the placebo arm.
At week 24, 46% of patients receiving ruxolitinib and 5% of
those receiving placebo experienced symptom alleviation by
at least 50%, as measured by the modified MF-SAF.53 Patients treated with ruxolitinib experienced relief of abdominal discomfort, early satiety, night sweats, itching, and
musculoskeletal pain. In the COMFORT-2 trial, the primary
endpoint (the same as the COMFORT-1 study but evaluated
at week 48) was reached in 28% of patients treated with
ruxolitinib and in 0% of those receiving BAT. Mean improvements in Functional Assessment of Cancer Therapy
Lymphoma System scores from baseline were greater in the
ruxolitinib arm.
Improved survival was observed for ruxolitinib versus
placebo (27 vs. 41 deaths), with a hazard ratio of 0.58.65 The
3-year analysis of the COMFORT-2 trial confirmed that dosedependent anemia and thrombocytopenia were the most
common adverse events in the ruxolitinib-treated group, but
these events rarely led to discontinuation.66 Other adverse
events of interest included leukopenia, bleeding, infections,
thromboembolic events, elevated transaminase levels, increased systolic blood pressure, and weight gain. The rate of
these events generally decreased with longer exposure to
ruxolitinib treatment, with the highest rates occurring within
the first 6 months of treatment. Among these events, infections occurred in 50% of patients between weeks 0 and 24
and included bronchitis, gastroenteritis, nasopharyngitis,
and urinary tract infections. The rate of infections was 25%
for weeks 144 to 168. Finally, patients randomly assigned to
ruxolitinib showed longer overall survival than those randomly assigned to BAT (hazard ratio 0.48).
Because all patients crossed over to ruxolitinib in both
studies, it is difficult to compare the effect on long-term
survival. In a recent study of COMFORT-2 participants,67
survival from diagnosis was compared for 100 patients with
intermediate-2 and high-risk primary myelofibrosis who
received ruxolitinib (the IPSS cohort) versus a comparable
group of 250 patients with primary myelofibrosis who received conventional treatment (namely, the DIPSS cohort)
when at the same risk. Patients treated with ruxolitinib at
some point during their disease history had better survival
compared with those who continued standard treatment for
the full duration of follow-up, ultimately suggesting that
ruxolitinib affects the natural history of primary myelofibrosis.
Five-year results of the COMFORT-2 trial were recently

presented.68 At a median follow-up of 4.3 years, 27% of
patients in the ruxolitinib arm and 24% who crossed over
from BAT completed 5 years of on-study treatment; primary
reasons for premature discontinuation were adverse events
(24%) and disease progression (22%). Overall, 54% of patients in the ruxolitinib arm achieved a 35% or more reduction in spleen volume from baseline at any time during
treatment. The probability of maintaining this reduction was
0.51 at 3 years and 0.48 at 5 years. One-third of evaluable
patients with a JAK2 V617F mutation had a greater than 20%
reduction in allele burden and 16% improved fibrosis.
Results of the RESPONSE trial were published in 2015. This
phase III open-label study evaluated the efficacy and safety
of ruxolitinib versus BAT in 110 and 112 patients with
splenomegaly, respectively, who had either hydroxyurearesistant or hydroxyurea-intolerant polycythemia vera.
RESPONSE-2 is an ongoing study with a similar design but
includes patients without splenomegaly.69 The primary
endpoint is the proportion of patients who achieve both
hematocrit control through week 32 and a 35% or greater
reduction in spleen volume from baseline shown on imaging
at week 32.
The primary endpoint was achieved in 21% of patients
treated with ruxolitinib versus 1% of patients who received
BAT. Of patients who received ruxolitinib or BAT, 60% and
20% achieved hematocrit control and 38% and 1%
achieved a 35% or greater spleen volume reduction, respectively. Complete hematologic remission was achieved in
24% and 9% of ruxolitinib-treated and BAT-treated patients,
respectively; 49% and 5% had a 50% or greater improvement
in MPN-SAF-TSS score at week 32. Herpes zoster (all grade 1
or 2) was reported in 6.4% and 0% of patients treated with
ruxolitinib and BAT, respectively. The rate of thromboembolic events was lower in the ruxolitinib arm versus the BATtreated arm (one and six events, respectively).
Pacritinib and Momelotinib: JAK Inhibition and

Improving Cytopenias
Pacritinib is a JAK2 and FLT3 inhibitor that distinguished itself in
early testing by alleviating myelofibrosis-associated splenomegaly and symptoms without drug-emergent worsening of
anemia or thrombocytopenia (Table 4). The PERSIST-1 trial
demonstrated that pacritinib was superior to the best alternative therapy, even in those with severe thrombocytopenia
(i.e., platelets less than 50 3 109/L), for control of splenomegaly. Subset analysis presented at the 2015 American Society of Hematology Annual Meeting showed that all subgroups
(in terms of disease features, risks, demographics, and baseline
blood counts) of patients with myelofibrosis seemed to benefit
from pacritinib therapy.70 On the basis of these results, the
manufacturers of pacritinib are seeking U.S. Food and Drug
Administration approval for the drug for use for patients with
myelofibrosis with severe thrombocytopenia, which is currently an unmet MPN need.
Investigators of a randomized phase III trial (a JAK1 and
JAK2 inhibitor) are currently enrolling participants to study
e331
MESA AND PASSAMONTI
TABLE 5. Serial Use of Myeloproliferative Neoplasm Symptom Instruments in Randomized Clinical Trials
Disease
Instrument
Trial/Agent
Reference
MF
MF-SAF 2.0
COMFORT-1 (ruxolitinib)
Verstovsek et al63
MF
EORTC QLQc30
COMFORT-2 (ruxolitinib)
Harrison et al64
MF
mMF-SAF
JAKARTA (fedratinib)
Pardanani et al62
MF
MPN-SAF TSS
PERSIST-1 (pacritinib)
Mesa et al77
MF
MPN-SAF TSS
PERSIST-2 (pacritinib); ongoing
NCT02055781
PV
MPN-SAF TSS
SIMPLIFY (momelotinib); ongoing
NCT0196838
PV
MPN-SAF
RESPONSE (ruxolitinib)
Vannucchi et al78
PV
MPN-SAF
RELIEF (ruxolitinib)
Mesa et al79
PV and ET
MPN-SAF
MPD-RC 112 (phase III) pegylated

interferon
NCT01259856
Abbreviations: MF, myelofibrosis; MF-SAF, Myelofibrosis Symptom Assessment Form; MPN-SAF TSS, Myeloproliferative Neoplasm Symptom Assessment Form Total Symptom Score;
mMF-SAF, Myeloproliferative Neoplasm Symptom Assessment Form; PV, polycythemia vera; ET, essential thrombocythemia.
momelotinib as a first-line treatment for myelofibrosis

compared with ruxolitinib. Building on the experience of
phase II trials, the SIMPLIFY 1 trial of momelotinib seeks to
demonstrate superiority of this agent, compared with ruxolitinib, for myelofibrosis-associated anemia and noninferiority in regard to reduction in splenomegaly and
symptoms.
Combination JAK Inhibition: Ongoing Investigations

Multiple combination approaches, with a base therapy of
ruxolitinib, have been tested during the past 2 years. The
goals of combination therapy in the setting of myelofibrosis
are to augment single-agent ruxolitinib benefits by any of the
following: improvement of cytopenias, reduction in marrow
fibrosis, or deeper or broader molecular responses (either in
driver mutations such as JAK2-V617F or CALR or associated
mutations such as ASXL1 or IDH1/2). Updates presented at
the 2015 American Society of Hematology Annual Meeting
were most promising with combinations (Table 1) including
danazol (with activity for cytopenias),72 azacitidine (decrease in blasts),73 pomalidomide (improved anemia),74 and
panobinostat (deeper responses perhaps in fibrosis and
molecular markers).75 The addition of ruxolitinib to pegylated interferon alfa-2a (IFN-a-2a) was also reported to help
overcome acquired resistance to single-agent interferon for
patients with a mixture of MPNs.76 Although all of these
trials showed benefits, given the significant activity of singleagent ruxolitinib, randomized trials that compare these
agents with single-agent ruxolitinib will likely be required
to change practice patterns to include combination
approaches.
EVOLVING OPTIONS IN MPN MANAGEMENT:

EXPERIENCE INCORPORATING SYMPTOM
ASSESSMENT
The current standard across evaluating patients with MPNs
participating in clinical trials includes an important baseline
assessment of symptom burden and potential improvement
with therapy (Table 5). In the inaugural phase III trials of
ruxolitinib in myelofibrosis, it was identified that ruxolitinib
e332
affected both individual and aggregate symptoms for patients with myelofibrosis.63 Participants ability to track MPN
symptoms with a daily diary also gave insight into issues
regarding symptomatic rebound with withdrawal of the
medication as well as durability of response. In a phase III
study of the JAK inhibitor fedratinib, investigators also used
patient-reported outcomes with the MF-SAF and similarly
showed improvement in individual and aggregate symptoms.62 The PERSIST-1 study reported the ability to determine symptomatic improvement across all subgroups
treated with pacritinib, and investigators subsequently analyzed those individuals with severe thrombocytopenia with
the greatest unmet needs and the resolution of symptoms
compared with the best alternative therapy.77 Use of JAK
inhibitors as second-line therapy for polycythemia vera
demonstrated improvement in symptom burden alone or in
aggregate for patients who received ruxolitinib compared
with individuals who received best alternative therapy (in
whom hydroxyurea previously failed).78 In addition, these
similar methodologies were used in the RELIEF study comparing patients who were treated with ruxolitinib versus hydroxyurea and those with inadequate symptomatic control.79
In the majority of ongoing clinical trials not only for myelofibrosis but also essential thrombocythemia and polycythemia
vera (e.g., the randomized phase III trial of hydroxyurea vs.
pegylated IFN-a-2a; NCT01259856), researchers analyzed
symptomatic response as an important component of disease
activity, drug efficacy, and clinical trial endpoints.
SYMPTOMS AND MPNs: MAIN COURSE OR SIDE

DISH?
Treatment for an MPN includes many potential areas of
clinical benefit, such as improved splenomegaly, blood
count, progression-free survival, and symptom resolution.
All of these areas contribute to morbidity, are related to
activity against the disease-associated clone, and may affect
patient suffering and even survival. Symptom response
should be considered as an integral part of assessment
of MPN therapy response that is as important as other
clinical features. Current response criteria for myelofibrosis,
polycythemia vera, and essential thrombocythemia have all

adopted symptomatic improvement as an important component of disease response.9,80 Symptomatic response for
therapy clearly must be weighed against the aggregate
impact of therapeutic efficacy. Clearly, the benefit of
symptomatic improvement can be diminished if significant
toxicity is experienced from a treatment. Individualizing care
for patients with MPNs requires an accurate assessment of
prognosis, a well-informed patient, good understanding of
the morbidity and mortality a patient may expect, a comprehensive therapeutic plan, and a nuanced assessment of
drug efficacy and toxicity.
INDIVIDUALIZED CARE FOR MPNs: FUTURE

LANDSCAPE
The landscape of MPN therapy continues to evolve and does
so around a better understanding of the effect of heterogeneous mutation profiles, widely variable disease burden
and prognosis, and an evolving range of options that currently have a clearly important role in JAK inhibition with
ruxolitinib. The efficacy of pacritinib, momelotinib, combination therapy with ruxolitinib, or new targets such as
pentraxin (PRM 151)81 or telomerase (imetelstat)82 will be
determined based on their effect on disease course, disease
burden, and individualized patient features.
References
1. Tefferi A, Thiele J, Vannucchi AM, et al. An overview on CALR and CSF3R
mutations and a proposal for revision of WHO diagnostic criteria for
myeloproliferative neoplasms. Leukemia. 2014;28:1407-1413.
2. Tefferi A, Thiele J, Orazi A, et al. Proposals and rationale for revision of
the World Health Organization diagnostic criteria for polycythemia
vera, essential thrombocythemia, and primary myelofibrosis: recommendations from an ad hoc international expert panel. Blood. 2007;
110:1092-1097.
3. Geyer H, Scherber R, Kosiorek H, et al. Symptomatic profiles of patients
with polycythemia vera: implications of inadequately controlled disease. J Clin Oncol. 2016;34:151-159.
4. Passamonti F, Rumi E, Pietra D, et al. A prospective study of 338 patients
with polycythemia vera: the impact of JAK2 (V617F) allele burden and
leukocytosis on fibrotic or leukemic disease transformation and vascular complications. Leukemia. 2010;24:1574-1579.
5. Passamonti F, Rumi E, Arcaini L, et al. Prognostic factors for thrombosis,
myelofibrosis, and leukemia in essential thrombocythemia: a study of
605 patients. Haematologica. 2008;93:1645-1651.
6. Passamonti F. How I treat polycythemia vera. Blood. 2012;120:275-284.
7. Passamonti F. How to manage polycythemia vera. Leukemia. 2012;26:
870-874.
8. Mesa RA, Tibes R. MPN blast phase: clinical challenge and assessing
response. Leuk Res. 2012;36:1496-1497.
9. Tefferi A, Cervantes F, Mesa R, et al. Revised response criteria for
myelofibrosis: International Working Group-Myeloproliferative Neoplasms Research and Treatment (IWG-MRT) and European LeukemiaNet (ELN) consensus report. Blood. 2013;122:1395-1398.
10. Cervantes F, Dupriez B, Passamonti F, et al. Improving survival trends in primary
myelofibrosis: an international study. J Clin Oncol. 2012;30:2981-2987.
11. Tefferi A, Guglielmelli P, Larson DR, et al. Long-term survival and blast
transformation in molecularly annotated essential thrombocythemia,
polycythemia vera, and myelofibrosis. Blood. 2014;124:2507-2513; quiz
2615.
12. Tefferi A, Vardiman JW. Classification and diagnosis of myeloproliferative neoplasms: the 2008 World Health Organization criteria and pointof-care diagnostic algorithms. Leukemia. 2008;22:14-22.
13. Mesa RA, Niblack J, Wadleigh M, et al. The burden of fatigue and quality
of life in myeloproliferative disorders (MPDs): an international Internetbased survey of 1179 MPD patients. Cancer. 2007;109:68-76.
14. Cervantes F, Dupriez B, Pereira A, et al. New prognostic scoring system
for primary myelofibrosis based on a study of the International Working
Group for Myelofibrosis Research and Treatment. Blood. 2009;113:
2895-2901.
15. Scherber R, Dueck AC, Johansson P, et al. The Myeloproliferative Neoplasm

Symptom Assessment Form (MPN-SAF): international prospective validation and reliability trial in 402 patients. Blood. 2011;118:401-408.
16. Verstovsek S, Kantarjian H, Mesa RA, et al. Safety and efficacy of
INCB018424, a JAK1 and JAK2 inhibitor, in myelofibrosis. N Engl J Med.
2010;363:1117-1127.
17. Passamonti F, Maffioli M, Caramazza D, et al. Myeloproliferative
neoplasms: from JAK2 mutations discovery to JAK2 inhibitor therapies.
Oncotarget. 2011;2:485-490.
18. Nangalia J, Massie CE, Baxter EJ, et al. Somatic CALR mutations in
myeloproliferative neoplasms with nonmutated JAK2. N Engl J Med.
2013;369:2391-2405.
19. Tefferi A, Lasho TL, Finke CM, et al. CALR vs JAK2 vs MPL-mutated or
triple-negative myelofibrosis: clinical, cytogenetic and molecular
comparisons. Leukemia. 2014;28:1472-1477.
20. Passamonti F, Caramazza D, Maffioli M. JAK inhibitor in CALR-mutant
myelofibrosis. N Engl J Med. 2014;370:1168-1169.
21. Maffioli M, Genoni A, Caramazza D, et al. Looking for CALR mutations in
familial myeloproliferative neoplasms. Leukemia. 2014;28:1357-1360.
22. Rumi E, Passamonti F, Della Porta MG, et al. Familial chronic myeloproliferative disorders: clinical phenotype and evidence of disease
anticipation. J Clin Oncol. 2007;25:5630-5635.
23. Milosevic Feenstra JD, Nivarthi H, Gisslinger H, et al. Whole-exome
sequencing identifies novel MPL and JAK2 mutations in triple-negative
myeloproliferative neoplasms. Blood. 2016;127:325-332.
24. Passamonti F, Elena C, Schnittger S, et al. Molecular and clinical features
of the myeloproliferative neoplasm associated with JAK2 exon 12
mutations. Blood. 2011;117:2813-2816.
25. Rumi E, Pietra D, Guglielmelli P, et al; Associazione Italiana per la Ricerca
sul Cancro Gruppo Italiano Malattie Mieloproliferative. Acquired copyneutral loss of heterozygosity of chromosome 1p as a molecular event
associated with marrow fibrosis in MPL-mutated myeloproliferative
neoplasms. Blood. 2013;121:4388-4395.
26. Klampfl T, Gisslinger H, Harutyunyan AS, et al. Somatic mutations of
calreticulin in myeloproliferative neoplasms. N Engl J Med. 2013;369:
2379-2390.
27. Passamonti F, Rumi E. Clinical relevance of JAK2 (V617F) mutant allele
burden. Haematologica. 2009;94:7-10.
28. Rumi E, Pietra D, Ferretti V, et al; Associazione Italiana per la Ricerca sul
Cancro Gruppo Italiano Malattie Mieloproliferative Investigators. JAK2 or
CALR mutation status defines subtypes of essential thrombocythemia with
substantially different clinical course and outcomes. Blood. 2014;123:
1544-1551.
e333
MESA AND PASSAMONTI
29. Tefferi A, Wassie EA, Guglielmelli P, et al. Type 1 versus type 2 calreticulin mutations in essential thrombocythemia: a collaborative study of
1027 patients. Am J Hematol. 2014;89:E121-E124.
30. Rotunno G, Mannarelli C, Guglielmelli P, et al; Associazione Italiana per
la Ricerca sul Cancro Gruppo Italiano Malattie Mieloproliferative Investigators. Impact of calreticulin mutations on clinical and hematological phenotype and outcome in essential thrombocythemia. Blood.
2014;123:1552-1555.
31. Passamonti F, Thiele J, Girodon F, et al. A prognostic model to predict
survival in 867 World Health Organization-defined essential thrombocythemia at diagnosis: a study by the International Working Group on
Myelofibrosis Research and Treatment. Blood. 2012;120:1197-1201.
32. Rumi E, Pietra D, Pascutto C, et al; Associazione Italiana per la Ricerca sul
Cancro Gruppo Italiano Malattie Mieloproliferative Investigators.
Clinical effect of driver mutations of JAK2, CALR, or MPL in primary
myelofibrosis. Blood. 2014;124:1062-1069.
33. Tefferi A, Guglielmelli P, Lasho TL, et al. CALR and ASXL1 mutationsbased molecular prognostication in primary myelofibrosis: an international study of 570 patients. Leukemia. 2014;28:1494-1500.
34. Tefferi A, Lasho TL, Finke C, et al. Type 1 vs type 2 calreticulin mutations
in primary myelofibrosis: differences in phenotype and prognostic
impact. Leukemia. 2014;28:1568-1570.
35. Tefferi A, Guglielmelli P, Lasho TL, et al. Driver mutations and prognosis
in 1118 patients with primary myelofibrosis. Blood. 2015;126 (suppl;
abstr 2801).
36. Passamonti F, Maffioli M, Caramazza D. New generation small-molecule
inhibitors in myeloproliferative neoplasms. Curr Opin Hematol. 2012;
19:117-123.
37. Vannucchi AM, Lasho TL, Guglielmelli P, et al. Mutations and prognosis
in primary myelofibrosis. Leukemia. 2013;27:1861-1869.
38. Tefferi A, Lasho TL, Finke C, et al. A 27-gene NGS panel in primary
myelofibrosis identifies ASXL1, CBL, RUNX1 and SRSF2 mutations as
being unfavorable and absence of any non-driver mutation as being
favorable to survival. Blood. 2015;126 (suppl; abstr 350).
39. Passamonti F, Cervantes F, Vannucchi AM, et al. A dynamic prognostic
model to predict survival in primary myelofibrosis: a study by the IWGMRT (International Working Group for Myeloproliferative Neoplasms
Research and Treatment). Blood. 2010;115:1703-1708.
40. Gangat N, Caramazza D, Vaidya R, et al. DIPSS Plus: a refined Dynamic
International Prognostic Scoring System for primary myelofibrosis that
incorporates prognostic information from karyotype, platelet count,
and transfusion status. J Clin Oncol. 2011;29:392-397.
41. Guglielmelli P, Lasho TL, Rotunno G, et al. The number of prognostically
detrimental mutations and prognosis in primary myelofibrosis: an international study of 797 patients. Leukemia. 2014;28:1804-1810.
42. Vannucchi AM, Guglielmelli P, Rotunno G, et al. Mutation-Enhanced
International Prognostic Scoring System (MIPSS) for primary myelofibrosis: an AGIMM & IWG-MRT project. Blood. 2014;124 (suppl; abstr
405).
43. Tefferi A, Guglielmelli P, Finke C, et al. Integration of mutations and
karyotype towards a Genetics-Based Prognostic Scoring System (GPSS)
for primary myelofibrosis. Blood. 2014;124 (suppl; abstr 406).
44. Cervantes F, Passamonti F, Barosi G. Life expectancy and prognostic
factors in the classic BCR/ABL-negative myeloproliferative disorders.
Leukemia. 2008;22:905-914.
45. Barbui T, Barosi G, Birgegard G, et al; European LeukemiaNet.
Philadelphia-negative classical myeloproliferative neoplasms: critical
concepts and management recommendations from European LeukemiaNet. J Clin Oncol. 2011;29:761-770.
46. Barbui T, Thiele J, Passamonti F, et al. Survival and disease progression in
essential thrombocythemia are significantly influenced by accurate
e334
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
morphologic diagnosis: an international study. J Clin Oncol. 2011;29:

3179-3184.
Carobbio A, Antonioli E, Guglielmelli P, et al. Leukocytosis and risk
stratification assessment in essential thrombocythemia. J Clin Oncol.
2008;26:2732-2736.
Landolfi R, Marchioli R, Kutti J, et al; European Collaboration on LowDose Aspirin in Polycythemia Vera Investigators. Efficacy and safety of
low-dose aspirin in polycythemia vera. N Engl J Med. 2004;350:114-124.
Passamonti F, Theile J, Barbui T, et al. A prognostic model to predict
survival in WHO-defined essential thrombocythemia: a study by the
IWG-MRT (International Working Group for Myeloproliferative Neoplasms Research and Treatment). Blood. 2011;118 (suppl; abstr 1746).
Barbui T, Finazzi G, Carobbio A, et al. Development and validation of an
International Prognostic Score of thrombosis in WHO-Essential
Thrombocythemia (IPSET-thrombosis). Blood. 2012;120:5128-5133;
quiz 5252.
Finazzi G, Carobbio A, Guglielmelli P, et al. Calreticulin mutation does
not modify the IPSET score for predicting the risk of thrombosis among
1150 patients with essential thrombocythemia. Blood. 2014;124:
2611-2612.
Tefferi A, Rumi E, Finazzi G, et al. Survival and prognosis among 1545
patients with contemporary polycythemia vera: an international study.
Leukemia. 2013;27:1874-1881.
Mesa RA, Schwager S, Radia D, et al. The Myelofibrosis Symptom
Assessment Form (MFSAF): an evidence-based brief inventory to
measure quality of life and symptomatic response to treatment in
myelofibrosis. Leuk Res. 2009;33:1199-1203.
Mesa RA, Gotlib J, Gupta V, et al. Effect of ruxolitinib therapy on
myelofibrosis-related symptoms and other patient-reported outcomes
in COMFORT-I: a randomized, double-blind, placebo-controlled trial.
J Clin Oncol. 2013;31:1285-1292.
Emanuel RM, Dueck AC, Geyer HL, et al. Myeloproliferative neoplasm
(MPN) symptom assessment form total symptom score: prospective
international assessment of an abbreviated symptom burden scoring
system among patients with MPNs. J Clin Oncol. 2012;30:4098-4103.
Geyer HL, Scherber RM, Dueck AC, et al. Distinct clustering of symptomatic burden among myeloproliferative neoplasm patients: retrospective assessment in 1470 patients. Blood. 2014;123:3803-3810.
Geyer HL, Scherber RM, Dueck AC, et al. Symptom severity and clinical
variables of polycythemia vera patients with splenomegaly, phlebotomy requirements and/or hydroxyurea use: a retrospective evaluation
of 1334 patients. Blood. 2014;124 (suppl; abstr 1848).
Kroger NM, Deeg JH, Olavarria E, et al. Indication and management of
allogeneic stem cell transplantation in primary myelofibrosis: a consensus process by an EBMT/ELN international working group. Leukemia.
2015;29:2126-2133.
Kroger N, Giorgino T, Scott BL, et al. Impact of allogeneic stem cell
transplantation on survival of patients less than 65 years of age with
primary myelofibrosis. Blood. 2015;125:3347-3350, quiz 3364.
Martnez-Trillos A, Gaya A, Maffioli M, et al. Efficacy and tolerability of
hydroxyurea in the treatment of the hyperproliferative manifestations
of myelofibrosis: results in 40 patients. Ann Hematol. 2010;89:
1233-1237.
Passamonti F. Balancing efficacy and safety of JAK inhibitors in myelofibrosis. Leuk Res. 2014;38:290-291.
Pardanani A, Harrison C, Cortes JE, et al. Safety and efficacy of fedratinib
in patients with primary or secondary myelofibrosis: a randomized
clinical trial. JAMA Oncol. 2015;1:643-651.
Verstovsek S, Mesa RA, Gotlib J, et al. A double-blind, placebocontrolled trial of ruxolitinib for myelofibrosis. N Engl J Med. 2012;
366:799-807.
64. Harrison C, Kiladjian JJ, Al-Ali HK, et al. JAK inhibition with ruxolitinib versus
best available therapy for myelofibrosis. N Engl J Med. 2012;366:787-798.
65. Verstovsek S, Mesa RA, Gotlib J, et al. Efficacy, safety and survival with
ruxolitinib in patients with myelofibrosis: results of a median 2-year
follow-up of COMFORT-I. Haematologica. 2013;98:1865-1871.
66. Cervantes F, Vannucchi AM, Kiladjian JJ, et al. Three-year efficacy,
safety, and survival findings from COMFORT-II, a phase 3 study comparing ruxolitinib with best available therapy for myelofibrosis. Blood.
2013;122:4047-4053.
67. Passamonti F, Maffioli M, Cervantes F, et al. Impact of ruxolitinib on the
natural history of primary myelofibrosis: a comparison of the DIPSS and
the COMFORT-2 cohorts. Blood. 2014;123:1833-1835.
68. Harrison CN, Vannucchi AM, Kiladjian JJ, et al. Long-term efficacy and
safety in COMFORT-II, a phase 3 study comparing ruxolitinib with best
available therapy for the treatment of myelofibrosis: 5-year final study
results. Blood. 2015;126 (suppl; abstr 59).
69. Passamonti F, Griesshammer M, Cavo M, et al. Demographics, baseline
characteristics, and disease symptom burden in RESPONSE-2: a randomized, phase 3 study of ruxolitinib in polycythemia vera patients (pts)
who are resistant to or intolerant of hydroxyurea (HU). Blood. 2015;126
70. Vannucchi AM, Mesa RA, Cervantes F, et al. Analysis of outcomes by
patient subgroups in patients with myelofibrosis treated with pacritinib
vs best available therapy (BAT) in the phase III Persist-1 Trial. Blood.
2015;126 (suppl; abstr 58).
71. Gupta V, Mesa RA, Deininger MW, et al. Circulating cytokines and
markers of iron metabolism in myelofibrosis patients treated with
momelotininb: correlatives from the Ym-387-II Study. Blood. 2015;126
72. Gowin KL, Kosiorek HE, Dueck AC, et al. Final analysis of a multicenter
pilot phase 2 study of ruxolitinib and danazol in patients with myelofibrosis. Blood. 2015;126 (suppl; abstr 1618).
73. Daver N, Garcia-Manero G, Cortes JE, et al. 5-Azacytidine (AZA) in
combination with ruxolitinib (RUX) as therapy for patients (pts) with
myelodysplastic/myeloproliferative neoplasms (MDS/MPNs). Blood.
2015;126 (suppl; abstr 823).
74. Stegelmann F, Bangerter M, Heidel FH, et al. A phase-Ib/II study of ruxolitinib
plus pomalidomide in myelofibrosis. Blood. 2015;126 (suppl; abstr 826).
75. Harrison CN, Kiladjian JJ, Heidel FH, et al. Efficacy, safety, and confirmation of the recommended phase 2 starting dose of the combination
of ruxolitinib (RUX) and panobinostat (PAN) in patients (pts) with
myelofibrosis (MF). Blood. 2015;126 (suppl; abstr 4060).
76. Mikkelsen SU, Kjr L, Skov V, et al. Safety and efficacy of combination
therapy of interferon-alpha2 + JAK1-2 inhibitor in the Philadelphianegative chronic myeloproliferative neoplasms. preliminary results
from the Danish Combi-Trial - an open label, single arm, nonrandomized multicenter phase II study. Blood. 2015;126 (suppl; abstr
824).
77. Mesa RA, Egyed M, Szoke A, et al. Results of the PERSIST-1 phase III
study of pacritinib (PAC) versus best available therapy (BAT) in primary
myelofibrosis (PMF), post-polycythemia vera myelofibrosis (PPV-MF),
or post-essential thrombocythemia-myelofibrosis (PET-MF). J Clin
Oncol. 2015;33 (suppl; abstr LBA7006).
78. Vannucchi AM, Kiladjian JJ, Griesshammer M, et al. Ruxolitinib versus
standard therapy for the treatment of polycythemia vera. N Engl J Med.
2015;372:426-435.
79. Mesa R, Vannucchi AM, Yacoub A, et al. The efficacy and safety of
continued hydroxyurea therapy versus switching to ruxolitinib in patients with polycythemia vera: a randomized, double-blind, doubledummy, symptom study (RELIEF). Blood. 2014;124 (suppl; abstr 3168).
80. Barosi G, Mesa R, Finazzi G, et al. Revised response criteria for polycythemia vera and essential thrombocythemia: an ELN and IWG-MRT
consensus project. Blood. 2013;121:4778-4781.
81. Verstovsek S, Mesa RA, Foltz LM, et al. PRM-151 in myelofibrosis:
durable efficacy and safety at 72 weeks. Blood. 2015;126 (suppl; abstr
56).
82. Tefferi A, Lasho TL, Begna KH, et al. A pilot study of the telomerase
inhibitor imetelstat for myelofibrosis. N Engl J Med. 2015;373:908-919.
83. Durrant ST, Nagler A, Vannucchi AM, et al. An open-label, multicenter,
2-arm, dose-finding, phase 1b study of the combination of ruxolitinib
and buparlisib (BKM120) in patients with myelofibrosis: results from
HARMONY Study. Blood. 2015;126 (suppl; abstr 827).
84. Gupta V, Harrison CN, Hasselbalch H, et al. Phase 1b/2 study of the
efficacy and safety of sonidegib (LDE225) in combination with ruxolitinib (INC424) in patients with myelofibrosis. Blood. 2015;126 (suppl;
abstr 825).
e335
ALLOTRANSPLANT
Why Are Myelodysplastic

Syndromes So Difficult to Cure?
CHAIR
Stephen Nimer, MD
Sylvester Comprehensive Cancer Center
Miami, FL
SPEAKERS
Gregory A. Abel, MD, MPH
Boston, MA
Rami S. Komrokji, MD
Tampa, FL
FRAILTY, COMORBIDITY, AND QOL IN MDS
Integrating Frailty, Comorbidity, and Quality of Life in the

Management of Myelodysplastic Syndromes
Gregory A. Abel, MD, MPH, and Rena Buckstein, MD, FRCPC
OVERVIEW
Myelodysplastic syndromes (MDS) are a group of acquired hematopoietic stem cell disorders that manifest with progressive
bone marrow failure and have a propensity to transform into leukemia. Although an increase in biologic understanding of
MDS has led to improved patient risk stratification and prognostication, advances in treatment have lagged behind. While
hematopoietic cell transplantation (HCT) is a potentially curative option for some, most affected patients continue to be
treated with supportive care or with drugs that offer temporary palliation such as hematopoietic growth factors, DNA
hypomethylating agents, or immunomodulatory therapy. For several groups, such as those with intermediate-risk disease as
classified by the Revised International Prognostic Scoring System (IPSS-R) or those with higher-risk disease for whom
hypomethylating agents have failed, optimal treatment remains uncertain. Inclusion of patient-related factors such as frailty
and comorbid conditions into risk assessment can improve prognostication beyond the disease-associated variables included in systems such as the IPSS-R. This article focuses on approaches to assessing and integrating frailty, comorbidities,
and quality of life into the treatment of patients with MDS.
yelodysplastic syndromes are a group of acquired

hematopoietic stem cell disorders that manifest with
progressive bone marrow failure and have a propensity to
transform into acute myeloid leukemia.1 Although supportive care had long been the standard for management of
MDS,2 there are now several medications widely used for its
treatment and many others under investigation.3 Still, only
about 50% of patients diagnosed with MDS are alive 3 years
after diagnosis, a figure even worse for those who develop
MDS related to prior treatment of cancer.4,5 The majority of
patients with MDS ultimately die of causes related to their
disease, such as infection, progression to acute myeloid
leukemia, and hemorrhage.6
For people living with MDS, the burden of disease can be
highly troublesome. For example, one study estimated the
3-year incidence of anemia, neutropenia, and thrombocytopenia to be 81%, 25%, and 41%, respectively, and 3-year
rates of hospitalizations (for infection or bleeding associated
with low blood counts), emergency department visits, and
transfusion to be 62%, 42%, and 45%.7 In addition, health
care costs are high, estimated at $51,066 yearly (2009 US$)
for patients who are transfusion-dependent (even without
accounting for the cost of current drug therapies).8 Moreover, patients with MDS suffer from impairment to their
quality of life (QOL),9 and, because the disease is incurable
except with allogeneic HCT, treatment decisions often focus

on improving QOL.
Determining the optimal treatment strategy is difficult for
patients with MDS due to the advanced age of affected
patients and heterogeneity in disease outcomes.10 Several
scoring systems have been developed to help guide clinical
decision making. These include the 1997 IPSS11 and its 2012
revision, the IPPS-R,12 the World Health Organization (WHO)
Prognostic Scoring System (WPSS),13 and the MD Anderson
Cancer Center MDS Model (MDACC Model).14 These systems use various combinations of histologic classification,
bone marrow blast percentage, cytogenetics, number and
severity of cytopenias, performance status, age, and red cell
transfusion dependence to assign patients to different risk
categories used by clinicians to tailor treatment approaches.
These include observation alone, supportive care with
transfusions, hypomethylating agents, immunomodulatory
drugs, HCT, and enrollment in clinical trials.
There are two groups of patients for whom current scoring
systems are of limited utility. First, the systems do not help
determine when patients with lower-risk disease at presentation should start therapy. Second, for patients in the
intermediate-risk categories, it is often unclear whether
treatments tailored to lower-risk or higher-risk patients are
more likely to be useful. Increasingly, molecular genetic
From the Department of Medical Oncology, Dana-Farber Cancer Institute and Harvard Medical School, Boston, MA; Department of Medical Oncology/Hematology, Odette Cancer and
Sunnybrook Health Sciences Center, Toronto, ON, Canada.
Corresponding author: Gregory A. Abel, MD, MPH, Dana-Farber Cancer Institute, 450 Brookline Ave., Dana 1106, Boston, MA 02215; email: gregory_abel@dfci.harvard.edu.
e337
ABEL AND BUCKSTEIN
testing can aid in risk stratification15 beyond the systems

mentioned above, and, in some cases, can identify candidates for specific clinical trials. Although such genomic
advances are promising, this article will address how three
additional concepts not specifically addressed in the risk
models abovefrailty, comorbidity, and QOLcan be formally assessed and potentially help with treatment decisions
for patients with MDS.
FRAILTY
Although many definitions exist, geriatricians largely agree
that frailty is a state of high vulnerability for adverse health
outcomes, including disability, dependency, falls, need for
long-term care, and mortality.16 Frailty is common in older
patients with cancer and has been associated with treatmentrelated toxicity, poor response to therapy, and worse overall
survival.17,18 Given the synergistic effects of chemotherapy
and underlying malignancy on the immune and hematopoietic
systems, older patients with blood cancers such as MDS are at
particular risk for the sequelae of frailty. For example, preliminary work in older patients with lymphoma,19 acute myeloid
leukemia,20 myeloma,21 and MDS22 has revealed that markers
of frailty are independent prognostic factors for morbidity and
death in models that included conventional risk factors such
as disease risk group, comorbidity, and performance status.
Importantly, older patients with MDS have been shown to
be less likely to receive active treatment23 and are also less
likely to receive high-quality care such as baseline marrow
cytogenetic testing.24 Although such decisions may be
reasonable for elderly patients who are frail (e.g., a diagnostic bone marrow assessment with cytogenetic testing may not be appropriate if a patient is too frail for
treatment), they are not reasonable for the elderly who are
robust. Stated another way, age is an imprecise proxy for
frailty.
A recent study assessed the impact of frailty on survival
in a large cohort of patients with MDS and chronic myelomonocytic leukemia (445 patients).25 Frailty was assessed by
clinical judgment and also with a combination of physical
measures such as hand grip strength26 and ability to get out
of a chair 10 times and walk 4 meters.27 Using a schema
KEY POINTS
e338
Frailty is distinct from age, can be measured with

precision, and is associated with survival.
The MDS-specific comorbidity index is a scale that
reliably measures comorbidity for patients with MDS.
Depth of comorbidity can be integrated into decision
making for all levels of disease risk.
Worse quality of life predicts poorer survival in MDS,
even when accounting for disease risk.
MDS-specific quality of life can be measured with
precision, but how to best integrate these data into
clinical decisions remains to be discovered.
developed by Rockwood and associates, known as the Clinical Frailty Scale (CFS), patients were assigned scores of 1 to 9.
A score 1 indicates very fit, 4 indicates vulnerable, and 8
indicates very severely frail (Fig. 1). The median age of enrolled patients was 73 (similar to the median age that MDS is
diagnosed in the United States),5 and 79% of patients had
IPSS-R scores of intermediate or lower. Frailty was found to
correlate only modestly with Eastern Cooperative Oncology
Group (ECOG) performance status (r = 0.39; p , .0001), less
with comorbidity as assessed by the MDS-specific comorbidity index (MDS-CI28; r = 0.33; p , .0001), and minimally
with age-adjusted IPSS-R (r = 0.12; p = .03). Frailty improved
the prognostication of the IPSS-R in all but the highest-risk
group. In a multivariate analysis that included IPSS-R and
comorbidity scores, frailty was independently associated
with survival (hazard ratio [HR] 2.7; 95% CI, 1.74.2).
Moreover, incorporation of frailty improved MDS risk stratification by 30%.
The goal of frailty screening is to estimate a patients
physiological age when considering treatment options and
goals of care. It can also help predict practical outcomes,
such as a tendency toward falls or hospital admission. Frailty
screening is not easily achieved with a self-administered
questionnaire, as cognitive issues may preclude completion,
and several aspects of frailty are best captured through
in-person functional examination. Many frailty assessment
instruments are based on Frieds phenotype model,29 which
assesses physiologic vulnerability: weight loss, poor grip
strength, slow gait speed, low physical activity, and selfreported exhaustion. An alternative model defines frailty as
the cumulative effect of individual deficits and considers
more than 30 additional factors (e.g., self-reported health,
questions about functioning in the home, and presence of
chronic illnesses), creating a continuous index.30 A cumulative frailty index may have improved discriminatory ability
compared with categorical measures and can incorporate
readily available clinical data.
Given its potential contribution to mortality, we suggest
that frailty be assessed in a rigorous manner for older
patients with MDS. Although the gold standard is comprehensive geriatric assessment with a trained geriatrician,31 such an approach is not practical at most MDS
treatment facilities. The above mentioned 9-point CFS is
assessed based on clinical judgment and is highly correlated with the more comprehensive Canadian Study of
Health and Aging Frailty Index (which assesses between
30-100 deficits).32 For those patients with MDS who are
of advanced age, or those who seem to lack physiologic
capacity, we suggest that clinicians consider patients as
fitting within one of three larger categories of the CFS: 1-3
(robust to prefrail), 4 (vulnerable), and 5-8 (frail). Clinicians
can assign a score by reviewing the patients history and
soliciting input directly from patients and caregivers,
asking Which do you think best describes you? and then
reading the descriptions from the most likely categories
(Fig. 1; we do not suggest specifying that this is for a frailty
assessment).
FIGURE 1. The Clinical Frailty Scale
Reproduced with permission from Rockwood et al.32
Routine geriatric screening assessment of newly presenting elderly patients with blood cancers by a trained
nonphysician clinic assistant has been shown to be feasible
and to correlate with the aggressiveness of subsequent
treatment decisions.33 Unfortunately, there are currently no
guidelines regarding the incorporation of frailty assessment
into MDS clinical practice, and consensus recommendations are needed. For example, elderly, robust patients
with intermediate-risk MDS may be referred for reducedintensity conditioning HCT, while vulnerable and frail
patients might be referred for comprehensive geriatrician
management. By addressing activities of daily living, mobility/
fall risk, nutrition, poly-pharmacy, and mood, geriatric consultation has been shown to not only impact oncologic
treatment decisions but also address many other comorbid
issues for older patients with cancer.34 Moreover, in a recent systematic review of 18 publications assessing geriatric
domains for patients with blood cancers, physical capacity,
nutritional status, and comorbidity were found to have independent predictive value for survival and chemotherapyrelated nonhematological toxicity and often performed better
than age and performance status.35
COMORBIDITY
Patients with MDS are often of advanced age at diagnosis
such that the majority have at least one additional comorbid
condition.24,28 Although frailty is associated with comorbid
burden,36,37 it is felt to be an overlapping but distinct
concept.16 One explanation is that much of the frailty seen in
patients with a syndrome such as MDS may be due to the
disease itself and cannot be attributed to additional

medical diagnoses. Thus, a patient with MDS and with no
comorbid conditionsor several comorbidities that are
well-managedmay still be considered frail due to the
effects of MDS alone (such as advanced anemia and/or
recurrent infections). Indeed, in the study of frailty among
patients with MDS cited above,25 both frailty and Charlson
Comorbidity Index (CCI)38 were independently prognostic
of overall survival in models that accounted for MDS
disease risk.
Analyses of MDS-related administrative data have demonstrated that comorbidities are associated with worse
survival. In one study using registry data linked to Medicare
claims (SEER-Medicare; 1,708 patients), patients with
comorbid conditions had greater risk of death than those
without comorbid conditions. Compared with patients with
a CCI of 0 (no comorbidity), those with a CCI of 1 or 2
(HR 1.19; 95% CI, 1.051.36) and those with a CCI of 3 or
greater (HR 1.77; 95% CI, 1.502.08) had worse overall
survival.39 Another study characterized the impact of
comorbidities on overall survival for patients with MDS
using a clinical database,40 assessing comorbidity with
the Comorbidity Evaluation-27 (ACE-27) scale,41 a measure
originally developed for patients with head and neck
cancer. Among the 600 patients with MDS, 77% had at
least one comorbid condition, and, although there was no
significant association between level of comorbidity and
leukemic transformation, patients with the highest level of
comorbidity had a 50% decrease in survival compared with
those with none, independent of age and IPSS risk group. A
follow-up study with the same cohort stratified by the
e339
ABEL AND BUCKSTEIN
revised IPSS found similar results.42 In both analyses,

comorbidity significantly impacted the median survival of
patients age 65 or younger but did not impact those older
than age 65.
One of the most important clinical situations in which
to formally assess comorbidity for patients with MDS is
when considering fitness for HCT. Many clinicians use the
HCT-comorbidity index (HCT-CI) for this purpose. The
HCT-CI is a measure developed by Sorror et al43 that has
been shown to have clinical utility specifically for patients
with MDS and acute myeloid leukemia.44 The HCT-CI was
developed to improve on the CCI, as some elements of the
CCIsuch as presence of AIDS, comorbid leukemia, and
comorbid lymphomawere not relevant for many patients who underwent HCT. The HCT-CI is comprised of
17 groups of disease, each weighted with points from 1 to
3. For example, the presence of diabetes merits 1 point,
and the presence of severe pulmonary disease merits
3 points. The resulting scores on the HCT-CI are: low
(0 points), intermediate (1 to 2 points), and high ($ 3
points).
In a retrospective analysis of patients with MDS from an
Austrian registry that used the HCT-CI (616 patients),45
the highest frequencies of specific comorbidities were
cardiovascular disease (28%), diabetes (12%), and prior
solid tumors (10%). Comorbidities were more frequent in
men and patients older than age 65; moreover, in multivariable models including age and IPSS, patients in higher
HCT-CI groups had worse overall survival (HR 1.26; 95% CI,
1.11.5). Building on the success of the HCT-CI for MDS
and other blood cancers, Della Porta et al created a
time-dependent, MDS-specific comorbidity index, the
MDS-CI.28
To create the MDS-CI, a learning cohort of 840 Italian
patients with MDS was assigned comorbidities using
HCT-CI definitions. Next, multivariable models were created
to assess risk of nonleukemic death, and each comorbidity
was given a score proportional to the regression coefficient of the resulting multivariable Cox proportional hazards model. Only five of the 17 HCT-CI variables were
significant: cardiac disease (2 points), moderate to severe
hepatic disease (1 point), severe pulmonary disease
(1 point), renal disease (1 point), and current or prior history
of solid tumor (1 point). Points were added together to
create low risk (0 points), intermediate risk (1 to 2 points), or
high risk (. 2 points) scores. These categories identified
three groups that showed significantly different probabilities of nonleukemic death and survival (p , .002 and
p = .005, respectively), a finding that was also reproduced in a validation cohort of German patients with MDS
(504 patients).
In a multivariable analysis including age, sex, WHO categories, cytogenetics, and transfusion-dependency, the MDSCI was an independent predictor of nonleukemic death (HR
1.89; p , .001) and survival (HR 1.67; p , .001).28 In additional data sets, the MDS-CI has been shown to add
prognostic information beyond established risk scoring
e340
systems.46-49 For example, one study (318 patients) found

the MDS-CI to be more effective than the HCT-CI in prognostication even when accounting for age, but only for
patients with lower-risk disease on the IPSS.46 Another study
(60 patients) found that the MDS-CI was able to identify
patients with improved outcomes after azacitidine,48 and
yet another (363 patients) found that the MDS-CI refined the
risk stratification of the IPSS-R.49
We suggest integrating the MDS-CI or the HCT-CI with
one of the commonly used MDS scoring systems such as the
IPSS-R or WPSS. A schema for practically doing so with the
WPSS is shown in Fig. 2.28 For example, for patients with
lower WPSS who are high risk on the MDS-CI, the focus
should be on preventing or ameliorating anemia, which
might exacerbate comorbid conditions. For patients who
are low risk on the MDS-CIparticularly those lacking
cardiac conditionsmore severe anemia might be tolerated. As another example, for patients with higher-risk
WPSS, clinical decisions regarding HCT can be helped with
the HCT-CI. If HCT is not available due to patient preference or lack of suitable donor, the MDS-CI may also
help in addressing potential benefits of hypomethylating
agents, although more research is needed to evaluate
how comorbidity scores can predict clinical outcomes and
tolerance after therapy.
QUALITY OF LIFE
Quality of life is defined by the WHO as the net consequence of life characteristics on a persons perception of
their position in life, in the context of the culture and value
systems in which they live, and in relation to their goals,
expectations, standards, and concerns.50,51 The key to
understanding patient-reported outcomes such as QOL is
that they are subjective and dynamic. Given that many
patients with MDS have a chronic illness with a relatively long survival and opportunity for many lived experiences with the disease and its consequence, QOL has been
demonstrated to be impaired for many affected patients.52,53 In a study of patients age 60 or older with acute
myeloid leukemia or advanced MDS (43 patients), although
initial QOL assessment was not associated with treatment
choice, 97% of patients reported that QOL was more important to them than length of life.54 Moreover, rigorous
measurement of QOL has been recognized as a priority for
MDS research.55-57
As an example, in one of the large clinical trials of azacitidine in MDS, QOL was found to be improved among
patients receiving therapy when assessed with a cancerrelated measure,58 data that undoubtedly helped lead to
regulatory approval of that medication. More recent work
by Efficace et al found that among patients with higher-risk
MDS (280 patients), self-reported fatigue (measured from
the cancer-specific EORTC QLQ-C30) had prognostic value
beyond standard MDS risk classification systems.59 Patients scoring equal or higher than the median on that
measures fatigue scale (34 or greater out of 100; higher
FIGURE 2. Schematic Representation of the Potential of Combined Use of WPSS and MDS-CI in Clinical Decision
Making in MDS
Reproduced with permission from Della Porta et al.28
scores or seen with more fatigue) had a worse median

overall survival than patients scoring lower than the median
(HR 1.6; p = .0013). In a multivariable analysis, high-risk IPSS
score (HR 2.5 vs. intermediate-2 risk; p = .004) and higher
fatigue score (HR 1.1 for every 10 points; p = .0007) were
both associated with poorer overall survival, a result also
seen when risk stratifying patients with the IPSS-R and the
WPSS.
Studies such as these argue that patient-reported outcomes should be a paramount consideration when making
treatment decisions for patients with MDS, not only because
they can help determine if patients QOL will potentially
improve (or be harmed) by treatment, but also because they
may have prognostic value. Another question that arises is
whether QOL should be a primary consideration when deciding on potentially curative therapy such as HCT. One
reason it might be a primary consideration is that pre-HCT
QOL likely adds prognostic information. Indeed, recent data
suggest that QOL prior to HCT for blood cancers is associated
with post-HCT mortality, even when controlling for performance status and comorbidity.60,61 Still, another study
demonstrated that among patients with higher-risk MDS age
60 to 75 with good performance status, scores on several
QOL measures were not associated with ultimate receipt of
reduced-intensity allogeneic HCT.62 Given that the best
treatment of HCT-eligible older adults with MDS is not
known, these results suggest that QOL is either not wellintegrated into HCT treatment decisions or not rigorously
assessed.
This may be because most studies assessing QOL in
MDSincluding most of the ones aboveuse general
measures such as the Short-Form Health Survey (SF-36)63

or cancer-specific scales such as the EOTC QLQ-C30
(EORTC-30)64 or the Functional Assessment of Cancer
Therapy-Anemia (FACT-An).65 Although these are useful,
they may not be specific enough to capture all of the
elements important to MDS-related QOL. Indeed, several
important questions about QOL in MDS remain unanswered.
For example, the impact of the erythropoiesis-stimulating
agents on patient-reported outcomes in MDS is still
unknown, with much contradictory data.66 In addition,
despite the existence of many associated studies, it is not
clear whether red cell transfusions improve MDS-related
QOL.67
Moreover, given the many ways that fatigue may affect
patient functioning, it has been shown that physicians
informal QOL assessment may not reveal what patients
actually experience. In a study pairing a total of 280 patients with higher-risk MDS with 68 physicians, patientphysician concordance regarding assessments of overall
health on a seven-point scale was found only 28% of the
time (weighted kappa, 0.270).68 With no consistent way of
measuring QOL in MDS clinical studies, and clinicians
imperfect assessments of patients well-being and experiences, an MDS-specific QOL instrument is needed to capture the experiences of patients with MDS in the modern
treatment era.
The Quality of Life in Myelodysplasia Scale (QUALMS)69
may help address this need. Developed after a series of
focus groups with patient with MDS and providers, and
validated by an international group of MDS researchers
(including the authors of this review),70 the QUALMS is a
e341
ABEL AND BUCKSTEIN
SIDEBAR 1. Topics Addressed in the QUALMS, an

MDS-Specific Measure of Quality of Life70
Too tired for prior responsibilities
Low energy changes schedule
Feeling weak
Unable to participate in activities
Take into account fatigue when planning
Worry about becoming burden
Felt hopelessness
Change in bowels
Shortness of breath
Changing long-term plans due to health
Trouble concentrating
Life organized around medical appointments
Nauseated
No energy for routine tasks
Family relationships strained
Feeling grateful
Getting quality information
Feel gratitude when prior took for granted
Bruising
Avoiding crowds
Could not do anything about disease
Disease feels unpredictable
Lack of concrete answers
No clear information
Afraid of dying
Difficulty explaining MDS to others
Worry about progressing/leukemia
Anxious about tests or laboratory results
Angry about diagnosis
Worried about infection
Feel limited emotional support available
Worried about bleeding
Concerned about being a financial burden
Concerned about losing job
Too tired to drive
Afraid to have sex due to blood counts
Worried MDS treatment will stop working
Too tired to take care of family
brief (less than 10 minute) questionnaire consisting of

38 items (Sidebar 1). Scored on a scale of 0 to 100 (higher
score is correlated with improved MDS-specific QOL), the
QUALMS recently underwent validation in United States,
Canada, and Italy (255 patients), where patients were
assessed twice with the instrument and several others. The
measure was internally consistent (a = 0.92) and moderately correlated with both the QLQ-C30 and FACT-An
scales. Moreover, patients with hemoglobin levels of
8 g/dL or lower scored lower than those with hemoglobin
higher than 10 g/dL (61.8 vs. 71.1; p , .001), and
transfusion-dependent patients scored lower than
transfusion-independent patients (62.4 vs. 69.7; p , .01).
There was good overall test-retest reliability among patients with stable hemoglobin (r = 0.81), and significant
changes for patients hospitalized or with infections
e342
between administrations (both p , .01). Principal components analysis revealed physical burden, benefit-finding,
and emotional burden subscales. Additional validation efforts are ongoing to determine the utility of the measure and subscales both for clinical trials and for clinical
decision-making.
We suggest that patients with MDS be asked specifically
about their QOLat diagnosis and after treatment has
begunusing a validated general, cancer, or MDS-specific
scale. If QOL is impaired, these data can be used together
with disease-specific risk factors and molecular genetic
studies to help inform treatment decisions. For example,
for patients with MDS who have low-risk-disease and
might otherwise be observed, reduced QOL may signify the need to start erythropoiesis-stimulating agents,
transfusions, or disease-modifying treatments earlier
than would otherwise be considered. Alternatively, for
patients with higher-risk disease who are being treated
with hypomethylating agents but do not seem to be
responding in terms of counts, formal documentation
of improved QOL may argue in favor of continuing
treatment.
CONCLUSION
Myelodysplastic syndromes are a group of diseases that
shorten life expectancy and negatively impact QOL. The
disease-related factors that predict survival and progression to acute myeloid leukemia are becoming increasingly
known but remain imprecise when patient-related factors
are excluded. Frailty and comorbidity have been convincingly shown to further refine current clinical prognostic risk scores. These factors are inextricably linked. We
believe a baseline frailty assessment does the best job at
staging the aging in MDS, and should be considered by all
clinicians at diagnosis and before initiating treatment. The
CFS is quick and simple to deploy in clinic. Patients deemed
vulnerable (score = 4) or mildly frail (score = 5) should have
poorly controlled comorbidities addressed, as these may be
contributing to frailty. Frankly frail patients (score = 6 or
higher) should be referred for comprehensive geriatric assessment and management, which includes comorbidity
assessment.
We realize that it is not practical to assess every patient
with MDS formally for frailty and comorbidity, but since a
good medical history and physical examination will solicit
most comorbidity data, given the choice, we would start
with frailty assessment. On the other hand, patients with
MDS who are considered for allogeneic HCT should have
both frailty and comorbidity formally assessed (CFS and
HCT-CI or MDS-CI), regardless of age. Finally, MDS clinicians
must aim to both extend life as well as to improve its
quality. In addition to guiding therapeutic decisions as
discussed above, periodic rigorous assessment of QOL has
been shown to enhance the experiences of patients
with cancer71 and promises to do the same for patients
with MDS.
References
1. Jaffe ES, Harris NL, Stein H, et al. World Health Organization Classification of Tumours: Pathology and Genetics, Tumors of Haematopoietic
and Lymphoid Tissues. Lyon, France: International Agency for Research
on Cancer Press; 2001.
2. Greenberg PL, Baer MR, Bennett JM, et al. Myelodysplastic syndromes clinical
practice guidelines in oncology. J Natl Compr Canc Netw. 2006;4:58-77.
3. Stone RM. How I treat patients with myelodysplastic syndromes. Blood.
2009;113:6296-6303.
4. Cogle CR, Craig BM, Rollison DE, et al. Incidence of the myelodysplastic
syndromes using a novel claims-based algorithm: high number of uncaptured cases by cancer registries. Blood. 2011;117:7121-7125.
5. Ma X, Does M, Raza A, et al. Myelodysplastic syndromes: incidence and
survival in the United States. Cancer. 2007;109:1536-1542.
6. Dayyani F, Conley AP, Strom SS, et al. Cause of death in patients with
lower-risk myelodysplastic syndrome. Cancer. 2010;116:2174-2179.
7. Lindquist KJ, Danese MD, Mikhael J, et al. Health care utilization and
mortality among elderly patients with myelodysplastic syndromes. Ann
Oncol. 2011;22:1181-1188.
8. Frytak JR, Henk HJ, De Castro CM, et al. Estimation of economic costs
associated with transfusion dependence in adults with MDS. Curr Med
Res Opin. 2009;25:1941-1951.
9. Thomas ML. The impact of myelodysplastic syndromes on quality of life:
lessons learned from 70 voices. J Support Oncol. 2012;10:37-44.
10. Klepin HD, Rao AV, Pardee TS. Acute myeloid leukemia and myelodysplastic syndromes in older adults. J Clin Oncol. 2014;32:2541-2552.
11. Greenberg P, Cox C, LeBeau MM, et al. International scoring system for
evaluating prognosis in myelodysplastic syndromes. Blood. 1997;89:
2079-2088.
12. Greenberg PL, Tuechler H, Schanz J, et al. Revised international
prognostic scoring system for myelodysplastic syndromes. Blood. 2012;
120:2454-2465.
13. Malcovati L, Germing U, Kuendgen A, et al. Time-dependent prognostic scoring system for predicting survival and leukemic evolution in
myelodysplastic syndromes. J Clin Oncol. 2007;25:3503-3510.
14. Kantarjian H, OBrien S, Ravandi F, et al. Proposal for a new risk model in
myelodysplastic syndrome that accounts for events not considered in
the original International Prognostic Scoring System. Cancer. 2008;113:
1351-1361.
15. Bejar R, Stevenson K, Abdel-Wahab O, et al. Clinical effect of point
mutations in myelodysplastic syndromes. N Engl J Med. 2011;364:
2496-2506.
16. Fried LP, Ferrucci L, Darer J, et al. Untangling the concepts of disability,
frailty, and comorbidity: implications for improved targeting and care.
J Gerontol A Biol Sci Med Sci. 2004;59:255-263.
17. Handforth C, Clegg A, Young C, et al. The prevalence and outcomes of frailty
in older cancer patients: a systematic review. Ann Oncol. 2014;26:1091-1101.
18. Puts MT, Tapscott B, Fitch M, et al. A systematic review of factors
influencing older adults decision to accept or decline cancer treatment.
Cancer Treat Rev. 2015;41:197-215.
19. Tucci A, Ferrari S, Bottelli C, et al. A comprehensive geriatric assessment
is more effective than clinical judgment to identify elderly diffuse large
cell lymphoma patients who benefit from aggressive therapy. Cancer.
2009;115:4547-4553.
20. Sherman AE, Motyckova G, Fega KR, et al. Geriatric assessment in older
patients with acute myeloid leukemia: a retrospective study of associated treatment and outcomes. Leuk Res. 2013;37:998-1003.
21. Palumbo A, Bringhen S, Mateos MV, et al. Geriatric assessment predicts
survival and toxicities in elderly myeloma patients: an International
Myeloma Working Group report. Blood. 2015;125:2068-2074.
22. Fega KR, Abel GA, Motyckova G, et al. Non-hematologic predictors of

mortality improve the prognostic value of the international prognostic
scoring system for MDS in older adults. J Geriatr Oncol. 2015;6:288-298.
23. Gattermann N, Kundgen A, Kellermann L, et al. The impact of age on the
diagnosis and therapy of myelodysplastic syndromes: results from a
retrospective multicenter analysis in Germany. Eur J Haematol. 2013;
91:473-482.
24. Abel GA, Cronin AM, Odejide OO, et al. Influence of patient and provider
characteristics on quality of care for the myelodysplastic syndromes. Br
J Haematol. Epub 2016 Feb 23.
25. Buckstein R, Wells R, Zhu N, et al. Patient-related factors independently
impact overall survival in patients with myelodysplastic syndromes: an
MDS-CAN prospective study. Br J Haematol. Epub 2016 March 15.
26. Bohannon RW. Reference values for extremity muscle strength obtained by hand-held dynamometry from adults aged 20 to 79 years.
Arch Phys Med Rehabil. 1997;78:26-32.
27. Guralnik JM, Ferrucci L, Simonsick EM, et al. Lower-extremity function in
persons over the age of 70 years as a predictor of subsequent disability.
N Engl J Med. 1995;332:556-561.
28. Della Porta MG, Malcovati L, Strupp C, et al. Risk stratification based on
both disease status and extra-hematologic comorbidities in patients
with myelodysplastic syndrome. Haematologica. 2010;96:441-449.
30. Rockwood K, Mitnitski A. Frailty in relation to the accumulation of
deficits. J Gerontol A Biol Sci Med Sci. 2007;62:722-727.
31. Dale W. Staging the aging when considering androgen deprivation
therapy for older men with prostate cancer. J Clin Oncol. 2009;27:
3420-3422.
32. Rockwood K, Song X, MacKnight C, et al. A global clinical measure of
fitness and frailty in elderly people. CMAJ. 2005;173:489-495.
33. Abel G, Hshieh T, Condron N, et al. Feasibility of routine frailty screening
assessment for patients in a hematologic oncology clinic: Results from a
pilot study. Presented at: 57th ASH Annual Meeting and Exposition;
2015; Orlando, FL.
34. Hamaker ME, Schiphorst AH, ten Bokkel Huinink D, et al. The effect of a
geriatric evaluation on treatment decisions for older cancer patients
a systematic review. Acta Oncol. 2014;53:289-296.
35. Hamaker ME, Prins MC, Stauder R. The relevance of a geriatric assessment
for elderly patients with a haematological malignancya systematic review. Leuk Res. 2014;38:275-283.
36. Garca-Garca FJ, Carcaillon L, Fernandez-Tresguerres J, et al. A new
operational definition of frailty: the Frailty Trait Scale. J Am Med Dir
Assoc. 2014;15:371.e7-371.e13.
37. Gharacholou SM, Roger VL, Lennon RJ, et al. Comparison of frail patients
versus nonfrail patients $65 years of age undergoing percutaneous
coronary intervention. Am J Cardiol. 2012;109:1569-1575.
38. Charlson ME, Pompei P, Ales KL, et al. A new method of classifying
prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373-383.
39. Wang R, Gross CP, Halene S, et al. Comorbidities and survival in a large
cohort of patients with newly diagnosed myelodysplastic syndromes.
Leuk Res. 2009;33:1594-1598.
40. Naqvi K, Garcia-Manero G, Sardesai S, et al. Association of comorbidities
with overall survival in myelodysplastic syndrome: development of a
prognostic model. J Clin Oncol. 2011;29:2240-2246.
41. Paleri V, Wight RG. Applicability of the adult comorbidity evaluation - 27
and the Charlson indexes to assess comorbidity by notes extraction in a
e343
ABEL AND BUCKSTEIN
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
cohort of United Kingdom patients with head and neck cancer: a retrospective study. J Laryngol Otol. 2002;116:200-205.
Daver N, Naqvi K, Jabbour E, et al. Impact of comorbidities by ACE-27 in
the revised-IPSS for patients with myelodysplastic syndromes. Am J
Hematol. 2014;89:509-516.
Sorror ML, Maris MB, Storb R, et al. Hematopoietic cell transplantation
(HCT)-specific comorbidity index: a new tool for risk assessment before
allogeneic HCT. Blood. 2005;106:2912-2919.
Sorror ML, Sandmaier BM, Storer BE, et al. Comorbidity and disease
status based risk stratification of outcomes among patients with acute
myeloid leukemia or myelodysplasia receiving allogeneic hematopoietic cell transplantation. J Clin Oncol. 2007;25:4246-4254.
Bammer C, Sperr WR, Kemmler G, et al. Clustering of comorbidities is
related to age and sex and impacts clinical outcome in myelodysplastic
syndromes. J Geriatr Oncol. 2014;5:299-306.
Balleari E, Salvetti C, Del Corso L, et al. Age and comorbidities deeply
impact on clinical outcome of patients with myelodysplastic syndromes.
Leuk Res. 2015;39:846-852.
Breccia M, Federico V, Latagliata R, et al. Evaluation of comorbidities at
diagnosis predicts outcome in myelodysplastic syndrome patients. Leuk
Res. 2011;35:159-162.
Breccia M, Salaroli A, Loglisci G, et al. MDS-specific comorbidity index is
useful to identify myelodysplastic patients who can have better outcome with 5-azacitidine. Haematologica. 2012;97:e2.
van Spronsen MF, Ossenkoppele GJ, Holman R, et al. Improved risk
stratification by the integration of the revised international prognostic
scoring system with the myelodysplastic syndromes comorbidity index.
Eur J Cancer. 2014;50:3198-3205.
World Health Organization. Study protocol for the World Health Organization project to develop a Quality of Life assessment instrument
(WHOQOL). Qual Life Res. 1993;2:153-159.
World Health Organization. The World Health Organization Quality
of Life assessment (WHOQOL): position paper from the World Health
Organization. Soc Sci Med. 1995;41:1403-1409.
Efficace F, Gaidano G, Breccia M, et al. Prevalence, severity and correlates of fatigue in newly diagnosed patients with myelodysplastic
syndromes. Br J Haematol. 2015;168:361-370.
Steensma DP, Heptinstall KV, Johnson VM, et al. Common troublesome
symptoms and their impact on quality of life in patients with myelodysplastic syndromes (MDS): results of a large internet-based survey.
Leuk Res. 2008;32:691-698.
Sekeres MA, Stone RM, Zahrieh D, et al. Decision-making and quality of
life in older adults with acute myeloid leukemia or advanced myelodysplastic syndrome. Leukemia. 2004;18:809-816.
Caocci G, La Nasa G, Efficace F. Health-related quality of life and
symptom assessment in patients with myelodysplastic syndromes. Expert
Rev Hematol. 2009;2:69-80.
Estey E. Acute myeloid leukemia and myelodysplastic syndromes in
older patients. J Clin Oncol. 2007;25:1908-1915.
Malcovati L, Hellstrom-Lindberg E, Bowen D, et al; European Leukemia
Net. Diagnosis and treatment of primary myelodysplastic syndromes in
e344
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
adults: recommendations from the European LeukemiaNet. Blood.

2013;122:2943-2964.
Kornblith AB, Herndon JE II, Silverman LR, et al. Impact of azacytidine on
the quality of life of patients with myelodysplastic syndrome treated
in a randomized phase III trial: a Cancer and Leukemia Group B study.
J Clin Oncol. 2002;20:2441-2452.
Efficace F, Gaidano G, Breccia M, et al. Prognostic value of self-reported
fatigue on overall survival in patients with myelodysplastic syndromes:
a multicentre, prospective, observational, cohort study. Lancet Oncol.
2015;16:1506-1514.
Hamilton BKR, Abounder L, Dabney DM, et al. Prognostic significance
of quality of life in patients undergoing allogeneic hematopoietic cell
transplantation. Presented at: 56th ASH Annual Meeting and Exposition; 2014; San Francisco, CA.
Wood WAL-R, Fei J, Logan MBR, et al. Patient-reported quality of life is
an independent predictor of survival after allogeneic hematopoietic
cell transplantation: a secondary analysis from the Blood and Marrow
Transplant Clinical Trials Network (BMT CTN) 0902. Presented at: 56th
ASH Annual Meeting and Exposition; 2014; San Francisco, CA.
El-Jawahri A, Kim H, Steensma D, et al. Does quality of life impact the
decision to pursue stem cell transplantation for elderly patients with
advanced MDS? BMT. Epub 2016 March 21.
McHorney CA, Ware JE Jr, Lu JF, et al. The MOS 36-item Short-Form
Health Survey (SF-36): III. Tests of data quality, scaling assumptions,
and reliability across diverse patient groups. Med Care. 1994;32:
40-66.
Aaronson NK, Ahmedzai S, Bergman B, et al. The European Organization
for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer
Inst. 1993;85:365-376.
Cella D. The Functional Assessment of Cancer Therapy-Anemia (FACTAn) Scale: a new tool for the assessment of outcomes in cancer anemia
and fatigue. Semin Hematol. 1997;34(3, Suppl 2)13-19.
Rizzo JD, Somerfield MR, Hagerty KL, et al. Use of epoetin and darbepoetin in patients with cancer: 2007 American Society of Hematology/
American Society of Clinical Oncology clinical practice guideline update.
Blood. 2008;111:25-41.
Pinchon DJ, Stanworth SJ, Doree C, et al. Quality of life and use of red cell
transfusion in patients with myelodysplastic syndromes. A systematic
review. Am J Hematol. 2009;84:671-677.
Caocci G, Voso MT, Angelucci E, et al. Accuracy of physician assessment
of treatment preferences and health status in elderly patients with
higher-risk myelodysplastic syndromes. Leuk Res. 2015;39:859-865.
Abel GA, Klaassen R, Lee SJ, et al. Patient-reported outcomes for the
myelodysplastic syndromes: a new MDS-specific measure of quality of
life. Blood. 2014;123:451-452.
Abel GA, Efficace F, Buckstein RJ, et al. Prospective international validation of the Quality of Life in Myelodysplasia Scale (QUALMS).
Haematologica. Epub 2016 March 4.
Cancer Quality Council of Ontario. Symptom Assessment and Management. Toronto, Canada: Cancer Quality Council of Ontario; 2016.
BEYOND HYPOMETHYLATING AGENTS IN MDS
Searching for a Light at the End of the Tunnel? Beyond

Hypomethylating Agents in Myelodysplastic Syndromes
Rami S. Komrokji, MD
OVERVIEW
Hypomethylating agents (HMAs) are the mainstay treatment of patients with myelodysplastic syndromes. Hypomethylating
agents remain the only treatment, other than allogeneic hematopoietic stem cell transplantation (AHSCT), that improves
overall survival (OS) for patients with higher-risk myelodysplastic syndromes. It is crucial to maximize the benefit of HMAs by
selecting the appropriate dosing and schedule and continuing therapy until clear evidence of lack of response or failure of
therapy. Strategies to improve outcome with HMAs include identifying tools and biomarkers for better patient selection,
namely the ability to identify those who achieve complete response (CR) or long duration of response, combination
strategies with HMAs aim to improve response rate or its duration. The outcome of patients with myelodysplastic syndromes
after HMA failure is poor for both patients with higher-risk and lower-risk disease and represents an unmet medical need.
The best outcomes after HMA failure are reported with AHSCT or novel agents in clinical trials. This article discusses the
maximizing benefit of HMAs, offers strategies to improve outcome with HMAs, and, finally, reviews selected novel agents in
development after HMA failure.
yelodysplastic syndromes are a heterogeneous group

of neoplastic hematopoietic stem cell disorders. The
common features of the disease include (1) clinical presentation with variant cytopenias and resultant complications; (2) the presence of dysplastic morphologic cytology
features; (3) evidence of clonal hematopoiesis for the majority of patients, either by detection of chromosomal abnormalities, often copy number variation changes, or by
identification of somatic gene mutations most commonly
involving epigenetic deregulation or splicing machinery; and
(4) a tendency to progress to acute myeloid leukemia
(AML).1-3
The first step in managing myelodysplastic syndromes is
risk stratification after ensuring the proper diagnosis. The
disease staging not only provides the needed prognostic
information for patients and caregivers, but it also allows
physicians to tailor therapy accordingly. The International
Prognostic Scoring System (IPSS) had been the most widely
used tool for risk assessment in myelodysplastic syndromes.
A lump score is based on the percentage of myeloblasts,
cytopenia(s), and karyotype.4 New risk models including the
revised IPSS5 and MD Anderson models6,7 refine the staging
system in myelodysplastic syndromes. The integration of
somatic gene mutations information adds independent
prognostic value to the clinical risk models.8 Patients are
classified as having lower-risk myelodysplastic syndromes,
with expected OS measured in years and the goal of therapy

being alleviation of cytopenia, or higher-risk myelodysplastic
syndromes, with a short expected OS and therapy attempting
to alter the natural history of disease. The treatment options
for myelodysplastic syndromes remain limited. AHSCT is the
only curative option, which is offered to patients with higherrisk myelodysplastic syndromes.9 The HMAs azacitidine and
decitabine are the only approved drugs for nondeletion 5q
myelodysplastic syndrome.
Epigenetic gene silencing is a well-described mechanism
by which tumor cells downregulate expression of tumor
suppression genes.10 Silencing is often accomplished by
methylation or histone deacylation. Hypomethylating agents
are a class of nucleoside analogs that induce hypomethylation,
allowing re-expression of epigenetically silenced genes. These
agents inhibit DNA methyl transferase by forming direct adducts after incorporation into DNA, resulting in global hypomethylation. Both agents are pyrimidine analogs; however,
decitabine as a deoxy-analog is incorporated solely into DNA,
whereas azacitidine has a ribose sugar structure and is
incorporated into RNA.11,12
This article aims to briefly summarize the current data
using HMAs for treatment of myelodysplastic syndromes,
discuss possible strategies to improve outcome using HMAs,
and focus on options for managing myelodysplastic syndromes after HMA failure.
From the Department of Malignant Hematology, Moffitt Cancer Center, Tampa, FL.
Corresponding author: Rami S. Komrokji, MD, Department of Malignant Hematology, Moffitt Cancer Center, 12902 Magnolia Dr., Tampa, FL 33612; email: rami.komrokji@moffitt.org.
e345
RAMI S. KOMROKJI
HOW TO OPTIMIZE CURRENT USE OF

HYPOMETHYLATING AGENTS FOR TREATMENT
OF MYELODYSPLASTIC SYNDROMES
Hypomethylating agents are currently the standard of care
for treating patients with higher-risk myelodysplastic syndromes. Azacitidine demonstrated an improvement in OS
compared with conventional care regimens in a randomized
phase III clinical trial. In the AZA-001 study, the median OS
was significantly prolonged with azacitidine treatment:
24.4 months versus 15 months with conventional care
regimens (p = .0001). The regimen consisted of 75 mg/m2
of azacitidine subcutaneously for 7 days every 28 days.13
For decitabine use in higher-risk myelodysplastic syndromes, a phase III study comparing the U.S. Food and
Drug Administrationapproved decitabine inpatient regimen with best supportive care failed to show an OS
advantage.14 The median OS with the commonly used
20 mg/m2 intravenously for a 5-day regimen every
28 days in higher-risk myelodysplastic syndromes is 19
to 20 months. 15,16
The key points in optimizing HMA use for patients with
higher-risk myelodysplastic syndromes include using a 7-day
azacitidine regimen, assessing initial response after four to
six cycles of therapy, and continuing therapy for patients
with stable disease or better response. The HMA emerged
as an acceptable alternative to intensive chemotherapy
prior to AHSCT.17 The role of HMA maintenance or treatment of post-AHSCT relapse is being explored and could be a
promising strategy.18
Objective clinical responses are seen for only approximately one-half of the patients treated with HMAs, and the
CR rate is only 10% to 20%.13,15 Even among patients with
initial responses to HMA, the median response duration
is only 9 to 15 months, with the vast majority of patients
losing response within 2 years.
KEY POINTS
e346
Hypomethylating agents are the only drugs that

potentially alter the natural history of myelodysplastic
syndromes and improve overall survival in higher-risk
disease.
Key points in maximizing benefit of HMAs include
using a 7-day regimen in higher-risk myelodysplastic
syndromes and continuing treatment for patients with
stable disease or better response.
Strategies to improve outcome with HMAs include
biomarker-driven patient selection, combination
strategies to improve response rates or duration, and
exploration of alternate doses.
The outcome after HMA failure is poor and represents
an unmet medical need.
The best outcomes after HMA failure are reported
among patients who proceed to allogeneic stem cell
transplantation or enroll in clinical trials studying novel
agents.
In lower-risk myelodysplastic syndromes, the goal of HMA

use is to alleviate cytopenia. The 5-day regimen of azacitidine every 28 days emerged as an acceptable regimen for
patients with lower-risk myelodysplastic syndromes. Lyons
et al19 randomly assigned patients with less than 10% myeloblasts to treatment with azacitidine at a dose of 75 mg/m2
for 5 days followed by 2 days of rest and then an additional
2 days of treatment (5-2-2), 50 mg/m2 for 5 days followed by
2 days of rest and then another 5 days (5-2-5), or 75 mg/m2
for 5 consecutive days. Among the 151 patients enrolled in
the study, no difference in response was observed among
the three regimens. Grade 3 or 4 hematologic toxicity was
lower among patients treated with the 5-day consecutive
regimen.19
POTENTIAL STRATEGIES TO IMPROVE OUTCOME

WITH HYPOMETHYLATING AGENTS IN
MYELODYSPLASTIC SYNDROMES
Biomarker-Driven Patient Selection
There are no clinical parameters that can clearly predict
response to HMAs. The MDS Clinical Research Consortium
explored the prognostic utility of the IPSS, the IPSS-R, the
MD Anderson Prognostic Scoring System, the World Health
Organizationbased Prognostic Scoring System, and the
French Prognostic Scoring System among 632 patients who
presented with higher-risk myelodysplastic syndromes and
were treated with azanucleosides as the first-line therapy.20 No prognostic tool predicted the probability of
achieving an objective response. Nonetheless, all five tools
were associated with OS. The Groupe-Francophone des
My e lodysplasies (GFM) proposed a prognostic score for
azacitidine-treated patients with higher-risk myelodysplastic syndromes. Among 282 patients, the study suggested that prior treatment with low-dose cytarabine
(p = .009), bone marrow blasts greater than 15% (p = .004),
and abnormal karyotype (p = .03) independently predicted
lower response rates. Complex karyotype predicted shorter
responses (p = .0003).21
Detection of certain somatic gene mutations may affect
response to HMAs. The presence of TET-2, DNMT3A, and
IDH-1/IDH-2 gene mutations was reported to be associated
with a higher response to HMAs. Among 213 patients with
myelodysplastic syndromes treated with HMAs, Bejar et al22
reported that clonal TET-2 mutations (allele fraction . 10%)
predicted response (odds ratio, 1.99; p = .036). Response
rates were highest in the subset of patients with a TET-2
mutant without clonal ASXL1 mutations (odds ratio, 3.65;
p = .009). Mutations of TP53 (hazard ration [HR] 2.01;
p = .002) and PTPN11 (HR 3.26; p = .006) were associated
with shorter OS but not drug response.22
Combination Strategies
The goals of this strategy would be to increase the frequency, quality, and duration of clinical responses and ultimately delay/prevent the development of resistance and
prolong survival.23
A promising overall response rate (ORR) of 72% was reported

in an early phase trial of 36 patients with higher-risk myelodysplastic syndromes who received azacitidine at a standard
dose of 75 mg/m2 per day concurrently with lenalidomide.24
The North American Intergroup Study SWOG S1117 unfortunately did not confirm those findings. In this phase II
study, 277 patients were randomly assigned to azacitidine,
azacitidine/lenalidomide, or azacitidine plus vorinostat. No
differences in response rate were observed; a tendency of
longer duration of response (. 6 on therapy) was noted for the
azacitidine/lenalidomide arm. Overall response rate for patients with chronic myelomonocytic leukemia (CMML) was
significantly higher for lenalidomide plus azacitidine versus
azacitidine (63% vs. 29%; p = .04), with a trend for longer
response duration for combinations (p = .06).25
Histone deacetylase (HDAC) inhibitors are another group
of agents that have been combined with azacitidine. As
mentioned, the azacitidine/vorinostat combination did not
improve the response rate or OS over azacitidine monotherapy.25 In a randomized phase II trial of 150 patients,
azacitidine/entinostat also did not improve outcomes.26
Finally, a randomized phase II study combining pracinostat
with azacitidine did not improve response rates.27 It is clear
that concomitant use of HMAs and HDAC inhibitors does not
improve responses and an alternative schedule or sequence
must be pursued if such a combination is to be further
explored.
There are ongoing or planned studies for other agents in
combination with HMAs. Vadastuximab talirine (SGN-CD33
A), a conjugated monoclonal antibody with a highly stable
linker, is a promising agent. An AML combination with
azacitidine or decitabine was explored in a phase I study in
which six of 23 patients achieved CR and nine other patients
had complete remission with incomplete count recovery
(CRi). Seventy-two percent of patients were alive at a median 7.7 months of follow-up. Prolonged myelosuppression
was observed with treatment.28 In subtypes of myelodysplastic syndromes, PD-L1 expression on myeloblasts has
been independently associated with myelodysplastic syndrome transformation to AML.29 Hypomethylating agents
can enhance the expression of CTLA-4, PD-1, and PD-L1
among patients with myelodysplastic syndromes and AML.30
Ongoing and planned studies are exploring combining PD-L1
and PD-1 inhibitors in higher-risk myelodysplastic syndromes. A study examining oral azacitidine combined with
darvalumab is ongoing. Pevonedistat (MLN4924; TAK-924)
is a first-in-class small molecule inhibitor of the neural
precursor cell expressed developmentally downregulated
protein 8activating enzyme.31 Pevonedistat had reported
single-agent clinical activity in a phase I study of patients
with relapsed/refractory AML.31 Based on nonclinical
studies in AML models that demonstrated a synergistic lethality in cell lines and tumor regression in murine xenografts when pevonedistat was combined with azacitidine,
pevonedistat is being studied in combination with azacitidine in treatment-naive elderly patients with AML and
higher-risk myelodysplastic syndromes.32
There is unmet need for treating patients with myelodysplastic syndromes with TP53 mutation. The outcome for
these patients is poor.33,34 The median OS for patients with
detectable TP53 mutation at the time of AHSCT is less than
1 year.35 Responses to HMAs, although not different, are
often short lived.22 APR-246 covalently binds to cysteines in
mutant TP53 or TP63. The drug reconstitutes wild-type
conformation and function in mutant proteins by stabilizing
protein folding.36 Preclinical data suggest intrinsic and additive in vitro schedule-dependent cytotoxicity with azacitidine when administered in sequential fashion (unpublished
data). The MDS Clinical Research Consortium is launching a
phase I/II study of APR-246 and azacitidine among patients
with higher-risk myelodysplastic syndromes with TP53
mutation. In addition, TP53 mutation is associated with
overexpression of PD-L1 and CTLA-4, which provides the
rationale to explore PD-L1 inhibitors and CTLA-4 inhibitors in this population (unpublished data).
Other selected current HMA combination clinical trials in
myelodysplastic syndromes are summarized in Table 1.
Alternative Dosing and Schedules

Recent data presented on the 3-day use of azacitidine or
decitabine in lower-risk myelodysplastic syndromes are
encouraging. Among 91 patients with lower-risk myelodysplastic syndromes who were treated with either azacitidine or decitabine in a 3-day regimen every 28 days, the
ORR was 59%, and 29% of patients became RBC transfusion
independent. Treatment was well tolerated and 1-year OS
was 86%.37 A randomized clinical trial comparing a 5-day
azacitidine regimen, a 3-day azacitidine regimen, or a 3-day
decitabine is currently ongoing. The study also includes
randomly assigning nontransfusion-dependent patients to
observation versus treatment.
TABLE 1. Selected Hypomethylating Agent

Combination Strategies
Drug
Target
Nivolumab
PD-1
Study Identifier
NCT02530463
Lirilumab and nivolumab
KIR and PD-1
NCT02599649
Atezolizumab
PD-L1
NCT02508870
Deferasirox (ICL670)
NF-kB
NCT02038816
Eltrombopag
TPO
NCT02158936
Ibrutinib
BTK
NCT02553941
PF-04449913
Hedgehog pathway
NCT02367456
LDE255
Hedgehog pathway
NCT02323139
Erismodegib
Hedgehog pathway
NCT02129101
Rigosertib
PLK-1
NCT01926587
Volasertib
PLK-1
NCT01957644
Sirolimus
mTOR
NCT01869114
Vosaroxin
Topoisomerase II
NCT01913951
Abbreviations: HMA, hypomethylating agent; KIR, killer-cell immunoglobulin-like

receptor; NF-kB, nuclear factor-kappa B; TPO, thrombopoietin; BTK, Bruton tyrosine
kinase; PLK, polo-like kinase.
e347
RAMI S. KOMROKJI
DEFINITION OF HYPOMETHYLATING AGENT

FAILURE AND THE EXPECTED OUTCOME
For higher-risk myelodysplastic syndromes, outcome after
HMA failure is poor.38-40 The median OS ranges from 4 to
6 months. Approximately 20% to 30% of patients progress to
AML. Hypomethylating agent treatment failure can be divided into primary or secondary HMA failure. Primary failure
is defined as lack of initial response in which there is clear
evidence of progressive disease. Stable disease is more
controversial where no definitive increase in myeloblasts is
observed, but on the other hand no hematologic improvement is achieved by the International Working Group
(IWG 2006) criteria. In a landmark analysis of the AZA-001
study, patients who achieved hematologic improvement or
better response at 3, 6, and 9 months had better OS while
receiving azacitidine treatment compared with conventional
care regimens. For patients with stable disease, OS was the
same for azacitidine and conventional care regimens.
However, 19% of patients treated with azacitidine who had
stable disease at 3 months achieved hematologic improvement or better at 6 months compared with 13% for
patients treated with conventional care regimens. At
9 months, only 14% of patients who had stable disease with
azacitidine at 6 months achieved hematologic improvement
or better, whereas none of the patients who received
conventional care regimens did.41 Among 291 patients
treated with HMA, Nazha et al42 reported that 55% of patients achieved their best response at 4 to 6 months. Among
patients with stable disease at 4 to 6 months, 29 (20%)
achieved a better response at a later treatment time point.
Younger patients with lower bone marrow blast percentages
and intermediate risk per the IPSS-R were more likely to
achieve a better response after stable disease at 4 to
6 months. Patients with stable disease who subsequently
achieved CR had superior OS compared with patients who
remained with stable disease (28.1 vs. 14.4 months, respectively; p = .04).42
The median OS for primary HMA failure was 5.5 months
in the rigosertib randomized clinical study.43 Secondary HMA failure is defined as a clear loss of initial response (hematologic improvement or better by IWG 2006
criteria).
Outcome after HMA failure is also reported to be poor for
patients with lower-risk myelodysplastic syndromes. The
MDS Clinical Research Consortium reported median OS of
17 months after HMA failure.44 The definition of treatment
failure in lower-risk myelodysplastic syndromes is strictly
based on hematologic improvement, where there is lack of
hematologic improvement in primary failure, and hematologic response is lost after treatment in secondary failure.
Primary failure should be assessed after at least four to six
cycles, in which treatment can be discontinued if there is no
clear evidence of hematologic improvement. There is
probably no value of achieving stable disease without hematologic improvement in lower-risk myelodysplastic syndromes, especially if myeloblasts are not increased.
e348
Acute myeloid leukemia arising from myelodysplastic

syndrome is also characterized by a particularly poor prognosis, especially if prior HMA has been received.38,45 The
median OS among 74 patients with higher-risk myelodysplastic syndromes who developed secondary AML after
azacitidine failure was 34 months with a 1-year survival
probability of only 8%.38 In another report, prior therapy with
HMAs for patients with myelodysplastic syndromes who
progressed to AML was independently associated with inferior response and survival after intensive chemotherapy.45
HOW TO TREAT PATIENTS WITH

MYELODYSPLASTIC SYNDROMES AFTER
HYPOMETHYLATING AGENT FAILURE WHEN
THERE IS NO CLINICAL TRIAL
There are no standard treatment options for patients with
myelodysplastic syndromes after HMA failure. In lower-risk
myelodysplastic syndromes, lenalidomide may be used for
treatment of anemia; however, responses are much lower
when lenalidomide is used as a second-line therapy after
HMA failure rather than as first-line treatment after erythroid stimulating agents in nondeletion 5q myelodysplastic
syndrome.46
In higher-risk myelodysplastic syndromes, few studies
have addressed sequential use of HMAs. In a study of 14
patients, Borthakur et al47 reported that 28% responded to
decitabine after azacitidine failure. In another study, 21
patients (19%) responded to decitabine after azacitidine
failure and 10 (40%) responded to azacitidine after decitabine failure.48 Harel et al reported response to decitabine
after azacitidine treatment among 7 of 37 patients (19%).49
Five of those responses were marrow CR or stable disease.49
These studies are limited by their retrospective nature, lack
of standard definition of HMA failure, and inclusion of patients who were nontolerant to the first agent.
The GFM reported no response to low-dose chemotherapy
after azacitidine failure among 18 evaluable patients. AHSCT
offers the best outcome for patients after HMA failure. An
analysis of 435 patients with higher-risk myelodysplastic
syndromes for whom azacitidine failed demonstrated that
participants who were offered an AHSCT or investigational
agents had better survival compared with those offered
supportive care or conventional chemotherapy, whether
low or intensive dose. The median OS after best supportive
care, low-dose chemotherapy, intensive-dose chemotherapy, investigational therapy, or AHSCT was 4.1, 7.3, 8.9, 13.2,
and 19.5 months, respectively.38
NOVEL AGENTS FOR TREATMENT OF

MYELODYSPLASTIC SYNDROMES AFTER
HYPOMETHYLATING AGENT FAILURE
A comprehensive review of all novel agents tested after HMA
failure is beyond the scope of this review. The following are
selected examples of current agents in development.
Oral Azacitidine
50
An oral formulation of azacitidine has been evaluated in

three clinical studies to date. Results from a pilot study
indicated that azacitidine in an oral formulation was bioavailable.51 This was confirmed when evaluating several
different formulations of the drug in a phase I study. Results
from another phase I study of CC-486 among patients with
myelodysplastic syndrome, CMML, or AML indicated that
administration on different treatment schedules (7 , 14 , and
21 days once a day and 14 and 21 days twice a day) was
feasible and generally well tolerated and exhibited biologic
and clinical activity.52 Responses for patients previously
treated with subcutaneous/intravenous azacitidine were reported. Oral azacitidine alone or in combination with a PD-L1
inhibitor is being tested for patients with HMA stable disease
or after failure.
Among patients with HMA failure, 40% CRi/partial response

was reported.57,58 In another study among patients with
AML receiving tosedostat in combination with low-dose
cytarabine, a CR rate of 48.5% was reported for 16 of 33
patients; 33% of patients were still in remission after a
median follow-up of 506 days.
Analysis of these patients cells identified a molecular
signature associated with clinical response (clinical response vs. no clinical response). Results suggested that
differentially expressed genes were associated with clinical
response, including among biologically relevant pathways
such as beta-catenin, tumor necrosis factor-alpha, nuclear
factor-kappa B, erythroblastosis oncogene B (ERB2), inflammatory response, and epithelial-mesenchymal transition pathways.59
Thrombopoietin Stimulants
SGI-110
SGI-110 is a dinucleotide of decitabine and deoxyguanosine
that protects it from deamination. In a phase I study that
included 14 patients with myelodysplastic syndromes after
HMA failure, SGI-110 had a 4.5-fold longer half-life than
decitabine. An equivalent or higher area under the curve was
reached with lower Cmax compared with reference levels
from intravenous decitabine (20 mg/m2). A dose-dependent
increase in demethylation was observed up to 60 mg/m2
daily for 5 days.53 In the phase II part of the study for
treatment-naive elderly patients with AML or refractory/
relapsed AML, 43% and 16% remission rates were reported,
respectively.54
Rigosertib
Rigosertib is a novel small molecule that targets pathways
including phosphoinositide-3 kinase and polo-like kinase.
Initial studies with both oral and intravenous rigosertib
indicated clinical activity in myelodysplastic syndromes and
AML.55,56 Results from ONTIME, a phase III randomized
clinical study comparing intravenous rigosertib with best
supportive care after HMA failure, were recently reported.
ONTIME enrolled 299 patients and rigosertib was administered as 1,800 mg/24 hours for 72 hours as a continuous
intravenous ambulatory infusion. The median OS was
8.2 months for patients who received rigosertib compared
with 5.9 months for best supportive care (p = .33). For
patients with primary HMA failure, the median OS was
8.6 months with rigosertib compared with 5.3 months for
best supportive care (p = .04).43 Further studies are being
conducted specifically for patients with higher-risk disease
and primary HMA failure. Rigosertib in combination with
azacitidine is being explored.
Tosedostat
Tosedostat is an aminopeptidase inhibitor that leads to
cellular amino acid deprivation. In a phase I study with 44
participants, the maximum tolerated dose was reached at
180 mg with hepatic toxicity as the dose-limiting toxicity.
Thrombocytopenia treatment is challenging in myelodysplastic syndromes. Severe thrombocytopenia is more

commonly encountered in higher-risk myelodysplastic
syndromes.60 Hypomethylating agents are currently the only
available therapy. Romiplostim is an Fc-fusion dipeptide
ligand of the thrombopoietin receptor. In a double-blind,
placebo-controlled study, 167 and 83 patients with lowor intermediate-risk myelodysplastic syndromes and
thrombocytopenia were treated with 750 mg of romiplostim
per week or placebo, respectively. Results showed a significant decrease in clinically relevant bleeding events for
romiplostim-treated patients with a baseline platelet count
of 20 to 50 3 109 cells/L (relative risk 0.35; p , .0001). In
addition, romiplostim produced a significant decrease in
platelet transfusion events versus placebo for patients with a
baseline platelet count of less than 20 3 109 cells/L (relative
risk 0.71; p , .0001). Originally there was a concern about
TABLE 2. Novel Agents in Myelodysplastic Syndromes

After HMA Failure
Drug
Target
Study Identifier
Durvalumab
PD-L1
NCT02281084
Atezolizumab
PD-L1
NCT02508870
Ipilimumab
CTLA-4
NCT01757639
Nivolumab
PD-1
NCT02530463
Lirilumab and
Nivolumab
KIR and PD-1
NCT02599649
FT-1101
BET
Omacetaxine
NCT02543879
NCT02159872
CPI-613
Ketoglutarate dehydrogenase
NCT01520805
TEN-010
Bromodomain
NCT02308761
Selinexor
Selective nuclear export
NCT02228525
ASTX727
Oralcytidinedeaminaseinhibitor
E7727 with oral decitabine
NCT02103478
Vosaroxin
Topoisomerase II
NCT01980056
Abbreviations: MDS, myelodysplastic syndrome; HMA, hypomethylating agent; KIR,

killer-cell immunoglobulin-like receptor; BET, bromodomain and extra-terminal family
of proteins.
e349
RAMI S. KOMROKJI
increased risk of AML transformation; however, this was not

confirmed on longer follow-up.61 Results of a multicenter,
randomized, double-blind, placebo-controlled phase II study
of romiplostim with azacitidine have been reported.62 Patients with IPSS low-risk and intermediate-risk myelodysplastic syndromes were randomly assigned to receive the
following in addition to azacitidine treatment: 500 mg of
romiplostim per week, 750 mg of romiplostim per week, or
placebo. The incidence of clinically relevant thrombocytopenic events for patients receiving 500 mg of romiplostim,
750 mg of romiplostim, or placebo was 62%, 71%, and 85%,
respectively. The rate of platelet transfusions was 46%, 36%,
and 69%, respectively.
The second approved thrombopoietin receptor agonist is
eltrombopag, an orally bioavailable agent that binds at a site
distinct from romiplostim.63 Preliminary results from a
phase II randomized study of eltrombopag versus placebo
for 70 patients with lower-risk myelodysplastic syndromes
were reported at the 2015 American Society of Hematology
Annual Meeting. Among 41 evaluable patients treated with
eltrombopag, 21 responded at week 8; nine of these patients
achieved CR (platelets . 100 3 109 cells/L). Twenty-four
patients completed 24 weeks of eltrombopag treatment at
the time of data reporting, in which 13 patients were still
responders (eight achieved CR).64
A phase I study of heavily treated patients with higher-risk
disease who were resistant to HMAs noted a transition to
transfusion independence for 30% of patients receiving
eltrombopag, with a median duration of response of
3.3 months. In higher-risk myelodysplastic syndromes,
concerns about increased myeloblasts, leukocytosis, and
bone marrow fibrosis were raised, particularly in CMML.65
H3B-8800
Splicing mutations are frequent among patients with
myelodysplastic syndromes. The splicing machinery mutations occur in an exclusive manner and in a heterozygous
pattern. At the plenary session of 2015 American Society of

Hematology Annual Meeting, Lee et al66 presented data
demonstrating that the presence of SRSF2 mutation with the
second allele knockout was lethal in mice. The presence of a
mutated allele along with wild copy translates in developing
hematologic malignancy. This provides a therapeutic opportunity in which using splicing inhibitors leads to synthetic
lethality for patients with splicing mutations. Mice with
SRSF-2 mutation treated with E7101, a splicing inhibitor,
lived longer.66 H3B-8800 is a potent oral and selective SF3b
complex pan modulator. Phase I/II studies of patients with
myelodysplastic syndromes/AML with splicing mutations
are planned to start in 2016.
CPX-351
In secondary AML from myelodysplastic syndromes after
HMA failure, responses to standard induction chemotherapy
is low with 14% to 29% CR rates reported.38,45 CPX-351 is a
liposomal formulation of daunorubicin and cytarabine in
fixed molar ratio (5:1) optimizing delivery and cytotoxicity.
Promising results were reported in a phase IIB clinical trial, in
which the median OS for secondary AML was 12.1 months
for the CPX351 arm compared with 6.1 months for the
standard induction (p = .01).67 A randomized phase III clinical
trial finished accrual of patients with secondary AML.
Table 2 summarizes other potential agents for myelodysplastic syndrome treatment after HMA failure not discussed in this review.
CONCLUSION
Hypomethylating agents remain the current standard
therapy for the majority of patients with myelodysplastic
syndromes. An international and collaborative effort is urgently needed to conduct clinical trials to improve outcomes
with HMAs and discover novel active new agents for
treatment of myelodysplastic syndromes after HMA failure.
References
1. Komrokji RS, Zhang L, Bennett JM. Myelodysplastic syndromes classification and risk stratification. Hematol Oncol Clin North Am. 2010;24:
443-457.
2. Komrokji R, Bennett JM: What is WHO?: myelodysplastic syndrome
classification and prognosis. Am Soc Clin Oncol Educ Book. 2009;29:
413-419.
3. Bennett JM, Komrokji RS. The myelodysplastic syndromes: diagnosis, molecular biology and risk assessment. Hematology.
2005;10:258-269.
4. Greenberg P, Cox C, LeBeau MM, et al. International scoring system for
evaluating prognosis in myelodysplastic syndromes. Blood. 1997;89:
2079-2088.
5. Greenberg PL, Tuechler H, Schanz J, et al. Revised international
prognostic scoring system for myelodysplastic syndromes. Blood. 2012;
120:2454-2465.
e350
6. Kantarjian H, OBrien S, Ravandi F, et al. Proposal for a new risk model in

myelodysplastic syndrome that accounts for events not considered in
the original International Prognostic Scoring System. Cancer. 2008;113:
1351-1361.
7. Garcia-Manero G, Shan J, Faderl S, et al. A prognostic score for patients
with lower risk myelodysplastic syndrome. Leukemia. 2008;22:538-543.
8. Bejar R, Stevenson K, Abdel-Wahab O, et al. Clinical effect of point
mutations in myelodysplastic syndromes. N Engl J Med. 2011;364:
2496-2506.
9. Cutler CS, Lee SJ, Greenberg P, et al. A decision analysis of allogeneic
bone marrow transplantation for the myelodysplastic syndromes:
delayed transplantation for low-risk myelodysplasia is associated with
improved outcome. Blood. 2004;104:579-585.
10. Issa JP. Epigenetic variation and human disease. J Nutr. 2002;132:
2388S-2392S.
11. Kumar A, List AF, Hozo I, et al. Decitabine versus 5-azacitidine for the
treatment of myelodysplastic syndrome: adjusted indirect metaanalysis. Haematologica. 2010;95:340-342, author reply 343-344.
12. Qin T, Castoro R, El Ahdab S, et al. Mechanisms of resistance to decitabine in the myelodysplastic syndrome. PLoS One. 2011;6:e23372.
13. Fenaux P, Mufti GJ, Hellstrom-Lindberg E, et al; International Vidaza
High-Risk MDS Survival Study Group. Efficacy of azacitidine compared
with that of conventional care regimens in the treatment of higher-risk
myelodysplastic syndromes: a randomised, open-label, phase III study.
Lancet Oncol. 2009;10:223-232.
14. Lubbert M, Suciu S, Baila L, et al. Low-dose decitabine versus best
supportive care in elderly patients with intermediate- or high-risk
myelodysplastic syndrome (MDS) ineligible for intensive chemotherapy: final results of the randomized phase III study of the European
Organisation for Research and Treatment of Cancer Leukemia Group
and the German MDS Study Group. J Clin Oncol. 2011;29:1987-1996.
15. Kantarjian H, Issa JP, Rosenfeld CS, et al. Decitabine improves patient
outcomes in myelodysplastic syndromes: results of a phase III randomized study. Cancer. 2006;106:1794-1803.
16. Steensma DP, Baer MR, Slack JL, et al. Multicenter study of decitabine
administered daily for 5 days every 4 weeks to adults with myelodysplastic syndromes: the alternative dosing for outpatient treatment
(ADOPT) trial. J Clin Oncol. 2009;27:3842-3848.
17. Damaj G, Duhamel A, Robin M, et al. Impact of azacitidine before allogeneic stem-cell transplantation for myelodysplastic syndromes:
a study by the Societe Franaise de Greffe de Moelle et de TherapieCellulaire and the Groupe-Francophone des Myelodysplasies. J Clin
Oncol. 2012;30:4533-4540.
18. de Lima M, Giralt S, Thall PF, et al. Maintenance therapy with low-dose
azacitidine after allogeneic hematopoietic stem cell transplantation for
recurrent acute myelogenous leukemia or myelodysplastic syndrome:
a dose and schedule finding study. Cancer. 2010;116:5420-5431.
19. Lyons RM, Cosgriff TM, Modi SS, et al. Hematologic response to three
alternative dosing schedules of azacitidine in patients with myelodysplastic syndromes. J Clin Oncol. 2009;27:1850-1856.
20. Zeidan AM, Sekeres MA, Garcia-Manero G, et al. Comparison of risk
stratification tools in predicting outcomes of patients with higher-risk
myelodysplastic syndromes treated with azanucleosides. Leukemia.
Epub 2015 Oct 14.
21. Itzykson R, Thepot S, Quesnel B, et al; Groupe Francophone des
Myelodysplasies (GFM). Prognostic factors for response and overall
survival in 282 patients with higher-risk myelodysplastic syndromes
treated with azacitidine. Blood. 2011;117:403-411.
22. Bejar R, Lord A, Stevenson K, et al. TET2 mutations predict response to
hypomethylating agents in myelodysplastic syndrome patients. Blood.
2014;124:2705-2712.
23. Zeidan AM, Kharfan-Dabaja MA, Komrokji RS. Beyond hypomethylating
agents failure in patients with myelodysplastic syndromes. Curr Opin
Hematol. 2014;21:123-130.
24. Sekeres MA, Othus M, List AF, et al. A randomized phase II study of
azacitidine combined with lenalidomide or with vorinostat vs. azacitidine monotherapy in higher-risk myelodysplastic syndromes (MDS)
and chronic myelomonocytic leukemia (CMML): North American Intergroup Study SWOG S1117. Blood. 2014;124 (abstr LBA-5).
25. Sekeres MA, Othus M, List AF, et al. Additional analyses of a randomized
phase II study of azacitidine combined with lenalidomide or with
vorinostat vs. azacitidine monotherapy in higher-risk myelodysplastic
syndromes (MDS) and chronic myelomonocytic leukemia (CMML):
North American Intergroup Study SWOG S1117. Blood. 2015;126 (abstr
908).
26. Prebet T, Sun Z, Figueroa ME, et al. Prolonged administration of azacitidine with or without entinostat for myelodysplastic syndrome and
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
acute myeloid leukemia with myelodysplasia-related changes: results of

the US Leukemia Intergroup trial E1905. J Clin Oncol. 2014;32:
1242-1248.
Garcia-Manero G, Berdeja JG, Komrokji RS, et al. A randomized,
placebo-controlled, phase II study of pracinostat in combination with
azacitidine (AZA) in patients with previously untreated myelodysplastic
syndrome (MDS). Blood. 2015;126 (abstr 911).
Fathi AT, Erba HP, Lancet JE, et al. SGN-CD33A plus hypomethylating
agents: a novel, well-tolerated regimen with high remission rate in
frontline unfit AML. Blood. 2015;126 (abstr 454).
Ogata K, Kakumoto K, Matsuda A, et al. Differences in blast immunophenotypes among disease types in myelodysplastic syndromes:
a multicenter validation study. Leuk Res. 2012;36:1229-1236.
Yang H, Bueso-Ramos CE, Parmar S, et al. Induction of PD-1 and PD-1
ligand expression by hypomethylating agents (HMA) in myelodysplastic
syndromes and acute myelogenous leukemia suggest a role for T cell
function in clinical resistance to Hmas. Blood. 2012;120 (abstr 3810).
Swords RT, Erba HP, DeAngelo DJ, et al. Pevonedistat (MLN4924), a firstin-class NEDD8-activating enzyme inhibitor, in patients with acute
myeloid leukaemia and myelodysplastic syndromes: a phase 1 study. Br
J Haematol. 2015;169:534-543.
Swords RT, Savona MR, Maris MB, et al. Pevonedistat (MLN4924), an
investigational, first-in-class NAE Inhibitor, in combination with azacitidine in elderly patients with acute myeloid leukemia (AML) considered
unfit for conventional chemotherapy: updated results from the phase 1
C15009 trial. Blood. 2014;124 (abstr 2313).
Sallman DA, Komrokji R, Vaupel C, et al. Impact of TP53 mutation variant
allele frequency on phenotype and outcomes in myelodysplastic syndromes. Leukemia. Epub 2015 Oct 30.
Bejar R, Papaemmanuil E, Haferlach T, et al. Somatic mutations in MDS
patients are associated with clinical features and predict prognosis
independent of the IPSS-R: analysis of combined datasets from the
International Working Group for Prognosis in MDS-Molecular Committee. Blood. 2015;126 (abstr 907).
Bejar R, Stevenson KE, Caughey B, et al. Somatic mutations predict poor
outcome in patients with myelodysplastic syndrome after hematopoietic stem-cell transplantation. J Clin Oncol. 2014;32:2691-2698.
Zawacka-Pankau J, Selivanova G. Pharmacological reactivation of p53
as a strategy to treat cancer. J Intern Med. 2015;277:248-259.
Short NJ, Garcia-Manero G, Montalban Bravo G, et al. Low-dose
hypomethylating agents (HMAs) are effective in patients (Pts) with
low- or intermediate-1-risk myelodysplastic syndrome (MDS): a report
on behalf of the MDS Clinical Research Consortium. Blood. 2015;126
(abstr 94).
Pre bet T, Gore SD, Esterni B, et al. Outcome of high-risk myelodysplastic
syndrome after azacitidine treatment failure. J Clin Oncol. 2011;29:
3322-3327.
Jabbour E, Garcia-Manero G, Batty N, et al. Outcome of patients with
myelodysplastic syndrome after failure of decitabine therapy. Cancer.
2010;116:3830-3834.
Duong VH, Lin K, Reljic T, et al. Poor outcome of patients with myelodysplastic syndrome after azacitidine treatment failure. Clin Lymphoma Myeloma Leuk. 2013;13:711-715.
Gore SD, Fenaux P, Santini V, et al. A multivariate analysis of the relationship between response and survival among patients with higherrisk myelodysplastic syndromes treated within azacitidine or conventional care regimens in the randomized AZA-001 trial. Haematologica.
2013;98:1067-1072.
Nazha A, Sekeres MA, Garcia-Manero G, et al; MDS Clinical Research
Consortium. Outcomes of patients with myelodysplastic syndromes
who achieve stable disease after treatment with hypomethylating
agents. Leuk Res. 2016;41:43-47.
e351
RAMI S. KOMROKJI
43. Garcia-Manero G, Fenaux P, Al-Kali A, et al. Overall survival and subgroup analysis from a randomized phase III study of intravenous rigosertib versus best supportive care (BSC) in patients (pts) with higherrisk myelodysplastic syndrome (HR-MDS) after failure of hypomethylating agents (HMAs). Blood. 2014;124 (abstr 163).
44. Jabbour EJ, Garcia-Manero G, Strati P, et al. Outcome of patients with
low-risk and intermediate-1-risk myelodysplastic syndrome after
hypomethylating agent failure: a report on behalf of the MDS Clinical
Research Consortium. Cancer. 2015;121:876-882.
45. Bello C, Yu D, Komrokji RS, et al. Outcomes after induction chemotherapy in patients with acute myeloid leukemia arising from myelodysplastic syndrome. Cancer. 2011;117:1463-1469.
46. Zeidan AM, Al Ali NH, Padron E, et al. Lenalidomide treatment for lower
risk nondeletion 5q myelodysplastic syndromes patients yields higher
response rates when used before azacitidine. Clin Lymphoma Myeloma
Leuk. 2015;15:705-710.
47. Borthakur G, Ahdab SE, Ravandi F, et al. Activity of decitabine in patients
with myelodysplastic syndrome previously treated with azacitidine.
Leuk Lymphoma. 2008;49:690-695.
48. Komrokji RS, Apuri S, Al Ali N, et al. Evidence for selective benefit of
sequential treatment with azanucleosides in patients with myelodysplastic syndromes (MDS). J Clin Oncol 2013;31 (suppl; abstr 7113).
49. Harel S, Cherait A, Berthon C, et al. Outcome of patients with high risk
myelodysplastic syndrome (MDS) and advanced chronic myelomonocytic leukemia (CMML) treated with decitabine after azacitidine failure.
Leuk Res. 2015;39:501-504.
50. Laille E, Shi T, Garcia-Manero G, et al. Pharmacokinetics and pharmacodynamics with extended dosing of CC-486 in patients with hematologic malignancies. PLoS One. 2015;10:e0135520.
51. Garcia-Manero G, Stoltz ML, Ward MR, et al. A pilot pharmacokinetic
study of oral azacitidine. Leukemia. 2008;22:1680-1684.
52. Garcia-Manero G, Gore SD, Kambhampati S, et al. Efficacy and safety of
extended dosing schedules of CC-486 (oral azacitidine) in patients with
lower-risk myelodysplastic syndromes. Leukemia. Epub 2015 Oct 7.
53. Kantarjian HM, Roboz GJ, Rizzieri DA, et al. Results from the dose
escalation phase of a randomized phase 1-2 first-in-human (FIH) study
of SGI-110, a novel low volume stable subcutaneous (SQ) second
generation hypomethylating agent (HMA) in patients with relapsed/
refractory MDS and AML. Blood. 2012;120 (abstr 414).
54. Jabbour E, Yee K, Kropf P, et al. First clinical results of a randomized
phase 2 study of SGI-110, a novel subcutaneous (SQ) hypomethylating
agent (HMA), in adult patients with acute myeloid leukemia (AML).
Blood. 2013;122 (abstr 497).
55. Raza A, Greenberg PL, Olnes MJ, et al. Final phase I/II results of rigosertib (ON 01910.Na) hematological effects in patients with
e352
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
myelodysplastic syndrome and correlation with overall survival. Blood.

2013;122 (abstr 3822).
Komrokji RS, Raza A, Lancet JE, et al. Phase I clinical trial of oral rigosertib
in patients with myelodysplastic syndromes. Br J Haematol. 2013;162:
517-524.
Cortes JE, Feldman EJ, Yee K, et al. Results of the OPAL study: a phase II
study to evaluate the efficacy, safety and tolerability of tosedostat
(CHR-2797) in elderly subjects with treatment refractory or relapsed
acute myeloid leukemia. Blood. 2011;118 (abstr 767).
Lowenberg B, Morgan G, Ossenkoppele GJ, et al. Phase I/II clinical
study of tosedostat, an inhibitor of aminopeptidases, in patients with
acute myeloid leukemia and myelodysplasia. J Clin Oncol. 2010;28:
4333-4338.
Visani G, Loscocco F, Fuligni F, et al. Tosedostat plus low dose cytarabine
induces a high rate of responses that can be predicted by genetic
profiling in elderly AML. Blood. 2015;126 (abstr 329).
Kantarjian H, Giles F, List A, et al. The incidence and impact of
thrombocytopenia in myelodysplastic syndromes. Cancer. 2007;109:
1705-1714.
Kantarjian HM, Mufti GJ, Fenaux P, et al. Treatment with the thrombopoietin (TPO)-receptor agonist romiplostim in thrombocytopenic
patients (Pts) with low or intermediate-1 (int-1) risk myelodysplastic
syndrome (MDS): follow-up AML and survival results of a randomized,
double-blind, placebo (PBO)-controlled study. Blood. 2012;120 (abstr
421).
Kantarjian HM, Giles FJ, Greenberg PL, et al. Phase 2 study of romiplostim in patients with low- or intermediate-risk myelodysplastic
syndrome receiving azacitidine therapy. Blood. 2010;116:3163-3170.
Bussel JB, Cheng G, Saleh MN, et al. Eltrombopag for the treatment of
chronic idiopathic thrombocytopenic purpura. N Engl J Med. 2007;357:
2237-2247.
Oliva EN, Santini V, Alati C, et al. Eltrombopag for the treatment of
thrombocytopenia of low and intermediate-1 IPSS risk myelodysplastic
syndromes: interim results on efficacy, safety and quality of life of an
international, multicenter prospective, randomized, trial. Blood. 2015;
126 (abstr 91).
Ramadan H, Duong VH, Al Ali NH, et al. Eltrombopag use in chronic
myelomonocytic leukemia (CMML) patients: a cautionary tale. Blood.
2015;126 (abstr 2897).
Lee SC-W, Dvinge H, Kim E, et al. Therapeutic targeting of spliceosomal
mutant myeloid leukemias through modulation of splicing catalysis.
Blood. 2015;126 (abstr 4).
Lancet JE, Cortes JE, Hogge DE, et al. Phase 2 trial of CPX-351, a fixed 5:1
molar ratio of cytarabine/daunorubicin, vs cytarabine/daunorubicin in
older adults with untreated AML. Blood. 2014;123:3239-3246.
AND CHRONIC LYMPHOCYTIC LEUKEMIA
Current Management Concepts:

Primary Central Nervous System
Lymphoma, Natural Killer T-Cell
Lymphoma Nasal Type,
Post-transplant
Lymphoproliferative Disorder
CHAIR
Thomas M. Habermann, MD
Mayo Clinic
Rochester, MN
SPEAKERS
Soon Thye Lim, MD
National Cancer Centre Singapore
Sinagpore
Tracy T. Batchelor, MD, MPH
Boston, MA
BATCHELOR, THYE, AND HABERMANN
Current Management Concepts: Primary Central Nervous

System Lymphoma, Natural Killer T-Cell Lymphoma Nasal
Type, and Post-transplant Lymphoproliferative Disorder
Tracy T. Batchelor, MD, MPH, Lim Soon Thye, MD, and Thomas M. Habermann, MD
OVERVIEW
Primary central nervous system lymphoma, natural killer T-cell lymphoma nasal type, and post-transplant lymphoproliferative disorder are uncommon and complex lymphoproliferative disorders. These disorders present with different risk
factors, have complex tumor characteristics, and require unique therapeutic interventions. These diseases require a
multidisciplinary complex team approach. This article will update current management approaches and concepts.
rimary central nervous system lymphoma (PCNSL) is

an extranodal non-Hodgkin lymphoma confined to
the brain, leptomeninges, eyes, or spinal cord. Natural
killer (NK)/T-cell lymphoma disorders, more prevalent in
Asian countries and parts of Latin America, are categorized as
extranodal NK T-cell lymphoma nasal type and as aggressive
NK-cell leukemia. Post-transplant lymphoproliferative disorder
is applied to a group of lymphoproliferative disorders arising in
a pharmacologically immunocompromised host after solid
organ or allogeneic stem cell transplantation. In this article,
we will update discuss current management approaches
and concepts for these diseases.
PRIMARY CENTRAL NERVOUS SYSTEM

LYMPHOMA
Epidemiology
Primary central nervous system lymphoma accounts for
approximately 2% of all primary central nervous system
(CNS) tumors, with median age 65 at diagnosis.1,2 Since
2000, there has been an increase in the overall incidence of
PCNSL, especially in elderly individuals.
Pathology
Approximately 90% of PCNSL cases are diffuse large B-cell
lymphomas (DLBCLs), with the remainder consisting of T-cell
lymphomas, poorly characterized low-grade lymphomas,
or Burkitt lymphomas.3 More than 90% of primary CNS
DLBCL cases consist of the activated B-celllike subtype. The
molecular mechanisms underlying transformation and localization to the CNS are poorly understood.4
Diagnosis and Prognostic Factors

Neurocognitive symptoms are the most common presenting
clinical features of PCNSL. The International PCNSL Collaborative Group has developed guidelines to determine extent
of disease.5 A gadolinium-enhanced brain MRI scan is the
most sensitive radiographic study for the detection of PCNSL
(Fig. 1). Most patients with PCNSL present with a single brain
mass. The diagnosis of PCNSL is typically established by a
stereotactic brain biopsy, cerebrospinal fluid (CSF) analysis,
or analysis of vitreous aspirate from patients with ocular
involvement. Given the possible delay in diagnosis and
treatment with the latter two methods, prompt stereotactic
biopsy is advised in almost all cases that are surgically accessible. Secondary CSF and ocular involvement occurs
among approximately 15%20% and 5%20% of patients
with PCNSL, respectively. Presenting symptoms of ocular
involvement include eye pain, blurred vision, and floaters.6
B symptoms such as weight loss, fevers, and night sweats are
infrequent in PCNSL. A thorough diagnostic evaluation is
needed to establish the extent of the lymphoma and to
confirm localization to the CNS. A lumbar puncture should
be performed if not contraindicated, and CSF should be
assessed by flow cytometry, cytology, and immunoglobulin
heavy chain gene rearrangement. Because extraneural
disease must be excluded to establish a diagnosis of primary CNS lymphoma, CT/PET scans of the chest, abdomen,
From the Division of Hematology/Oncology, Departments of Neurology and Radiation Oncology, Massachusetts General Hospital Cancer Center and Harvard Medical School, Boston,
MA; Department of Medical Oncology, National Cancer Centre, Duke-National University of Singapore Medical School, Singapore; Division of Hematology, Department of Medicine,
Mayo Clinic, Rochester, MN.
Corresponding author: Thomas M. Habermann, MD, Division of Hematology, Department of Medicine, Mayo Clinic, 200 First St. SW, Rochester, MN 55905;
email: habermann.thomas@mayo.edu.
e354
POST-TRANSPLANT LYMPHOPROLIFERATIVE DISORDER
FIGURE 1. MRIs From a Patient With Primary Central Nervous System Lymphoma
A T1-weighted, postcontrast, sequence (left) demonstrates homogenous enhancement of the tumor in the region of the left caudate nucleus. A T2/fluid-attenuated inversion
recovery sequence (right) demonstrates a hyperintense signal surrounding the tumor, reflecting vasogenic cerebral edema.
Courtesy of Priscilla Brastianos, MD.
and pelvis and a bone marrow biopsy and aspirate analysis

should be performed to exclude occult systemic disease.
Involvement of the optic nerve, retina, or vitreous humor
should be excluded with a comprehensive eye evaluation by
an ophthalmologist that includes a slit-lamp examination.
Blood tests should include serum lactate dehydrogenase
(LDH) and HIV serology.5
Two prognostic scoring systems have been developed
specifically for PCNSL.7,8 In a retrospective review of 105
patients with PCNSL, the International Extranodal Lymphoma Study Group identified the following as independent
predictors of poor prognosis: age older than 60, Eastern
KEY POINTS
The prognostic scoring systems in PCNSL, ENKL nasal

type, and PTLD are different than standard non-Hodgkin
lymphomas.
The most effective treatment in PCNSL is high-dose
methotrexate in combination with other agents, but
standardized induction and consolidation regimens
have yet to be defined.
ENKL is an Epstein-Barr virusrelated disorder. The
incorporation of radiation therapy with chemotherapy
is crucial.
Advanced ENKL should be treated with L-asparaginase
regimens.
Immunosuppression reduction and rituximab are key
management approaches in the initial approach to
PTLD.
Cooperative Oncology Group performance status greater

than 1, elevated serum LDH level, elevated CSF protein
concentration, and involvement of deep regions of the brain.
For patients with zero to one factor, two to three factors, and
four to five factors, the 2-year survival proportions were
80%, 48%, and 15%, respectively. In another prognostic
model, patients with PCNSL were divided into three groups
based on age and performance status: (1) age younger than
50, (2) age 50 and older with a Karnofsky performance score
(KPS) of 70 or greater, and (3) age 50 and older with a KPS less
than 70. Based on these three divisions, significant differences in overall survival (OS) and failure-free survival were
observed (p = .0001).
Treatment
Defining response to treatment in PCNSL requires assessment of all sites involved by the disease. The International
PCNSL Collaborative Group has established response criteria
that have been adopted into most prospective clinical trials
(Table 1).5
Corticosteroids decrease tumor-associated edema and
may result in partial radiographic regression of tumors. An
initial response to corticosteroids is associated with a favorable outcome in PCNSL.9 However, after they have an
initial response to corticosteroids, almost all patients quickly
experience relapse. Corticosteroids should be avoided prior
to a biopsy if possible, given the risk of disrupting cellular
morphology, which can result in a nondiagnostic pathologic
specimen.
Surgical resection is not part of the standard treatment
approach for PCNSL, given the multifocal nature of this
e355
TABLE 1. International PCNSL Collaborative Group Consensus Guidelines for the Assessment of Response in
Primary Central Nervous System Lymphoma5
Response
Brain Imaging
Steroid
Dose
Ophthalmologic Examination
CSF Cytology
Negative
Complete Response
No contrast-enhancing disease
None
Normal
Unconfirmed Complete
Response
Any
Normal
Negative
Minimal enhancing disease
Any
Minor RPE abnormality
Negative
Partial Response
50% decrease in enhancement
NA
Normal or minor RPE abnormality
Negative
NA
Decrease in vitreous cells or retinal

infiltrate
Persistent or
suspicious
25% increase in enhancing disease
NA
Recurrent or new disease
Recurrent or positive
Progressive Disease
Any new site of disease

Stable Disease
All scenarios not covered by responses

above
Abbreviations: PCNSL, primary central nervous system lymphoma; NA, not applicable; RPE, retinal pigment epithelium; CSF, cerebral spinal fluid.
tumor and potential long-term morbidities.10 The role of

neurosurgery in PCNSL is to establish a diagnosis via stereotactic biopsy.
Standardized induction and consolidation treatment of
PCNSL has yet to be defined. Historically, PCNSL was treated
only with whole-brain radiotherapy (WBRT) at doses ranging
from 36 to 45 Gy, which resulted in a high proportion of
radiographic responses but early relapse. In a multicenter
phase II trial, 41 patients were treated with WBRT to 40 Gy
plus a 20-Gy tumor boost and achieved a median OS of only
12 months.11 Given the lack of durable responses to radiation
and the risk of neurotoxicity associated with this therapeutic
modality, WBRT alone is no longer a recommended initial
treatment for most patients with PCNSL.
The most effective treatment of PCNSL is high-dose
methotrexate (HD-MTX) administered intravenously at
variable doses (18 g/m2), typically used in combination with
other chemotherapeutic agents and/or WBRT. However,
there is no consensus on the optimal dose of HD-MTX or on
the role of radiation in combination with methotrexate in
the management of PCNSL. A number of randomized trials
are ongoing to address these issues. Methotrexate doses of
3 g/m2 or greater result in therapeutic concentrations in
the brain parenchyma and CSF and lead to more durable
treatment responses when combined with WBRT.12-14 In a
phase II trial, 79 patients with PCNSL were randomly assigned
to receive either HD-MTX (3.5 g/m2, day 1) or HD-MTX
(3.5 g/m2, day 1) plus cytarabine (2 g/m2 twice a day, days
23). Each chemotherapy cycle was 21 days. All patients
underwent consolidative WBRT after induction chemotherapy. The HD-MTX plus cytarabine arm had a higher
proportion of complete radiographic responses and a superior 3-year OS.14 However, it is now widely recognized
that there is a high incidence of neurotoxicity with combined modality treatment that includes WBRT.15 The latter
observation prompted studies utilizing lower doses of
WBRT. In a multicenter phase II study, no significant
neurocognitive decline was observed after consolidative
reduced-dose WBRT (23.4 Gy) and cytarabine were
e356
administered to patients who had achieved a complete response (CR) to induction chemotherapy including HD-MTX.16
However, longer neuropsychologic follow-up of these patients is necessary to definitively assess the safety of this
regimen because numerous studies have demonstrated the
delayed neurotoxic effects of WBRT in the PCNSL population
and the reduced risk of neurotoxicity in regimens consisting
of chemotherapy alone.17,18 Given the risk of clinical neurotoxicity, other studies have assessed whether WBRT can be
eliminated from the initial management of PCNSL. In a
multicenter phase III trial, patients were randomly assigned
to receive HD-MTXbased chemotherapy with or without
WBRT.19 The trial enrolled 551 patients; of these, 318 were
treated per protocol. Intent-to-treat analysis revealed that
patients treated in the combined modality arm (chemotherapy plus WBRT) achieved prolonged progression-free
survival (PFS) but had no improvement in OS, demonstrating that the elimination of WBRT from the treatment regimen
did not compromise OS. This has led to deferral of WBRT and
chemotherapy-alone approaches for patients with newly
diagnosed PCNSL. These approaches are based on a foundation of HD-MTX. Variable doses and schedules of HD-MTX
have been used, but in general, a dose of 3 g/m2 or greater
delivered as an initial bolus followed by an infusion over
3 hours administered every 1021 days is recommended.20
Multiple phase II studies have reported on the safety, efficacy, and relatively preserved cognition of HD-MTXbased
chemotherapy regimens.21,22 Moreover, a longer duration
of induction chemotherapy with HD-MTX (more than six
cycles) results in higher CR proportions.16,21
Several first-generation chemotherapy regimens for
PCNSL included intrathecal chemotherapy. However, in nonrandomized studies that included intrathecal chemotherapy,
there was no improvement in outcomes versus regimens that
did not include intrathecal injections of chemotherapy.23,24
Rituximab is being incorporated in combination regimens
for PCNSL. When rituximab is administered intravenously at
doses of 375800 mg/m2, CSF levels from 0.1% to 4.4% of
serum levels are achieved. Despite limited CSF penetration,
radiographic responses have been observed for patients with

relapsed PCNSL treated with rituximab monotherapy.25 In a
cooperative group phase II study, 44 patients with PCNSL were
treated with induction chemotherapy consisting of 8 g/m2 of
HD-MTX (day 1), 375 mg/m2 of rituximab (day 3), and 150 mg/m2
of temozolomide (days 711), all drugs with demonstrated
efficacy as monotherapy in PCNSL.22 This induction chemotherapy was followed by consolidation chemotherapy consisting of 5 mg/kg of intravenous etoposide as a continuous
infusion over 96 hours and 2 g/m2 of cytarabine every 12 hours
for 8 doses. Sixty-six percent of these patients achieved CR, the
median PFS of the entire group was 2.4 years, and the estimated 4-year OS was 65%. These results are comparable to
regimens that include WBRT.
Given the limited durability of responses observed in many
studies of PCNSL, there is increasing interest in high-dose
chemotherapy (HDT) followed by autologous stem cell
transplantation (ASCT) as first-line consolidative therapy
for PCNSL. Conditioning regimens including thiotepa have
demonstrated the most encouraging results. In a multicenter phase II study, 79 patients were treated with induction HD-MTX, cytarabine, rituximab, and thiotepa, followed
by carmustine and thiotepa conditioning prior to ASCT. The
overall response rate (ORR) was 91%, 2-year OS was 87%, and
treatment-related deaths occurred among less than 10% of
enrolled patients. The toxicities, mostly cytopenias, were
manageable.26 There are three ongoing multicenter randomized trials comparing the efficacy of consolidative
HDT/ASCT versus chemotherapy or WBRT for newly diagnosed
PCNSL (Table 2).
EXTRANODAL NATURAL KILLER/T-CELL

LYMPHOMA NASAL TYPE
Natural killer/T-cell lymphoma is more prevalent in Asian
countries and parts of Latin America.27 The World Health Organization (WHO) classification categorized these disorders as
extranodal NK/T-cell lymphoma nasal type (ENKL) and as aggressive NK-cell leukemia.27,28 Histologically and phenotypically,
they are similar. The neoplastic cells are surface CD32 but
cytoplasmic CD3e+, CD56+, and cytotoxic molecule (granzyme B,
TIA1, and perforin) positive. Epstein-Barr virusencoded RNA
(EBER) in situ hybridization (ISH) is invariably positive.
Clinical Presentation, Diagnosis, and Staging

ENKL commonly involves the nasal cavity and upper aerodigestive tract.28 Rarely, patients may present with nonnasal lesions in the skin, gastrointestinal tract, and/or
testis. In such cases, it is important to exclude occult nasal involvement using either more sensitive radiologic
modalities such as PET/CT or nasal panendoscopy.27,29,30
This distinction is important because it affects staging and
treatment.
The minimum diagnostic criteria for ENKL include appropriate histopathologic features, surface CD32, cytoplasmic CD3e+, CD56+, and EBER+.27,28 If CD56 is negative,
both cytotoxic molecules and EBER should be positive.
A PET/CT scan is the preferred imaging modality.29-31 Bone

marrow biopsy is recommended, and EBER-ISH should be
considered to detect occult marrow involvement.32,33 Quantitative measurement of Epstein-Barr virus (EBV) DNA in cellfree plasma, which has been shown to correlate with tumor
load and adverse outcomes, is recommended.34,35
Prognostic Factors
Several prognostic models for ENKL based on pretreatment
characteristics have been developed, including the International Prognostic Index (IPI) and the Korean Prognostic
Index.36-39 Nonetheless, these models were based on data
from patients treated with cyclophosphamide, doxorubicin,
vincristine, and prednisolone (CHOP) or CHOP-like regimens.
Because prognostic factors are dependent on the efficacy
of the regimen used, they may not be applicable to patients who are increasingly treated with nonanthracyclinecontaining regimens. Recently, combined evaluation of the
post-treatment Deauville score on a PET/CT scan as well as
EBV DNA was shown to be very effective in predicting risk of
treatment failure and may be more relevant.40
Management of Localized Nasal NK/T-Cell Lymphoma

Radiotherapy is a crucial treatment modality in localized ENKL,
and a dose in excess of 50 Gy is recommended.41 This requires
careful planning to cover the entire nasal cavity and involved
areas.42,43 Radiotherapy alone is insufficient to achieve high
cure rates, owing to both local and systemic relapses.44,45
However, results from early studies incorporating standard
CHOP and dose-intensified CHOP were disappointing.46,47
This high failure rate is attributed to the expression of the
multidrug resistance (MDR) gene in NK cells, a feature that is
recapitulated in neoplastic NK cells.
Subsequent approaches incorporated non-MDRrelated
agents and etoposide with radiotherapy (Table 3).48-52
However, the optimal sequence of chemotherapy and radiotherapy remains to be determined.
In a phase II study of 27 patients with localized ENKL,
concurrent radiotherapy alone (50 Gy) with three cycles of
2/3 dexamethasone, etoposide, ifosfamide, and carboplatin
(DeVIC) resulted in an ORR of 81% and a CR rate of 77%.53
With a median follow-up of 67 months, the updated 5-year OS
and PFS were 73% and 67%, respectively.52 Acute toxicities
associated with this regimen were generally mild. In another
phase II study, 30 patients received concurrent radiotherapy
with cisplatin followed by three cycles of etoposide, ifosfamide, cisplatin, and dexamethasone (VIPD).50 The CR and ORR
at best response were 80% and 83.3%, whereas the estimated
3-year PFS and OS were 85.2% and 86.3%, respectively. Acute
toxicities during concurrent chemotherapy were mild, but
62% of patients developed grade 3/4 febrile neutropenia
during VIPD therapy. A recent study of concurrent cisplatin
with radiotherapy followed by gemcitabine, dexamethasone,
and cisplatin (GDP) chemotherapy reported comparable
treatment outcomes and less toxicity.49
Because arranging timely radiotherapy may be challenging, sandwich protocols were explored.48,51 In one study,
e357
26 patients received two cycles of L-asparaginase, vincristine, and prednisone (LVP) followed by radiotherapy (56 Gy)
and two to four additional cycles of LVP.48 The ORR and CR
rate after initial chemotherapy were 92% and 88.5%, respectively. The corresponding ORR and CR after completion
of radiotherapy and subsequent chemotherapy were 92%
and 88.5%, respectively. The reported 2-year OS and PFS
were 88.5% and 80.6%. Frontline use of sandwich gemcitabine, oxaliplatin, and L-asparaginase (GELOX) with radiotherapy yielded similar ORR and CR rates of 96.3% and
74.1%.51 At present, the most active regimen for patients
with advanced ENKL is the steroid, methotrexate, ifosfamide,
54,55
L-asparaginase, and etoposide (SMILE) regimen.
The ORR
and CR rates for patients with untreated NK/T-cell lymphoma
after two to three cycles of therapy were 81% and 60%,
respectively.54 The high induction response rate observed
suggests that SMILE with sandwich radiotherapy may be an
attractive approach but requires confirmation in a prospective clinical trial.
A risk-adapted approach to the treatment of patients with

localized ENKL was proposed.56 Patients were classified as
low or high risk using five independent prognostic factors:
age older than 60, Eastern Cooperative Oncology Group
performance status of 2 or more, stage II disease, elevated
LDH level, and presence of primary tumor invasion. For highrisk patients (defined as the presence of one or more adverse
factors), radiotherapy followed by chemotherapy was found
to be the most optimal sequence because it was associated
with the best OS.56 For low-risk patients (defined as no
adverse factors), radiotherapy alone achieved favorable
long-term survival.56 Neither induction nor consolidation
chemotherapy was beneficial. Although these results are intriguing, the conclusions are limited by the fact that more than
80% of the patients were treated with anthracycline-based
regimens, and these findings may not be applicable to patients
treated with contemporary regimens.
In summary, patients with localized ENKL should receive combination chemotherapy and radiotherapy. Both
TABLE 2. Randomized Trials in Primary Central Nervous System Lymphoma

Completed Trials
Trial
Type
Phase
No. of Patients (Age)
Description
Medical Research Council39
Induction
II
53
Stopped early; CHOP vs. WBRT followed by CHOP
IELSG 20 (NCT00210314)14
Induction
II
79 (1875)
Induction arm 1: methotrexate 1 cytarabine 0 WBRT
ANOCEF-GOELAMS
(NCT00503594)31
Induction
II
95 ($ 60)
Arm 1: methotrexate, procarbazine, vincristine, cytarabine
G-PCNSL-SG-1 (NCT00153530)19
Consolidation
III
551 ($ 18)
Arm 1: methotrexate 6 ifosfamide 0 WBRT
Induction arm 2: methotrexate 0 WBRT

Arm 2: methotrexate, temozolomide
Arm 2: methotrexate 6 ifosfamide
Ongoing Trials
Trial
Type
Phase
No. of Patients (Age)
IESLG 32 (NCT01011920)
Induction
II
200 (1870)
Description
Induction arm 1: methotrexate, cytarabine
Induction arm 2: methotrexate, cytarabine, rituximab
Induction arm 3: methotrexate, cytarabine, rituximab,
thiotepa
ALLG/HOVON (EudraCT 2009014722-42)
Induction
III
200 (1870)
Arm 1: methotrexate, BCNU, teniposide, prednisone 0

cytarabine, WBRT
Arm 2: methotrexate, BCNU, teniposide, prednisone 0
cytarabine, WBRT
IESLG 32 (NCT01011920)
Consolidation
II
104 (1870)
Consolidation arm 1: WBRT
ANOCEF-GOELAMS (NCT00863460)
Consolidation
II
100 (1860)
R-MBVP 0
Consolidation arm 2: HDT/ASCT

Consolidation arm 2: WBRT
RTOG 1114 (NCT01399372)
Consolidation
II
84 ($ 18)
Methotrexate, procarbazine, vincristine, rituximab 0

Consolidation arm 1: WBRT (lower dose) 0 cytarabine
Consolidation arm 2: cytarabine
Alliance 51101 (NCT01511562)
Consolidation
II
160 (1875)
Methotrexate, temozolomide, rituximab, cytarabine 0

Consolidation arm 2: etoposide, cytarabine
Abbreviations: CHOP, cyclophosphamide, doxorubicin, vincristine, and prednisolone; WBRT, whole-brain radiotherapy; BCNU, bis-chloroethylnitrosourea; HDT/ASCT, high-dose
chemotherapy and autologous stem cell transplantation.
e358
TABLE 3. Combined Chemotherapy and Radiotherapy for Localized ENKL

No. of
Patients ORR, % CR, % Survival Outcomes
Combination Therapy
Reference
Concurrent RT (50 Gy) with 2/3 DeVIC alone
Rubenstein et al and
Khan et al22,23
27
81
77
5-year PFS, 67%; 5-year

OS, 73%
30 mg/m2 of concurrent weekly intravenous cisplatin for 35 weeks with Sierra del Rio et al24
RT (4052.8 Gy) followed by 3 cycles of VIPD
30
83.3
80
3-year PFS, 85.2%; 3-year

OS, 86.3%
30 mg/m2 of concurrent cisplatin weekly for 35 weeks with IMRT (56 Gy)
followed by 3 cycles of GDP
Batchelor et al25
32
90.6
84.4

OS, 87.5%
2 cycles of LVP RT (56 Gy) 24 cycles LVP
Illerhaus et al26
26
92.0
88.5

OS, 88.5%
2 cycles of GELOX RT (56 Gy) 24 cycles GELOX
Kwong et al27
27
96.3
74.1
2-year PFS, 86%; 2-year

OS, 86%
Abbreviations: ENKL, extranodal natural killer/T-cell lymphoma, nasal type; RT, radiotherapy; DeVIC, dexamethasone, etoposide, ifosfamide, and carboplatin; VIPD, etoposide,
ifosfamide, cisplatin, and dexamethasone; IMRT, intensive modulated radiotherapy; GDP, gemcitabine, dexamethasone, and cisplatin; LVP, L-asparaginase, vincristine, and
prednisolone; GELOX, gemcitabine, L-asparaginase, and oxaliplatin; ORR, overall response rate; CR, complete remission; PFS, progression-free survival; OS, overall survival.
concurrent chemoradiotherapy and sandwiched radiotherapy approaches are appropriate (Table 4). Chemotherapy
regimens should incorporate non-MDRrelated agents and
etoposide. In comprehensive cancer centers where administering concurrent chemoradiation is feasible, radiotherapy
concurrent with 2/3 DeVIC is an attractive option based on its
robust clinical evidence and toxicity profile. There is a trend to
adopt sandwich radiotherapy or sequential approaches with
L-asparaginasebased regimens, which may be logistically less
complex to administer.
Management of Newly Diagnosed Advanced ENKL

Based on earlier studies57-59 that demonstrated the efficacy of L-asparaginase, a phase II trial involving the
SMILE regimen (Table 2)55 was initiated among 38 patients with newly diagnosed or relapsed/refractory disease. The primary endpoint of the study was ORR after
two courses of SMILE. The ORR and CR rates after 2 cycles
of SMILE were 79% and 45%, respectively. Interestingly,
there were no differences in response rates between
patients with untreated disease or those with a first relapse. With a median follow-up of 24 months, the 1-year
OS rate was 45%. Of the 28 patients who completed
at least two courses of SMILE treatment, four received
further autologous hematopoietic stem cell transplantation
(HSCT), 17 received allogeneic HSCT, and seven received
chemotherapy alone. Although the 1-year OS and PFS of patients who received autologous HSCT seemed better compared
with those who received allogeneic HSCT or chemotherapy
alone, the differences were not statistically significant. Of note,
the regimen was associated with significant toxicities, with
grade 4 neutropenia occurring among 92% of patients.
Outside the context of a clinical trial, the efficacy and
safety of the SMILE regimen was studied in 87 patients
with either untreated or relapsed/refractory disease.54 The
median number of courses of SMILE administered was
three (range, 06). The ORR and CR rates were 81% and 66%,
respectively. Of note, the estimated 5-year OS for newly
diagnosed patients was similar to the 5-year OS for patients
with relapsed/refractory disease (47.3% vs. 52.3%, p = .17).
Intriguingly, among the patients with relapsed/refractory

disease who attained CR after induction SMILE chemotherapy, eight patients (28%) who went on to receive
consolidation HSCT remained in remission compared with
12 (41%) who received only chemotherapy. Consistent with
the SMILE regimen, the hematologic toxicities encountered
were significant, and care is required in its administration.
One recent study reported that the combination of gemcitabine, oxaliplatin, L-asparagine, and dexamethasone may
yield similar efficacious results with fewer toxicities.60
Although consolidation treatment with autologous or allogenic HSCT for patients with advanced ENKL after induction
chemotherapy is feasible, its role remains contentious.61-63
Recent data suggest that the results of autologous HSCT are
comparable or inferior to L-asparaginasecontaining regimens.54,64 The Asian Lymphoma Study Group examined the
outcome of allogeneic HSCT among 18 patients, 14 of whom
received SMILE prior to transplantation.65 With a median
follow-up of 20.5 months, 5-year OS and event-free survival
were 57% and 51%, respectively, comparable to the results
observed with SMILE.54 Four patients (22%) died of treatmentrelated infective complications. Interestingly, the survival rates
of patients with CR1 and CR2 were similar, implying that allogeneic HSCT may not be necessary for a significant proportion of patients achieving CR1 with L-asparaginase therapy.
Taken together, patients with newly diagnosed advanced
ENKL should be treated with L-asparaginasebased regimens
such as SMILE (Fig. 2). The role of consolidation autologous or
allogenic HSCT for patients attaining complete remission remains debatable and best administered in the context of a
clinical trial.
Management of Relapsed/Refractory NK/T-Cell

Lymphoma
Patients who experience relapse after receiving L-asparaginase
based therapy are particularly difficult to treat. Gemcitabinecontaining regimens may be potentially effective.66 Several agents
are now approved for use in peripheral T-cell lymphoma
not otherwise specified, including romidepsin, pralatrexate,
and brentuximab vedotin.67-69 Their roles in ENKL remain to
e359
TABLE 4. Selected Chemotherapy Regimens for ENKL

Drug
Regimen
VIPD (repeat every

21 days)24
Days 13: 100 mg/m2 of intravenous etoposide

Days 13: 1,200 mg/m2 of intravenous
ifosfamide
can identify patients who are at risk for treatment failure.40

Future studies should examine the best consolidation approaches in this group of patients. Finally, recent molecular
studies have identified aberrations involving the JAK/STAT
cascade and histone modificationrelated genes that may
offer opportunities for targeted therapy in future.71,72
Days 13: 33 mg/m2 of intravenous cisplatin

Days 13: 40 mg of intravenous or oral
dexamethasone
GDP (repeat every
21 days)25
Day 1 and 8: 1,000 mg/m2 of intravenous

gemcitabine
Days 14: 40 mg of oral dexamethasone
Day 1: 75 mg/m2 of cisplatin
LVP (repeat every

21 days)26
Days 15: 6,000 IU/m2 of intravenous

L-asparaginase
Day 1: 1.4 mg/m2 of intravenous vincristine
Days 15: 100 mg of oral prednisolone
GELOX (repeat every Day 1 and 8: 1,000 mg/m2 of intravenous

gemcitabine
21 days)27
Day 1: 130 mg/m2 of intravenous oxaliplatin
Days 17: 6,000 U/m2 of intravenous
L-asparaginase
SMILE (repeat every
21 days)28,29
Day 1: 2 g/m2 of intravenous methotrexate

Days 24: 40 mg of oral dexamethasone (days
14)
Days 24: 15 mg 3 4 of intravenous or oral
leucovorin
Days 24: 1,500 mg/m2 of intravenous
ifosfamide
Days 24: 100 mg/m2 of intravenous etoposide
Days 8, 10, 12, 14, 16, 18, and 20: 6,000 U/m2 of
intravenous L-asparaginase
GOLD (repeat every

14 days)34
Days 1: 1,000 mg/m2 of gemcitabine

Day 1: 100 mg/m2 of intravenous oxaliplatin
Days 15: 10,000 U/m2 of intravenous
L-asparaginase
Days 14: 20 mg of intravenous or oral dexamethasone twice a day
Abbreviations: ENKL, extranodal natural killer/T-cell lymphoma nasal type; VIPD,

etoposide, ifosfamide, cisplatin, and dexamethasone; LVP, L-asparaginase, vincristine,
and prednisone; GELOX, gemcitabine, oxaliplatin, and L-asparaginase; SMILE, steroid,
methotrexate, ifosfamide, L-asparaginase, and etoposide; GOLD, emcitabine,
oxaliplatin, L-aspariginase, and dexamethasone.
be determined, and caution is required when used outside the

context of a clinical trial. In one pilot study, romidepsin increased
the risk of EBV reactivation in patients with ENKL.70 Allogeneic
transplantation for patients achieving CR2 may be potentially
curable, but its role has not been confirmed in prospective trials.
An individual patient-based decision approach is suggested.
Future Directions
Current management strategies are summarized (Table 4).
Significant progress has been achieved over the years with the
adoption of combined modality approaches and intensified
L-asparaginasebased therapy. However, disease progression
remains a risk. Recent data showed that the post-treatment
Deauville score on a PET/CT scan and the presence of EBV DNA
e360
POST-TRANSPLANT LYMPHOPROLIFERATIVE
DISORDERS
The term post-transplant lymphoproliferative disorder (PTLD)
is applied to a group of lymphoproliferative disorders arising
in a pharmacologically immunocompromised host after solid
organ or allogeneic stem cell transplantation.73 PTLD is related to EBV, but the number of EBV cases has increased from
10% (1990 to 1995) to 48% (2008 to 2013).74 Lymphoma after
solid organ transplantation represents about 21% of all
malignancies, whereas lymphoma represents 5% of all malignancies among the normal population. The biology, diagnosis, and management of this heterogeneous group of
disorders is different from other lymphoproliferative disorders. There are many variables in PTLD presentation, including
EBV serostatus prior to organ transplantation in the donor and
the recipient, histology, location of disease, stage of disease,
types and doses of immunosuppression agents in the course
of the transplant, initial management strategies (surgery,
immunosuppression reduction strategies), types of treatment (rituximab monotherapy or chemotherapy), presentation
from the time of transplantation, locations of extranodal disease, comorbid conditions (other vital organ involvement),
performance score, and active infections (cytomegalovirus,
fungal, etc.). International consensus development meetings
held in 1997 and 1998 addressed PTLD in working groups and
recommended management guidelines that set the stage for
future directions.75
Risk Factors for the Development of PTLD

PTLD has been highly associated with immunosuppression,
which creates a state of impaired T-cell immunity and EBV
infection. EBV has been implicated in 48%93% of all cases of
PTLD. Early PTLD occurs in the first 612 months, and late
PTLD occurs after 12 months. ISH studies on paraffin section
tissues determine EBV status. Late PTLD is associated with
negative EBV status and has an inferior prognosis.
The spectrum of PTLDs in the WHO classification include
early-type PTLD lesions (reactive plasmacytic hyperplasia
that is infectious mononucleosislike), polymorphic PTLD,
monomorphic PTLD (DLBCL, Burkitt lymphoma, plasmablastic
lymphoma, T-cell lymphomas, and composite lymphomas),
and Hodgkin lymphoma. Immunochemotherapy approaches
are histology dependent.76 There may be histologic overlap,
and there may be progressive transition from early PTLD
through polymorphic PTLD to monomorphic PTLD.
The Risk Factors for Surviving PTLD

The risk factors for surviving PTLD have been reported by
different groups. Leblond et al reported that a performance
FIGURE 2. Management of Newly Diagnosed Natural Killer/T-Cell Lymphoma
score of 2 or more, number of extranodal sites (one vs. more

than one), primary CNS location, T-cell origin, monoclonality,
EBV-negative status, and treatment with chemotherapy
were poor prognostic factors for 61 patients.77 Ghobrial et al
reported on 107 patients in a multivariate analysis and
identified monomorphic disease, graft organ involvement,
and a performance score of 3 to 4 as adverse; patients with a
score of 0 to 1 versus 2 to 3 had superior outcomes (hazard
ratio 5.31; p , .0001).78 CNS involvement, bone marrow
involvement, and hypoalbuminemia were predictive of a
negative outcome among 81 patients.79 The PTLD prognostic index was predictive of outcome. In a series of 70
patients from the PTLD-1 trial, an IPI score less than 3, lung
transplant, and response to rituximab were the strongest
predictors of outcome.80 The PTLD-2 trial and organspecific studies will define risk groups based on response
to rituximab monotherapy, IPI score, and the type of organ
transplanted.81,82
Management
The treatment of PTLD is complex and may include multiple
approaches.
Local therapy. A minority of patients with very limited
disease might be treated with surgical extirpation of localized radiation and minor immunosuppression reduction.
Emergent or palliative local radiation therapy has been
reported.
Immunosuppression reduction. The hallmark of the initial

management of PTLD has been to reduce immunosuppression with a goal of restoring EBV-specific cellular immunity without inducing graft rejection. The initial
consideration for all patients should be a reduction in immunosuppression, irrespective of EBV status. Regression of
monoclonal and polyclonal lymphoproliferative disorders
after reduction of the doses of immunosuppressive agents
has been reported to be 20%85%.83 The approaches to
immunosuppression reduction have been empirical, prospective clinical trials have not been reported, and approaches
are dependent on the organ transplanted. Consensus guidelines recommended the following and have evolved.75 For
critically ill patients with excessive disease, prednisone should
be decreased to 7.510 mg/day and all other immunosuppressive agents should be discontinued. If there is no response (objective reduction in tumor mass within a period of
1020 days), then further interventions must be considered. In
less critically ill patients (the majority of patients), the initial
management strategy should include reduction of cyclosporine, tacrolimus, prednisone, and related immunosuppressive
agents by at least 50%, and azathioprine or mycophenolate
mofetil should be discontinued. After a 14-day trial of decreased immunosuppression, a further decrease can be considered. This is one of the few examples in lymphoproliferative
disorders that patients are restaged with scans in such a short
period of time. Serial biopsies or other forms of graft organ
function monitoring are performed during this time period of
immunosuppression reduction to allow early detection of
e361
allograft rejection. This is especially important in the rejection

of vital organs such as the heart, liver, and lung, which may
carry a risk of death. Clinical improvement occurs within
1.54 weeks. EBV-negative disease is less responsive to immunosuppression reduction, but successful outcomes have
been reported. In a series of 42 patients treated with
immunosuppression reduction alone or in conjunction with
surgical excision, the complete remission rate was 74%.84
The response rate was 89% if no risk factors were present
(elevated LDH, organ dysfunction, or multiorgan disease),
which is in contrast with two to three risk factors and a 0%
response rate. The long-term remission rates in two series
were 20% and 25%.78,84
Rituximab. Rituximab after reduction in immunosuppression is the standard of care for most CD20+ PTLD subtypes,
including polymorphic and monomorphic DLBCL disease.
The ORRs to rituximab therapy, in conjunction with optimization of immunosuppression agents, are 44%75%, with
CR rates of 35%69%.85-88 The variable response rates relate
to whether the PTLD was early of late, the presence of EBV in
the tissue, LDH level, the type of organ transplanted, histology, other comorbid conditions, retrospective versus prospective reports, and duration of follow-up. There are no
randomized clinical trials comparing rituximab with other interventions. Monotherapy trials have reported a median OS
of 2.4 years. An international phase II trial, PTLD-1, enrolled
152 patients to evaluate rituximab consolidation among patients
who had a CR after rituximab consolidation introducing riskstratified sequential treatment to PTLD management in a third
amendment to the trial. Patients were treated with 375 mg/m2
of intravenous rituximab on days 1, 8, 15, and 22. After restaging,
patients in CR were treated with 3-weekly courses of rituximab
monotherapy. The 70 patients who were treated with rituximab
followed by cyclophosphamide, adriamycin, vincristine, and
prednisone (CHOP-21) served as the control population. Of
148 patients, 37 (25%) achieved a CR with four cycles of rituximab and went on to rituximab consolidation.89 The median
time to progression (TTP) was significantly longer than in the
corresponding group in the PTLD-1 trial (37 patients vs. 14
patients; p , .05). The addition of four doses of rituximab was a
predictor of OS, TTP, and PFS. Rituximab consolidation among
early rituximab responders resulted in significantly better
disease control and fewer toxicities compared with CHOP
consolidation. The PD-1 trial provides information to tailor
treatment.90
Immunochemotherapy in DLBCL PTLD. The response rates
to systemic chemotherapy during the 1980s were reported to
be less than 20%. This has changed over time. The PTLD-1 trial
demonstrated the efficacy and safety of 375 mg/m2 of rituximab weekly for four doses followed by four cycles of CHOP21 with a median OS of 6.6 years.91 This study underwent three
amendments, and the protocol was amended. All patients
were initially treated with 375 mg/m2 of rituximab on days 1, 8,
15, and 22. Patients who were not in a complete remission were subsequently treated with rituximab and CHOP
e362
combination therapy (R-CHOP-21) with granulocyte-colony

stimulating factor for four cycles. There were 111 patients
who received R-CHOP consolidation, with an ORR of 85% (78 of
92) and a CR rate of 60% (55 of 92). The 3-year TTP was 73% in
the PTLD-1 trial among patients who were refractory to rituximab induction. The treatment-related mortality was 7%.91
Chemotherapy. Immunosuppression reduction should be
considered, but, in late PTLD, doses are low. Indications for
initial treatment with chemotherapy include specific histologies, such as peripheral T-cell lymphoma not otherwise
specified, Hodgkin lymphoma, late PTLD, and other uncommon
lymphomas. These lymphomas must be treated with the
standard-of-care approaches for these specific histologies.
Stem cell transplantation. Patients who have failed other
approaches have been in long-term remission with autologous or allogeneic stem cell transplantation.92
Cellular immunotherapy. The transfer of selected or unselected EBV-specific cytotoxic T cells to patients with
EBV-positive PTLD to restore EBV-specific cellular immunity
as prevention or treatment has been under investigation. Of
114 patients who received infusions of EBV-specific cytotoxic T lymphocytes, none of 101 patients who received
prophylaxis developed PTLD, and 11 of 13 patients with
biopsy-proven or probable PTLD achieved a sustained
CR.93-95
Unique PTLD Scenarios

Burkitt lymphoma PTLD is rare and may be MYC negative, but it
is associated with 11q- gain/loss abnormality.95,96 In a retrospective review, 25 of 26 evaluable patients had a proven MYC
rearrangement. Intensive therapy was associated with toxicityrelated deaths in three of five patients.96 The German Study
Group on PTLD reported that four of five patients reached a
complete remission, and the authors concluded that sequential
immunochemotherapy was safe and effective.97
Primary CNS PTLD represents a distinct clinicopathologic entity that occurs in 7%15% of patients with PTLD.
An international collaboration reported on 84 patients
treated over a 14-year period between 1997 and 2010.98
This entity is more commonly a late PTLD with median
time of solid organ transplantation to PTLD of 54 months;
94% of patients had EBV-positive disease the histology
was monomorphic for 83% (DLBCL in 66), and 79% had
received renal transplants. In addition, 37% of patients
had multifocal disease, and deep brain involvement was
seen in 33% of patients. Immunosuppression was decreased
in 93% of patients. Additional first-line treatment consisted
of HD-MTX (48%), high-dose cytarabine (33%) alone or in
combination with 20% receiving after methotrexate, brain
radiation (24%), and/or rituximab with 10% rituximab
alone. Methotrexate and rituximab was the most common
regimen administered to 12 patients (10%). The ORR was
60%. The median PFS was 8 months and median OS was
17 months. At a median follow-up of 43 months, the 3-year
FIGURE 3. Proposed Initial Management of PTLD
PFS and OS were 32% and 43%, respectively. In a multivariate analysis, lack of response to first-line therapy
(p = .0005) and increased LDH were associated with an
inferior OS (p = .02). The treatment-related mortality was
13%. Proposed initial management strategies for PTLD are

outlined in Fig. 3. Dierickx et al have published a proposed
treatment algorithm PTLD following hematopoietic stem
cell transplantation.81
References
1. Dolecek TA, Propp JM, Stroup NE, et al. CBTRUS statistical report:
primary brain and central nervous system tumors diagnosed in the
United States in 2005-2009. Neuro Oncol. 2012;14:v1-v49.
2. Villano JL, Koshy M, Shaikh H, et al. Age, gender, and racial differences in
incidence and survival in primary CNS lymphoma. Br J Cancer. 2011;105:
1414-1418.
3. Swerdlow SH, Campo E, Harris NL, et al (eds). WHO Classification of
Tumours of the Haematopoietic and Lymphoid Tissues. Lyon, France:
International Agency for Research on Cancer; 2008.
4. Ponzoni M, Issa S, Batchelor TT, et al. Beyond high-dose methotrexate
and brain radiotherapy: novel targets and agents for primary CNS
lymphoma. Ann Oncol. 2014;25:316-322.
5. Abrey LE, Batchelor TT, Ferreri AJ, et al; International Primary CNS
Lymphoma Collaborative Group. Report of an international workshop
to standardize baseline evaluation and response criteria for primary
CNS lymphoma. J Clin Oncol. 2005;23:5034-5043.
6. Chan CC, Rubenstein JL, Coupland SE, et al. Primary vitreoretinal
lymphoma: a report from an International Primary Central Nervous
System Lymphoma Collaborative Group symposium. Oncologist. 2011;
16:1589-1599.
7. Ferreri AJ, Blay JY, Reni M, et al. Prognostic scoring system for primary
CNS lymphomas: the International Extranodal Lymphoma Study Group
experience. J Clin Oncol. 2003;21:266-272.
8. Abrey LE, Ben-Porat L, Panageas KS, et al. Primary central nervous

system lymphoma: the Memorial Sloan-Kettering Cancer Center
prognostic model. J Clin Oncol. 2006;24:5711-5715.
9. Mathew BS, Carson KA, Grossman SA. Initial response to glucocorticoids. Cancer. 2006;106:383-387.
10. Bellinzona M, Roser F, Ostertag H, et al. Surgical removal of primary
central nervous system lymphomas (PCNSL) presenting as space
occupying lesions: a series of 33 cases. Eur J Surg Oncol. 2005;31:
100-105.
11. Nelson DF, Martz KL, Bonner H, et al. Non-Hodgkins lymphoma of the
brain: can high dose, large volume radiation therapy improve survival?
Report on a prospective trial by the Radiation Therapy Oncology Group
(RTOG): RTOG 8315. Int J Radiat Oncol Biol Phys. 1992;23:9-17.
12. Glantz MJ, Cole BF, Recht L, et al. High-dose intravenous methotrexate
for patients with nonleukemic leptomeningeal cancer: is intrathecal
chemotherapy necessary? J Clin Oncol. 1998;16:1561-1567.
13. DeAngelis LM, Seiferheld W, Schold SC, et al; Radiation Therapy
Oncology Group Study 93-10. Combination chemotherapy and radiotherapy for primary central nervous system lymphoma: Radiation Therapy Oncology Group Study 93-10. J Clin Oncol. 2002;20:
4643-4648.
14. Ferreri AJ, Reni M, Foppoli M, et al; International Extranodal Lymphoma
Study Group (IELSG). High-dose cytarabine plus high-dose methotrexate versus high-dose methotrexate alone in patients with primary
e363
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
CNS lymphoma: a randomised phase 2 trial. Lancet. 2009;374:

1512-1520.
Correa DD, Maron L, Harder H, et al. Cognitive functions in primary
central nervous system lymphoma: literature review and assessment
guidelines. Ann Oncol. 2007;18:1145-1151.
Morris PG, Correa DD, Yahalom J, et al. Rituximab, methotrexate,
procarbazine, and vincristine followed by consolidation reduced-dose
whole-brain radiotherapy and cytarabine in newly diagnosed primary
CNS lymphoma: final results and long-term outcome. J Clin Oncol. 2013;
31:3971-3979.
Doolittle ND, Korfel A, Lubow MA, et al. Long-term cognitive function,
neuroimaging, and quality of life in primary CNS lymphoma. Neurology.
2013;81:84-92.
Juergens A, Pels H, Rogowski S, et al. Long-term survival with favorable
cognitive outcome after chemotherapy in primary central nervous
system lymphoma. Ann Neurol. 2010;67:182-189.
Thiel E, Korfel A, Martus P, et al. High-dose methotrexate with or
without whole brain radiotherapy for primary CNS lymphoma (GPCNSL-SG-1): a phase 3, randomised, non-inferiority trial. Lancet
Oncol. 2010;11:1036-1047.
Ferreri AJ, Guerra E, Regazzi M, et al. Area under the curve of methotrexate and creatinine clearance are outcome-determining factors in
primary CNS lymphomas. Br J Cancer. 2004;90:353-358.
Batchelor T, Carson K, ONeill A, et al. Treatment of primary CNS
lymphoma with methotrexate and deferred radiotherapy: a report of
NABTT 96-07. J Clin Oncol. 2003;21:1044-1049.
Rubenstein JL, Hsi ED, Johnson JL, et al. Intensive chemotherapy and
immunotherapy in patients with newly diagnosed primary CNS lymphoma: CALGB 50202 (Alliance 50202). J Clin Oncol. 2013;31:
3061-3068.
Khan RB, Shi W, Thaler HT, et al. Is intrathecal methotrexate necessary in
the treatment of primary CNS lymphoma? J Neurooncol. 2002;58:
175-178.
Sierra del Rio M, Ricard D, Houillier C, et al. Prophylactic intrathecal
chemotherapy in primary CNS lymphoma. J Neurooncol. 2012;106:
143-146.
Batchelor TT, Grossman SA, Mikkelsen T, et al. Rituximab monotherapy
for patients with recurrent primary CNS lymphoma. Neurology. 2011;
76:929-930.
Illerhaus G, Fritsch K, Egerer G, et al. Sequential high dose immunochemotherapy followed by autologous peripheral blood stem cell
transplantation for patients with untreated primary central nervous
system lymphoma - a multicentre study by the Collaborative PCNSL
Study Group Freiburg. Blood. 2012;120 (abstr 302).
Kwong YL, Anderson BO, Advani R, et al; Asian Oncology Summit.
Management of T-cell and natural-killer-cell neoplasms in Asia: consensus statement from the Asian Oncology Summit 2009. Lancet Oncol.
2009;10:1093-1101.
Chan JKC, Jaffe ES, Ralfkiaer E, et al. Aggressive NK-cell leukemia. In
Swerdlow SH, Campo E, Harris NL, et al (eds). WHO Classification of
Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France: International Agency for Research on Cancer, 2008; 276-277.
Chan WK, Au WY, Wong CY, et al. Metabolic activity measured by F-18
FDG PET in natural killer-cell lymphoma compared to aggressive B- and
T-cell lymphomas. Clin Nucl Med. 2010;35:571-575.
Khong PL, Pang CB, Liang R, et al. Fluorine-18 fluorodeoxyglucose
positron emission tomography in mature T-cell and natural killer cell
malignancies. Ann Hematol. 2008;87:613-621.
Moon SH, Cho SK, Kim WS, et al. The role of 18F-FDG PET/CT for initial
staging of nasal type natural killer/T-cell lymphoma: a comparison with conventional staging methods. J Nucl Med. 2013;54:
1039-1044.
e364
32. Ito Y, Kimura H, Maeda Y, et al. Pretreatment EBV-DNA copy number is

predictive of response and toxicities to SMILE chemotherapy for
extranodal NK/T-cell lymphoma, nasal type. Clin Cancer Res. 2012;18:
4183-4190.
33. Lee J, Suh C, Huh J, et al. Effect of positive bone marrow EBV in situ
hybridization in staging and survival of localized extranodal natural
killer/T-cell lymphoma, nasal-type. Clin Cancer Res. 2007;13:3250-3254.
34. Au WY, Pang A, Choy C, et al. Quantification of circulating Epstein-Barr
virus (EBV) DNA in the diagnosis and monitoring of natural killer cell and
EBV-positive lymphomas in immunocompetent patients. Blood. 2004;
104:243-249.
35. Suzuki R, Yamaguchi M, Izutsu K, et al; NK-cell Tumor Study Group.
Prospective measurement of Epstein-Barr virus-DNA in plasma and
peripheral blood mononuclear cells of extranodal NK/T-cell lymphoma,
nasal type. Blood. 2011;118:6018-6022.
36. Chim CS, Ma SY, Au WY, et al. Primary nasal natural killer cell lymphoma:
long-term treatment outcome and relationship with the International
Prognostic Index. Blood. 2004;103:216-221.
37. Lee J, Suh C, Park YH, et al. Extranodal natural killer T-cell lymphoma,
nasal-type: a prognostic model from a retrospective multicenter study.
J Clin Oncol. 2006;24:612-618.
38. Suzuki R, Suzumiya J, Yamaguchi M, et al; NK-cell Tumor Study Group.
Prognostic factors for mature natural killer (NK) cell neoplasms: aggressive NK cell leukemia and extranodal NK cell lymphoma, nasal type.
Ann Oncol. 2010;21:1032-1040.
39. Yang Y, Zhang YJ, Zhu Y, et al. Prognostic nomogram for overall
survival in previously untreated patients with extranodal NK/T-cell
lymphoma, nasal-type: a multicenter study. Leukemia. 2015;29:
1571-1577.
40. Kim SJ, Choi JY, Hyun SH, et al; Asia Lymphoma Study Group. Risk
stratification on the basis of Deauville score on PET-CT and the presence
of Epstein-Barr virus DNA after completion of primary treatment for
extranodal natural killer/T-cell lymphoma, nasal type: a multicentre,
retrospective analysis. Lancet Haematol. 2015;2:e66-e74.
41. Li YX, Yao B, Jin J, et al. Radiotherapy as primary treatment for stage IE
and IIE nasal natural killer/T-cell lymphoma. J Clin Oncol. 2006;24:
181-189.
42. Isobe K, Uno T, Tamaru J, et al. Extranodal natural killer/T-cell lymphoma, nasal type: the significance of radiotherapeutic parameters.
Cancer. 2006;106:609-615.
43. Koom WS, Chung EJ, Yang WI, et al. Angiocentric T-cell and NK/T-cell
lymphomas: radiotherapeutic viewpoints. Int J Radiat Oncol Biol Phys.
2004;59:1127-1137.
44. Cheung MM, Chan JK, Lau WH, et al. Early stage nasal NK/T-cell lymphoma: clinical outcome, prognostic factors, and the effect of treatment modality. Int J Radiat Oncol Biol Phys. 2002;54:182-190.
45. Kim GE, Cho JH, Yang WI, et al. Angiocentric lymphoma of the head and
neck: patterns of systemic failure after radiation treatment. J Clin Oncol.
2000;18:54-63.
46. Kim WS, Song SY, Ahn YC, et al. CHOP followed by involved field radiation: is it optimal for localized nasal natural killer/T-cell lymphoma?
Ann Oncol. 2001;12:349-352.
47. Lee SH, Ahn YC, Kim WS, et al. The effect of pre-irradiation dose intense
CHOP on anthracyline resistance in localized nasal NK/T-cell lymphoma.
Haematologica. 2006;91:427-428.
48. Jiang M, Zhang H, Jiang Y, et al. Phase 2 trial of sandwich
L-asparaginase, vincristine, and prednisone chemotherapy with radiotherapy in newly diagnosed, stage IE to IIE, nasal type, extranodal natural
killer/T-cell lymphoma. Cancer. 2012;118:3294-3301.
49. Ke QH, Zhou SQ, Du W, et al. Concurrent IMRT and weekly cisplatin
followed by GDP chemotherapy in newly diagnosed, stage IE to IIE,
nasal, extranodal NK/T-Cell lymphoma. Blood Cancer J. 2014;4:e267.
50. Kim SJ, Kim K, Kim BS, et al. Phase II trial of concurrent radiation and
weekly cisplatin followed by VIPD chemotherapy in newly diagnosed,
stage IE to IIE, nasal, extranodal NK/T-cell lymphoma: Consortium for
Improving Survival of Lymphoma study. J Clin Oncol. 2009;27:
6027-6032.
51. Wang L, Wang ZH, Chen XQ, et al. First-line combination of gemcitabine,
oxaliplatin, and L-asparaginase (GELOX) followed by involved-field radiation therapy for patients with stage IE/IIE extranodal natural killer/Tcell lymphoma. Cancer. 2013;119:348-355.
52. Yamaguchi M, Tobinai K, Oguchi M, et al. Concurrent chemoradiotherapy for localized nasal natural killer/T-cell lymphoma: an
updated analysis of the Japan clinical oncology group study JCOG0211.
J Clin Oncol. 2012;30:4044-4046.
53. Yamaguchi M, Tobinai K, Oguchi M, et al. Phase I/II study of concurrent
chemoradiotherapy for localized nasal natural killer/T-cell lymphoma:
Japan Clinical Oncology Group Study JCOG0211. J Clin Oncol. 2009;27:
5594-5600.
54. Kwong YL, Kim WS, Lim ST, et al. SMILE for natural killer/T-cell lymphoma: analysis of safety and efficacy from the Asia Lymphoma Study
Group. Blood. 2012;120:2973-2980.
55. Yamaguchi M, Kwong YL, Kim WS, et al. Phase II study of SMILE
chemotherapy for newly diagnosed stage IV, relapsed, or refractory extranodal natural killer (NK)/T-cell lymphoma, nasal type:
the NK-Cell Tumor Study Group study. J Clin Oncol. 2011;29:
4410-4416.
56. Yang Y, Zhu Y, Cao JZ, et al. Risk-adapted therapy for early-stage
extranodal nasal-type NK/T-cell lymphoma: analysis from a multicenter study. Blood. 2015;126:1424-1432, quiz 1517.
57. Matsumoto Y, Nomura K, Kanda-Akano Y, et al. Successful treatment
with Erwinia L-asparaginase for recurrent natural killer/T cell lymphoma. Leuk Lymphoma. 2003;44:879-882.
58. Nagafuji K, Fujisaki T, Arima F, et al. L-asparaginase induced durable
remission of relapsed nasal NK/T-cell lymphoma after autologous
peripheral blood stem cell transplantation. Int J Hematol. 2001;74:
447-450.
59. Obama K, Tara M, Niina K. L-Asparaginase-based induction therapy for
advanced extranodal NK/T-cell lymphoma. Int J Hematol. 2003;78:
248-250.
60. Guo HQ, Liu L, Wang XF, et al. Efficacy of gemcitabine combined with
oxaliplatin, L-asparaginase and dexamethasone in patients with newlydiagnosed extranodal NK/T-cell lymphoma. Mol Clin Oncol. 2014;2:
1172-1176.
61. Kim HJ, Bang SM, Lee J, et al. High-dose chemotherapy with autologous
stem cell transplantation in extranodal NK/T-cell lymphoma: a retrospective comparison with non-transplantation cases. Bone Marrow
Transplant. 2006;37:819-824.
62. Murashige N, Kami M, Kishi Y, et al. Allogeneic haematopoietic stem cell
transplantation as a promising treatment for natural killer-cell neoplasms. Br J Haematol. 2005;130:561-567.
63. Suzuki R, Suzumiya J, Nakamura S, et al; NK-cell Tumor Study Group.
Hematopoietic stem cell transplantation for natural killer-cell lineage
neoplasms. Bone Marrow Transplant. 2006;37:425-431.
64. Kwong YL. High-dose chemotherapy and hematopoietic SCT in the
management of natural killer-cell malignancies. Bone Marrow Transplant. 2009;44:709-714.
65. Tse E, Chan TS, Koh LP, et al. Allogeneic haematopoietic SCT for
natural killer/T-cell lymphoma: a multicentre analysis from the Asia
Lymphoma Study Group. Bone Marrow Transplant. 2014;49:
902-906.
66. Ahn HK, Kim SJ, Hwang DW, et al. Gemcitabine alone and/or containing
chemotherapy is efficient in refractory or relapsed NK/T-cell lymphoma.
Invest New Drugs. 2013;31:469-472.
67. Coiffier B, Pro B, Prince HM, et al. Results from a pivotal, open-label,
phase II study of romidepsin in relapsed or refractory peripheral Tcell lymphoma after prior systemic therapy. J Clin Oncol. 2012;30:
631-636.
68. OConnor OA, Pro B, Pinter-Brown L, et al. Pralatrexate in patients with
relapsed or refractory peripheral T-cell lymphoma: results from the
pivotal PROPEL study. J Clin Oncol. 2011;29:1182-1189.
69. Younes A, Bartlett NL, Leonard JP, et al. Brentuximab vedotin (SGN-35)
for relapsed CD30-positive lymphomas. N Engl J Med. 2010;363:
1812-1821.
70. Kim SJ, Kim JH, Ki CS, et al. Epstein-Barr virus reactivation in extranodal
natural killer/T-cell lymphoma patients: a previously unrecognized
serious adverse event in a pilot study with romidepsin. Ann Oncol. Epub
2015 Dec 9.
71. Koo GC, Tan SY, Tang T, et al. Janus kinase 3-activating mutations
identified in natural killer/T-cell lymphoma. Cancer Discov. 2012;2:
591-597.
72. Lee S, Park HY, Kang SY, et al. Genetic alterations of JAK/STAT cascade
and histone modification in extranodal NK/T-cell lymphoma nasal type.
Oncotarget. 2015;6:17764-17776.
73. Murray JE, Wilson RE, Tilney NL, et al. Five years experience in renal
transplantation with immunosuppressive drugs: survival, function,
complications, and the role of lymphocyte depletion by thoracic duct
fistula. Ann Surg. 1968;168:416-435.
74. Luskin MR, Heil DS, Tan KS, et al. The impact of EBV status on characteristics and outcomes of posttransplantation lymphoproliferative
disorders. Am J Transplant. 2015;15:2665-2673.
75. Paya CV, Fung JJ, Nalesnik MA, et al. Epstein-Barr virus-induced posttransplant lymphoproliferative disorders. AST/ASTP EBV-PTLD Task
Force and the Mayo Clinic Organized International Consensus Development Meeting. Transplantation. 1999;68:1517-1525.
76. Swerdlow SH, Webber SA, Chadbum A, et al. Post-transplant lymphoproliferative disorders. In Swerdlow SH, Campo E, Harris NL, et al
(eds). WHO Classification of Tumours of Haematopoietic and Lymphoid
Tissues. Lyon, France: International Agency for Research on Cancer
Press, 2008; 343-349.
77. Leblond V, Dhedin N, Mamzer Bruneel MF, et al. Identification of
prognostic factors in 61 patients with posttransplantation lymphoproliferative disorders. J Clin Oncol. 2001;19:772-778.
78. Ghobrial IM, Habermann TM, Maurer MJ, et al. Prognostic analysis for
survival in adult solid organ transplant recipients with posttransplantation lymphoproliferative disorders. J Clin Oncol. 2005;23:
7574-7582.
79. Evens AM, David KA, Helenowski I, et al. Multicenter analysis of 80 solid
organ transplantation recipients with post-transplantation lymphoproliferative disease: outcomes and prognostic factors in the modern
era. J Clin Oncol. 2010;28:1038-1046.
80. Dierickx D, Tousseyn T, Sagaert X, et al. Single-center analysis of biopsyconfirmed posttransplant lymphoproliferative disorder: incidence,
clinicopathological characteristics and prognostic factors. Leuk Lymphoma. 2013;54:2433-2440.
81. Dierickx D, Tousseyn T, Gheysens O. How I treat posttransplant lymphoproliferative disorders. Blood. 2015;126:2274-2283.
82. Hourigan MJ, Doecke J, Mollee PN, et al. A new prognosticator for posttransplant lymphoproliferative disorders after renal transplantation. Br
J Haematol. 2008;141:904-907.
83. Habermann TM. Posttransplant lymphoproliferative disorders. Cancer
Treat Res. 2008;142:273-292.
84. Tsai DE, Hardy CL, Tomaszewski JE, et al. Reduction in immunosuppression as initial therapy for posttransplant lymphoproliferative disorder: analysis of prognostic variables and long-term follow-up of 42
adult patients. Transplantation. 2001;71:1076-1088.
e365
85. Oertel SH, Verschuuren E, Reinke P, et al. Effect of anti-CD 20 antibody

rituximab in patients with post-transplant lymphoproliferative disorder
(PTLD). Am J Transplant. 2005;5:2901-2906.
86. Blaes AH, Peterson BA, Bartlett N, et al. Rituximab therapy is effective for posttransplant lymphoproliferative disorders after solid
organ transplantation: results of a phase II trial. Cancer. 2005;104:
1661-1667.
87. Choquet S, Leblond V, Herbrecht R, et al. Efficacy and safety of rituximab in B-cell post-transplantation lymphoproliferative disorders:
results of a prospective multicenter phase 2 study. Blood. 2006;107:
3053-3057.
88. Gonzalez-Barca E, Domingo-Domenech E, Capote FJ, et al; GEL/TAMO
~ol de Linfomas); GELCAB (Grupo para el Estudio de los
(Grupo Espan
Linfomas Catalano-Balear); GOTEL (Grupo Oncologico
para el Tratamiento y Estudio de los Linfomas). Prospective phase II trial of extended

treatment with rituximab in patients with B-cell post-transplant lymphoproliferative disease. Haematologica. 2007;92:1489-1494.
89. Trappe RU, Dierickx D, Zimmermann H, et al. Response to rituximab
induction is a predictive biomarker in post-transplant lymphoproliferative disorder (PTLD) and allows successful treatment stratification
in an international phase II trial including 152 patients. Blood. 2015;126
(abstr 816).
90. Trappe RU, Choquet S, Dierickx D, et al; German PTLD Study Group and
the European PTLD Network. International prognostic index, type of
transplant and response to rituximab are key parameters to tailor
treatment in adults with CD20-positive B cell PTLD: clues from the PTLD1 trial. Am J Transplant. 2015;15:1091-1100.
e366
91. Trappe R, Oertel S, Leblond V, et al; German PTLD Study Group; European PTLD Network. Sequential treatment with rituximab followed by
CHOP chemotherapy in adult B-cell post-transplant lymphoproliferative
disorder (PTLD): the prospective international multicentre phase 2
PTLD-1 trial. Lancet Oncol. 2012;13:196-206.
92. Komrokji RS, Oliva JL, Zand M, et al. Mini-BEAM and autologous hematopoietic stem-cell transplant for treatment of post-transplant
lymphoproliferative disorders. Am J Hematol. 2005;79:211-215.
93. Choquet S, Varnous S, Deback C, et al. Adapted treatment of EpsteinBarr virus infection to prevent posttransplant lymphoproliferative
disorder after heart transplantation. Am J Transplant. 2014;14:857-866.
94. Omar H, Hagglund H, Gustafsson-Jernberg A, et al. Targeted monitoring
of patients at high risk of post-transplant lymphoproliferative disease by
quantitative Epstein-Barr virus polymerase chain reaction. Transpl Infect Dis. 2009;11:393-399.
95. Heslop HE, Slobod KS, Pule MA, et al. Long-term outcome of EBVspecific T-cell infusions to prevent or treat EBV-related lymphoproliferative disease in transplant recipients. Blood. 2010;115:925-935.
96. Ferreiro JF, Morscio J, Dierickx D, et al. Post-transplant molecularly defined
Burkitt lymphomas are frequently MYC-negative and characterized by the
11q-gain/loss pattern. Haematologica. 2015;100:e275-e279.
97. Xicoy B, Ribera JM, Esteve J, et al. Post-transplant Burkitts leukemia or
lymphoma. Study of five cases treated with specific intensive therapy
(PETHEMA ALL-3/97 trial). Leuk Lymphoma. 2003;44:1541-1543.
98. Evens AM, Choquet S, Kroll-Desrosiers AR, et al. Primary CNS posttransplant lymphoproliferative disease (PTLD): an international report
of 84 cases in the modern era. Am J Transplant. 2013;13:1512-1522.
PET-Directed Strategies in
Lymphoma
CHAIR
John A. Radford, MD
Christie Hospital NHS Foundation Trust
Manchester, United Kingdom
SPEAKERS
Franco Cavalli, MD
Oncology Institute of Southern Switzerland
Bellinzona, Switzerland
Ranjana H. Advani, MD
Stanford, CA
CAVALLI, CERIANI, AND ZUCCA
Functional Imaging Using 18-Fluorodeoxyglucose PET in the

Management of Primary Mediastinal Large B-Cell Lymphoma:
The Contributions of the International Extranodal Lymphoma
Study Group
Franco Cavalli, MD, Luca Ceriani, MD, and Emanuele Zucca, MD
OVERVIEW
Primary mediastinal large B-cell lymphoma (PMLBCL) is recognized as a distinct disease entity. Treatment outcomes appear better
than in other diffuse large B-cell lymphoma (DLBCL) types, partly because of their earlier stage at presentation and the younger
age of most patients. If initial treatment fails, however, the results of salvage chemotherapy and myeloablative treatment are
poor. The need to avoid relapses after initial therapy has led to controversy over the extent of front-line therapy, particularly
whether consolidation radiotherapy to the mediastinum is always required and whether the 18-fluorodeoxyglucose (18F-FDG)
uptake detected by PET-CT scan can be used to determine its requirements. Functional imaging using PET-CT generally allows
distinguishing of residual mediastinal masses containing active lymphoma from those with only sclerotic material remaining. The
International Extranodal Lymphoma Study Group (IELSG) conducted the prospective IELSG-26 study, which showed that a fivepoint visual scale can be used to define metabolic response after immunochemotherapy and that a cut point based on liver uptake
discriminates effectively between high or low risk of failure, with 5-year progression-free survival (PFS) of 99% versus 68% and 5year overall survival (OS) of 100% versus 83%. This study also showed that a baseline quantitative PET parameter, namely the total
lesion glycolysis describing the metabolic tumor burden, can be a powerful predictor of PMLBCL outcomes and warrants further
validation as a biomarker. The ongoing IELSG-37 randomized study addresses the need for consolidation mediastinal radiotherapy
in patients in whom a complete metabolic response (CMR) can be seen on PET scans after standard immunochemotherapy.
rimary mediastinal large B-cell lymphoma is recognized

by the World Health Organizations classification of
hematopoietic and lymphoid tissue tumors as a distinct
entity on clinicopathologic1,2 and molecular3-5 criteria. It
accounts for less than 5% of non-Hodgkin lymphomas and
comprises approximately 5%10% of the DLBCL.1,2,6 Patients
with PMLBCL tend to be younger than the other DLBCL cases
with a median age at diagnosis in the third to fourth decade
and a female preponderance.2,7 This lymphoma is clinically
characterized by a rapidly progressive anterior mediastinal
mass, often with local invasion and compressive syndromes,
and by recurrence at unusual sites, such as the liver, kidneys,
and central nervous system.6
Cell of Origin and Immunophenotype

Immunohistochemical characterization and molecular studies
strongly suggest that PMLBCL is of germinal center or postgerminal center origin.2,5 Phenotypic analysis shows positivity
for CD45, CD79a, CD22, CD19, and CD20, with negativity for
CD3, CD10, CD21, and class I/II major histocompatibility
antigens. Expression of surface immunoglobulins is lost
despite the presence of isotype-switched immunoglobulin
genes.1,8,9 CD30 staining is observed in most cases but is
weaker and less homogeneous than in Hodgkin lymphoma
or anaplastic large cell lymphoma1,5; BCL2 and BCL6 proteins
are often expressed.1
Molecular Genetics
BIOLOGIC PECULIARITIES OF PMLBCL

Primary mediastinal large B-cell lymphoma is considered to
typically arise from a small population of B cells within the
thymus and, thymic components such as Hassall corpuscles
may be identified in the pathology specimen.2
Primary mediastinal large B-cell lymphoma is characterized

by peculiar genetic features that make this entity closer to
classic Hodgkin lymphoma than to DLBCL.3-5,10 The main
characteristics are represented by deregulated immune
checkpoint and JAK/STAT signaling, via DNA gains and
From the Oncology Institute of Southern Switzerland, Lymphoma UnitOspedale San Giovanni, Bellinzona, Switzerland.
Corresponding author: Franco Cavalli, MD, Ospedale Regionale Bellinzona e Valli, Bellinzona, CH-6500, Switzerland; email: franco.cavalli@eoc.ch.
e368
ROLE OF PET IN THE MANAGEMENT OF PMLBCL
amplifications at 9p, affecting PDL2 and JAK2, and recurrent

somatic mutations of the STAT6 and SOCS1 genes.1,5,9,11-15
There are also recurrent point mutations of the XPO1 gene,
coding for the Exportin-1 cargo protein that mediates the
nuclear export of multiple tumor suppressor proteins, including p53. Although their biologic significance is still under
investigation, XPO1 mutations appear to be of potential
clinical relevance in terms of outcome prediction and differential diagnosis with classic Hodgkin lymphoma and grayzone lymphomas.16,17
CURRENT CLINICAL CONTROVERSIES

The outcome of therapy for patients with PMLBCL is generally better than for those with other DLBCL types, possibly
also as a result of the younger age of the patients and earlier
stage at presentation. In untreated PMLBCL, the response
to combination chemotherapy is very good and is commonly
followed by consolidation involved-field radiotherapy (IFRT).
After the addition of rituximab to doxorubicin, cyclophosphamide, vincristine, and prednisone (CHOP) or CHOP-like
regimens, the reported PFS and OS rates at 3 years are
approximately 75%80% and 85%90%, respectively.18,19
However, there is a subset of patients in whom early refractory disease develops, often during the initial courses of
treatment, with a high failure rate2,7: in the case this initial
treatment fails, the results of salvage chemotherapy and
myeloablative treatment are often very poor.20 The need to
maximize cure rates with initial therapy has led to controversy over the utility of consolidation radiotherapy to the
mediastinum. The optimal chemotherapy regimen also remains controversial.
The outcomes reported in the prerituximab era by several
European groups have suggested that third-generation
regimens (e.g., MACOP-B or VACOP-B) may be superior to
the CHOP regimen.21-25 Improved PFS and OS with more
aggressive chemotherapy regimens compared with CHOPlike regimens were confirmed by a large European survey
conducted by the IELSG.26 This multinational retrospective
study (IELSG-9) compared the outcomes of patients with
KEY POINTS
PMLBCL is a distinct clinicopathologic and biologic type

of large-cell lymphoma.
Survival is usually excellent with intensive front-line
immunochemotherapy regimens, but the few patients
in whom it fails have a particularly dismal outcome.
18
F-FDG PET-CT is a very important staging tool for
patients with PMLBCL.
The metabolic activity on PET scans is a powerful
predictor of PMLBCL outcome.
An ongoing randomized study is addressing the need of
consolidation mediastinal radiotherapy in patients who
have shown a complete metabolic response on PET
scans.
PMLBCL after first-generation (standard CHOP or CHOP-like

regimens), third-generation (dose-intensive/alternating regimens such as MACOP-B, VACOP-B, ProMACE, CytaBOM), and
high-dose chemotherapy strategies with autologous stem cell
transplant, frequently including adjuvant radiation therapy. A
total of 426 previously untreated patients with confirmed
diagnosis were enrolled by 20 institutions. Projected 10-year
PFS rates were 35%, 67%, and 78%, after first generation, third
generation, and high-dose chemotherapy strategies, respectively (p , .001). Projected 10-year OS rates were 44%,
71%, and 77%, respectively (p , .001), after median followups from diagnosis of 52.3 months, 54.9 months, and
35.8 months, respectively. Very similar results were reported
by other retrospective studies from Italy27 and North
America.18,28,29
In the last decade, the inclusion of rituximab as part of
initial therapy for CD20+ B-cell lymphomas has clearly improved response rates and OS in several DLBCL randomized
trials.30,31 Although direct comparison between different
nonrandomized clinical studies with or without rituximab
is problematic, the rituximab regimen of CHOP (R-CHOP)
is likely to have also improved the outcome in PMLBCL.
Indeed, in a Canadian retrospective study, the addition of
rituximab to CHOP chemotherapy resulted in the same 5-year
OS obtained with more intensive regimens of MACOP-B/
VACOP-B.18 In the subgroup analysis of the MINT international randomized trial, the subgroup of patients with PMLBCL
treated with rituximab plus chemotherapy (CHOP in most
cases) showed a significantly better 3-year PFS than did patients treated with chemotherapy alone (87.7% vs. 64.1%;
p = .005).32 Conversely, in a phase II Italian trial on MACOP-B/
VACOP-B plus rituximab followed by mediastinal IFRT, the PFS
improvement was not significant compared with historical
controls.33
Overall, the combination of immunochemotherapy plus
IFRT results in a 3-year PFS rate of 80%85%.32-34 Recent
guidelines35 recommended chemotherapy with a CHOP/
MACOP-Blike regimen plus rituximab, followed by mediastinal IFRT as standard front-line treatment of PMLBCL.
However, the role of adjuvant mediastinal irradiation in
patients who achieve CMR with chemotherapy remains
unclear. Retrospective series suggest that the best outcomes
are seen when consolidation radiotherapy is given to the
mediastinum,26,27 particularly in the large proportion of
patients with a residual mediastinal mass at completion of
chemotherapy.36,37 Indeed, before rituximab introduction,
the use of consolidation radiotherapy for PMLBCL was a
historical standard of care, based on poor results after CHOP
chemotherapy alone compared with the excellent results in
series where almost all patients underwent irradiation.2,7
This combined modality strategy was also preferred by many
to maximize cures at the first attempt because of the dismal
outcomes for patients in whom recurrent disease develops.7
However, concerns about long-term effects of mediastinal
radiotherapy, including coronary heart disease, valvular
disorders, heart failure, and risk of second tumors (mainly
cancers of the breast, lung, and thyroid), justify studies
e369
aiming to investigate scenarios where radiotherapy may be

omitted.38-40 Because modern radiotherapy techniques may
minimize the morbidity to other tissues and/or reduce the
risk of long-term side effects,41-43 but do not completely
address the risks of late effects, safety remains a central
issue.44,45 For this reason, in a patient group dominated by
young women, some investigators have elected to exclude
consolidation radiotherapy. In a phase II study at the National Cancer Institute, the use of dose-adjusted REPOCH
combined with rituximab seemed to allow omission of radiotherapy without compromising cure rates (the event-free
survival rate was 93%, and the OS rate was 97%),46 and in the
BC Cancer Agency, the decision to spare radiotherapy in
patients with a negative 18F-FDG PET-CT at the completion of
R-CHOP, reduced the use of radiotherapy in patients from
80% to 38%, apparently with good outcomes.47 Unfortunately,
to date there are no published randomized trials in which this
question has been definitely answered.
FIGURE 1. IELSG-37 Study Design
THE IELSG-37 RANDOMIZED TRIAL

The ongoing IELSG-37 (NCT01599559) study was designed to
address the issue of the role of irradiation in the current
rituximab era offering randomization to standard radiotherapy versus no further treatment for patients who were
found on PET scan to be tumor free at the end of initial
rituximab-chemotherapy treatment. The study design is
summarized in Fig. 1.
This is the first study on a large international basis that
aimed to define in a prospective, controlled way the role of
radiotherapy in first-line treatment of these patients. This
very ambitious study, to which more than 500 patients will
be enrolled (more than half of them have been enrolled so
far), aims at personalizing treatment and, it is hoped, will
determine whether radiotherapy can be safely spared in
patients achieving CMR indicated on PET scans after induction immunochemotherapy. The use of PET in this trial is
based on the result achieved by the IELSG-26 study, which
will be described in the following section.
THE ROLE OF FUNCTIONAL IMAGING:

CONTRIBUTIONS FROM THE IELSG-26 STUDY
PET-CT is currently considered the best imaging tool for
staging and response assessment in 18F-FDGavid lymphomas,48 as defined by the recommendations of the Lugano
Classification for initial evaluation, staging, and response
assessment of Hodgkin lymphoma and non-Hodgkin lymphoma.49 PET-CT scans are increasingly used as prognostic
indicators in classic Hodgkin lymphoma50-53 and, to a minor
extent, in DLBCL.54-57 Although PMLBCL is an 18F-FDGavid
lymphoma,48,49,58 very few studies on the use of PET-CT have
focused specifically on PMLBCL.
Establishing whether residual masses, which are almost
invariably visible on CT scans at the completion of initial
chemotherapy, represent viable lymphoma or merely fibrotic scar tissue is a major problem in the clinical management of PMLBCL.59 Studies of functional imaging using
e370
This is a prospective, randomized, noninferiority phase III trial that requires the
random assignment of 376 patients. Patients are treated with standard
immunochemotherapy regimens currently in use for PMLBCL (e.g., R-CHOP, DAEPOCH-R, R- ACVBP, R-VACOP-B, or R-MACOP-B). Restaging PET-CT scans will be
performed at 56 weeks after the last immunochemotherapy administration. Central
review of PET-CT scans is mandatory before randomization. All patients with
a negative PET-CT scan, either with complete or partial radiologic regression of the
mediastinal mass, will be randomly assigned to receive consolidation IFRT (30 Gy) or
observation. Based on the results of the previous IELSG-26 study, a negative PET scan
is defined using the liver uptake as the cutoff point (Figs. 2 and 3).
Abbreviation: IFRT, involved-field radiotherapy.
Gallium scans24 or, more recently, 18F-FDG PET-CT scans60

suggest that it may be possible to distinguish residual mediastinal masses, which contain active lymphoma, from
those in which only sclerotic material remains. However,
they have not clarified whether radiotherapy could be
avoided solely on the basis of a negative PET-CT scan.59,61
In a retrospective study of 54 patients with PMLBCL who
were treated with the R-CHOP/ICE dose-dense regimen
without mediastinal radiotherapy, the Memorial Sloan
Kettering Cancer Center group reported 3-year OS and PFS
rates of 88% and 78%, respectively, in patients who had
negative PET-CT scans at the end of the chemotherapy
regimen. Furthermore, an abnormal result in an interim PETCT scan was still evident in 47% of patients and did not
predict a worse PFS rate.62
In a BC Cancer Agency retrospective study of 196 patients
with DLBCL treated with R-CHOP (25 with PMLBCL) with a
residual mass of greater than 2 cm, post-therapy PET-CT scan
was negative in 62% of patients who were then only observed, whereas 34% had a positive post-therapy PET-CT
scan and received radiotherapy (IFRT, 3040 Gy) to sites
noted as positive on PET-CT scan. No differences were
observed in terms of 3-year PFS (80% vs. 75%; p = .41)
between the patients with negative PET-CT versus positive
PET-CT scans.63
PET-CT Response After Immunochemotherapy

Currently, the introduction of rituximab has changed the
therapeutic scenario, whereas the general use of 18F-FDG
PET-CT to evaluate mediastinal residual disease after chemotherapy has led to the assumption that a lymphoma can
also be considered in complete remission in the presence
of a negative PET-CT scan showing residual lesion.64
Only scanty and mainly retrospective studies on the validity
of PET in response evaluation were specifically focused on
PMLBCL.65 The IELSG-26 study of PMLBCL represents a first,
and thus far unique, attempt to obtain data on this issue
in a prospective manner.66 This is the largest study on the
results of 18F-FDG PET-CT scanning in PMLBCL. The outcome
after induction treatment in this series, in which a majority of
patients received consolidation radiotherapy of treatment
with rituximab and anthracycline-containing chemotherapy,
is very good, with more than 90% of patients projected to
be alive and progression-free at 5 years despite a lower
proportion of patients classified as having CMR. Among 125
patients prospectively enrolled in this study, 115 were eligible
for central review of PET-CT scans at the completion of
standard immunochemotherapy, using the Deauville fivepoint scale (Fig. 2).67 Consolidation radiotherapy was given
to 102 patients. Fifty-four patients (47%) achieved CMR,
defined as a completely negative scan or with residual
18
F-FDG activity equal to or below the mediastinal blood
pool (MBP) uptake. In the remaining 61 patients (53%), the
residual uptake was higher than the MBP but below the
liver uptake in 27 (23%), slightly higher than the liver uptake
in 24 (21%), and markedly higher in 10 (9%). Complete
metabolic response after immunochemotherapy predicted
higher 5-year PFS (98% vs. 82%; p = .0044) and OS (100% vs.
91%; p = .0298). Patients with residual uptake higher than
the MBP but below liver uptake had equally good outcomes,
without any recurrence. Using the liver uptake as the cutoff
for PET positivity (boundary of score 34) discriminated
most effectively between high or low risk of failure, with
5-year PFS of 99% versus 68% (p , .0001) and 5-year OS of
100% versus 83% (p = .0003). Moreover, moving the cut
point for the definition of CMR from the MBP to the liver
uptake increased the specificity, and the positive predictive
value rose from 18% to 32%, without loss of sensitivity (Fig. 3).
The use of liver uptake as the cutoff for PET positivity resulted
in a clearer distinction between risk subgroups, both in terms
of PFS (p , .0001) and OS (p = .0003). The International
Prognostic Index (IPI) and age-adjusted IPI at diagnosis and
the chemotherapy regimen did not significantly correlate with
the PET response, whereas initial bulky disease (. 10 cm) was
significantly associated with a persistent postchemotherapy
residual uptake above the liver cutoff (p = .005). This correlation with tumor bulk was only of borderline significance
for the MBP cut point (p = .05).
This study is also the first attempt to validate the use of
PET-CT scanning in restaging an aggressive lymphoma at the
completion of chemotherapy using the Deauville criteria
originally developed for interim PET-CT evaluation during
FIGURE 2. Deauville Criteria for the Visual

Interpretation of PET/CT Scan Using a Five-Point Scale
(Deauville Score)
Interim and end-of-treatment PET/CT scans are scored according to uptake in

sites involved at baseline PET/CT by lymphoma as: (1) no uptake, (2) 18F-FDG
uptake equal to or below the mediastinum blood pool uptake, (3) 18F-FDG uptake
equal to or below the liver uptake, (4) 18F-FDG uptake moderately higher than the
liver uptake, or (5) 18F-FDG uptake markedly higher (i.e., more than two-fold) than the
uptake in the liver and/or new lesions.67 At the end of treatment, a score of 13 is
currently considered to indicate a complete metabolic response in most instances,48
whereas a score of 4 or 5 is considered to indicate lymphoma persistence or
recurrence. When new areas of uptake are considered unlikely to be caused by
lymphoma, they are usually reported as Score X.48
Abbreviations: SUVmax, maximum standardized uptake value;18FDG,
18-fluorodeoxyglucose.
the treatment of DLBCL and Hodgkin lymphoma.67 A cutoff

point for PET positivity below the uptake in the liver discriminates most effectively between groups of PMLBCL at
high or low risk of treatment failure.
The Problem of False-Positive PET-CT Scans After

Immunochemotherapy
Although the outcome of treatment was very good, after
immunochemotherapy a persistently positive PET-CT scan
was seen in more than half of the patients (53%, with
Deauville score 35), which contributed to a positive
predictive value substantially lower than that reported for
other types of DLBCL and Hodgkin lymphoma. With the use
of the liver cutoff point to define a CMR, the rate of patients
with PET-positive scans decreases (30% with score 4 or 5).
The results of IELSG-26 appear to be in line with those
observed in a series of patients treated with the infusional
doseadjusted R-EPOCH (DA-R-EPOCH) regimen, which
showed 50% PET positivity, defined by 18F-FDG uptake
greater than the MBP at the completion of chemotherapy
but only 3 of 18 progressions without the use of consolidation radiotherapy in this group.46 Because all of the cases
that recurred had a standardized uptake value (SUV) of at
least 5, the authors suggested that radiotherapy might
reasonably be omitted for nearly all patients treated with
this regimen.46 The IELSG-26 study likely carries a similarly
high false-positive rate, although PMLBCL is clearly radiosensitive, so it is also possible that the almost universal
use of radiotherapy in this study may have eliminated
residual lymphoma in some instances. However, the IELSG26 study design did not allow us to draw any conclusions
on the benefit of radiotherapy.
e371
FIGURE 3. IELSG-26 Study Results Using the Five-Point

Scale (Deauville Score) for the Visual Interpretation of
PET-CT Scans Obtained at 34 Weeks After
Immunochemotherapy
(A) PET-CT interpretation in a blind central review performed on 115 evaluable

patients and change in the positive predictive value induced by a shift of the
cutoff point for the definition of CMR from Deauville score 2 (MBP uptake) to
score 3 (liver uptake). (B) Kaplan-Meier estimates of OS in PMLBCL according to
positive PET using the MBP uptake (Deauville score 35) or the liver uptake as
a cutoff point (Deauville score 45).
Abbreviations: MBP, mediastinal blood pool; NPV, negative predictive value; PPV,
positive predictive value; OS, overall survival; CMR, complete metabolic response.
Modified from Martelli et al 201466.
PET-CT Response After Consolidation Radiotherapy

The evaluation of postirradiation PET-CT scans in the IELSG26 study has thus far been reported only as a meeting
abstract.68 In this study, 88 of 125 patients were eligible for
central review of PET-CT scans at 8 weeks after the completion of radiotherapy. The CMR (using the liver cutoff
point) rate increased from 74% after immunochemotherapy
to 89% after irradiation. At a median follow-up of 60 months,
no patients with CMR after radiotherapy have relapsed.
Notably, in this study the few patients with a Deauville score
of 4 also had an excellent outcome, suggesting that they do
not necessarily require additional therapy because the residual 18F-FDG uptake may be caused by an inflammatory
reaction. In this context, it seems worth mentioning that the
18
F-FDG uptake in untreated PMLBCL is likely a result of the
lymphoma cells that are the main component of the tumor,
even though a contributing microenvironment uptake cannot
be excluded58 given its relevant role in PMLBCL.5
In post-treatment evaluation of PMLBCL, because of their
thymic derivation and anatomic localization and young age
e372
of most patients, thymic rebound should also be considered

as a possible explanation of a positive PET scan.
Utility of Baseline 18F-FDG PET-CT Functional

Parameters in Defining Prognosis
Because a successful primary treatment is crucial for
PMLBCL, the early identification of high-risk patients at risk
for relapse is critical to allow for alternative therapeutic
strategy. Valuable prognosis factors are therefore warranted
for tailoring therapy and possibly reducing treatment toxicity, as IPI and age-adjusted IPI have proved not useful for
this purpose. Other proposed clinical predictors of survival
have not been validated prospectively, and biomarkers
useful for risk stratification are not yet available.58
In this sense, the IELSG-26 study also aimed at exploring
the potential of quantitative 18F-FDG PET-CT as a biomarker
in PMLBCL by investigating whether the main quantitative
baseline PET-CTderived metabolic parameters can predict
prognosis and aimed to compare them with the commonly
used clinical indices.69
The metabolic activity defined by maximum SUV (SUVmax),
the metabolic tumor volume, and the total lesion glycolysis
(TLG), an index of metabolic tumor burden, were measured in
103 patients on baseline 18F-FDG PET-CT using a practical and
reproducible method to segment the mediastinal mass, based
on the 25% thresholding of the tumor SUVmax.69 This method
was developed considering the specific presentation features
of PMLBCL, where the technical difficulties of quantitative
PET-CT are reduced because the tumor burden is mainly
limited to a unique bulky mass.58
All patients received combination immunochemotherapy
with doxorubicin and rituximab-based regimens; 93 received
consolidation radiotherapy. Cutoff points were determined
using the receiver operating characteristic curve.69
At a median follow-up of 36 months, PFS and OS rates were
87% and 94%, respectively. In univariate analysis, elevated
metabolic tumor volume and TLG were significantly associated with worse PFS and OS. Only TLG retained statistical
significance for both OS (p = .001) and PFS (p , .001) in
multivariate analysis. At 5 years, OS was 100% for patients
with low TLG versus 80% for those with high TLG (p = .0001),
whereas PFS was 99% versus 64%, respectively (p , .0001).69
This study represents the largest prospective series of
PMLBCL investigated with central review of PET-CT scans, in
which the quantitative functional parameters were analyzed. The PET-CT imaging was obtained by following a
strictly defined technical procedure, although detailed crosscalibration for equipment at the different centers was not
possible. Total lesion glycolysis on baseline PET-CT appeared
to be a powerful predictor of PMLBCL outcomes and warrants further validation as a biomarker. Indeed, this easily
and early accessible parameter can help identify the few
patients who will eventually experience relapse and may be
useful for designing trials aimed at improving their poor
prognosis (e.g., using TLG to distinguish patients being
considered for elective front-line consolidation with myeloablative therapies). However, despite the very high sensitivity
(92%) of TLG with a 98% negative predictive value for PFS, the
positive predictive value was only 36%. Hence, a better selection of high-risk patients would be useful to choose those to
submit to intensive treatment.58
Integrative PET: A Novel Promising Tool

Integrative 18F-FDG PET in lymphoma is a new concept that
can be defined as a method of prognostic imaging, where
imaging data obtained with different techniques are combined together or with data from other fields (clinical, biologic, molecular) to improve the risk stratification of
patients.70 In classic Hodgkin lymphoma, it has been suggested that baseline metabolic tumor volume may be
combined with the interim PET-CT score to identify patients
at different levels of risk.71 Preliminary data from the IELSG26 series of PMLBCL showed that the combination of
baseline TLG with end-treatment PET-CT more accurately
identified patients at risk, improving the positive predictive
value without a detrimental effect on negative predictive

value.72 However, because early stratification is preferred,
the integration of other prognostic indicators (such as interim PET-CT data or baseline molecular and clinical features) with baseline quantitative 18F-FDG PET-CT should be
addressed by future studies to provide an earlier and clinically meaningful selection of patients with high-risk PMLBCL
suitable for risk-adapted therapeutic strategies.58
ACKNOWLEDGMENT
We thank Rita Gianascio Gianocca and Sarah Ortelli Giannakis
for the secretarial and editorial assistance; the chairpersons
(Prof. Peter Johnson and Prof. Maurizio Martelli), data
managers and investigators of the IELSG-26 and IELSG-37
studies on PET-CT use in PMLBCL; Dr. Francesco Bertoni for his
critical suggestions; and Swiss Cancer Research for the support to the IELSG activities.
References
1. Gaulard P, Harris NL, Pileri SA, et al. Primary mediastinal (thymic) large
B-cell lymphoma. In Swerdlow S, Campo E, Harris NL, et al (eds).
WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. Lyon: IARC, 2008;250-253.
2. Johnson PW, Davies AJ. Primary mediastinal B-cell lymphoma. Hematology (Am Soc Hematol Educ Program). 2008;2008:349-358.
3. Savage KJ, Monti S, Kutok JL, et al. The molecular signature of mediastinal large B-cell lymphoma differs from that of other diffuse large Bcell lymphomas and shares features with classical Hodgkin lymphoma.
Blood. 2003;102:3871-3879.
4. Rosenwald A, Wright G, Leroy K, et al. Molecular diagnosis of primary
mediastinal B cell lymphoma identifies a clinically favorable subgroup of
diffuse large B cell lymphoma related to Hodgkin lymphoma. J Exp Med.
2003;198:851-862.
5. Steidl C, Gascoyne RD. The molecular pathogenesis of primary mediastinal large B-cell lymphoma. Blood. 2011;118:2659-2669.
6. van Besien K, Kelta M, Bahaguna P. Primary mediastinal B-cell lymphoma:
a review of pathology and management. J Clin Oncol. 2001;19:1855-1864.
7. Martelli M, Di Rocco A, Russo E, et al. Primary mediastinal lymphoma:
diagnosis and treatment options. Expert Rev Hematol. 2015;8:173-186.
8. Pileri SA, Gaidano G, Zinzani PL, et al. Primary mediastinal B-cell
lymphoma: high frequency of BCL-6 mutations and consistent expression of the transcription factors OCT-2, BOB.1, and PU.1 in the
absence of immunoglobulins. Am J Pathol. 2003;162:243-253.
9. Pileri SA, Zinzani PL, Gaidano G, et al; International Extranodal Lymphoma Study Group. Pathobiology of primary mediastinal B-cell lymphoma. Leuk Lymphoma. 2003;44(suppl 3):S21-S26.
10. Feuerhake F, Kutok JL, Monti S, et al. NFkappaB activity, function, and
target-gene signatures in primary mediastinal large B-cell lymphoma and
diffuse large B-cell lymphoma subtypes. Blood. 2005;106:1392-1399.
11. Ritz O, Guiter C, Castellano F, et al. Recurrent mutations of the STAT6
DNA binding domain in primary mediastinal B-cell lymphoma. Blood.
2009;114:1236-1242.
12. Hao Y, Chapuy B, Monti S, et al. Selective JAK2 inhibition specifically
decreases Hodgkin lymphoma and mediastinal large B-cell lymphoma
growth in vitro and in vivo. Clin Cancer Res. 2014;20:2674-2683.
13. Weniger MA, Gesk S, Ehrlich S, et al. Gains of REL in primary mediastinal
B-cell lymphoma coincide with nuclear accumulation of REL protein.
Genes Chromosomes Cancer. 2007;46:406-415.
~o-Joos MI, Ziegler S, et al. Primary mediastinal (thymic) B-cell
14. Joos S, Otan
lymphoma is characterized by gains of chromosomal material including
9p and amplification of the REL gene. Blood. 1996;87:1571-1578.
15. Melzner I, Bucur AJ, Bruderlein S, et al. Biallelic mutation of SOCS-1
impairs JAK2 degradation and sustains phospho-JAK2 action in the
MedB-1 mediastinal lymphoma line. Blood. 2005;105:2535-2542.
16. Mareschal S, Dubois S, Viailly PJ, et al. Whole exome sequencing of relapsed/
refractory patients expands the repertoire of somatic mutations in diffuse
large B-cell lymphoma. Genes Chromosomes Cancer. 2016;55(3):251-267.
17. Jardin F, Pujals A, Pelletier I, et al. Whole exome sequencing of refractory
aggressive B-cell lymphomas identified recurrent mutations of the
exportin 1 gene (XPO1) in primary mediastinal B-cell lymphoma subtype, a
LYSA study. Hematol Oncol. 2015;33:124-125 (suppl; abstr 048).
18. Savage KJ, Al-Rajhi N, Voss N, et al. Favorable outcome of primary
mediastinal large B-cell lymphoma in a single institution: the British
Columbia experience. Ann Oncol. 2006;17:123-130.
19. Rieger M, Osterborg A, Pettengell R, et al; MabThera International Trial
(MInT) Group. Primary mediastinal B-cell lymphoma treated with
CHOP-like chemotherapy with or without rituximab: results of the
Mabthera International Trial Group study. Ann Oncol. 2011;22:664-670.
20. Kuruvilla J, Pintilie M, Tsang R, et al. Salvage chemotherapy and autologous stem cell transplantation are inferior for relapsed or refractory
primary mediastinal large B-cell lymphoma compared with diffuse large
B-cell lymphoma. Leuk Lymphoma. 2008;49:1329-1336.
21. Cazals-Hatem D, Lepage E, Brice P, et al. Primary mediastinal large B-cell
lymphoma. A clinicopathologic study of 141 cases compared with 916
nonmediastinal large B-cell lymphomas, a GELA (Groupe dEtude des
Lymphomes de lAdulte) study. Am J Surg Pathol. 1996;20:877-888.
22. Bieri S, Roggero E, Zucca E, et al. Primary mediastinal large B-cell
lymphoma: the need for prospective controlled clinical trials. Leuk
Lymphoma. 1999;35:537-544.
23. Lazzarino M, Orlandi E, Paulli M, et al. Primary mediastinal B-cell
lymphoma with sclerosis: an aggressive tumor with distinctive clinical and pathologic features. J Clin Oncol. 1993;11:2306-2313.
e373
24. Zinzani PL, Martelli M, Magagnoli M, et al. Treatment and clinical

management of primary mediastinal large B-cell lymphoma with
sclerosis: MACOP-B regimen and mediastinal radiotherapy monitored
by (67)Gallium scan in 50 patients. Blood. 1999;94:3289-3293.
25. Zinzani PL, Martelli M, Bendandi M, et al. Primary mediastinal large Bcell lymphoma with sclerosis: a clinical study of 89 patients treated with
MACOP-B chemotherapy and radiation therapy. Haematologica. 2001;
86:187-191.
26. Zinzani PL, Martelli M, Bertini M, et al; International Extranodal Lymphoma Study Group (IELSG). Induction chemotherapy strategies for
primary mediastinal large B-cell lymphoma with sclerosis: a retrospective multinational study on 426 previously untreated patients.
Haematologica. 2002;87:1258-1264.
27. Todeschini G, Secchi S, Morra E, et al. Primary mediastinal large B-cell
lymphoma (PMLBCL): long-term results from a retrospective multicentre Italian experience in 138 patients treated with CHOP or MACOPB/VACOP-B. Br J Cancer. 2004;90:372-376.
28. Hamlin PA, Portlock CS, Straus DJ, et al. Primary mediastinal large B-cell
lymphoma: optimal therapy and prognostic factor analysis in 141
consecutive patients treated at Memorial Sloan Kettering from 1980 to
1999. Br J Haematol. 2005;130:691-699.
29. Dunleavy K, Pittaluga S, Janik J, et al. Primary mediastinal large B-cell
lymphoma (PMBL) outcome is significantly improved by the addition of
rituximab to dose adjusted (DA)-EPOCH and overcomes the need for
radiation. Blood. 2005;106:929.
30. Pfreundschuh M, Schubert J, Ziepert M, et al; German High-Grade NonHodgkin Lymphoma Study Group (DSHNHL). Six versus eight cycles of
bi-weekly CHOP-14 with or without rituximab in elderly patients with
aggressive CD20+ B-cell lymphomas: a randomised controlled trial
(RICOVER-60). Lancet Oncol. 2008;9:105-116.
31. Coiffier B, Thieblemont C, Van Den Neste E, et al. Long-term outcome of
patients in the LNH-98.5 trial, the first randomized study comparing
rituximab-CHOP to standard CHOP chemotherapy in DLBCL patients:
a study by the Groupe dEtudes des Lymphomes de lAdulte. Blood.
2010;116:2040-2045.
32. Trneny M, Osterborg A, Pettengel R, et al. The addition of rituximab
eliminates the negative prognostic impact of PMLBCL compared to
DLBCL in young patients with CD20-positive aggressive lymphomas
receiving a CHOP-like chemotherapy: results of a subgroup analysis of
the Mabthera International Trial Group (MInT) study. Blood. 2008;112:
839.
33. Zinzani PL, Stefoni V, Finolezzi E, et al. Rituximab combined with
MACOP-B or VACOP-B and radiation therapy in primary mediastinal
large B-cell lymphoma: a retrospective study. Clin Lymphoma Myeloma.
2009;9:381-385.
34. Vassilakopoulos TP, Angelopoulou MK, Galani Z, et al. Rituximab-CHOP
(R-CHOP) and radiotherapy (RT) for primary mediastinal large B-cell
lymphoma (PMLBCL): an update. Haematologica. 2006;91:260 (abstr
0704).
35. Zinzani PL, Martelli M, Poletti V, et al; Italian Society of Hematology;
Italian Society of Experimental Hematology; Italian Group for Bone
Marrow Transplantation. Practice guidelines for the management of
extranodal non-Hodgkins lymphomas of adult non-immunodeficient
patients. Part I: primary lung and mediastinal lymphomas. A project of
the Italian Society of Hematology, the Italian Society of Experimental
Hematology and the Italian Group for Bone Marrow Transplantation.
Haematologica. 2008;93:1364-1371.
36. Mazzarotto R, Boso C, Vianello F, et al. Primary mediastinal large Bcell lymphoma: results of intensive chemotherapy regimens
(MACOP-B/VACOP-B) plus involved field radiotherapy on 53 patients. A single institution experience. Int J Radiat Oncol Biol Phys.
2007;68:823-829.
e374
37. De Sanctis V, Finolezzi E, Osti MF, et al. MACOP-B and involved-field

radiotherapy is an effective and safe therapy for primary mediastinal
large B cell lymphoma. Int J Radiat Oncol Biol Phys. 2008;72:1154-1160.
38. Hancock SL, Hoppe RT. Long-term complications of treatment and
causes of mortality after Hodgkins disease. Semin Radiat Oncol. 1996;6:
225-242.
39. van Leeuwen FE, Klokman WJ, Veer MB, et al. Long-term risk of second
malignancy in survivors of Hodgkins disease treated during adolescence or young adulthood. J Clin Oncol. 2000;18:487-497.
40. Henderson TO, Amsterdam A, Bhatia S, et al. Systematic review: surveillance for breast cancer in women treated with chest radiation for
childhood, adolescent, or young adult cancer. Ann Intern Med. 2010;
152:444-455.
41. Yoon M, Ahn SH, Kim J, et al. Radiation-induced cancers from modern
radiotherapy techniques: intensity-modulated radiotherapy versus
proton therapy. Int J Radiat Oncol Biol Phys. 2010;77:1477-1485.
42. Bucci MK, Bevan A, Roach M III. Advances in radiation therapy: conventional to 3D, to IMRT, to 4D, and beyond. CA Cancer J Clin. 2005;55:
117-134.
43. Tsang RW, Gospodarowicz MK. Radiation therapy for localized lowgrade non-Hodgkins lymphomas. Hematol Oncol. 2005;23:10-17.
44. van Nimwegen FA, Schaapveld M, Cutter DJ, et al. Radiation Doseresponse relationship for risk of coronary heart disease in survivors of
Hodgkin lymphoma. J Clin Oncol. 2016;34:235-243.
45. Nieder C, Schill S, Kneschaurek P, et al. Comparison of three different
mediastinal radiotherapy techniques in female patients: Impact on
heart sparing and dose to the breasts. Radiother Oncol. 2007;82:
301-307.
46. Dunleavy K, Pittaluga S, Maeda LS, et al. Dose-adjusted EPOCHrituximab therapy in primary mediastinal B-cell lymphoma. N Engl J
Med. 2013;368:1408-1416.
47. Savage KJ, Yenson PR, Shenkier T, et al. The outcome of primary mediastinal large B-cell lymphoma (PMLBCL) in the R-CHOP treatment era.
Blood. 2012;120:303.
48. Barrington SF, Mikhaeel NG, Kostakoglu L, et al. Role of imaging in the
staging and response assessment of lymphoma: consensus of the International Conference on Malignant Lymphomas Imaging Working
Group. J Clin Oncol. 2014;32:3048-3058.
49. Cheson BD, Fisher RI, Barrington SF, et al; Alliance, Australasian Leukaemia and Lymphoma Group; Eastern Cooperative Oncology Group;
European Mantle Cell Lymphoma Consortium; Italian Lymphoma
Foundation; European Organisation for Research; Treatment of Cancer/
~ol de Medula Osea;
Dutch Hemato-Oncology Group; Grupo Espan

German High-Grade Lymphoma Study Group; German Hodgkins Study
Group; Japanese Lymphorra Study Group; Lymphoma Study Association; NCIC Clinical Trials Group; Nordic Lymphoma Study Group;
Southwest Oncology Group; United Kingdom National Cancer Research
Institute. Recommendations for initial evaluation, staging, and response
assessment of Hodgkin and non-Hodgkin lymphoma: the Lugano classification. J Clin Oncol. 2014;32:3059-3068.
50. Adams HJ, Nievelstein RA, Kwee TC. Prognostic value of interim FDG-PET
in Hodgkin lymphoma: systematic review and meta-analysis. Br J
Haematol. 2015;170:356-366.
51. Sickinger MT, von Tresckow B, Kobe C, et al. Positron emission
tomography-adapted therapy for first-line treatment in individuals
with Hodgkin lymphoma. Cochrane Database Syst Rev. 2015;1:
CD010533.
52. Gallamini A, Hutchings M, Rigacci L, et al. Early interim 2-[18F]fluoro-2deoxy-D-glucose positron emission tomography is prognostically superior to international prognostic score in advanced-stage Hodgkins
lymphoma: a report from a joint Italian-Danish study. J Clin Oncol. 2007;
25:3746-3752.
53. Gallamini A, Barrington SF, Biggi A, et al. The predictive role of interim
positron emission tomography for Hodgkin lymphoma treatment
outcome is confirmed using the interpretation criteria of the Deauville
five-point scale. Haematologica. 2014;99:1107-1113.
54. Mamot C, Klingbiel D, Hitz F, et al. Final Results of a prospective
evaluation of the predictive value of interim positron emission tomography in patients with diffuse large B-cell lymphoma treated with
R-CHOP-14 (SAKK 38/07). J Clin Oncol. 2015;33:2523-2529.
55. Mikhaeel NG, Timothy AR, ODoherty MJ, et al. 18-FDG-PET as a
prognostic indicator in the treatment of aggressive Non-Hodgkins
Lymphoma-comparison with CT. Leuk Lymphoma. 2000;39:543-553.
56. Spaepen K, Stroobants S, Dupont P, et al. Prognostic value of positron
emission tomography (PET) with fluorine-18 fluorodeoxyglucose ([18F]FDG)
after first-line chemotherapy in non-Hodgkins lymphoma: is [18F]FDG-PET a
valid alternative to conventional diagnostic methods? J Clin Oncol. 2001;19:
414-419.
57. Zinzani PL, Fanti S, Battista G, et al. Predictive role of positron emission
tomography (PET) in the outcome of lymphoma patients. Br J Cancer.
2004;91:850-854.
58. Meignan M. Quantitative FDG-PET: a new biomarker in PMBCL. Blood.
2015;126:924-926.
59. Johnson PWM. IV. Masses in the mediastinum: primary mediastinal
lymphoma and intermediate types. Hematol Oncol. 2015;33(suppl 1):
29-32.
60. Zinzani PL, Tani M, Trisolini R, et al. Histological verification of positive
positron emission tomography findings in the follow-up of patients with
mediastinal lymphoma. Haematologica. 2007;92:771-777.
61. Kahn ST, Flowers C, Lechowicz MJ, et al. Value of PET restaging after
chemotherapy for non-Hodgkins lymphoma: implications for consolidation radiotherapy. Int J Radiat Oncol Biol Phys. 2006;66:961-965.
62. Moskowitz C, Hamlin PA Jr, Maragulia J, et al. Sequential dose-dense
RCHOP followed by ICE consolidation (MSKCC protocol 01-142) without
radiotherapy for patients with primary mediastinal large B cell lymphoma. Blood. 2010;116:420.
63. Sehn LH, Hoskins P, Klasa R, et al. FDG-PET scan guided consolidative
radiation therapy optimizes outcome In patients with advanced-stage
diffuse large B-cell lymphoma (DLBCL) with residual abnormalities on CT
scan following R-CHOP. Blood. 2010;-116:854.
64. Cheson BD, Pfistner B, Juweid ME, et al; International Harmonization
Project on Lymphoma. Revised response criteria for malignant lymphoma. J Clin Oncol. 2007;25:579-586.
65. Zinzani PL, Broccoli A, Casadei B, et al. The role of rituximab and positron
emission tomography in the treatment of primary mediastinal large Bcell lymphoma: experience on 74 patients. Hematol Oncol. 2015; 33:
145-150.
66. Martelli M, Ceriani L, Zucca E, et al. [18F]fluorodeoxyglucose positron
emission tomography predicts survival after chemoimmunotherapy for
primary mediastinal large B-cell lymphoma: results of the International
Extranodal Lymphoma Study Group IELSG-26 Study. J Clin Oncol. 2014;
32:1769-1775.
67. Meignan M, Gallamini A, Meignan M, et al. Report on the First International Workshop on Interim-PET-Scan in Lymphoma. Leuk Lymphoma. 2009;50:1257-1260.
68. Ceriani L, Martelli M, Zinzani PL, et al. Use of the Lugano classification
criteria for PET/CT assessment of primary mediastinal B-cell lymphoma
after immunochemotherapy and irradiation in the IELSG-26 study.
Hematol Oncol. 2015;33:143 (abstr 081).
69. Ceriani L, Martelli M, Zinzani PL, et al. Utility of baseline 18FDG-PET/CT
functional parameters in defining prognosis of primary mediastinal
(thymic) large B-cell lymphoma. Blood. 2015;126:950-956.
70. Meignan M, Cottereau AS. Integrative PET: a new concept for outcome
prediction in lymphoma. Clin Transl Imaging. 2015;3:343-344.
71. Meignan M, Itti E, Gallamini A, et al. FDG PET/CT imaging as a biomarker
in lymphoma. Eur J Nucl Med Mol Imaging. 2015;42:623-633.
72. Zucca E, Martelli M, Zinzani PL, et al. Prognostic models for primary
mediastinal B-cell lymphoma derived from 18-FDG PET/CT quantitative
parameters in the IELSG-26 study. Haematologica. 2014;99:524-525
(suppl; abstr S1346).
e375
LYNCH AND ADVANI
Risk-Adapted Treatment of Advanced Hodgkin Lymphoma

With PET-CT
Ryan C. Lynch, MD, and Ranjana H. Advani, MD
OVERVIEW
Although patients with advanced-stage classic Hodgkin lymphoma have excellent outcomes with contemporary therapy,
the outcomes of patients with refractory disease is suboptimal. Identification of these high-risk patients at diagnosis is
challenging as the differences in outcomes using clinical criteria are less marked using current modern therapy. Data suggest
that an interim PET-CT may be a powerful tool in risk-stratifying patients. Retrospective studies show that a negative interim
PET-CT after two to four cycles of ABVD (doxorubicin, bleomycin, vinblastine, dacarbazine) is predictive of favorable outcome
independent of IPS score. Currently, there are several ongoing trials that aim to determine whether early-response assessment can be used to select patients who might benefit from modifications of subsequent therapy, either by intensifying
or abbreviating regimens and/or omitting radiotherapy with promising early results. Longer follow-up is required to assess
whether this strategy impacts overall survival (OS). Herein, we review the results of recent trials using interim PET-CT-based
adaptive design in the treatment of advanced HL.
atients with advanced-stage classic Hodgkin lymphoma

(defined in North America as clinical stage IIIIV disease
and early-stage disease with bulk or B symptoms) have
excellent outcomes with contemporary therapy with an OS
of greater than 80%.1,2
In the National Comprehensive Cancer Network guidelines, therapy choices for advanced-stage disease include
ABVD, Stanford V regimen (a combined modality approach
including mechlorethamine, doxorubicin, vincristine, vinblastine, bleomycin, etoposide, and prednisone), and the
German Hodgkin Study Group (GHSG) regimen escalated
BEACOPP ( escBEACOPP; bleomycin, etoposide, doxorubicin,
cyclophosphamide, vincristine, procarbazine, and prednisone) followed by radiotherapy given to PET-positive areas
greater than 2.5 cm after completion of chemotherapy.3 In
North America, ABVD is the most commonly used regimen
based on a balance of efficacy and toxicity, and escBEACOPP
is used more commonly in Europe. The latter has not been
widely accepted in North America largely because of increased
toxicity as well as an increased risk of sterility and secondary
cancers.1,4,5 Several randomized clinical trials have compared
some version of escBEACOPP to ABVD and have shown an
improved progression-free survival (PFS) without any impact
on OS.6-8 A recent long-term follow-up from the HD2000 trial
failed to confirm the PFS benefit with escBEACOPP mainly
because of higher mortality rates resulting from second
malignancies observed after treatment with escBEACOPP.9
Therefore, there is an ongoing debate regarding how to best

achieve a balance between aggressively curing the disease
with front-line therapy versus a gentler approach that may be
slightly less efficacious in the short term but with no long-term
OS difference. Attempts to identify patients who are defined
as high risk and thus might potentially benefit from a more
aggressive initial approach is also an evolving concept.
A commonly used index is the International Prognostic
Score (IPS), which has been used in some trials to stratify
patients (score $ 3 identifies a high-risk group10). However,
as the benefit of escBEACOPP has been observed across all
IPS risk groups, its utility in patient selection is limited.11
Additionally, the range of differences in outcomes between
groups in the contemporary era has also narrowed, and
other methods for patient selection are required.12,13
18-Fluorodeoxyglucose-(FDG)-PET/CT (or PET-CT) is an
important tool in the management of Hodgkin lymphoma
and has a well-established role in staging as well as response
assessment at the completion of therapy.14,15 Several studies
also suggest that a negative interim PET-CT after two to four
cycles of ABVD is predictive of favorable outcomes independent of the IPS score.16-20 The initial studies used
variable criteria to define a positive scan with a sensitivity of
67% to 100% and a specificity of 95% to 100% because of
inter-reader variability.14,21,22 Efforts to standardize response
criteria led to the development of the numeric five-point scale
(5PS), also known as the Deauville criteria. In this scale, FDG
From the Division of Oncology, Department of Medicine, Stanford University School of Medicine, Stanford, CA.
Corresponding author: Ranjana H. Advani, MD, Stanford University Medical Center, 875 Blake Wilbur Dr., Suite CC-2338, Stanford, CA 94305-5821; email: radvani@stanford.edu.
e376
RISK-ADAPTED TREATMENT OF ADVANCED HL
uptake at sites of disease is compared with uptake in the liver

and the mediastinal blood pool. The specific criteria consist
of a score of the following: (1) no uptake; (2) uptake higher
than or equal to mediastinal blood pool; (3) uptake higher
than mediastinal blood pool, but uptake is lower than or equal
to liver; (4) uptake moderately higher than liver; (5) uptake
markedly higher than liver and/or new lesions; (X) new areas
of uptake unlikely to be related to lymphoma.23 A score of 13 is
considered to represent a complete metabolic response in
advanced Hodgkin lymphoma. The prognostic utility value of this
score was confirmed in an international study that showed a
3-year failure-free survival (FFS) of 28% and 95% for patients
with early PET-positive versus PET-negative disease respectively
(p , .0001).24 Additionally, the interobserver agreement was
also high between six independent reviewers. The reproducibility
has resulted in the Deauville criteria being endorsed as the
current recommended method for response assessment.25,26
This has resulted in numerous trials in Hodgkin lymphoma
based on interim PET-CTadapted strategies. These studies aim
to determine whether early-response assessment can be used
to select patients who might benefit from modifications of
subsequent therapy, either by intensifying or abbreviating
regimens and/or omitting radiotherapy. Herein, we review the
results of recent trials using interim PET-CTbased adaptive
design in the treatment of advanced Hodgkin lymphoma.
STUDIES EVALUATING THERAPY ESCALATION IN

PATIENTS WITH A POSITIVE INTERIM PET-CT
SCAN
Results of an early interim PET-CT is now a widely accepted
prognostic tool for patients with advanced Hodgkin lymphoma
KEY POINTS
Retrospective studies in advanced Hodgkin lymphoma

suggest that a negative interim PET-CT (variably defined)
after two to four cycles of ABVD is predictive of
a favorable outcome.
Several prospective trials aim to determine whether
early response assessment after two cycles of ABVD
using the five-point scale can be used to select patients
who might benefit from modifications of subsequent
therapy.
Results of several trials suggest that an interim
PETadapted approach of escalating therapy to a more
aggressive regimen is feasible and possibly associated
with better progression-free survival in patients with
PET-2positive disease after ABVD compared with
historical controls.
A major consideration for the escalation strategies is the
lack of a control arm in which ABVD therapy is continued
regardless of interim PET-CT results.
Interim results of the RATHL trial suggest that for
patients with a negative PET-CT after two cycles of
ABVD, bleomycin can be safely omitted from
subsequent cycles.
treated with ABVD. Therefore, an adaptive treatment strategy

based on interim PET-CT results may help distinguish patients
with high-risk disease who could potentially benefit from a
change in therapy to an escalated regimen despite the increased toxicity. This was evaluated in a retrospective analysis
of prospectively collected data in a study in which all patients
with PET-2positive disease after two cycles of ABVD received
four escBEACOPP plus four standard BEACOPP cycles, whereas
patients with PET-2negative disease continued on ABVD (total
of six cycles) followed by consolidative radiotherapy.27 At a
median follow-up of 34 months and a central radiology review
of PET-2 data using the Deauville criteria, the FFS in PET2negative versus PET-2positive group was 95% versus
62%, respectively (p , .001). The outcomes of patients with
PET-2positive disease (approximately 20% of patients) was
better than historical controls using ABVD.20
The efficacy of intensifying therapy to escBEACOPP for
patients with PET-2positive disease has also been tested in
several prospectively designed trials in advanced Hodgkin
lymphoma described below and summarized in Table 1.
Although most of the interim PET-positive arms do not
have a control group without escalated therapy, the statistical design is essentially based on the historical PFS data
described above.
The U.S. intergroup phase II trial S0816 led by the
Southwest Oncology Group (SWOG) reported preliminary
results in 357 patients with a median follow-up 28 months in
patients with advanced Hodgkin lymphoma. Patients received two cycles of ABVD followed by a PET-CT with central
review.28 Those with a PET-2negative scan (82% of patients) defined as Deauville 13 continued with four more
cycles of ABVD, whereas patients with PET-2positive disease (Deauville 45) had a change in therapy to six cycles
of escBEACOPP. An interim analysis presented in 2013
reported a 1-year PFS for the PET-2positive group of 72%
(95% CI, 55%84%) and 85% in the PET-2negative group
(95% CI, 79%90%).
The U.K. RATHL trial included 1,214 patients and tested
a similar concept as the U.S. intergroup study. Patients with
advanced-stage Hodgkin lymphoma (stages IIBIV, or IIA
with bulk or three or more involved sites) received two
cycles of ABVD followed by a PET-CT. Patients with PET2negative disease (defined as Deauville 13) based on a
central review were randomly assigned to either ABVD or
AVD for four additional cycles.29,30 Patients with PET2positive disease (Deauville 45) received intensified
therapy with four cycles of BEACOPP-14 or three cycles of
escBEACOPP followed by another PET-CT. Patients whose
disease remained PET-3positive (performed after four
additional cycles of BEACOPP-14 or three of escBEACOPP)
were taken off study and given salvage treatment. Patients
with PET-3negative disease received two additional cycles
of BEACOPP-14 or one more cycle of escBEACOPP. Patients
who had a positive PET-2 scan (16% of all patients) had a
3-year PFS and OS of 68% and 86%, respectively, with no
difference in the nonrandomized comparison between
escBEACOPP and BEACOPP-14.
e377
LYNCH AND ADVANI
TABLE 1. Comparison of PET-Adapted Prospective Clinical Trials in Advanced Hodgkin Lymphoma
Study
Trial Design
Median
Follow-up
PET-2 Results
Clinical Stage
16.1% positive
IIBIV, IIA with 32 months

bulk or $ 3
involved
sites
Results
PFS/EFS/FFS
OS
Trials With Treatment Escalated If PET-2 Positive

RATHL
Two cycles ABVD interim PET-CT
(Johnson
et al,29 2015,
1,214
patients)
PET2+: six cycles BEACOPP-14 or four
cycles escBEACOPP (taken off study
if PET+ after four cycles BEACOPP-14
or three cycles escBEACOPP)
AHL 2011 LYSA Standard arm
Study
(Casanovas
et al,41 2015,
823 patients)
12.4% positive
IIBIV
3-year PFS
3-year OS
PET2+: 68%
PET2+: 86%
16.3 months 2-year PFS
Six cycles escBEACOPP (regardless of

result of interim PET-CT)
Standard arm: 91.6%
Experimental arm
Experimental arm:
88.3%
Two cycles escBEACOPP interim

PET-CT
Overall study cohort

PET2+: 72.9%
PET2+: four cycles escBEACOPP
Overall study cohort

PET22: 92.8%
(p , .0001)*
PET22: four cycles ABVD
Standard PET22:
75.1%
Not reported
Experimental
PET22: 70.8%
Israeli H2 (Dann IPS $ 3
et al,35 2014,
183 patients)
14.2% positive
IB, IIB, IIIIV
36 months
Two cycles escBEACOPP, interim

PET-CT
3-year PFS (all advanced HL regardless of IPS)
Not reported
PET2+: 75%
PET2+: four cycles escBEACOPP + RT to

bulky mediastinal disease at EOT
IPS , 3
PET2+: four cycles escBEACOPP + RT to
bulky mediastinal disease at EOT
Israeli Group
(Kedmi
PET-CT
et al,39 2015,
69 patients;
24 patients
were treated
on the same
protocol
off-study)
24.6% positive
IIBIV (IPS $ 3) 5.6 years
5-year PFS
5-year OS
PET2+: 60%
PET2+: 79%
PET22: 80%
PET22: 98%
3-year PFS
3-year OS

HD18 (BorchPET-CT
mann et al,34
2014, 1,100
patients)
40.0% positive
PET2+: randomized to four cycles

escBEACOPP vs. one cycle escBEACOPP, then four cycles R-BEACOPP
IIB with bulky 35 months

mediastinal
mass or EN
involvement,
IIIIV
PET2+: escBEACOPP, PET2+: escBEACOPP

91.4%;
96.5%,
R-BEACOPP, 93%
R-BEACOPP 94.4%
Continued
e378

(Contd)
Study
Trial Design
U.S. Intergroup Two cycles ABVD, interim PET-CT

S0816 (Press
et al,28 2013,
357 patients)
PET-2 Results
Clinical Stage
77.7% negative IIIIV
Median
Follow-up
28 months
Results
PFS/EFS/FFS
1-year PFS
PET2+: six cycles escBEACOPP
PET2+: 72%
PET22: 85%
OS
Not reported
Trials With Treatment De-escalated If PET-2 Negative

RATHL
(Johnson
et al,29 2015;
1,214
patients)
83.9% negative IIBIV, IIA with 32 months

bulk or three
or more involved sites
PET22: randomized to four cycles

ABVD vs. four cycles AVD
HD 0607
(Gallamini
et al,31 2015,
773 patients)
80.5% negative IIBIV
3-year PFS
3-year OS
PET22: ABVD,
85.5%;
PET22: ABVD 97%,
AVD, 84.5%
AVD 97.5%
34.3 months No interim data on

randomized arms
No interim data on
randomized arms
PET2+: randomized to escBEACOPP +

BEACOPP (four + four) vs. four + four
and rituximab
Overall study cohort Overall study cohort
PET22: four cycles ABVD. If CR, then

randomized to RT vs. observation
4-year FFS
4-year OS
PET2+: 62%
PET2+: 86%
PET22: 85%
PET22: 95%
HD 0801
(Zinzani
32
et al, 2015,
519 patients)
Not reported PET2+: 78.6%

Not reported
in abstract
(81/103 patients)
received
HDC-ASCT,
PET2+: ifosfamide-containing salvage

regimen HDC-ASCT
2-year PFS
PET2+: 74%
PET22: 81%
Israeli H2 (Dann IPS $ 3

et al,35 2014,
183 patients)
85.8% negative IB, IIB, IIIIV
Two cycles escBEACOPP, interim

PET-CT
36 months
3-year PFS (All

Not reported
advanced HL
regardless of IPS)
PET22: 86%
(p = .012)
PET22: four cycles ABVD, omit RT to

bulky sites
IPS , 3
PET22: four cycles ABVD, omitting RT
to bulky sites
NCT013048490 Two cycles ABVD interim PET-CT
(Ganesan
et al,36 2015,
50 patients)
24.7 months 2-year EFS
PET2+: 50%
PET22: 82%
(p = .013)
Not reported
Continued
e379
LYNCH AND ADVANI

(Contd)
Study
Trial Design
HD18
(Borchmann
PET-CT
et al,34 2014,
1,100
patients)
PET-2 Results
Clinical Stage
Median
Follow-up
60.0% negative IIB with bulky 35 months

mediastinal
mass or EN
involvement,
IIIIV
PET22: randomized to Two cycles

escBEACOPP vs. four cycles
escBEACOPP
Results
PFS/EFS/FFS
OS
3-year PFS
3-year OS
PET22: no data
available
PET22: no data
available
Abbreviations: ABVD, doxorubicin, bleomycin, vinblastine, and dacarbazine; AVD, doxorubicin, vinblastine, and dacarbazine; BEACOPP, bleomycin, etoposide, doxorubicin,
cyclophosphamide, vincristine, procarbazine, and prednisone; CR, complete remission; EFS, event-free survival; EN, extranodal; EOT, end of therapy; escBEACOPP, escalated-dose
BEACOPP; FFS, failure-free survival; HDC-ASCT, high-dose chemotherapy-autologous stem cell transplantation; HL, Hodgkin lymphoma; IPS, International Prognostic Score; LYSA,
Lymphoma Study Association; NCT, National Clinical Trial number; OS, overall survival; PET2+, interim PET-CT positive after 2 cycles of therapy; PET22, interim PET-CT negative after
two cycles of therapy; PFS, progression-free survival; R-BEACOPP, rituxumab plus BEACOPP; RT, radiotherapy.
* If comparison is statistically significant, p value is given.
A phase II multicenter Italian HD 0607 trial evaluated

773 patients with advanced Hodgkin lymphoma who received
two cycles of ABVD followed by an interim PET-CT. Patients
with PET-2negative disease (Deauville 13) continued with
four additional cycles of ABVD.31 Those in complete remission
(CR) after six cycles of ABVD were subsequently randomly
assigned to radiotherapy to sites of initial bulky disease versus
observation. Patients with PET-2positive disease (Deauville
45) were randomly assigned to either alternating cycles of
standard BEACOPP and escBEACOPP (four plus four) or the
four-plus-four regimen plus rituximab. Although there are no
data currently available comparing the randomized arms, at a
median follow-up 34.3 months, the overall 4-year FFS and OS
were 62% and 86% for patients with PET-2positive disease
and 85% and 95% for patients with PET-2negative disease
respectively.
The Italian HD 0801 trial tested an alternative aggressive
strategy in 519 patients. Treatment of patients with PET2positive disease after two cycles of ABVD was changed
to IGEV (ifosfamide, gemcitabine, vinorelbine) in preparation
for transplant.32 Patients in CR by PET-CT after IGEV received
an autologous transplant, whereas those not in CR received a
tandem autologous-allogeneic transplant. Of the patients
with PET-2positive disease 78.6% (81 of 103) received a
salvage regimen containing at least one autologous transplant; 14.6% (15 of 103) of patients with PET-2positive
disease continued on ABVD per investigator decision due
to a minimally positive PET-CT. There was no significant
difference in 2-year PFS in the PET-2positive intention-totreat group (76%) and the patients with PET-2positive
disease who subsequently underwent salvage chemotherapy and transplant (74%) with a 2-year PFS for patients with
PET-2negative disease of 81%, which was similar to that
observed in patients with PET-2positive disease.
The GHSG HD18 study evaluated 1,100 patients and
also used a PET-adaptive strategy intensifying therapy
based on interim PET results after two initial cycles
of escBEACOPP.33,34 Patients with PET-2negative disease
were randomly assigned to two versus four additional cycles
e380
of escBEACOPP (four or six total). No data from the PET2negative arm have been presented to date. In the PET2positive group, patients were randomly assigned to four
additional cycles of escBEACOPP versus one cycle of escBEACOPP followed by four cycles of rituximab-escBEACOPP
(R-escBEACOPP). Patients with PET-CTpositive disease
greater than 2.5 cm at the end of treatment received radiotherapy. Interim results of this trial reported no difference
in outcomes between the two arms. At a median follow-up of
35 months, the 3-year PFS for escBEACOPP was 91.4% (95%
CI, 87.0%95.7%) versus 93% for R-escBEACOPP (95% CI,
89.4%96.6%). Moreover, 3-year OS was not significantly
different between escBEACOPP and R-escBEACOPP (96.5% vs.
94.4%; p = .31).
The Israeli H2 study included 183 patients and stratified initial
treatment of patients with advanced-stage Hodgkin lymphoma
or early-stage Hodgkin lymphoma with B symptoms based
on the IPS score. 35 Patients with IPS scores lower than 3
(106 patients) were treated with two cycles of ABVD followed by interim PET-CT. Patients with PET-2positive disease had therapy escalated to escBEACOPP. In contrast,
patients with IPS scores of 3 or lower started out with two
cycles of escBEACOPP, with patients with PET-2positive
disease continuing with escBEACOPP and patients with
PET-2negative disease deescalating therapy to four cycles
of ABVD (discussed in the next section). Among the 162
patients with advanced-stage disease, 26 had a positive PET2 scan. Interim results report a 3-year PFS for PET-2 negative
and PET-2 positive disease of 86% and 75%, respectively
(p = .012), with no difference in PFS based on IPS score.
An Indian phase II study evaluated two cycles of ABVD
followed by interim PET-CT in 50 patients with stage IIB-IV
Hodgkin lymphoma.36 Patients with PET-2positive disease were escalated to escBEACOPP and patients with
PET-2negative disease continued therapy with four additional cycles of ABVD. At a median follow-up 24.7 months,
the 2-year EFS for patients with PET-2negative disease
and patients with PET-2positive disease was 82% and 50%,
respectively (p = .013).
In addition to the above trials, there is an ongoing study

examining very early PET-adapted escalation with interim
PET-CT after one cycle of ABVD.37 No interim results have
been presented so far.
Cumulatively, results of all of the above trials suggest that
an interim PET-adapted approach of escalating therapy to a
more aggressive regimen is feasible and possibly associated
with better outcomes in patients with PET-2positive disease after ABVD compared with historical controls, albeit
with more toxicity.
A major issue with most of these trials is the lack of a control
arm in which ABVD therapy is continued regardless of interim
PET-CT results. Additionally, as discussed previously, the Deauville scoring system has led to less inter-reader variability and
differences in outcomes between patients with PET-CTnegative
or PET-CTpositive disease are not as disparate as in the original
report by Gallamini et al.20,21,38
Therefore, long-term follow-up is critical to determine if
treatment modification translates into an OS advantage or
not. Until that is clearer, escalating therapy based on interim PET-CT results should ideally be done in the context of a clinical trial with strict definitions of PET-CT
interpretation.
STUDIES EVALUATING DE-ESCALATION OF

CHEMOTHERAPY IN PATIENTS WITH A NEGATIVE
INTERIM PET-CT SCAN
An alternate strategy for patients with advanced Hodgkin
lymphoma is to de-escalate chemotherapy in patients with
a negative interim PET-CT to mitigate some short- and longterm toxicities.
The Israeli group was among the first to report on a deescalation strategy. Patients with stage IIBIV disease with
IPS of 3 or higher started therapy with two cycles of escBEACOPP.39,40 For patients who had PET-2negative disease,
chemotherapy was de-escalated to four additional cycles of
ABVD, whereas patients with PET-2positive disease continued on escBEACOPP. Radiotherapy was administered to
initial sites of bulk ($ 10 cm) or mediastinal bulk at physician
discretion. The results of their prospective phase II study
were combined with additional patients who were treated
on the same protocol off-study (24 patients) for a total of 69
patients. With this de-escalation strategy, an interim PET-CT
predicted 5-year OS (PET-2 positive, 79%, vs. PET-2 negative,
98%; p = .015). The same group is currently evaluating in the H2
trial a similar strategy as used in the earlier study, in which
patients with PET-2negative disease after two cycles of escBEACOPP had therapy de-escalated to four additional cycles of
ABVD without radiotherapy.35 Interim results for 183 patients
at a median follow-up of 36 months show a 3-year PFS for
patients with PET-2negative disease and those with PET2positive disease of 86% and 75%, respectively (p = .012), with
no difference in PFS based on IPS score.
The AHL 2011 study randomly assigned 823 patients with
stage IIBIV Hodgkin lymphoma to a standard arm that
received six cycles total of escBEACOPP regardless of results
of interim PET after two cycles versus an experimental arm that

evaluated a PET-adapted de-escalation strategy.41 Patients with
PET-2positive disease continued with four additional cycles of
escBEACOPP, whereas for patients with PET-2negative disease, therapy was de-escalated to four cycles of ABVD. At a
median follow-up of 16.3 months, the 2-year PFS was similar in
the standard and experimental arms (91.6% vs. 88.3%; p = .79).
Overall in both arms, patients with PET-2positive disease
(97 patients) had a lower 2-year PFS than patients with
PET-2negative disease (72.9% vs. 92.8%; p , .0001). Standardarm patients with PET-2negative disease had a similar 2-year
PFS as the de-escalated experimental-arm patients with
PET-2negative disease (75.1% vs. 70.8%; not significant).
The U.K. RATHL trial randomly assigned patients with
Hodgkin lymphoma with a negative PET-CT after two cycles
of ABVD to continue with four additional cycles of ABVD
versus AVD (omitting bleomycin).29 Interim results of the
PET-2negative group (944 patients) with a median follow-up
of 32 months showed similar 3-year PFS for ABVD (85.45%;
95% CI, 83.4289.70) versus AVD (84.48%; 95% CI,
82.4788.97) with no difference in 3-year OS (ABVD, 97.0%;
95% CI, 94.598.4; AVD, 97.5%; 95% CI, 95.198.7). In addition, there were more grade 3/4 clinical pulmonary events
in the ABVD arm (12 total doses of bleomycin; 13 of 468
patients, 3%) versus the AVD arm (four total doses of
bleomycin; 6 of 440 patients, 1%).42 There was also a greater
decrease in lung diffusing capacity in patients who received
additional doses of bleomycin (8.7%; 95% CI, 6.1011.36).
As described earlier, the GHSG HD18 trial also randomly
assigned patients with PET-2negative disease to two versus
four additional cycles of escBEACOPP (four or six total).33,34
Although there have been no data presented yet from the
PET-2negative arm, the 3-year PFS for patients with PET2positive disease in this study exceeded 90%, suggesting
that the PET-2negative cohort is at lower risk with room for
de-escalation.
Cumulatively, the results of studies on continuing with
standard therapy (i.e., ABVD) after a negative interim PET-CT
scan report excellent outcomes and also support the
omission of bleomycin.
END OF THERAPY PET: WHEN SHOULD

RADIOTHERAPY BE GIVEN
Another ongoing debate is regarding the role of radiotherapy
in advanced Hodgkin lymphoma specifically because of concern
regarding late effects.43-46 Although the risk of late effects with
contemporary radiotherapy are lower, attempts to eliminate
radiotherapy are ongoing.1,47-49 Although radiotherapy is well
studied in the management of early-stage disease, its role in
patients with advanced Hodgkin lymphoma is less precise.
Historic data suggest that, with consolidative radiotherapy,
patients who achieve a partial remission (PR) using standard
definitions based on CT scans to assess response have outcomes similar to those in CR.50-52 Therefore, radiotherpy has
commonly been used to consolidate sites of initial bulky disease especially in the mediastinum with improved outcomes
e381
LYNCH AND ADVANI
in some series.52 Because current response criteria are now

based on PET-CT, the role of radiotherapy needs to be reevaluated. The only trial that has prospectively studied this
is the GHSG HD15 trial, in which patients with a residual mass
larger than 2.5 cm mass and PET positivity after six cycles of
escBEACOPP were treated with 30 Gy.1 The addition of PETCT to response assessment decreased the total number of
patients receiving radiotherapy when compared with the prior
GHSG HD9 trial (11% vs. 66%).53 The 4-year PFS curves for PETnegative PR and CR/CRu post-treatment were similar at 92.1%.
Patients with residual PET positivity had excellent outcomes
with a PFS of 86.2% after radiotherapy.
When ABVD is used as the chemotherapy backbone, the
data are less clear as there are no trials that have evaluated
radiotherapy versus no radiotherapy. The GITL/IIL study
used radiotherapy at the completion of treatment25.2 Gy
in all patients to sites of initial bulk and 30.6 Gy to areas of
residual disease.6 However, the use of PET was not mandated by the protocol. No patients in the RATHL or U.S.
Intergroup trials received radiotherapy.28,29
The ongoing HD 0607 trial is randomly assigning patients
with PET-2negative disease who have received a total of six
cycles of ABVD to radiotherapy to initially bulky sites versus
observation, and results are awaited.31 Since July 2005, the
BC Cancer Agency has adopted a treatment strategy of six
cycles of ABVD for patients with advanced-stage Hodgkin
lymphoma (including stage I with bulk and stage II with B
symptoms and/or bulk) followed by observation with radiotherapy used only for patients with PET-positive disease
at the end of therapy.54 Recent results suggest that posttreatment patients with PET-negative disease had superior
5-year freedom-from-treatment failure (FFTF) compared
with those with PET-positive disease (89% vs. 56%; p , .00001).
In addition, there was no difference in 5-year FFTF in the
PET-negative group between bulky (112 patients) and nonbulky (152 patients; 89% vs. 88.5%; p = .50) with similar OS
(96% vs. 94%, p = .51). Although the BC Cancer Agency data
support the lack of a benefit of radiotherapy in patients
who achieve CR by PET-CT at completion of chemotherapy,
the outcomes of patients with PET-positive disease after
ABVD is suboptimal with use of radiotherapy. Every attempt
should be made to rebiopsy such patients and consider salvage
therapy followed by autologous stem cell transplantation if
feasible.
For patients with stage III unfavorable disease (often treated
as advanced disease in the United States if bulky disease or B
symptoms), the EORTC/GELA/IIT H10 trial randomly assigned
patients to a standard arm (four total cycles of ABVD followed by
involved-node radiotherapy [INRT]), or an experimental arm
based on interim PET results.55 After two cycles of ABVD, patients
with PET-2positive disease received two cycles of escBEACOPP
followed by INRT. Patients with PET-2negative disease received
four additional cycles of ABVD (six total) without radiotherapy.
Interim analysis found an increase in events in the experimental
arm (16 vs. seven), which met prespecified significance for
futility, and therefore the experimental arm omitting radiotherapy
was closed to further accrual. The 1-year PFS for early-stage
e382
unfavorable patients with PET-2negative disease with

and without radiotherapy was 97.3% and 94.9%, respectively
(p = .026), with no difference in OS.
CONCLUSION
Interim PET-CT has emerged as a surrogate to better select
patients for therapy intensification or de-escalation. The
goal of these trials is to limit intensive chemotherapy/
radiotherapy to a subset of patients who might benefit
from an aggressive approach and spare the vast majority of
patients the associated toxicities. Early PFS and FFS data
suggest that there may be fewer earlier treatment failures in
patients with PET-2positive disease with escalation of
therapy compared with historical controls, although longer
follow-up is required to assess whether these strategies
actually translate to an OS advantage. Results with escalation or de-escalation strategies appear largely similar, and
the two have not been compared. The U.K. RATHL trial
supports the omission of bleomycin in patients with a
negative PET-CT after two cycles of ABVD.
Although patients with interim or end-of-therapy PET2positive disease and PET-2negative disease have different outcomes with standard therapy, these are likely a
reflection of variable biology at diagnosis. The limitations of
clinical prognostic models have led to the development of
molecular and immune-based markers.56 Gene expression
profiling identified a signature of tumor-associated CD68+
macrophages, whose frequency correlated with treatment
failure and OS and outperformed the IPS as a prognostic
biomarker.57 Increase in CD68+ macrophages at diagnosis
has also been shown to correlate with positive interim PETCT using the Deauville scale.58,59 Correlative data from the
E2496 trial found that tumor-associated macrophages on
initial pathology was associated with a worse FFS and OS.56
An analysis of a cohort of GHSG patients found that elevated serum levels of thymus and activation-regulated
chemokine/CCL17 at diagnosis were associated with an
increased risk of chemotherapy failure with multivariate
analysis (odds ratio, 3.052; 95% CI, 1.6055.804, p , .007)
when treated based on standard of care at the time of
diagnosis.60 A 23-gene panel has been developed that may
identify at diagnosis patients who are at increased risk of
treatment failure and death.61 All of these biomarkers hold
promise but require prospective validation. Additional
studies are also required to assess how they compare with
prognosis based on interim PET results.
Although the trials discussed in the risk-adapted strategies
described have largely used ABVD and escBEACOPP, there
are other novel combinations that are currently being
evaluated in phase III randomized trials due to excellent
initial results. A phase I pilot study showed an excellent CR
rate with the anti-CD30 antibody-drug conjugate, brentuximab vedotin plus ABVD (21 of 22 patients, 95%) or for
brentuximab plus AVD (24 of 25 patients, 95%) when used
up front for stage IIAX, IIBIVB Hodgkin lymphoma.62 Because of increased pulmonary toxicity seen when bleomycin
was used in conjunction with brentuximab vedotin,
the ongoing phase III multicenter study is evaluating

brentuximab plus AVD versus ABVD in Hodgkin lymphoma.63
A variant of BEACOPP called BrECADD has been tested
in an early-phase clinical trial with excellent results.64
BrECADD eliminates bleomycin, vincristine, procarbazine,
and prednisone and replaces them with brentuximab
vedotin, dacarbazine, and dexamethasone. BrECADD has an
18-month PFS of 89% (95% CI, 77%100%) with 18-month
OS of 100% in an early-phase trial. BrECADD is currently being compared with escBEACOPP in a randomized
frontline trial for advanced Hodgkin lymphoma (HD21 trial)
by the GHSG. Emerging data suggest that immune
checkpoint inhibitors are very effective in the relapsed/
refractory setting, and trials are underway in the frontline
setting.65-67 Whether these novel combinations will result in

better outcomes than standard therapies or interim PETadapted strategies is unknown. Additionally, the utility of
the Deauville criteria in the setting of these agents is largely
unknown. It is likely that some modification may be required
to account for tumor flare associated with checkpoint
inhibitors to eliminate over interpretation due to a pseudoprogression.68
Careful attention will need to be made toward late
follow-up with these regimens to assess for OS benefit,
as well as secondary malignancies. The future will likely
incorporate biologic and clinical factors along with tumor
burden defined by PET-CT to better tailor treatment to
each individual patient.
References
1. Engert A, Haverkamp H, Kobe C, et al; German Hodgkin Study Group;
Swiss Group for Clinical Cancer Research; Arbeitsgemeinschaft Medikamentose Tumortherapie. Reduced-intensity chemotherapy and
PET-guided radiotherapy in patients with advanced stage Hodgkins
lymphoma (HD15 trial): a randomised, open-label, phase 3 non-inferiority
trial. Lancet. 2012;379:1791-1799.
2. Gordon LI, Hong F, Fisher RI, et al. Randomized phase III trial of ABVD
versus Stanford V with or without radiation therapy in locally extensive
and advanced-stage Hodgkin lymphoma: an intergroup study coordinated by the Eastern Cooperative Oncology Group (E2496). J Clin
Oncol. 2013;31:684-691.
3. Hoppe RT, Advani RH, Ai WZ, et al. Hodgkin lymphoma, version 2.2015.
4. Wongso D, Fuchs M, Plutschow A, et al. Treatment-related mortality in
patients with advanced-stage hodgkin lymphoma: an analysis of the
german hodgkin study group. J Clin Oncol. 2013;31:2819-2824.
5. Behringer K, Mueller H, Goergen H, et al. Gonadal function and fertility in
survivors after Hodgkin lymphoma treatment within the German Hodgkin
Study Group HD13 to HD15 trials. J Clin Oncol. 2013;31:231-239.
6. Viviani S, Zinzani PL, Rambaldi A, et al; Michelangelo Foundation;
Gruppo Italiano di Terapie Innovative nei Linfomi; Intergruppo Italiano
Linfomi. ABVD versus BEACOPP for Hodgkins lymphoma when highdose salvage is planned. N Engl J Med. 2011;365:203-212.
7. Federico M, Luminari S, Iannitto E, et al; HD2000 Gruppo Italiano per lo
Studio dei Linfomi Trial. ABVD compared with BEACOPP compared with
CEC for the initial treatment of patients with advanced Hodgkins
lymphoma: results from the HD2000 Gruppo Italiano per lo Studio dei
Linfomi Trial. J Clin Oncol. 2009;27:805-811.
8. Carde PP, Karrasch M, Fortpied C, et al. ABVD (8 cycles) versus
BEACOPP (4 escalated cycles =. 4 baseline) in stage III-IV high-risk
Hodgkin lymphoma (HL): First results of EORTC 20012 Intergroup
randomized phase III clinical trial. J Clin Oncol. 2012;30 (suppl; abstr
8002).
9. Merli F, Luminari S, Gobbi PG, et al. Long-term results of the HD2000
trial comparing ABVD versus BEACOPP versus COPP-EBV-CAD in untreated patients with advanced Hodgkin lymphoma: a study by Fondazione Italiana Linfomi. J Clin Oncol. Epub 2015 Dec 28.
10. Hasenclever D, Diehl V. A prognostic score for advanced Hodgkins
disease. International Prognostic Factors Project on Advanced Hodgkins Disease. N Engl J Med. 1998;339:1506-1514.
11. Engert A, Diehl V, Franklin J, et al. Escalated-dose BEACOPP in the

treatment of patients with advanced-stage Hodgkins lymphoma: 10
years of follow-up of the GHSG HD9 study. J Clin Oncol. 2009;27:
4548-4554.
12. Moccia AA, Donaldson J, Chhanabhai M, et al. International Prognostic
Score in advanced-stage Hodgkins lymphoma: altered utility in the
modern era. J Clin Oncol. 2012;30:3383-3388.
13. Diefenbach CS, Li H, Hong F, et al. Evaluation of the International Prognostic
Score (IPS-7) and a Simpler Prognostic Score (IPS-3) for advanced Hodgkin
lymphoma in the modern era. Br J Haematol. 2015;171:530-538.
14. Juweid ME, Stroobants S, Hoekstra OS, et al; Imaging Subcommittee of
International Harmonization Project in Lymphoma. Use of positron
emission tomography for response assessment of lymphoma: consensus of the Imaging Subcommittee of International Harmonization
Project in Lymphoma. J Clin Oncol. 2007;25:571-578.
15. El-Galaly TC, dAmore F, Mylam KJ, et al. Routine bone marrow biopsy
has little or no therapeutic consequence for positron emission
tomography/computed tomography-staged treatment-naive patients
with Hodgkin lymphoma. J Clin Oncol. 2012;30:4508-4514.
16. Cerci JJ, Pracchia LF, Linardi CC, et al. 18F-FDG PET after 2 cycles of ABVD
predicts event-free survival in early and advanced Hodgkin lymphoma.
J Nucl Med. 2010;51:1337-1343.
17. Hutchings M, Loft A, Hansen M, et al. FDG-PET after two cycles of
chemotherapy predicts treatment failure and progression-free survival
in Hodgkin lymphoma. Blood. 2006;107:52-59.
18. Hutchings M, Mikhaeel NG, Fields PA, et al. Prognostic value of interim
FDG-PET after two or three cycles of chemotherapy in Hodgkin lymphoma. Ann Oncol. 2005;16:1160-1168.
19. Hutchings M, Kostakoglu L, Zaucha JM, et al. In vivo treatment sensitivity testing with positron emission tomography/computed tomography after one cycle of chemotherapy for Hodgkin lymphoma. J Clin
Oncol. 2014;32:2705-2711.
20. Gallamini A, Hutchings M, Rigacci L, et al. Early interim 2-[18F]fluoro-2deoxy-D-glucose positron emission tomography is prognostically superior
to international prognostic score in advanced-stage Hodgkins lymphoma:
a report from a joint Italian-Danish study. J Clin Oncol. 2007;25:3746-3752.
21. Le Roux PY, Gastinne T, Le Gouill S, et al. Prognostic value of interim FDG
PET/CT in Hodgkins lymphoma patients treated with interim responseadapted strategy: comparison of International Harmonization Project
(IHP), Gallamini and London criteria. Eur J Nucl Med Mol Imaging. 2011;
38:1064-1071.
e383
LYNCH AND ADVANI
22. Terasawa T, Lau J, Bardet S, et al. Fluorine-18-fluorodeoxyglucose

positron emission tomography for interim response assessment of
advanced-stage Hodgkins lymphoma and diffuse large B-cell lymphoma: a systematic review. J Clin Oncol. 2009;27:1906-1914.
23. Meignan M, Gallamini A, Meignan M, et al. Report on the First International Workshop on Interim-PET-Scan in Lymphoma. Leuk Lymphoma. 2009;50:1257-1260.
24. Biggi A, Gallamini A, Chauvie S, et al. International validation study for
interim PET in ABVD-treated, advanced-stage Hodgkin lymphoma: interpretation criteria and concordance rate among reviewers. J Nucl
Med. 2013;54:683-690.
25. Cheson BD, Fisher RI, Barrington SF, et al; Alliance, Australasian Leukaemia and Lymphoma Group; Eastern Cooperative Oncology Group;
European Mantle Cell Lymphoma Consortium; Italian Lymphoma
Foundation; European Organisation for Research; Treatment of Cancer/
~ol de Medula Osea;
Dutch Hemato-Oncology Group; Grupo Espan
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
German High-Grade Lymphoma Study Group; German Hodgkins Study

Group; Japanese Lymphorra Study Group; Lymphoma Study Association; NCIC Clinical Trials Group; Nordic Lymphoma Study Group;
Southwest Oncology Group; United Kingdom National Cancer Research
Institute. Recommendations for initial evaluation, staging, and response assessment of Hodgkin and non-Hodgkin lymphoma: the Lugano
classification. J Clin Oncol. 2014;32:3059-3068.
Barrington SF, Mikhaeel NG, Kostakoglu L, et al. Role of imaging in the
staging and response assessment of lymphoma: consensus of the International Conference on Malignant Lymphomas Imaging Working
Group. J Clin Oncol. 2014;32:3048-3058.
Gallamini A, Patti C, Viviani S, et al; Gruppo Italiano Terapie Innovative
nei Linfomi (GITIL). Early chemotherapy intensification with BEACOPP in
advanced-stage Hodgkin lymphoma patients with a interim-PET positive after two ABVD courses. Br J Haematol. 2011;152:551-560.
Press OW, LeBlanc M, Rimsza LM, et al. A phase II trial of response-adapted
therapy of stage III-IV Hodgkin lymphoma using early interim FDG-PET
imaging: U.S. Intergroup S0816. Paper presented at: 12th International
Conference on Malignant Lymphoma; June 2013; Lugano, Switzerland.
Johnson PW, Federico M, Fossa A, et al. Response-adapted therapy
based on interim FDA-PET scans in advanced Hodgkin lymphoma: first
analysis of the safety of de-escalation and safety of escalation in the
international RATHL study. Paper presented at: 13th International
Barrington SF, Kirkwood AA, Franceschetto A, et al. PET-CT for staging and
early response: results from Response Adapted Therapy in Advanced
Hodgkin Lymphoma (RATHL) (CRUK/07/033). Blood. Epub 2016 Jan 8.
Gallamini A, Rossi A, Patti C, et al. Interim PET-adapted chemotheapy in
advanced Hodgkin lymphoma (HL): Results of the second interim analysis
of the Italian GITIL/FIL DH0607 trial. Paper presented at: 13th International
Zinzani PL, Broccoli A, Gioia D, et al. Interim PET response-adapted
therapy in advanced stage Hodgkins lymphoma: final results of the
Phase II part of the Fondazione Italiana Linfomi (FIL) HD0801 Study.
Paper presented at: 13th International Conference on Malignant
Lymphoma; June 2015; Lugano, Switzerland.
Borchmann P, Eichenauer DA, Engert A. State of the art in the treatment
of Hodgkin lymphoma. Nat Rev Clin Oncol. 2012;9:450-459.
Borchmann P, Haverkamp H, Lohri A, et al. Addition of rituximab to BEACOPPescalated to improve the outcome of early interim PET positive advanced stage Hodgkin lymphoma patients: second planned interim analysis
of the HD18 study. Paper presented at: 56th American Society of Hematology Annual Meeting and Exposition; December 2014; San Francisco, CA.
Dann EJ, Bairer O, Bar-Shalom R, et al. Tailored therapy in Hodgkin lymphoma, based on predefined risk factors and early interim PET/CT: Israeli H2
e384
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
study. Paper presented at: 56th American Society of Hematology Annual

Meeting and Exposition; December 2014; San Francisco, CA.
Ganesan P, Rajendranath R, Kannan K, et al. Phase II study of interim
PET-CT-guided response-adapted therapy in advanced Hodgkins
lymphoma. Ann Oncol. 2015;26:1170-1174.
Waxman JH, Terry YA, Wrigley PF, et al. Gonadal function in Hodgkins
disease: long-term follow-up of chemotherapy. Br Med J (Clin Res Ed).
1982;285:1612-1613.
Gallamini A, Barrington SF, Biggi A, et al. The predictive role of interim
positron emission tomography for Hodgkin lymphoma treatment
outcome is confirmed using the interpretation criteria of the Deauville
five-point scale. Haematologica. 2014;99:1107-1113.
Kedmi M, Apel A, Davidson T, et al. High-risk, advanced-stage Hodgkin
lymphoma: the impact of combined escalated BEACOPP and ABVD
treatment in patients who rapidly achieve metabolic complete remission on
interim FDG-PET/CT scan. Acta Haematol. 2016;135:156-161.
Dann EJ, Blumenfeld Z, Bar-Shalom R, et al. A 10-year experience with
treatment of high and standard risk Hodgkin disease: six cycles of
tailored BEACOPP, with interim scintigraphy, are effective and female
fertility is preserved. Am J Hematol. 2012;87:32-36.
Casasnovas O, Brice P, Bouabdallah R, et al. Randomized phase III study
comparing an early PET driven treatment de-escalation to a not PETmonitored strategy in patients with advanced stages Hodgkin lymphoma: interim analysis of the AHL2011 Lysa study. Paper presented at:
57th American Society of Hematology Annual Meeting and Exposition;
December 2015; Orlando, FL.
Hague C, Johnson P, Federico M, et al. Pulmonary function and grade 3/
4 clinical events in PET negative patients taking part in the international
rathl trial (CRUK/07/033): a comparison of 12 vs 4 doses of bleomycin.
Paper presented at: 13th International Conference on Malignant
Lymphoma; June 2015; Lugano, Switzerland.
van Leeuwen FE, Klokman WJ, Stovall M, et al. Roles of radiation dose,
chemotherapy, and hormonal factors in breast cancer following
Hodgkins disease. J Natl Cancer Inst. 2003;95:971-980.
Aleman BM, van den Belt-Dusebout AW, Klokman WJ, et al. Long-term
cause-specific mortality of patients treated for Hodgkins disease. J Clin
Oncol. 2003;21:3431-3439.
Myrehaug S, Pintilie M, Tsang R, et al. Cardiac morbidity following
modern treatment for Hodgkin lymphoma: supra-additive cardiotoxicity of doxorubicin and radiation therapy. Leuk Lymphoma. 2008;49:
1486-1493.
Hodgson DC, Grunfeld E, Gunraj N, et al. A population-based study of followup care for Hodgkin lymphoma survivors: opportunities to improve surveillance for relapse and late effects. Cancer. 2010;116:3417-3425.
De Bruin ML, Sparidans J, vant Veer MB, et al. Breast cancer risk in
female survivors of Hodgkins lymphoma: lower risk after smaller radiation volumes. J Clin Oncol. 2009;27:4239-4246.
Picardi M, De Renzo A, Pane F, et al. Randomized comparison of
consolidation radiation versus observation in bulky Hodgkins lymphoma with post-chemotherapy negative positron emission tomography scans. Leuk Lymphoma. 2007;48:1721-1727.
Hodgson DC. Long-term toxicity of chemotherapy and radiotherapy in
lymphoma survivors: optimizing treatment for individual patients. Clin
Adv Hematol Oncol. 2015;13:103-112.
Aleman BM, Raemaekers JM, Tirelli U, et al; European Organization for
Research and Treatment of Cancer Lymphoma Group. Involved-field
radiotherapy for advanced Hodgkins lymphoma. N Engl J Med. 2003;
348:2396-2406.
Aleman BM, Raemaekers JM, Tomisic R, et al; European Organization for
Research and Treatment of Cancer (EORTC) Lymphoma Group.
Involved-field radiotherapy for patients in partial remission after
52.
53.
54.
55.
56.
57.
58.
59.
chemotherapy for advanced Hodgkins lymphoma. Int J Radiat Oncol

Biol Phys. 2007;67:19-30.
Johnson PW, Sydes MR, Hancock BW, et al. Consolidation radiotherapy
in patients with advanced Hodgkins lymphoma: survival data from the
UKLG LY09 randomized controlled trial (ISRCTN97144519). J Clin Oncol.
2010;28:3352-3359.
Diehl V, Franklin J, Hasenclever D, et al. BEACOPP, a new dose-escalated
and accelerated regimen, is at least as effective as COPP/ABVD in
patients with advanced-stage Hodgkins lymphoma: interim report
from a trial of the German Hodgkins Lymphoma Study Group. J Clin
Oncol. 1998;16:3810-3821.
Savage KJ, Connors JM, Villa DR, et al. Advanced stage classical Hodgkin
lymphoma patients with a negative PET-scan following treatment with
ABVD have excellent outcomes without the need for consolidative
radiotherapy regardless of disease bulk at presentation. Paper presented at: 57th American Society of Hematology Annual Meeting and
Exposition; December 2015; Orlando, FL.
Raemaekers JM, Andre MP, Federico M, et al. Omitting radiotherapy in
early positron emission tomography-negative stage I/II Hodgkin lymphoma is associated with an increased risk of early relapse: clinical
results of the preplanned interim analysis of the randomized EORTC/
LYSA/FIL H10 trial. J Clin Oncol. 2014;32:1188-1194.
Tan KL, Scott DW, Hong F, et al. Tumor-associated macrophages predict
inferior outcomes in classic Hodgkin lymphoma: a correlative study
from the E2496 Intergroup trial. Blood. 2012;120:3280-3287.
Steidl C, Lee T, Shah SP, et al. Tumor-associated macrophages and survival in
classic Hodgkins lymphoma. N Engl J Med. 2010;362:875-885.
Cuccaro A, Annunziata S, Cupelli E, et al. CD68+ cell count, early
evaluation with PET and plasma TARC levels predict response in Hodgkin
lymphoma. Cancer Med. Epub 2016 Jan 13.
Touati M, Delage-Corre M, Monteil J, et al. CD68-positive tumorassociated macrophages predict unfavorable treatment outcomes in
classical Hodgkin lymphoma in correlation with interim fluorodeoxyglucose-
60.
61.
62.
63.
64.
65.
66.
67.
68.
positron emission tomography assessment. Leuk Lymphoma. 2015;56:

332-341.
Sauer M, Plutschow A, Jachimowicz RD, et al. Baseline serum TARC
levels predict therapy outcome in patients with Hodgkin lymphoma. Am
J Hematol. 2013;88:113-115.
Scott DW, Chan FC, Hong F, et al. Gene expression-based model using
formalin-fixed paraffin-embedded biopsies predicts overall survival in
advanced-stage classical Hodgkin lymphoma. J Clin Oncol. 2013;31:692-700.
Younes A, Connors JM, Park SI, et al. Brentuximab vedotin combined
with ABVD or AVD for patients with newly diagnosed Hodgkins lymphoma: a phase 1, open-label, dose-escalation study. Lancet Oncol.
2013;14:1348-1356.
National Institutes of Health. Phase 3 Frontline Therapy Trial in Patients
With Advanced Classical Hodgkin Lymphoma. https://clinicaltrials.gov/
ct2/show/nct01712490?Term=c25003&rank=1. Accessed January 6,
2016.
Borchmann P, Eichenauer DA, Plutschow A, et al. Targeted BEACOPP
variants in patients with newly diagnosed advanced stage classical
Hodgkin lymphoma: final analysis of a randomized phase II study. Paper
presented at: 57th American Society of Hematology Annual Meeting
and Exposition; December 2015; Orlando, FL.
Ansell SM, Lesokhin AM, Borrello I, et al. PD-1 blockade with nivolumab
in relapsed or refractory Hodgkins lymphoma. N Engl J Med. 2015;372:
311-319.
National Institutes of Health. A Study of Brentuximab Vedotin Combined With Nivolumab for Relapsed or Refractory Hodgkin Lymphoma.
https://clinicaltrials.gov/ct2/show/study/NCT02572167. Accessed January 6, 2016.
National Institutes of Health. ACP-196 in Combination With Pembrolizumab, for Treatment of Hematologic Malignancies (KEYNOTE145).
https://clinicaltrials.gov/ct2/show/NCT02362035. Accessed January 6,
2016.
Chiou VL, Burotto M. Pseudoprogression and Immune-Related Response in Solid Tumors. J Clin Oncol. 2015;33:3541-3543.
e385
Chemoimmunotherapy Versus
Targeted Treatment in Chronic
Lymphocytic Leukemia: When,
How Long, How Much, and in
Which Combination?
CHAIR
John M. Pagel, MD, PhD
Seattle, WA
SPEAKERS
Jennifer R. Brown, MD, PhD
Boston, MA
Michael J. Hallek, MD
University Hospital Cologne
Cologne, Germany
MANAGEMENT OF CLL IN THE ERA OF NOVEL TARGETED THERAPIES
Chemoimmunotherapy Versus Targeted Treatment in Chronic

Lymphocytic Leukemia: When, How Long, How Much, and in
Which Combination?
Jennifer R. Brown, MD, PhD, Michael J. Hallek, MD, and John M. Pagel, MD, PhD
OVERVIEW
During the past 5 years, rapid therapeutic advances have changed the landscape of chronic lymphocytic leukemia (CLL)
therapy. This disease has traditionally been treated using cytotoxic chemotherapy regimens in combination with antiCD20 antibody treatment, and recent long-term follow-up data from multiple centers suggest that fit patients with CLL
with favorable disease featuresparticularly mutated immunoglobulin heavy chain variable region (IGHV) genes
derive very long-term benefit from the most potent of these regimens, namely the fludarabine, cyclophosphamide, and
rituximab (FCR) regimen. The advent of oral targeted therapies, particularly ibrutinib and idelalisib, has provided
generally well-tolerated and highly effective additional options that have come into widespread use in the relapsed
setting. Additional agents are advancing in clinical development, with the BCL-2 inhibitor venetoclax likely to be approved by the U.S. Food and Drug Administration (FDA) in 2016. With the development of these novel therapies for
patients with relapsed CLL, many unanswered questions remain, including the optimal sequence (first vs. second line),
duration, discontinuation, and combination of these agents. In addition, recent publications show the emergence of a
pattern of treatment resistance in certain subgroups of patients with del(17p) and complex karyotype that needs further
study and improvement. Because the field of CLL management has become much more complex, we focus here on
understanding the recent data and discuss many of the questions and controversies important for how we approach
patients with CLL.
hronic lymphocytic leukemia is a heterogeneous neoplasm. Some patients never require treatment, whereas
others face an aggressive treatment course similar to
acute leukemia. In general practice, newly diagnosed
patients with asymptomatic early- or intermediate-stage
disease (Rai 0, I-II, Binet A or B) should be monitored
without therapy unless they have evidence of disease
progression or show clear symptoms caused by CLL.1
Several studies on early-stage CLL, as well as a metaanalysis,2 have not been able to demonstrate a potential
benefit of early intervention therapy to date, and thus the
International Workshop on CLL guidelines1 remain the
standard determinant of therapy initiation. In CLL, the
absolute lymphocyte count should not be used as the sole
indicator for treatment because even markedly elevated
leukocyte counts rarely cause symptoms or complications.
It should be stressed that these recommendations have
not changed in the era of new targeted agents. Clinical
trials that test the early use of novel inhibitors are currently underway.3
USE OF CHEMOIMMUNOTHERAPY FOR PATIENTS

WITH FRONT-LINE AND RELAPSED DISEASE
Potential Advantages of Chemoimmunotherapy
Currently there remains a general consensus that no curative
treatment of CLL exists, with the exception of allogeneic
stem cell transplantation.4 Chemoimmunotherapy achieves
disease control and survival prolongation and has therefore
been the current standard treatment for patients with
previously untreated CLL.5-7 Additionally, though the outcome
following chemoimmunotherapy may be highly variable, important subgroups of patients show an excellent outcome,
corroborating its use as standard first-line therapy. On the
other hand, secondary malignancies are one of the most
concerning unwanted effects following chemoimmunotherapy
and the true long-term incidence in comparison with patients
with CLL not receiving chemoimmunotherapy needs to be
better defined.
The German CLL Study Group has previously shown in their
CLL8 study that the addition of the anti-CD20 monoclonal
antibody rituximab (R) to fludarabine and cyclophosphamide
From the Dana-Farber Cancer Institute, Boston, MA; University Hospital of Cologne, Cologne, Germany; Swedish Cancer Institute, Seattle, WA.
Corresponding author: John M. Pagel, MD, PhD, Swedish Cancer Institute, 1221 Madison St., Suite 1000, Seattle, WA 98104; email: john.pagel@swedish.org.
e387
BROWN, HALLEK, AND PAGEL
(FC) improved both the progression-free survival (PFS) and the

overall survival (OS) of physically fit patients with previously
untreated symptomatic CLL.5 On the basis of these findings,
FCR has become a standard therapy for physically fit patients
with previously untreated CLL. In an updated report of this trial,
Fischer et al presented the results of an extended observation
time with a median follow-up of 5.9 years, with particular
emphasis on long-term follow-up for survival and adverse
events.8 This follow-up analysis of the CLL8 study continues to
demonstrate an improvement in PFS and OS in physically fit
patients treated with FCR when compared with FC. Patients
with del(17p) showed a significantly shorter OS than those in all
of the other cytogenetic subgroups. Conversely, of the patients
with del(11q), which is considered an adverse prognostic
group,9 patients with IGHV mutation responded very well to
FCR and, as a consequence, showed an outcome similar to
other patients without del(17p). FCR induced a higher rate of
prolonged neutropenia during the first year after the end of
treatment.
In multivariate analyses including the most commonly
used clinical and biologic prognostics, several predicted
OS. Each of these factors were shown to correlate with OS
in previous publications.9-16 The PFS for patients with
IGHV-mutated disease showed a significant difference in
favor of the IGHV mutation subgroup compared with unmutated disease. Most importantly, the IGHV mutation
subgroup treated with FCR showed a significantly longer PFS
and OS than those treated with FC. The median OS for
patients with IGHV-mutated CLL was not reached. More than
83% of the patients with an IGHV mutation treated with FCR
were still alive after almost 6 years of observation time.
Notably, cytogenetic subgroup analysis of patients with
IGHV mutation treated with FCR confirmed this excellent
outcome for all cytogenetic subgroups including del(11q),
except for patients with del(17p) and normal karyotype.
In a similar way, the University of Texas MD Anderson
Cancer Center group reviewed the phase II FCR study to
identify long-term disease-free survivors among the 300
patients included in the trial.17 At the median follow-up of
12.8 years, PFS was 30.9% (median PFS, 6.4 years). The
KEY POINTS
e388
Chemoimmunotherapy remains an appropriate option

for many patients with CLL and in particular those with
most favorable features in the front-line setting.
The use of targeted novel agents has changed the
therapeutic landscape for treating relapsed, refractory
CLL.
Many unanswered questions remain for further
investigation, including those assessing the optimal
length of therapy, the sequencing of agents, and
understanding responses with class switching, as well as
questions focusing on the mechanisms of overcoming
resistance.
12.8-year PFS was 53.9% for patients with mutated IGHV

(IGHV-M) and 8.7% for patients with unmutated IGHV (IGHVUM). A plateau was seen on the PFS curve for patients
with IGHV-M, with no relapses beyond 10.4 years in 42
patients (total follow-up, 105.4 patient-years). On multivariable analysis, IGHV-UM (hazard ratio [HR] 3.37; 95% CI,
2.185.21; p , .001) and del(17p) by conventional karyotyping (HR 7.96; 95% CI, 1.0261.92; p = .048) were significantly associated with inferior PFS. In a retrospective
report on 404 patients with CLL receiving front-line FCR,
Rossi et al also found that patients with CLL with mutated
IGHV and neither del(11q) nor del(17p) deletion had a very
high rate of long-term PFS, with OS similar to an agematched normal control population at 5 years.18
Taken together, most patients with IGHV-mutated CLL
seem to benefit substantially from FCR chemoimmunotherapy,
resulting in long-term control of the disease. This observed
benefit of FCR for the majority of patients with IGHV mutation
may announce the emergence of a therapeutic approach that
is guided by the IGHV mutational status. Secondly, given the
excellent PFS following first-line FCR therapy, it may be more
difficult to replace first-line chemoimmunotherapy with FCR
with novel agents in this specific subset of patients with CLL.
To test the value of newer targeted therapies against FCR
in previously untreated, fit patients with CLL, several phase
III trials have started that compare the potential of
chemotherapy-free treatment strategies with FCR or other
chemoimmunotherapies.19,20
Potential Disadvantages of Chemoimmunotherapy

One of the potential risks of chemoimmunotherapy is myeloand immunosuppression and, in its consequence, the onset
of myeloid malignancies (acute myeloid leukemia [AML] and
myelodysplastic syndrome [MDS]). Moreover, other cancers
may also be increased in CLL. Therefore, Benjamini et al
investigated the characteristics, incidence, outcomes, and
factors associated with second cancers for 234 patients
receiving FCR-based regimens in the front-line setting.21
They showed that the risk of second cancers was 2.38 times
higher than the expected risk for the general population.
Ninety-three patients (40%) had other cancers before FCR,
and 66 patients (28%) had them after FCR. Rates of therapyrelated acute AML/MDS (t-AML/MDS; 5.1%) and Richter
transformation (9%) were high, while solid tumors were not
increased.
In some contrast, in their follow-up report on the CLL8
protocol, Fischer et al were not able to report increased rates
of t-AML/MDS and Richter transformation following FCR
chemoimmunotherapy when compared with FC.22,23 The
frequency of secondary malignancies was also similar in both
arms. An increased frequency of Richter transformations
was noted in the FC but not FCR arm, in line with the current literature.24,25 In addition, achievement of a response
seemed to correlate with a lower incidence of secondary
malignancies and Richter transformations.26 As CLL itself
appears to increase the risk of second malignancies, very
long-term follow-up studies of patients treated with

chemoimmunotherapy or with novel agents will be required
to accurately assess how either type of therapy influences
this risk.
PARAMETERS TO BE CONSIDERED FOR THE

CHOICE OF THE BEST THERAPY
Given the impressive number of choices, the selection of the
optimal treatment of a given CLL patient has become a task
that requires experience, good clinical judgment, and appropriate use of diagnostic tools. The following parameters
should be considered before recommending a treatment of
CLL: (1) the clinical stage of the disease, (2) the symptoms of
the patient, (3) the fitness of the patient, (4) the genetic risk
of the leukemia, and (5) the treatment situation (first vs.
second line, response vs. nonresponse to the last treatment).27 Using the assessment of these five parameters, the
following recommendations can be given.
First-Line Treatment
Treatment should be initiated for patients with advanced
(Binet C, Rai III-IV) or active symptomatic disease. In this
situation, patients must be evaluated for their physical
condition (or comorbidity). For patients in good physical
condition (go-go) as defined by a normal creatinine clearance and a low score on the cumulative illness rating scale
(CIRS),28 patients should be offered combination therapies
such as FCR. Patients with a somewhat impaired physical
condition (slow-go) may be offered either chlorambucil in
combination with an anti-CD20 antibody (with obinutuzumab being the most active agent). The aim of therapy in this
situation is symptom control. For patients with symptomatic
disease and with del(17p) or TP53 mutations, it has been
generally agreed that patients should receive ibrutinib as
first-line treatment.29,30
Second-Line Treatment
As a general rule, the first-line treatment may be repeated, if
the duration of the first remission exceeds 24 to 36 months.
The choice is more difficult in treatment-refractory CLL (as
defined by an early relapse within 612 months after the
last treatment) or in cases with the chromosomal aberration del(17p). In principle, the initial regimen should be
changed. The following treatment options should be discussed: (1) kinase inhibitors such as idelalisib or ibrutinib,
(2) experimental protocols using other nonapproved drugs,
and (3) cellular therapies (e.g., chimeric antigen receptor [CAR] T-cell therapy) and allogeneic stem cell transplantation with curative intent in select patients.31 The
choice of one of these options may depend on the fitness of
the patient, the availability of the drugs, and the molecular
cytogenetics. Finally, it is important to emphasize that
patients with refractory disease should be treated within
clinical trials whenever possible.
NOVEL TARGETED THERAPIES IN CHRONIC

LYMPHOCYTIC LEUKEMIA
The recent development of targeted therapies has revolutionized the treatment landscape for patients with CLL,
particularly relapsed disease. The most advanced of these
molecules are FDA approved and include the first-generation
Bruton tyrosine kinase (BTK) inhibitor, ibrutinib, and the PI3K
delta inhibitor, idelalisib. The BCL-2 inhibitor venetoclax has
been also expected to receive FDA approval in the first half of
2016. To understand how we might use these agents and
others as the field evolves, we will consider each target and
inhibitor in turn (Table 1 and Fig. 1).
BTK Inhibitors: Ibrutinib

Ibrutinib is the first-in-class inhibitor of BTK, which is critical
for B-cell survival and function but whose loss otherwise
causes few problems for patients with Bruton X-linked
agammaglobulinemia. Ibrutinib is a covalent inhibitor that
binds to Cys481 in BTK, resulting in prolonged target inhibition despite a short circulating half-life.32,33 Ibrutinib is a
potent inhibitor but not a specific one, inhibiting 19 other
kinases, several covalently, with IC50 greater than 100 nM.33
In fact, ibrutinib has been shown to covalently inhibit ITK,34 a
kinase important for T and natural killer cell activity that may
lead to impaired antibody-dependent cellular cytotoxicity.35
Nonetheless, the clinical activity of ibrutinib in CLL is
impressive. In the initial CLL-specific phase IB/II study,
nodal responses were noted to be very high initially, often
with increase in lymphocytosis, which then resolved
slowly over time.29 This observation, which is universal
with B-cell receptortargeting therapies, led to the definition of a new response categorypartial response (PR)
with lymphocytosisfor patients who meet all criteria for
PR except that their lymphocyte count is persistently elevated.36 Over time as the lymphocyte count decreases,
these responses have evolved to PR by standard criteria.1
Thus, at the time of the first single-agent report, the overall
response rate (ORR) was 71%, with an estimated 26-month
PFS of 75% for patients with a median of four prior therapies.29 This study led to accelerated FDA approval in
February 2014 of ibrutinib for patients with CLL after at
least one prior therapy. The confirmatory registration trial,
RESONATE, randomly assigned patients with relapsed CLL
to ibrutinib or ofatumumab and demonstrated a significant
prolongation of both PFS (HR 0.22; p , .001) and OS (HR
0.43; p = .005) in the ibrutinib arm.37 The RESONATE-17
study was a phase II registration trial focused on patients
with relapsed del(17p) with one to four prior therapies and
early results demonstrated an estimated 12-month PFS of
79%.38 Based on these data, full FDA approval was granted
in July 2014 for patients with CLL after at least one prior
therapy, or patients with CLL with del(17p) in any line of
therapy.
Longer-term 3-year follow-up of the original phase IB/II
study was recently published and demonstrated that the
ORR has increased in the relapsed refractory cohort to 90%
e389
TABLE 1. Current Novel Agents for Chronic Lymphocytic Leukemia

Agent
Unique Features
Select Key Trials
Progression-Free Survival
First-in-class BTK
inhibitor
RESONATE, R/R CLL, ibrutinib vs. ofatumumab, FDA

approved37
R/R CLL single-agent: 26 mos = 75%39
RESONATE-17, R/R CLL and front-line 17p deletion

under FDA review38
del17p R/R CLL single-agent:

12 mos = 79%38
RESONATE-2, first-line, ibrutinib vs. chlorambucil

FDA approved49
First-line single-agent:
18 mos = 94%49
BTK Inhibitors
Ibrutinib
Acalabrutinib
Lacks targeting of
ITK or TEC kinase
Phase I-II single-agent trial with ORR of 95%51
R/R CLL single-agent with median

3 prior regimens: 12 mos = 100%51
BGB-3111
Targets BTK and TEC

but not ITK kinase
Phase I trial in first 14 patients with ORR of 93%52
N/A
First-in-class PI3K
inhibitor targeting
d isoform
Study 116, R/R CLL, idelalisib/rituximab vs.

rituximab, FDA approved.57
R/R CLL with rituximab (45% with

del(17p)): median 19.4 mos57
Study 115, R/R CLL BR/idelalisib vs. BR under FDA

review59
R/R CLL with BR: median 23.1 mos59
Phase II first line, idelalisib and rituximab in 64

patients older than age 65 with ORR of 97%64
First line with rituximab: 36

mos = 83%64
Pi3K Inhibitors
Idelalisib
Duvelisib
Dual PI3K d and g

inhibitor
Phase III R/R CLL, duvelisib vs. ofatumumab,

under FDA review
R/R CLL phase I single-agent (52%

with del(17p)): 24 mos = 59%.66
TGR-1202
PI3K d inhibitor with

once-daily dosing
Phase I, single-agent TGR-1202 in 20 patients

with R/R CLL with 88% nodal response67
N/A
Lacks targeting of
Bcl-xL
Phase I/expansion R/R CLL, venetoclax single

agent, under FDA review72
R/R CLL single agent: median

25 mos72
Phase Ib R/R CLL, venetoclax and rituximab in

49 patients with ORR of 86%73
R/R CLL with rituximab (33% with

del(17p)): 24 mos = 83%.73
Bcl-2 Inhibitor
Venetoclax
Monoclonal Antibodies/Engineered Antibody-Like Molecules

Obinutuzumab
Humanized antiCD20 antibody
CLL11 first line, R/R CLL, obinutuzumab and

chlorambucil, FDA approved7
First line with chlorambucil: median

26.7 mos.7
Ofatumumab
Humanized antiCD20 antibody
First line, ofatumumab and chlorambucil

vs. chlorambucil, FDA approved87
First line with chlorambucil: median

22.4 mos.87
CLL refractory to fludarabine (F) and alemtuzumab

(A), FDA approved88
Refractory to FA duration of
response = 6.5 mos.88
Phase II R/R (40 patients) CLL with ibrutinib with

ORR = 88% and 95% in high risk (20 patients)67
N/A
Ublituximab
Otlertuzumab
IgG1 antibody with

unique binding
sequence
Humanized antiCD37 Ig fusion
protein
Phase III R/R CL, ublituximab and TGR-1202

vs. obinutuzumab and chlorambucil ongoing
Phase I, single agent in 83 patients with R/R CLL
with 23% ORR89
R/R CLL with bendamustine:

median 16.1 mos90
Phase II R/R CLL, otlertuzumab and bendamustine

vs. bendamustine90
Abbreviations: N/A, not available; R/R, relapsed/refractory; BR, bendamustine and rituximab; BTK, Bruton tyrosine kinase; CLL, chronic lymphocytic leukemia; FDA, U.S. Food and Drug
Administration; mos, months; ORR, overall response rate.
with 7% complete remissions (CR) with a median time on

treatment of 23 months and 53% of patients remaining
on treatment.39 The estimated PFS at 30 months was
69%, indicating that although most patients remain in PR,
nonetheless they can maintain long-term remissions. In
addition, cytogenetic aberrations were also profoundly
predictive of PFS. Those patients with del(17p) had a median
PFS of 28 months, while the PFS for those with del(11q) was
74% at 30 months and for those with neither high-risk
deletion, it was 87%. In the RESONATE study, at present
with 16-month follow-up on the ibrutinib arm, no difference
e390
in PFS has been seen between patients with and without

del(17p).40 However, among patients with disease that relapsed on ibrutinib, both del(17p) and complex karyotype are
quite common, with the latter suggested to be more important
in two studies.41,42 Among patients with relapsing disease,
Richter transformation had a cumulative incidence of 4.5%
at 1 year in the largest study to date and was seen in 18 of
31 patients with relapsed disease with median survival
of 3.5 months.41 Chronic lymphocytic leukemia relapse
tended to occur later but was also aggressive with median
OS of 17.5 months. In a second study, median OS after
FIGURE 1. Schematic Representation of CLL Lymphocyte Depicting the B-Cell Signaling Cascade and Novel
Therapeutic Agents and Their Associated Targets
discontinuing ibrutinib was only 3.1 months.43 Thus, disease

progression on ibrutinib has become a new unmet need for
patients showing aggressive disease that may be very difficult to treat. Ongoing work has focused on elucidating the
mechanism(s) of relapse; among patients with relapsing CLL,
mutations have been identified in the target Cys residue in
BTK that results in a change to Ser that abrogates covalent
binding, as well as activating mutations in PLCg, the immediate downstream target of BTK.44,45 Little is yet known
about the biology of the Richter transformations.
Ibrutinib has been generally well tolerated with the most
common adverse events including mild diarrhea, nausea,
ecchymosis, and arthralgia; in most clinical trials, the rate of
discontinuation for adverse events has been 10%12%.37,39
However, this risk has been shown to be age-dependent,
with higher rates among those over age 70 and even more-so
among patients over age 80.41 Medically important adverse
events include atrial fibrillation that was increased in the
ibrutinib arm in RESONATE and occurred at 5%10% frequency in most studies,37,38 as well as risk of major hemorrhage, which may be related, at least in part, to inhibition
of collagen-induced platelet aggregation by ibrutinib through
its effect on BTK and possibly TEC kinase.46,47 Because of the
latter risk, ibrutinib should be cautiously combined with
anticoagulation, if at all.
Increasing data have become available on the use of
ibrutinib as initial therapy for CLL. In the initial phase IB/II
study, 31 treatment-naive patients over age 65 with generally low-risk disease were enrolled, and, at the most recent
30-month follow-up, 81% remained on study treatment with
an ORR of 84% and 23% patients were in CR at this longer
follow-up.48 The PFS was 96% at 30 months with one relapse
and death in a patient with del(17p).39 RESONATE-2 was a
registration trial for patients with previously untreated CLL

older than age 65 without del(17p) who were randomly
assigned to either single-agent ibrutinib or chlorambucil.
Ibrutinib significantly improved PFS (HR 0.16; p = .001), with
an estimated 18-month PFS of 94%, as well as ORR and OS,
and 87% of patients continue to take ibrutinib.49 Notable
serious adverse events among the 136 patients in the
ibrutinib arm included five major hemorrhages, two unexplained sudden deaths, hypertension in 14%, and atrial
fibrillation in 6%. This study led to FDA approval of singleagent ibrutinib for patients with previously untreated CLL in
early March 2016.
Next-Generation BTK Inhibitors

Given the success of ibrutinib, much interest has naturally
focused on efforts for continued improvement. To date, this
attention has largely been aimed at developing an equally
potent inhibitor that may be more specific for inhibition of
BTK in hopes of reducing the bleeding and cardiac toxicities
that appear to be because of ibrutinib-mediated off-target
effects. The first more specific inhibitor was CC-292, which
showed lower response rates and less durability of response
despite escalating to a high-dose and achieving high-levels of
BTK occupancy.50 The reason for this reduced activity has
been unclear but seems likely due to reduced bioavailability
or delivery of the drug to tumor. Within the past year,
however, two additional specific BTK inhibitors that bind
covalently to the same Cys target as ibrutinib have shown
promising results. Acalabrutinib (ACP-196) targets BTK but
not ITK or TEC and has been studied in a phase I and II
relapsed CLL single-arm study among patients with a median of three prior regimens and showed an ORR of 95%,
e391
including 10% PR with lymphocytosis (PR-L) at 14.3 months,

with only two PFS events thus far (one death from pneumonia
and one CLL progression).51 With current early follow-up, no
events of major hemorrhage or atrial fibrillation have occurred.
Acalabrutinib has entered two registration trials, including a trial
head-to-head against ibrutinib in relapsed high-risk CLL. The
second drug, BGB-3111, targets BTK and TEC but not ITK. In a
phase I study across B-cell malignancies presented at the 2015
American Society of Hematology (ASH) Annual Meeting, 39 patients have been enrolled, and complete BTK occupancy has been
demonstrated in lymph node biopsies.52 The overall response
rate was 93% (13/14 patients) to date, and this study is ongoing.
PI3K Inhibitors: Idelalisib

The other critical target that led to an FDA-approved drug in
CLL has been the delta isoform of PI3 kinase (PI3K) that is
targeted by idelalisib. Expression of the delta isoform has
been shown to be limited to hematopoietic cells, and its
primary physiologic function appears to be in B cells based
on the phenotype of the knockout mouse.53 Most downstream PI3K signaling in CLL cells has been shown to go
through delta54,55 and can be inhibited in vitro and in vivo by
idelalisib.54-56 In the phase I relapsed CLL study, idelalisib
had a 72% ORR, including 33% PR-L.56 Median PFS across all
cohorts in this heavily pretreated population (median of five
prior regimens) was 15.8 months, while the median PFS at
the recommended phase II dose of 150 mg or more two
times per day was 32 months. Because of the persistence of
lymphocytosis over time in the phase I patients and the
ongoing evolution of the response criteria, it was decided to
study combination therapies with idelalisib in registration
trials. Three phase III randomized trials were designed for
patients with relapsed CLL: idelalisib/rituximab versus rituximab (study 116), idelalisib/ofatumumab versus ofatumumab (study 119), and idelalisib/bendamustine/rituximab
(BR) versus bendamustine/rituximab (BR) (study 115). Study
116 enrolled patients with relapsed CLL with increased
numbers of comorbidities, 45% of whom had del(17p), and
demonstrated significant improvements in both PFS (HR
0.25; median, 19.4 vs. 7.3 months at the second interim
analysis following crossover; p , .0001) and OS.57 No difference in PFS was seen based on the adverse prognostic
factors of unmutated IGHV or del(17p)/TP53 mutation.
These results led to the FDA approval of idelalisib with
rituximab in July 2014 for patients with relapsed CLL for
whom rituximab may be appropriate therapy. Subsequently,
study 119 results showed improved PFS for idelalisib with
ofatumumab (median, 16.3 months) versus ofatumumab
alone (median, 8 months; HR 0.27; p , .0001).58 The results
of idelalisib plus BR versus BR (study 115) were reported at
the ASH Annual Meeting in December 2015 and showed
significant improvement in median PFS 11.1 to 23.1 months
(HR 0.33; p , .0001).59 In this patient population with 33% of
patients with del(17p) and 30% of patients with refractory
CLL, OS was also significantly improved. Both of these studies
are expected to lead to additional indications for idelalisib.
e392
Idelalisib shows a characteristic pattern of toxicity that

includes transaminitis at weeks 4 to 8, delayed colitis occurring at a median of 7 months after starting treatment,
and a low-rate of drug-related pneumonitis occurring at a
median of 4 months. A recent summary of toxicity data
across eight trials of patients primarily with relapsed disease
demonstrated that grade 3 or greater transaminitis and
colitis were each seen in 14% of patients, and pneumonitis
was seen in 3%. These toxicities have been responsive to
drug hold and to corticosteroids, and recent data have
demonstrated CD8+ T-cell infiltrates in the liver60 and colon61,62 of affected patients. An immunologic mechanism
of toxicity may not be surprising given that the delta isoform of PI3K has been shown to be critical for the survival
and function of regulatory T cells,63 and recent data have
demonstrated that early hepatotoxicity is associated with a
decrease in T regulatory cells 1 month after starting idelalisib.60 These toxicities do resolve with drug hold and/or
steroid use, and, following that, the majority of patients are
able to resume dosing at the same or lower dose. Colitis can
prove most troublesome but can be commonly managed by
oral budesonide with a prolonged slow taper.
Use of idelalisib in the up-front setting has been more
limited than ibrutinib. The study with the longest follow-up
was a phase II study of 64 patients older than age 65 treated
front line with idelalisib with rituximab. These results
showed a 97% ORR with 19% CR and 83% PFS at 36 months.64
Among nine patients with del(17p)/TP53 mutation, the ORR
and PFS were both 100%. However, toxicity was also higher,
with rates of 42% for grade 3 or higher diarrhea with longterm follow-up and 23% for grade 3-4 transaminitis. These
higher rates of toxicity appear consistent with an immunologic mechanism and recently reported high rates of grade
3-4 transaminitis among younger patients treated up front
with idelalisib.60 Randomized registration trials in previously
untreated patients were recently stopped due to excess
infectious deaths on the idelalisib arms, albeit without
mandatory prophylaxis.
Other PI3K Inhibitors

Two other PI3K inhibitors are in advanced clinical development. Duvelisib is a dual PI3K delta and gamma inhibitor.
The gamma isoform of PI3K is expressed in CLL cells but also
in cells of the microenvironment, including T cells and
macrophages.65 The phase I study of duvelisib enrolled 55
patients with CLL with a median of four prior regimens, 89%
unmutated IGHV, and 52% with del(17p) or TP53 mutation.66
The ORR was 58% with an estimated 24-month PFS of 59%.
The recommended phase II dose was selected as 25 mg two
times per day, which reaches the IC90 for delta and the IC50
for gamma inhibition in vivo. A phase III registration trial in
relapsed CLL that randomly assigned patients to duvelisib or
ofatumumab has now completed accrual. TGR-1202 is a
next-generation PI3K delta inhibitor that has a different
chemistry and has been postulated to reduce the transaminitis seen with other delta inhibitors. A phase I study
using TGR-1202 enrolled patients across multiple B-cell

malignancies, including 20 patients with CLL.67 Among the
patients with CLL, the nodal response rate was 88% (14/16)
and the PR rate was 63%. Thus far, the rates of grade 3-4
transaminitis (2%) and colitis (1%) have been much lower
than with the other delta inhibitors, but follow-up remains
shorter. TGR-1202 is being further evaluated in many ongoing phase I and II combination trials, with recent initiation
of a phase III registration trial.
Few studies of pan-PI3K inhibitors have been performed in CLL. Pilaralisib has been reported to have a
similar pattern of response as delta inhibitors and a 50%
ORR in CLL, with PFS of 7.4 to 22 months and a 20% overall
response in lymphoma.68 This level of activity may not
be as high as the delta-specific inhibitors described
above. Voxtalisib had activity in follicular lymphoma but
was limited in other histologies. 69 This more limited
activity may reflect the pharmacologic properties of
these compounds in comparison with the delta-specific
inhibitors, so additional studies of pan-PI3K inhibitors with better pharmacodynamic properties are still
warranted.
BCL-2 Inhibition
The BCL-2 specific inhibitor venetoclax is expected to receive
FDA approval in the first half of 2016. This drug represents
the culmination of 2 decades of structure-based design
targeted at BCL-2. Its predecessor navitoclax had clinical
activity in CLL70 but showed dose-limiting thrombocytopenia
because of the direct effect of navitoclax on BCL-XL in circulating platelets, leading to their rapid clearance from the
circulation.71 Venetoclax only has activity against BCL-2 and
has not shown significant clinical thrombocytopenia.72 The
most substantial adverse event with venetoclax has been
tumor lysis syndrome, which led to several revisions of the
dosing schema during the phase I study. Ultimately, the
selected dosing schedule that has moved forward into all
other trials includes a risk stratification for tumor lysis that
dictates whether patients are admitted to the hospital for
first dose and starts all patients at a low dose of 20 mg daily
for a week, with increases to 50 mg, 100 mg, 200 mg, and
eventually to 400 mg daily, which is the recommended phase
II dose for patients with CLL.72 The phase I and expansion
cohort study enrolled 116 patients with a median of four
prior regimens and showed an ORR of 79% with 20% CR.
Median PFS was 25 months in the dose-escalation cohorts.
The CR rate suggests that venetoclax can more effectively
clear blood and bone marrow than the kinase inhibitors, and
this observation has been further supported by data from
the venetoclax combination study with rituximab, which
enrolled less heavily pretreated patients with a median of
two prior regimens.73 The overall response rate was 86%
with 47% CR, and, perhaps most impressively, the estimated
24-month PFS was 83%. Furthermore, half of both CR and PR
were minimal residual disease (MRD)negative in bone
marrow, for an overall rate of 55%.
Venetoclax has also been investigated in very high-risk CLL

groups. A phase II potential registration study focused on
patients with relapsed disease with del(17p) who received
continuous single-agent venetoclax. The study enrolled 107
patients with a median of two prior regimens, and PFS at
12 months was estimated at 72% with OS estimated at 87%.
These data appear comparable to those of ibrutinib in a
similar study.38
UNANSWERED QUESTIONS
Despite the seminal advances seen to date with these novel
agents in CLL therapy, many practical questions remain. The
appropriate sequencing of these agents remains in question,
and the utility of class switching after drug failure has not
been fully elucidated. Questions also persist about the
unknown ability to stop one of these novel agents, particularly in more favorable, less-heavily pretreated patients
who may not require indefinite long-term therapy. At the
other end of the disease spectrum, will these agents be
meaningful options for patients with very aggressive disease, such as those with Richter transformation? Or will use
of these drugs result in increased rates of aggressive Richter
transformation? Recently there has also been keen interest
in combinations of targeted agents with standard chemoimmunotherapeutic regimens. Will these combination approaches be critical to overcome drug resistance and lead to
major improvements in survival? What about combinations
of two (or even more) targeted agents? Furthermore, how
will these important advances translate from the major
academic centers, where there has been the vast majority of
experience with these agents, to the community setting and
standard clinical practice? Lastly, given this new landscape of
exciting agents, where do we use cellular therapies and
potentially curative stem cell transplantation as we move
forward? We should recognize, moreover, that follow-up
observation after beginning use with a kinase inhibitor has
been short relative to their projected duration of therapy,
and, thus, the full toxicity profiles of these agents have not
yet become completely known.
Can We Stop Therapy With the New Targeted Agents?

How long should patients remain on therapy with a kinase or
BCL-2 inhibitor? Although patients universally favor the use
of an oral regimen, the use of these agents indefinitely may
be less appealing to a large proportion of patients, especially
those who are treatment-naive or relatively early in their
treatment encounters. We previously acknowledged that
many patients with high-risk disease who have failed multiple lines of therapy will relapse very quickly after stopping
one of the kinase inhibitors. However, for those who may
have lower-risk disease and particularly those who might be
receiving single agents as first-line therapy, can, or should,
we stop therapy at some point and observe? Specifically, is
there a clinical marker such as the absence of any evidence of
MRD that would suggest appropriate discontinuation of a
novel drug therapy?
e393
Stopping therapy has the advantage of reducing toxicity

and reducing the selective pressure for resistance. It has
become widely accepted that assessment of MRD can be a
highly sensitive measure of disease burden, especially following CLL-directed chemoimmunotherapy, and has been
commonly defined as having less than one CLL cell in 10,000
leukocytes using standard four-color flow cytometric analysis.1 Compared with patients with MRD-positive disease,
achieving MRD negativity after treatment correlates with
longer disease-free intervals.74 If a patients disease becomes MRD negative, would this result allow for discontinuation of the targeted drug therapy? Data addressing this
question are limited, but Ma et al presented a 2015 ASH
report in a small number of patients using venetoclax in
combination with rituximab where patients were permitted
to stop the BCL-2 inhibitor at any point desired. The data
demonstrated that 27 of 49 treated patients became MRD
negative in the bone marrow, and nine of those patients
stopped venetoclax between 10 and 36 months after beginning
therapy.73 All nine patients with MRD-negative disease remain
in CR with a median follow-up of 16 months. Of note, two
patients in CR but who had evidence of MRD in the marrow also
stopped treatment between 5 and 9 months, but both had
asymptomatic progression while off of therapy. Interestingly,
at least one of these patients responded to rechallenge with
the same regimen. These data clearly set the foundation for
early stopping studies and in particular for patients with
favorable-risk disease, although it must be noted that venetoclax has been much more potent in inducing MRD negativity than the kinase inhibitors that may require combination
therapy to achieve appreciable rates of MRD negativity.
Can Further Advances Be Made With Combinations of

Therapies Using Targeted Agents?
Combining established therapies with new complementary,
perhaps synergistic, approaches to improve response rates
and deepen the quality of remissions remains a common
theme in oncology care. The promise of inducing more CR or
perhaps becoming MRD negative may translate to longer
durations of remission and improvements in survival. The
conclusions are not clear, but the rationale may be solid,
albeit primarily based on preclinical models. For example,
venetoclax was demonstrated in lymphoma xenograft
models to significantly control tumor growth and provide
synergy when combined with BR.71 In clinical practice,
several single-arm combination studies have been reported
with these novel agents and anti-CD20 antibodies, including
rituximab and ofatumumab or with chemoimmunotherapy,
particularly BR. The HELIOS trial randomly assigned patients
with relapsed CLL without del(17p) to BR versus BR with
ibrutinib and reported a significant improvement in PFS (HR
0.20; p , .0001) in the ibrutinib arm, but no difference in
OS.75 The CR rate was 21% by investigator assessment in
the investigational arm as compared with 5.9% with BR
alone. Whether BR with ibrutinib will prove to be significantly better than ibrutinib alone remains unclear and may
e394
ultimately depend on whether the increased early CR rate

translates into improved PFS.
With the use of new targeted agents and combinations
with cytotoxic regimens such as BR, however, it is well
understood that there are increased associated treatmentrelated toxicities. Perhaps a chemotherapy-free regimen,
even in the front-line setting, may be an attractive option for
patients who are older and who may have multiple comorbidities. The long-term complication of developing an MDS and
acute leukemia from the use of cytotoxic chemotherapy regimens must also be considered. Therefore, specifically in
patients with a low to moderate tumor burden and nonlifethreatening CLL, the improved responses and important
advances with combination therapies must result in large
improvements in long-term survival; otherwise, a more aggressive regimen may not be warranted. Studies of combinations of two novel agents have just begun, so neither the
efficacy nor toxicity of those regimens is yet known. Whether
concurrent combination therapy truly represents an advancement over sequential therapy with the multiple novel
targeted agents discussed here remains unclear, although it
has the theoretical benefit of reducing selection pressure for
resistance and may increase the rate of MRD negativity,
thereby allowing consideration of stopping therapy.
How Do We Sequence These Novel Agents?

Is switching drug class required or is an alternative kinase
inhibitor appropriate when a patient fails a targeted agent?
Recent data suggest that patients who discontinue a targeted agent for adverse events have better outcomes than
those who progress, but because of the rapid evolution of
the targeted agents, little data are available addressing
approaches to sequencing these therapeutic agents once a
patient may have failed treatment. Limited data exist regarding practice patterns following ibrutinib or idelalisib
discontinuation. The response to subsequent therapies is
also largely unknown, particularly outside the context of
clinical research. A phase II, open-label, two-arm study
evaluating venetoclax monotherapy for patients with CLL
that had relapsed after or was refractory to either ibrutinib
or idelalisib, was presented at the 2015 ASH Annual
Meeting.76 In this early report, 41 patients had failed prior
ibrutinib, and 13 failed treatment with idelalisib, with three
patients who had failed both of these kinase inhibitors.
Approximately one-third of patients discontinued treatment
because of kinase inhibitor intolerance. About 8 weeks after
starting venetoclax, the majority of patients had resolution
of B-symptoms, with approximately 80% of patients having
normalization of their absolute lymphocyte count, which
occurred quickly at a median less than 1 month after starting
the BCL-2 inhibitor. In addition, a 50% or greater reduction in
nodal mass was achieved in almost three-quarters of patients after switching drug class. Overall, venetoclax monotherapy demonstrated an ORR of 61% in the ibrutinib
cohort and 50% in the idelalisib cohort, albeit with small
numbers of patients studied to date. Importantly, responses
were similar for patients who discontinued kinase inhibitors

for refractory disease compared with intolerance. Clearly
continued treatment and extended follow-up will be required to fully determine response rates and durability of
responses after class switching.
Can switching to an alternative kinase inhibitor provide
efficacy after failure from the first kinase inhibitor? A
multicenter retrospective cohort analysis among patients
with CLL who discontinued ibrutinib or idelalisib therapy was
also recently reported at the 2015 ASH Annual Meeting.77 A
total of 143 patients with CLL who had previously taken
ibrutinib and another 35 who had failed idelalisib switched
to the alternate kinase inhibitor. Two-thirds of these patients had unmutated IGHV and more than one-third had
del(17p). More than 50% who failed either drug discontinued the kinase inhibitor because of toxicity, and
another one-third of patients stopped therapy because of
disease progression, while 6%8% discontinued because of
Richter transformation. The ORR was 50% (all PR) after
switching to the alternative kinase inhibitor. Therefore,
patients who discontinue a kinase inhibitor because of
toxicity can achieve a durable response with an alternate
kinase inhibitor, although one shorter than the response
durations reported when these drugs are used first; the
median PFS from the start of the alternative kinase inhibitor was 11.9 months, and the alternative kinase inhibitor choice did not affect PFS. Importantly the median
PFS was only 7 months for those who had failed the first
kinase inhibitor because of CLL progression, while a median
PFS was not reached after 6 months of median follow-up
for those who failed because of kinase inhibitor intolerance. These data suggest that turning to an alternate
kinase inhibitor therapy following kinase inhibitor discontinuation can be efficacious, particularly for patients
who discontinue kinase inhibitor because of intolerance. A
clinical trial targeting the kinase inhibitorintolerant patient population will be undertaken to validate these
findings prospectively.
Will This New Area of Targeted Therapy Lead to the

Development of an Increased Rate of Richter
Syndrome?
An obvious unmet need remains how to treat Richter
transformation, which occurs in about 5%10% of patients
with CLL and transforms into a fast-growing lymphoma,
typically diffuse large B-cell lymphoma.78 It is well understood that Richter transformation carries a very poor
prognosis and is commonly highly refractory to standard
treatments. Unfortunately, there is a paucity of data to help
understand how these new targeted drugs will impact the
incidence of Richter transformation. At the 2015 ASH Annual
Meeting, the first report of outcomes after failure of venetoclax was reported in 28 patients with CLL who had a
median of four prior regimens and 59% with deletion of
17p.79 Discontinuations occurred after a median 7.5 months
on venetoclax, with a median 12-month follow-up after
discontinuation. Reasons for discontinuation included 57%

with Richter, 25% with CLL progression, and 18% with other
events. The high rate of Richter appears reminiscent of the
reports to date of patients progressing on ibrutinib where
there is also a 30%50% rate of transformation.41,43
Whether these Richter events are the natural history of the
disease, if we are able to keep patients alive long enough or
they represent a biologic evolution under strong selection
pressure from the drugs remains to be seen. Survivals reported with venetoclax were somewhat better than previously reported with ibrutinib, however, with 1-year OS of
69% for those progressing with CLL and a median 1-year OS
for those progressing with Richter.
Will Novel Immunotherapies Overcome Resistance in

Patients With CLL?
Treatment of relapsed disease continues to be clinically
challenging, primarily because of eventual drug resistance.
There has been tremendous recent interest in novel immunotherapeutic approaches to treat patients with advanced disease and overcome resistance. Blocking the
immune checkpoint PD-1 receptor to activate cytotoxic
T cells has been shown to have remarkable clinical results
in a variety of advanced malignancies but as yet limited
evaluation in CLL.80-82 PD-1 has been shown to be highly
expressed in CLL cells and is also expressed in the CLL microenvironment.83 Preclinical data have demonstrated that
blocking PD-L1 in Tcl-1Tg mice can delay CLL progression.84
Data from an early efficacy report from a phase II trial
presented at the 2015 ASH Annual Meeting explored PD-1
blockade in 20 patients with relapsed/refractory CLL, including seven with Richter transformation, of which five had
progressed after ibrutinib therapy.85 The ORR was limited at
about 25%, yet interestingly, one patient with Richter
achieved a CR and two achieved a PR, suggesting that further
assessment of PD-1 blockade may be appropriate for patients with highly aggressive disease and probably drug
resistance.
In addition, the use of CAR T cells has recently been investigated in an attempt to provide long-term immune
surveillance and decrease relapse in aggressive B-cell malignancies.86 Briefly, the CAR has been introduced into T cells
by genetic modification designed to redirect T-cell specificity
to the CAR-targeted antigen. In particular, CD19-specific
CAR-modified T cells have shown activity in aggressive
B-cell lymphoma and leukemia, although toxicity can be
high. Currently, however, given the other options in CLL, CAR
T-cell therapy remains an investigational approach that
should be limited to patients with aggressive relapsed, refractory disease.
CONCLUSION
Many unresolved and controversial issues remain with
regard to the treatment of patients with CLL, including the
identification of appropriate candidates for each category of
therapy. Patients with low-risk IGHV-mutated disease in the
e395
front-line setting derive long-term benefit from chemoimmunotherapy (e.g., FCR). Chemoimmunotherapy in general has the benefit of planned short duration therapy with
long remission, in contrast to the targeted agents that to
date have largely been studied as continuous single agents
or continuously in combination with chemoimmunotherapy.
Randomized trials are comparing targeted agents with
chemoimmunotherapy, particularly in the front-line setting,
while in the relapse setting, targeted agents have been
preferred for most patients, although it remains unclear
whether single-agent therapy or combinations are optimal.
It appears likely that patients of all stages in disease evolution may benefit from targeted kinase or BCL-2 inhibitor
agents, but questions regarding the use of these agents with
regard to sequencing, class switching, length of therapy, and
how to avoid or overcome resistance remain at the forefront. Nonetheless, the exciting clinical trial data with these
rational targeted therapies will soon translate to clinical
practice experience and set the stage for continued profound improvements in outcomes for patients with CLL, with
major promise for eventual long-term improvements in
survival.
References
1. Hallek M, Cheson BD, Catovsky D, et al; International Workshop on
Chronic Lymphocytic Leukemia. Guidelines for the diagnosis and
treatment of chronic lymphocytic leukemia: a report from the International Workshop on Chronic Lymphocytic Leukemia updating the
National Cancer Institute-Working Group 1996 guidelines. Blood. 2008;
111:5446-5456.
2. Chemotherapeutic options in chronic lymhocytic leukemia. CLL Trialists
Collaborative Group. J Natl Cancer Inst. 1999;91:861-868.
3. Langerbeins P, Gro-Ophoff-Muller C, Herling CD. Risk-adapted therapy in
early-stage chronic lymphocytic leukemia. Oncol Res Treat. 2016;39:18-24.
4. Dreger P. Allotransplantation for chronic lymphocytic leukemia. ASH
Education Book. 2009;2009:602-609.
5. Hallek M, Fischer K, Fingerle-Rowson G, et al; International Group of
Investigators; German Chronic Lymphocytic Leukaemia Study Group.
Addition of rituximab to fludarabine and cyclophosphamide in patients
with chronic lymphocytic leukaemia: a randomised, open-label, phase 3
trial. Lancet. 2010;376:1164-1174.
6. Fischer K, Cramer P, Busch R, et al. Bendamustine in combination with
rituximab for previously untreated patients with chronic lymphocytic
leukemia: a multicenter phase II trial of the German Chronic Lymphocytic Leukemia Study Group. J Clin Oncol. 2012;30:3209-3216.
7. Goede V, Fischer K, Busch R, et al. Obinutuzumab plus chlorambucil in
patients with CLL and coexisting conditions. N Engl J Med. 2014;370:
1101-1110.
8. Fischer K, Bahlo J, Fink AM, et al. Long-term remission after FCR
chemotherapy in previously untreated patients with CLL: updated
results of the CLL8 trial. Blood. 2016;127:208-215.
9. Dohner H, Stilgenbauer S, Benner A, et al. Genomic aberrations and
survival in chronic lymphocytic leukemia. N Engl J Med. 2000;343:
1910-1916.
10. Zenz T, Eichhorst B, Busch R, et al. TP53 mutation and survival in chronic
lymphocytic leukemia. J Clin Oncol. 2010;28:4473-4479.
11. Pflug N, Bahlo J, Shanafelt TD, et al. Development of a comprehensive
prognostic index for patients with chronic lymphocytic leukemia. Blood.
2014;124:49-62.
12. Ghia EM, Jain S, Widhopf GF II, et al. Use of IGHV3-21 in chronic
lymphocytic leukemia is associated with high-risk disease and reflects
antigen-driven, post-germinal center leukemogenic selection. Blood.
2008;111:5101-5108.
13. Rassenti LZ, Jain S, Keating MJ, et al. Relative value of ZAP-70, CD38, and
immunoglobulin mutation status in predicting aggressive disease in
chronic lymphocytic leukemia. Blood. 2008;112:1923-1930.
14. Stilgenbauer S, Schnaiter A, Paschka P, et al. Gene mutations and
treatment outcome in chronic lymphocytic leukemia: results from the
CLL8 trial. Blood. 2014;123:3247-3254.
e396
15. Hallek M, Wanders L, Ostwald M, et al. Serum beta(2)-microglobulin

and serum thymidine kinase are independent predictors of progressionfree survival in chronic lymphocytic leukemia and immunocytoma. Leuk
Lymphoma. 1996;22:439-447.
16. Hamblin TJ, Orchard JA, Ibbotson RE, et al. CD38 expression and immunoglobulin variable region mutations are independent prognostic
variables in chronic lymphocytic leukemia, but CD38 expression may
vary during the course of the disease. Blood. 2002;99:1023-1029.
17. Thompson PA, Tam CS, OBrien SM, et al. Fludarabine, cyclophosphamide, and rituximab treatment achieves long-term disease-free
survival in IGHV-mutated chronic lymphocytic leukemia. Blood. 2016;
127:303-309.
18. Rossi D, Terzi-di-Bergamo L, De Paoli L, et al. Molecular prediction of
durable remission after first-line fludarabine-cyclophosphamiderituximab in chronic lymphocytic leukemia. Blood. 2015;126:
1921-1924.
19. Hallek M. Signaling the end of chronic lymphocytic leukemia: new
frontline treatment strategies. Blood. 2013;122:3723-3734.
20. Cramer P, Hallek M, Eichhorst B. State-of-the-art treatment and novel
agents in chronic lymphocytic leukemia. Oncol Res Treat. 2016;39:
25-32.
21. Benjamini O, Jain P, Trinh L, et al. Second cancers in patients with
chronic lymphocytic leukemia who received frontline fludarabine, cyclophosphamide and rituximab therapy: distribution and clinical outcomes. Leuk Lymphoma. 2015;56:1643-1650.
22. Strati P, Wierda W, Burger J, et al. Myelosuppression after frontline
fludarabine, cyclophosphamide, and rituximab in patients with chronic
lymphocytic leukemia: analysis of persistent and new-onset cytopenia.
Cancer. 2013;119:3805-3811.
23. Gill S, Carney D, Ritchie D, et al. The frequency, manifestations, and
duration of prolonged cytopenias after first-line fludarabine combination chemotherapy. Ann Oncol. 2010;21:331-334.
24. Tam CS, OBrien S, Wierda W, et al. Long-term results of the fludarabine,
cyclophosphamide, and rituximab regimen as initial therapy of chronic
lymphocytic leukemia. Blood. 2008;112:975-980.
25. Woyach JA, Ruppert AS, Heerema NA, et al. Chemoimmunotherapy with
fludarabine and rituximab produces extended overall survival and
progression-free survival in chronic lymphocytic leukemia: long-term
follow-up of CALGB study 9712. J Clin Oncol. 2011;29:1349-1355.
26. Langerbeins P. Impact of first-line therapy on secondary malignancies in
CLL: a meta-analysis of four prospective trials of the German CLL Study
Group. Presented at: XVth International Workshop on CLL (IwCLL);
September 2013; Sydney, Australia.
27. Cramer P, Hallek M. Prognostic factors in chronic lymphocytic leukemiawhat do we need to know? Nat Rev Clin Oncol. 2011;8:38-47.
28. Extermann M, Overcash J, Lyman GH, et al. Comorbidity and functional

status are independent in older cancer patients. J Clin Oncol. 1998;16:
1582-1587.
29. Byrd JC, Furman RR, Coutre SE, et al. Targeting BTK with ibrutinib in
relapsed chronic lymphocytic leukemia. N Engl J Med. 2013;369:32-42.
30. Burger JA, Keating MJ, Wierda WG, et al. Safety and activity of ibrutinib
plus rituximab for patients with high-risk chronic lymphocytic leukaemia: a single-arm, phase 2 study. Lancet Oncol. 2014;15:1090-1099.
31. Dreger P, Schetelig J, Andersen N, et al; European Research Initiative on
CLL (ERIC) and the European Society for Blood and Marrow Transplantation (EBMT). Managing high-risk CLL during transition to a new
treatment era: stem cell transplantation or novel agents? Blood. 2014;
124:3841-3849.
32. Advani RH, Buggy JJ, Sharman JP, et al. Bruton tyrosine kinase inhibitor
ibrutinib (PCI-32765) has significant activity in patients with relapsed/
refractory B-cell malignancies. J Clin Oncol. 2013;31:88-94.
33. Honigberg LA, Smith AM, Sirisawad M, et al. The Bruton tyrosine kinase
inhibitor PCI-32765 blocks B-cell activation and is efficacious in models
of autoimmune disease and B-cell malignancy. Proc Natl Acad Sci USA.
2010;107:13075-13080.
34. Dubovsky JA, Beckwith KA, Natarajan G, et al. Ibrutinib is an irreversible
molecular inhibitor of ITK driving a Th1-selective pressure in
T lymphocytes. Blood. 2013;122:2539-2549.
35. Kohrt HE, Sagiv-Barfi I, Rafiq S, et al. Ibrutinib antagonizes rituximabdependent NK cell-mediated cytotoxicity. Blood. 2014;123:1957-1960.
36. Cheson BD, Byrd JC, Rai KR, et al. Novel targeted agents and the need to
refine clinical end points in chronic lymphocytic leukemia. J Clin Oncol.
2012;30:2820-2822.
37. Byrd JC, Brown JR, OBrien S, et al; RESONATE Investigators. Ibrutinib
versus ofatumumab in previously treated chronic lymphoid leukemia.
N Engl J Med. 2014;371:213-223.
38. OBrien S, Jones JA, Coutre S, et al. Efficacy and safety of ibrutinib in
patients with relapsed or refractory chronic lymphocytic leukemia or
small lymphocytic leukemia with 17p deletion: results from the phase II
RESONATE-17 trial. Blood. 2014;124 (abstr 327).
39. Byrd JC, Furman RR, Coutre SE, et al. Three-year follow-up of treatmentnave and previously treated patients with CLL and SLL receiving singleagent ibrutinib. Blood. 2015;125:2497-2506.
40. Brown JR, Hillmen P, OBrien S, et al. Updated efficacy including genetic
and clinical subgroup analysis and overall safety in the phase 3 RESONATE trial of ibrutinib versus ofatumumab in previously treated
chronic lymphocytic leukemia/small lymphocytic lymphoma. Blood.
2014;124 (abstr 3331).
41. Maddocks KJ, Ruppert AS, Lozanski G, et al. Etiology of ibrutinib therapy
discontinuation and outcomes in patients with chronic lymphocytic
leukemia. JAMA Oncol. 2015;1:80-87.
42. Thompson PA, OBrien SM, Wierda WG, et al. Complex karyotype is a
stronger predictor than del(17p) for an inferior outcome in relapsed or
refractory chronic lymphocytic leukemia patients treated with
ibrutinib-based regimens. Cancer. 2015;121:3612-3621.
43. Jain P, Keating M, Wierda W, et al. Outcomes of patients with chronic
lymphocytic leukemia after discontinuing ibrutinib. Blood. 2015;125:
2062-2067.
44. Woyach JA, Furman RR, Liu TM, et al. Resistance mechanisms for the
Brutons tyrosine kinase inhibitor ibrutinib. N Engl J Med. 2014;370:
2286-2294.
45. Albitar A, Ma W, De Dios I, et al. High sensitivity testing shows multiclonal mutations in patients with CLL treated with BTK inhibitor and
lack of mutations in ibrutinib-naive patients. Blood. 2015;126 (abstr
716).
46. Kamel S, Horton L, Ysebaert L, et al. Ibrutinib inhibits collagen-mediated
but not ADP-mediated platelet aggregation. Leukemia. 2014:1-5.
47. Levade M, David E, Garcia C, et al. Ibrutinib treatment affects collagen

and von Willebrand factor-dependent platelet functions. Blood. 2014;
124:3991-3995.
48. OBrien S, Furman RR, Coutre SE, et al. Ibrutinib as initial therapy for
elderly patients with chronic lymphocytic leukaemia or small lymphocytic lymphoma: an open-label, multicentre, phase 1b/2 trial.
Lancet Oncol. 2014;15:48-58.
49. Burger JA, Tedeschi A, Barr PM, et al; RESONATE-2 Investigators.
Ibrutinib as initial therapy for patients with chronic lymphocytic leukemia. N Engl J Med. 2015;373:2425-2437.
50. Harb WA, Hill BT, Gabrilove J, et al. Phase 1 study of single agent CC292, a highly selective Brutons tyrosine kinase (BTK) inhibitor, in
relapsed/refractory chronic lymphocytic leukemia (CLL). Blood. 2013;
122:1630.
51. Byrd JC, Harrington B, OBrien S, et al. Acalabrutinib (ACP-196) in relapsed chronic lymphocytic leukemia. N Engl J Med. 2016;374:323-332.
52. Tam C, Grigg AP, Opat S, et al. The BTK inhibitor, Bgb-3111, is safe,
tolerable, and highly active in patients with relapsed/refractory B-cell
malignancies: initial report of a phase 1 first-in-human trial. Blood.
2015;126:832.
53. Vanhaesebroeck B, Ali K, Bilancio A, et al. Signalling by PI3K isoforms:
insights from gene-targeted mice. Trends Biochem Sci. 2005;30:
194-204.
54. Herman SE, Gordon AL, Wagner AJ, et al. Phosphatidylinositol 3-kinased inhibitor CAL-101 shows promising preclinical activity in chronic
lymphocytic leukemia by antagonizing intrinsic and extrinsic cellular
survival signals. Blood. 2010;116:2078-2088.
55. Lannutti BJ, Meadows SA, Herman SE, et al. CAL-101, a p110delta
selective phosphatidylinositol-3-kinase inhibitor for the treatment of
B-cell malignancies, inhibits PI3K signaling and cellular viability. Blood.
2011;117:591-594.
56. Brown JR, Byrd JC, Coutre SE, et al. Idelalisib, an inhibitor of phosphatidylinositol 3-kinase p110d, for relapsed/refractory chronic lymphocytic leukemia. Blood. 2014;123:3390-3397.
57. Sharman JP, Coutre SE, Furman RR, et al. Second interim analysis of a
phase 3 study of idelalisib (ZYDELIG) plus rituximab (R) for relapsed
chronic lymphocytic leukemia (CLL): efficacy analysis in patient subpopulations with Del(17p) and other adverse prognostic factors. Blood.
2014;124 (abstr 33).
58. Jones JA, Wach M, Robak T, et al. Results of a phase III randomized,
controlled study evaluating the efficacy and safety of idelalisib (IDELA)
in combination with ofatumumab (OFA) for previously treated chronic
lymphocytic leukemia (CLL). J Clin Oncol. 2015;33:15 (suppl; abstr 7023).
59. Zelenetz AD, Robak T, Coiffier B, et al. Idelalisib plus bendamustine and
rituximab (BR) is superior to BR alone in patients with relapsed/
refractory chronic lymphocytic leukemia: results of a phase 3 randomized double-blind placebo-controlled study. Blood. 2015;126 (abstr
LBA-5).
60. Lampson BL, Matos T, Kim HT, et al. Idelalisib given front-line for the
treatment of chronic lymphocytic leukemia results in frequent and
severe immune-mediated toxicities. Blood. 2015;126:497-497.
61. Louie CY, DiMaio MA, Matsukuma KE, et al. Idelalisib-associated enterocolitis: clinicopathologic features and distinction from other
enterocolitides. Am J Surg Pathol. 2015;39:1653-1660.
62. Weidner AS, Panarelli NC, Geyer JT, et al. Idelalisib-associated colitis:
histologic findings in 14 patients. Am J Surg Pathol. 2015;39:1661-1667.
63. Patton DT, Garden OA, Pearce WP, et al. Cutting edge: the phosphoinositide 3-kinase p110 delta is critical for the function of
CD4+CD25+Foxp3+ regulatory T cells. J Immunol. 2006;177:6598-6602.
64. OBrien SM, Lamanna N, Kipps TJ, et al. A phase 2 study of idelalisib plus
rituximab in treatment-nave older patients with chronic lymphocytic
leukemia. Blood. 2015;126:2686-2694.
e397
65. Douglas M, Allen K, Sweeney J, et al. Serum chemokines and cytokines in

CLL patients treated with duvelisib, a PI3K-{delta},{gamma} inhibitor.
66. OBrien SM, Patel M, Kahl BS, et al. Duvelisib (IPI-145), a PI3K-d,g inhibitor, is clinically active in patients with relapsed/refractory chronic
lymphocytic leukemia. Blood. 2014;124 (abstr 334).
67. Burris HA, Patel MR, Fenske TS, et al. TGR-1202, a novel once daily PI3Kdelta inhibitor, demonstrates clinical activity with a favorable safety
profile, lacking hepatotoxicity, in patients with CLL and B-cell lymphoma. Haematologica. 2015;100:154.
68. Brown JR, Davids MS, Rodon J, et al. Phase I trial of the pan-PI3K inhibitor pilaralisib (SAR245408/XL147) in patients with chronic lymphocytic leukemia (CLL) or relapsed/refractory lymphoma. Clin Cancer
Res. 2015;21:3160-3169.
69. Brown JR, Hamadani M, Arnason J, et al. SAR245409 monotherapy in
relapsed/refractory follicular lymphoma: preliminary results from the
phase II ARD12130 study. Blood. 2013;122 (abstr 86).
70. Roberts AW, Seymour JF, Brown JR, et al. Substantial susceptibility of
chronic lymphocytic leukemia to BCL2 inhibition: results of a phase I
study of navitoclax in patients with relapsed or refractory disease. J Clin
Oncol. 2012;30:488-496.
71. Souers AJ, Leverson JD, Boghaert ER, et al. ABT-199, a potent and
selective BCL-2 inhibitor, achieves antitumor activity while sparing
platelets. Nat Med. 2013;19:202-208.
72. Roberts AW, Davids MS, Pagel JM, et al. Targeting BCL2 with venetoclax in
relapsed chronic lymphocytic leukemia. N Engl J Med. 2016;374:311-322.
73. Ma S, Brander DM, Seymour JF, et al. Deep and durable responses
following venetoclax (ABT-199/GDC-0199) combined with rituximab in
patients with relapsed/refractory chronic lymphocytic leukemia: results
from a phase 1b study. Blood. 2015;126 (abstr 830).
74. Moreton P, Kennedy B, Lucas G, et al. Eradication of minimal residual
disease in B-cell chronic lymphocytic leukemia after alemtuzumab
therapy is associated with prolonged survival. J Clin Oncol. 2005;23:
2971-2979.
75. Chanan-Khan A, Cramer P, Demirkan F, et al. Ibrutinib combined with
bendamustine and rituximab (BR) in previously treated chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL): First results
from a randomized, double-blind, placebo-controlled, phase III study. J
Clin Oncol. 2015;33:15s (suppl; LBA7005).
76. Jones J, Mato AR, Coutre S, et al. Preliminary results of a phase 2, openlabel study of venetoclax (ABT-199/GDC-0199) monotherapy in patients
with chronic lymphocytic leukemia relapsed after or refractory to
ibrutinib or idelalisib therapy. Blood. 2015;126 (abstr 715).
77. Mato A, Nabhan C, Barr PM, et al. Favorable outcomes in CLL pts
with alternate kinase inhibitors following ibrutinib or idelalisib
e398
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
discontinuation: results from a large multi-center study. Blood.

2015;126 (abstr 719).
Jain P, OBrien S. Richters transformation in chronic lymphocytic
leukemia. Oncology (Williston Park). 2012;26:1146-1152.
Tam C, Anderson MA, Ritchie DS, et al. Favorable patient survival after
failure of venetoclax (ABT-199/ GDC-0199) therapy for relapsed or
refractory chronic lymphocytic leukemia (CLL). Blood. 2015;126 (abstr
2939).
Ansell SM, Lesokhin AM, Borrello I, et al. PD-1 blockade with nivolumab
in relapsed or refratory Hodgkins lymphoma. N Engl J Med. 2015;372:
311-319.
Hamid O, Robert C, Daud A, et al. Safety and tumor responses with
lambrolizumab (anti-PD-1) in melanoma. N Engl J Med. 2013;369:
134-144.
Grzywnowicz M, Zaleska J, Mertens D, et al. Programmed death-1 and
its ligand are novel immunotolerant molecules expressed on leukemic
B cells in chronic lymphocytic leukemia. PLoS One. 2012;7:e35178.
Riches JC, Davies JK, McClanahan F, et al, T cells from CLL patients
exhibit features of T-cell exhaustion but retain capacity for cytokine
production. Blood. 2013;121:1612-1621.
McClanahan F, Riches JC, Miller S, et al. Mechanisms of PD-L1/PD-1mediated CD8 T-cell dysfunction in the context of aging-related immune
defects in the Em-TCL1 CLL mouse model. Blood. 2015;126:212-221.
Ding W, Dong H, Call T, et al. PD-1 blockade with pembrolizumab (MK3475) in relapsed/refractory CLL including Richter transformation: an
early efficacy report from a phase 2 trial (MC1485). Blood. 2015;126
(abstr 834).
Turtle CJ, Hudecek M, Jensen MC, et al. Engineered T cells for anticancer therapy. Curr Opin Immunol. 2012;24:633-639.
Hillmen P, Robak T, Janssens A, et al; COMPLEMENT 1 Study Investigators. Chlorambucil plus ofatumumab versus chlorambucil alone
in previously untreated patients with chronic lymphocytic leukaemia
(COMPLEMENT 1): a randomised, multicentre, open-label phase 3 trial.
Lancet. 2015;385:1873-1883.
Wierda WG, Kipps TJ, Mayer J, et al; Hx-CD20-406 Study Investigators.
Ofatumumab as single-agent CD20 immunotherapy in fludarabinerefractory chronic lymphocytic leukemia. J Clin Oncol. 2010;28:
1749-1755.
Byrd JC, Pagel JM, Awan FT, et al. A phase 1 study evaluating the safety
and tolerability of otlertuzumab, an anti-CD37 mono-specific ADAPTIR
therapeutic protein in chronic lymphocytic leukemia. Blood. 2014;123:
1302-1308.
Robak T, Hellmann A, Kloczko J, et al. Randomized phase 2 study of
otlertuzumab (TRU-016) and bendamustine vs bendamustine in patients with relapsed chronic lymphocytic leukemia. In press.
HEMATOLOGIC MALIGNANCIESPLASMA CELL

DYSCRASIA
Consolidation and Maintenance:

Moving Beyond Autotransplant
CHAIR
Parameswaran Hari, MD, MRCP
Medical College of Wisconsin
Milwaukee, WI
SPEAKERS
Amrita Krishnan, MD
City of Hope
Duarte, CA
Ravi Vij, MBBS, MD
Washington University School of Medicine
St. Louis, MO
KRISHNAN ET AL
Moving Beyond Autologous Transplantation in Multiple

Myeloma: Consolidation, Maintenance, Allogeneic
Transplant, and Immune Therapy
Amrita Krishnan, MD, Ravi Vij, MD, MBA, Jesse Keller, MD, Binod Dhakal, MD, and Parameswaran Hari,
MD, MRCP
OVERVIEW
For multiple myeloma, introduction of novel agents as part of the front-line treatment followed by high-dose chemotherapy
and autologous hematopoietic stem cell transplantation (ASCT) induces deep responses in a majority of patients with this
disease. However, disease relapse is inevitable, and, with each relapse, the remission duration becomes shorter, ultimately
leading to a refractory disease. Consolidation and maintenance strategy after ASCT is one route to provide sustained disease
control and prevent repeated relapses. Though the consolidation strategy remains largely confined to clinical trials, significant data support the efficacy of consolidation in improving the depth of response and outcomes. There are also increasing rates of minimal residual diseasenegativity with additional consolidation therapy. On the other hand,
maintenance with novel agents post-transplant is well established and has been shown to improve both progression-free
and overall survival. Evolving paradigms in maintenance include the use of newer proteasome inhibitors, immunotherapy
maintenance, and patient-specific maintenancea concept that utilizes minimal residual disease as the primary driver of
decisions regarding starting or continuing maintenance therapy. The other approach to overcome residual disease is immune
therapeutic strategies. The demonstration of myeloma-specific alloimmunity from allogeneic transplantation is well
established. More sophisticated and promising immune approaches include adoptive cellular therapies, tumor vaccines, and
immune checkpoint manipulations. In the future, personalized minimal residual diseasedriven treatment strategies
following ASCT will help overcome the residual disease, restore multiple myelomaspecific immunity, and achieve sustained
disease control while minimizing the risk of overtreatment.
utologous hematopoietic cell transplantation with highdose chemotherapy and autologous cell rescue is a
mainstay of therapy for patients with multiple myeloma and
induces a response in a large proportion of patients.
However, relapse is near universal as a result of either
measurable disease or minimal residual disease (MRD) after
ASCT, and multiple myeloma remains incurable.1 Effective
therapies for sustained disease control after ASCT and
prevention of repeated relapses in high-risk subgroups are
unmet needs. Consolidation and maintenance after ASCT is
one route to overcome residual disease, whereas immune
approaches such as allogeneic transplantation (allo-HCT),
adoptive cellular therapies, vaccines, or antibody-based
immune manipulations represent another approach. AlloHCT, despite its toxicities, has been known to cure a
minority of patients by establishing myeloma-specific
alloimmunity.2-4 Herein, potential treatments beyond
ASCT (planned consolidation or maintenance), allo-HCT, and

emerging immune therapies, are discussed.
CONSOLIDATION THERAPY IN MULTIPLE

MYELOMA
Consolidation therapy following ASCT for multiple myeloma
implies a short period of intensive treatment with a singleagent or a combination of agents. Depth of response is
widely accepted as prognostic for outcomes with multiple
myeloma, and this strategy further reduces disease burden
following ASCT.5,6 Yet, compared with lower-dose long-term
maintenance, consolidation mostly remains confined to
clinical trials. Historically, efforts of post-transplant consolidation have ranged from aggressive attempts to eradicate disease with tandem autologous transplantation and
consolidation cytotoxic chemotherapy, to modern novel
agentbased approaches6,7
From the Judy and Bernard Briskin Center for Myeloma, City of Hope Cancer Center, Duarte, CA; Division of Medical Oncology, Washington University School of Medicine in St. Louis,
St Louis, MO; Division of Hematology Oncology, Medical College of Wisconsin, Milwaukee, WI.
Corresponding author: Parameswaran Hari, MD, MRCP, Froedtert Hospital and Medical College of Wisconsin, 9200 West Wisconsin Ave., Milwaukee, WI 53226; email: phari@mcw.
edu.
210
BEYOND AUTOLOGOUS TRANSPLANTATION IN MULTIPLE MYELOMA
TABLE 1. Lenalidomide and Bortezomib Monotherapy Consolidation

Consolidation
Regimen
Induction
Regimen
Comparison
Arm
Attal et al
(614 patients)
Lenalidomide
VAD (46%)
Uy et al13
(40 patients)
Bortezomib
Mellqvist et al14
(187 patients)
Bortezomib
Study
11
Duration
Before
Consolidation
After
Consolidation
PFS
OS
None
(all treated)
2 cycles
$ VGPR: 58%
$ VGPR: 69%
(p , .001)
4-year: 43%
vs. 22%
4-year : 73%
vs. 75%
Bortezomib-naive
None
6 cycles
CR + VGPR: 43%
CR + VGPR: 43%
NR
3-year: 63.1%
Bortezomib-naive
Placebo
6 cycles
$ nCR: 20.1%
$ nCR: 45.1%
$ VGPR: 39.7%
$ VGPR: 70.9%
27 vs.
20 months
3-year: 80%
vs. 80%
VD (46%)
Abbreviations: PFS, progression-free survival; OS, overall survival; VAD, vincristine/doxorubicin/dexamethasone; VD, bortezomib/dexamethasone; VGPR, very good partial response;
nCR, near CR; NR, not reported.
TANDEM TRANSPLANTATION AND CYTOTOXIC

CHEMOTHERAPY
IMMUNOMODULATORY AGENT
CONSOLIDATION THERAPY
Initial efforts of post-ASCT consolidation therapy evolved

from the University of Arkansas Total Therapy trials and
mirrored prolonged treatment regimens in pediatric acute
lymphocytic leukemia. Total Therapy 1 (TT1) (which did not
include consolidation) used aggressive induction chemotherapy and tandem ASCT combined with maintenance
interferon (IFN) to produce results superior to contemporary
standard therapy in a SWOG cooperative group trial.8 Total
Therapy 2 (TT2), which expanded on TT1, evaluated aggressive induction chemotherapy, tandem ASCT followed
by four cycles of dexamethasone, cisplatin, doxorubicin,
cyclophosphamide, and etoposide (DPACE) consolidation
chemotherapy; patients were then randomly assigned to
receive maintenance therapy with IFN with or without
thalidomide.9 Total Therapy 3 (TT3A) included the addition
of bortezomib-based chemotherapy induction, tandem
ASCT followed by DPACE plus thalidomide and bortezomib
(V-DTPACE) consolidation and then maintenance.10 The role
of maintenance in these trials was evaluated in a nonrandomized fashion, making its individual contribution difficult to discern; however, they spawned further interest for
improved post-ASCT strategies.
Novel agents such as proteasome inhibitors and immunomodulatory agents (IMIDs) have advanced all aspects of
multiple myeloma treatment. Evaluation of consolidation
therapy with IMID monotherapy was undertaken in a phase III
study by Intergroupe Francophone du Myelome (IFM 05-02;
Table 1).11 Newly diagnosed patients with multiple myeloma
who underwent post-ASCT consolidation therapy with lenalidomide were randomly assigned to receive maintenance
lenalidomide versus placebo. Two cycles of lenalidomide
(25 mg daily) improved the rate of complete response (CR) or
very good partial responses (PR) from 58% to 69% (p , .001),
respectively, among all participants, but maintenance was not
associated with improvement of overall survival (OS). A
similarly designed phase III U.S. study (CALGB 100104) lacked
the lenalidomide consolidation phase, but randomly assigned
patients to receive maintenance lenalidomide versus placebo
and reported an OS benefit in the maintenance arm.12 One of
the possible explanations for the lack of OS benefit in the
former study is that a short period of consolidation given to all
subjects in IFM 0502 may have abrogated the survival benefit
for protracted maintenance therapy.
PROTEASOME INHIBITORBASED
CONSOLIDATION
KEY POINTS
Induction with novel agent combinations followed by highdose chemotherapy ASCT is the standard of care in patients
with multiple myeloma who are eligible for transplant.
Relapse is inevitable in the majority of the patients
despite recent improvements in progression-free
survival.
Effective strategies to overcome residual disease and
prevent relapse is an unmet need.
Post-transplant consolidation and minimal residual
diseasedirected maintenance are promising new
avenues.
Emerging post-ASCT immunotherapy to establish
myeloma-specific immunity and allotransplant are tools
for long-term disease control, especially for young highrisk patients.
Bortezomib
Uy et al13 evaluated pre- and post-ASCT bortezomib consolidation in a phase II study with 40 patients (all newly diagnosed with multiple myeloma) who received two cycles of
intravenous bortezomib followed by ASCT, and six cycles of
intravenous bortezomib after ASCT. Only 28 patients received
post-ASCT bortezomib, with two upgraded responsesone
from very good PR to CR and one from PR to very good PR. The
3-year OS and disease-free survival (DFS) were 63.1% and
38.2%, respectively.13
The NORDIC Myeloma Study Group randomly assigned
bortezomib-naive patients post-ASCT to receive either no
further treatment or bortezomib consolidation for six cycles.
With a median follow-up of 38 months, PFS for the bortezomibtreatment group was 27 months compared with 20 months
for the control group (p = .05), and no difference in OS.
Patients who experienced at least a very good PR
211
KRISHNAN ET AL
TABLE 2. Bortezomib and Immunomodulatory Combination Consolidation Therapy

Study
72
Cavo et al
Consolidation
Regimen
Induction Comparator
Regimen Arm
Before
Duration Consolidation
After
Consolidation
PFS
OS
VTD
(160 patients)
VTD x 3
2 cycles
CR: 48.7%
CR: 60.6%
3-year: 60%
3-year: 90%
$ VGPR: 86.2%
$ VGPR: 91.9%
3-year: 48%
3-year: 88%
NR
3-year: NR vs.
22 months
3-year: 77%
3-year: 100%
32 months
3-year: 93%
TD
(161 patients)
p = NS (VGPR)
72
TD
(161 patients)
Cavo et al
TD x 3
VTD
(160 patients)
2 cycles
CR: 40.4%
CR: 46.6%
$ VGPR: 81.4%
$ VGPR: 88.2%
p = NS (VGPR)
17
VTD
(121 patients)
Leleu et al
VTD
No consolidation
(96 patients)
2 cycles
CR: 33%
CR: 52%
$ VGPR: 43%
$ VGPR: 31%
p , .001 (CR)
73
Roussel et al
Nooka et al18
Moreau et al19
RVD
(31 patients)
RVD
RVD
RVD
RVD
RVD
None
None
No transplant
2 cycles
3 years
2 cycles
CR: 47%
CR: 50%
$ VGPR: 70%
$ VGPR: 87%
sCR approx. 20%
sCR: 51%
$ VGPR approx. 85%
$ VGPR: 96%
$ VGPR:73%
$ VGPR: 81%
34 vs.
4-year: 81%
43 months
vs. 83%
Abbreviations: PFS, progression-free survival; OS, overall survival; CR, complete response; NR, not reported; NS, not significant; RVD, lenalidomide/bortezomib/dexamethasone; TD,
thalidomide/dexamethasone; VGPR, very good partial response; VTD, bortezomib/thalidomide/dexamethasone.
demonstrated a prolonged PFS irrespective of bortezomib

consolidation. The beneficial effect of bortezomib consolidation was confined to patients who did not have at least a
very good PR following ASCT. More importantly, PFS among
patients whose disease had a very good PR to therapy at
randomization was similar to patients whose disease response improved to the very good PR category or better
during consolidation. These results provide substantial
support to the role of post-transplant consolidation to
further deepen disease response.14
Combination Therapy Bortezomib and

Immunomodulatory Agents
In a phase III trial, Cavo et al15 assessed the efficacy of
consolidation with thalidomide and dexamethasone (TD)
compared with bortezomib, thalidomide, and dexamethasone (VTD; Table 2). The patients were randomly assigned at
diagnosis to induction therapy with either VTD or TD, and all
underwent tandem ASCT followed by two 35-day consolidation cycles with the same regimen as their induction
therapy (VTD vs. TD). CR rates after second ASCT was 48.7%
for the VTD group and 40.4% for the TD group, improving to
60.6% and 46.6%, respectively, following consolidation. PFS
was superior for the VTD group compared with the TD group
(60% vs. 48% at 3 years; p = .042).15 Notably, rates of CR and
at least very good PR favored VTD treatment, but were not
significant. The isolated impact of consolidation is difficult to
determine from this study.
Roussel et al16 investigated lenalidomide, bortezomib, and
dexamethasone (RVD) induction and consolidation therapy
in 31 patients with newly diagnosed multiple myeloma who
received three RVD induction cycles, followed by ASCT and
two RVD consolidation cycles with subsequent lenalidomide
maintenance for 1 year. The estimated 3-year PFS and OS
212
were 77% and 100%, respectively, and CR rates were 47%

following ASCT and 50% after RVD consolidation. Therapy
was tolerable, with no treatment-related mortality (TRM),
and 97% completed the planned sequence.16 Leleu et al17
retrospectively analyzed 121 patients with newly diagnosed
multiple myeloma across nine IFM centers who had received
three cycles of VTD induction, followed by ASCT and two
cycles of VTD consolidation. Complete response rates increased from 33% after induction and ASCT to 52% following
consolidation. Compared with 96 similar patients treated
with VTD induction and ASCT without consolidation, CR rates
were superior and relapse risk were lower in the consolidation arm.17
Nooka et al18 evaluated prolonged therapy with RVD in
45 patients with high-risk multiple myeloma (p53 deletion,
1p deletion, immunoglobulin heavy chain gene translocations, or plasma cell leukemia). Patients received RVD
as consolidation/maintenance therapy post-ASCT for up to
3 years, followed by single-agent lenalidomide maintenance
thereafter. RVD consolidation/maintenance achieved at least
very good PR status in 96% of patients, and a stringent CR
(sCR) in 51%. Median PFS was 32 months with a 3-year OS of
93%a promising improvement in PFS and OS in this high-risk
group. Despite its length and intensity, therapy was well
tolerated overall.18
More recently, the second interim analysis of the IFM 2009
trial was reported.19 In this trial, RVD was administered for
eight cycles followed by 12 months of lenalidomide maintenance or RVD induction with RVD consolidation post-ASCT
followed by lenalidomide maintenance. Response to treatment continued to deepen with progressive therapy, and at
least very good PR rates improved from 73% to 81% after RVD
consolidation. This study also showed an improvement in PFS
(median of 34 vs. 43 months) in the ASCT arm.
TABLE 3. Carfilzomib-Based Consolidation Therapy

Induction
Regimen
Comparator
Arm
Duration
Before
Consolidation
After
Consolidation
PFS
OS
KTd (Sonneveld et al ;
91 patients)
KTd
None
4 cycles
CR: 33%
CR: 63%
3-year: 60%
3-year: 90%
$ VGPR: 76%
$ VGPR: 89%
KRd (Jakubowiak et al21;

71 patients)
KRd
$ CR: 27%
$ CR: 77%
11 months: 99%
11 months: 100%
$ sCR: 22%
$ sCR: 70%
Consolidation Regimen
20
None
4 cycles
Abbreviations: PFS, progression-free survival; OS, overall survival; KTd, carfilzomib/thalidomide/dexamethasone; KRd, carfilzomib/lenalidomide/dexamethasone; CR, complete
response; VGPR, very good partial response; sCR, stringent complete response.
Carfilzomib
In a multicenter phase II study from the European Myeloma
Network, the carfilzomib, thalidomide, and dexamethasone
(KTd) combination was investigated as induction and consolidation therapy in patients with multiple myeloma who
were previously untreated (Table 3). KTd was given in 28-day
cycles for up to four cycles and followed by ASCT. After ASCT
patients received four cycles of KTd consolidation therapy.
The at least very good PR rate increased from 68% after
induction to 76% after ASCT, and finally to 89% after
four cycles of consolidation. Progression-free survival at
36 months was 72%, and only 5% of patients discontinued
therapy.20 Pre- and post-transplant carfilzomib, lenalidomide, and dexamethasone (KRd) was studied in a phase II
trial. After KRd induction, ASCT and KRd consolidation (four
cycles) were administered followed by KRd maintenance
for nine cycles, and then lenalidomide maintenance offprotocol. Rates of CR increased from 27% following ASCT to
77% after consolidation. At the conclusion of KRd treatment,
90% of patients demonstrated an at least CR with a 3-year
PFS and OS of 79% and 100%.21
CONSOLIDATION THERAPY AND MINIMAL

RESIDUAL DISEASE
The role of MRD evaluation in prediction of outcomes and
duration of therapy is increasing. Ladetto et al22 enrolled
patients whose disease had a very good PR following ASCT
and monitored MRD after four cycles of VTD consolidation using polymerase chain reactionbased techniques
(Table 4).22 Molecular remission rates were 3% after ASCT
compared with 18% after VTD consolidation. Achievement
of MRD-negative response was significantly associated

with improved median PFS (68 months vs. 23 months,
p , .0001).
In the study by Roussel et al,16 the outcomes of patients
who were MRD-negative (evaluated by marrow flow cytometry) were impressive with a 3-year PFS of 100%.16 MRD
negativity rates steadily increased through induction to the
completion of consolidation. Jakubowiak et al21 assessed MRD
in a subset of patients whose disease had a CR using 10-color
flow cytometry and noted substantial improvements in MRD
negativity rates with consolidation therapy.21
CONSOLIDATION VERSUS MAINTENANCE

Current evidence does not enable a clear recommendation
of consolidation therapy for all patients after they have
undergone ASCT. Substantial data support its efficacy in
improving depth of response of multiple myeloma, which
has historically translated into improvements of survival
outcomes. The awaited results of the Blood and Marrow
Transplant Clinical Trials Network (BMT CTN) 0702 STAMINA
trial (NCT 1109004), compares three competing postASCT strategies after initial ASCT: patients will be randomly assigned to receive a second (tandem) ASCT, RVD
consolidation, or immediate initiation of lenalidomide
maintenance (Fig. 1). All arms will ultimately receive
lenalidomide maintenance based on the paradigm that
continuous therapy or maintenance is sufficient to control
residual disease. Additionally, an ongoing multicenter study
(NCT02253316) is evaluating the role of ixazomib-based
consolidation therapy followed by maintenance with MRD
post-ASCT (Fig. 2).
TABLE 4. Minimal Residual Disease With Consolidation Therapy

Consolidation Regimen
Induction Regimen
Comparator Arm
Duration
Before
Consolidation
After
Consolidation
VTD (Ladetto et al22;

39 patients)
VAD
None
4 cycles
CR: 15%
CR: 49%
MRD: 4.15log
reduction*
MRD: 10.09log
reduction*
VRD (Roussel et al16;

26 patients)
VRD
None
2 cycles
MRD: 54%
negative
MRD: 58%
negative
KRd (Jakubowiak et al21;

71 patients)
KRd
None
4 cycles
MRD: 79%
negative
MRD: 90%
negative
*Reduction from baseline levels.

Abbreviations: VTD, bortezomib/thalidomide/dexamethasone; VAD, vincristine/doxorubicin/dexamethasone; CR, complete response; MRD, minimal residual disease; KRd,
carfilzomib/lenalidomide/dexamethasone.
213
KRISHNAN ET AL
FIGURE 1. CTN 0702 Study (STAMINA) Schema
Newly Diagnosed
Transplant Eligible
Mulple Myeloma
Mel 200 Transplant
Mel 200 Transplant &

Lenalidomide Maintenance
Mel 200 Transplant
VRD Consolidaon &

Lenalidomide Maintenance
Lenalidomide
Maintenance
Mel 200 Transplant
MAINTENANCE THERAPY AFTER

TRANSPLANTATION: THE PAST AND THE FUTURE
The concept of sustained low intensity post-ASCT therapy
(or maintenance) for patients with multiple myeloma dates
back to IFN, which was associated with improved PFS and
OS in several trials.23 The advent of novel agents with improved efficacy and toxicity supplanted IFN. Given emerging
immuno-oncologic approaches, this may also have been
prescient in harnessing the immune system as the major
mechanism to maintain control against residual disease.
Maintenance Agents: Immunomodulatory Agents

Thalidomide. Probably the largest body of clinical trial experience has been with thalidomide maintenance therapy,
although these trials differed with some having used thalidomide as part of induction continuing through maintenance, whereas others used it in conjunction with steroids,
and with major variation in thalidomide doses chosen.
Table 5 outlines the major trials of thalidomide maintenance
following ASCT. All studies of thalidomide maintenance
suggest improved PFS, but toxicity and tolerability remain
issues. In the IFM 99 trial, 39% of patients discontinued
thalidomide mainly due to neuropathy.24 In the MRC IX trial
with thalidomide maintenance after intensive (transplant) or
nonintensive (chemotherapy only) pathways, 52% stopped
therapy due to side effects, and thalidomide maintenance
did not prolong PFS or OS in those with adverse cytogenetics.
FIGURE 2. Design of NCT02253316: Ixazomib,

Lenalidomide, and Dexamethasone Consolidation
Therapy and Minimal Residual Disease
Autologous Stem
Cell Transplant
Completion of
Consolidation
Day 80-120:
MRD Assessment
Consolidation with
IRD Begins
4 Cycles of IRD:
MRD Assessment
Randomized to
Lenalidomide
Maintenance
Randomized to
Ixazomib
Maintenance
Assess Tolerability, PFS, OS
Abbreviations: MRD, minimal residual disease; IRD, ixazomib, lenalidomide, and

dexamethasone; PFS, progression-free survival; OS, overall survival.
214
In the United States, therefore, interest shifted to lenalidomide as a better-tolerated IMID maintenance strategy.25
Lenalidomide. Three published phase III trials explored
lenalidomide maintenance after ASCT (Table 6). The CALGB
100104 Intergroup trial randomly assigned patients to receive lenalidomide (10 mg daily escalated to 15 mg) or
observation after ASCT. The median time-to-progression
(TTP) was 46 months in the lenalidomide arm compared
with 27 months with placebo, which translated into an
improved OS (88% vs. 80% at 3 years).12 A recently updated
analysis showed a median TTP of 53 months compared with
27 months as well as continued OS advantage.26 The phase III
IFM 05-02 trial, as noted previously, had the subtle, but
important, difference that all patients received lenalidomide
consolidation followed by lenalidomide maintenance or
placebo.11 In this trial, although median TTP was superior for
the lenalidomide arm (41 months vs. 23 months p , .001),
there was no difference in OS. The lack of OS benefit may in
part be attributed to lenalidomide consolidation and or an
imbalance of high-risk patients between arms. An Italian trial
(GIMEMA RV-MM-PI209) assigned patients to melphalan,
prednisone, and lenalidomide (MPR) or ASCT followed by
second random assignment to maintenance lenalidomide or
no maintenance. The maintenance lenalidomide cohort in
both the MPR and ASCT groups had superior PFS (but not
OS).27 In all trials, the lenalidomide cohorts had a higher
incidence of neutropenia. Even more important was the
higher incidence of second primary malignancies (SPM) at
around 8% in the IFM and CALGB trials compared with 3% to
4% in the placebo arm. Strategies to minimize SPM risk
without losing the benefit of maintenance are areas of study,
such as limiting the duration of maintenance, pre-assessing
SPM risk, and MRD-based discontinuation.
A comparison of the recently reported IFM DFCI 2009 trial
with the ongoing DFCI (Dana-Farber Cancer Institute) BMT
CTN Determination trial is expected to shed light on the
duration of maintenance lenalidomide and the impact of
MRD. As described previously, the IFM version of this phase
III trial incorporated 1 year of lenalidomide maintenance.19
In contrast, the DFCI version (DFCI10-106; NCT1208662)
follows the same overall schema, but lenalidomide maintenance continues until disease progression. The incidence
of SPM as well as sequential evaluation of MRD by flow
cytometry and next-generation sequencing is being evaluated in both trials.28
Maintenance Agents: Proteasome Inhibitors

Bortezomib. Although there is a large body of evidence
surrounding bortezomib maintenance (Table 5), most protocols used bortezomib administered intravenously, which
renders the toxicity data less relevant in the era of subcutaneous dosing. A large Dutch-German trial randomly
assigned patients to receive bortezomib, doxorubicin, and
dexamethasone (PAD) or vincristine, doxorubicin, and
dexamethasone (VAD)29 induction followed by ASCT. The
PAD induction arm subsequently received bortezomib
TABLE 5. Thalidomide and Bortezomib Maintenance Studies

Schedule
Control
CR
PFS/EFS/TTP
Reference
PAM + Thal 400 mg/day
PAM or OBS
NR
52% (3-year)
Attal et al24
Thal 400 mg/day
None
50%
72% (6-year)
Barlogie et al74
Thal 200 mg/day + PSE 50 mg QOD
PSE
NR
42% (3-year)
Spencer et al75
Thal 50 mg/day
IFN
31%
50% (34-month)
Lokhorst et al76
Thal 100 mg/day
OBS
NR
50% (30-month)
Morgan et al25
Thal 200 mg/day + PSE 50 mg QOD
OBS
NR
32% (4-year)
Stewart et al77
Bort 1.3 mg/m Q2 weeks
Thal
NR
50% (35-month)
Sonneveld et al29
Bort + Thal
Thal or IFN
19%
50% (45-month)
Rosinol et al31
Abbreviations: CR, complete response; PFS, progression-free survival; EFS, event-free survival; TTP, time-to-progression; PAM, pamidronate; Thal, thalidomide; OBS, observation, NR,
not reported; PSE, prednisone; QOD, every other day; IFN, interferon; Bort, bortezomib; Q2, every two.
maintenance for 2 years, whereas the VAD arm received

thalidomide maintenance. Median PFS was 35 months in the
bortezomib-treated arm compared with 28 months in the
VAD/thalidomide-treated arm. A landmark analysis suggested an OS benefit, although it is not clear if the benefit
was derived solely from the bortezomib maintenance or
induction, or from the higher discontinuation rates with
thalidomide maintenance. There was particular improvement in the high-risk subgroup with 17p deletion that received bortezomib (PFS of 26 months vs. 12 months in the
thalidomide arm).30 Another three-arm Spanish trial compared bortezomib plus thalidomide, thalidomide, or IFN as
maintenance with a planned duration of 3 years. PFS was
superior with the bortezomib/thalidomide combination, but
there was no OS benefit.31
valuable. In another phase II trial using ixazomib and

lenalidomide combination maintenance,32 ixazomib dose
was reduced to 3 mg or less (from the standard 4 mg) in
some patients because of the development of neuropathy.
Overall, the combination was well-tolerated, with the majority of patients still receiving treatment at 30-plus cycles
with an estimated 2-year PFS of 83%.
In the allogeneic setting, the BMT CTN 1302 phase II trial
for patients with protocol-defined high-risk multiple myeloma (NCT02440464) randomly assigned patients to receive
ixazomib maintenance or placebo starting at 60- to 120-days
after allo-HCT. Because bortezomib has been shown to
reduce graft-versus-host disease (GVHD), the effects of
ixazomib on GVHD and relapse rates are being evaluated.
Immunotherapy
Carfilzomib and ixazomib. Other proteasome inhibitors
could offer the potential advantages of minimizing neuropathy with possibly higher potency. Carfilzomib-based
combinations with lenalidomide have been described previously, whereas ixazomib maintenance is another attractive
option, given its oral administration and lower neuropathy
risk. As shown in Fig. 2, a current trial (NCT02253316) is
evaluating ixazomib, lenalidomide, and dexamethasone
consolidation followed by random assignment to ixazomib
or lenalidomide maintenance until disease progression.
Although not powered to show a benefit with either of the
maintenance arms, the tolerability and feasibility of ixazomib maintenance and ongoing MRD assessments will be
Immunotherapy for multiple myeloma encompasses approved

myeloma-directed passive antibody or active approaches that
enhance tumor-specific immune responses. The post-ASCT
maintenance setting is the ideal platform for immunotherapy, given low disease burden as well as a favorable immune
milieu. Multiple immunotherapy options are being evaluated
in trials designed to elicit antitumor immune responses, as well
as augment T-cell function (Table 7).
Daratumumab,33 the newly approved anti-CD38directed
antibody, has a half-life of 21 days, which makes it an attractive maintenance option. An ongoing IFM phase III trial
(NCT02541383) for front-line therapy randomly assigns
patients to receive daratumumab maintenance for 2 years
TABLE 6. Lenalidomide Maintenance Studies

Maintenance
Study
Outcome
Comparison
Planned Length
of Maintenance
McCarthy et al12
(CALGB 100104)
Lenalidomide vs. placebo
Until progression
Median PFS (46 vs.

27 months; p , .001)
3-year OS (88%
vs. 80%; p = .03)
Attal et al75
(IFM 0502)
Lenalidomide vs. placebo

after 2 months lenalidomide
consolidation
Until progression, but

terminated early for SPM
Median PFS (41 vs.

23 months; p , .001)
4-year OS (73%
vs. 75%; p = NS)
Palumbo et al27
MPR vs. tandem ASCT followed

by lenalidomide vs. placebo
Until progression
Median PFS 41.9 vs.

21.6 months; (p , .001)
3-year OS (88%
vs. 79.2%; p = .14)
PFS
OS
Abbreviations: PFS, progression-free survival; OS, overall survival; SPM, secondary primary malignancies; NS, not significant; MPR, melphalan/prednisone/lenalidomide; ASCT,
autologous hematopoietic cell transplantation.
215
KRISHNAN ET AL
TABLE 7. Comparison of Immunotherapy Maintenance Strategies

Therapy
Advantage
Disadvantage
Vaccine
Ease of administration
Low toxicity
Patient-specific (manufacturing)
T-cellbased (Including CAR)
Cytolytic trafficking to extramedullary sites
Toxicity manufacturing
Antibodies
Long half-life
Commercially available
Infusional toxicity
Cytopenias
Checkpoint Blockade
Commercially available
Low single-agent response rates in MM
Abbreviations: CAR, chimeric antigen receptor; MM multiple myeloma.
after ASCT. Elotuzumab, a natural killer (NK) celldirected

antibody requires lenalidomide to augment its activity. An
ongoing phase II trial (NCT 02420860) explores the combination of elotuzumab with lenalidomide following ASCT.
observed in recipients of T-cell replete allografts compared

with recipients of T-cell depleted or syngeneic grafts.
Modern Randomized Trials of Allo-HCT

Emerging Paradigms in Maintenance
Patient-specific maintenance is a concept that utilizes MRD
as the primary driver of decisions regarding starting or
continuing maintenance therapy. Most ongoing phase III and
phase II trials are collecting MRD data at various points in
time, although the optimal method for MRD assessment is
debated (multicolor flow cytometry vs. sequencing). To a
smaller extent, decisions regarding choice of maintenance
drug are also patient- and disease-specific (e.g., bortezomib
for patients with t[4,14] or 17p deletion). Further refinement
of this concept with more targeted therapy based on genetic
sequencing is key. Lastly, augmenting immune function
through a variety of methods including adoptive cell therapy, vaccines, checkpoint blockade, and myeloma-specific
antibodies will likely become the backbone of post-ASCT
interventions.
IMMUNOTHERAPY FOR MULTIPLE MYELOMA:

ALLOGENEIC TRANSPLANTATION AND BEYOND
Allogeneic transplantation is a well-established immunotherapy strategy for multiple myeloma with evidence of the existence of a potent and often sustained graft-versus-myeloma
effect.4 Conventional myeloablative regimens resulted in prohibitive TRM (40% to 60%), but apparent plateaus in transplant
survival curves indicated long-term disease control. The allogeneic arm of U.S. intergroup trial (S93-21) was prematurely
closed as a result of high early TRM (53%), but showed long
relapse-free survivorship of 39% at 7 years.34 Development of
nonmyeloablative and reduced-intensity conditioning regimens
has decreased TRM,35,36 but the lower TRM was negated by an
increase in relapse risk during later years.37 The risks of chronic
GVHD and the need for long-term immunosuppression remain
major challenges. The success of allo-HCT as a relapse prevention strategy resides in the ability of donor-derived alloreactive T cells to eliminate or suppress multiple myeloma
propagating residual cells. Several lines of evidence point to this,
including the correlation of chronic GVHD with protection from
relapse,2,4 the ability of donor derived lymphocytes to eliminate
residual or relapsing disease,38 and lower relapse rates
216
After promising phase II data, a series of randomized,

prospective studies explored the concept of decoupling
myeloablation and immune therapy by a tandem approach
involving an initial autologous myeloablative transplant
followed (within 3 to 6 months) by an allogeneic nonmyeloablative transplant from a matched sibling or unrelated donor. The results of this strategy have been
discordant in the front-line treatment of multiple myeloma.
Two major European studies39,40 reported favorable longterm OS.41,42 In these studies, compared with tandem autologous transplantation, a second (tandem) allo-HCT from
an HLA-matched donor (after a prior autograft) results in
superior PFS and OS despite an increase in early TRM. This is
in contrast to the U.S. study led by the BMT CTN in which no
survival or progression benefit was seen in the allo-HCT
group.2 Interestingly, in the larger European Group on Blood
and Marrow Transplantation (EBMT) study, in addition to
relapse risk being lower in the allograft arm, post-relapse
survival was also superior for the allo-HCT cohort compared
with tandem autograft recipients.42 This latter phenomenon
has been attributed to a synergy between novel agents and
the immune benefits of an allo-HCT.
As discussed in the maintenance section, more recent
studies incorporated maintenance agents (lenalidomide,
bortezomib, and others) in the post allo-HCT setting. These
results were discordant with the HOVON group,43 reporting
substantial GVHD that threatened the feasibility of planned
lenalidomide maintenance. In contrast, a multicenter U.S.
study,44 in a predominantly high-risk population, indicated
promising PFS and OS of 63% and 78%, respectively, at
18 months. A current multicenter U.S. study (BMT CTN 1302;
NCT02440464) uses the newly approved agent ixazomib in
maintenance after an allogeneic transplant in patients with
genetically defined high-risk multiple myeloma, plasma cell
leukemia, or early relapse after an ASCT.
On a practical basis, if allo-HCT is considered part of the
therapeutic armamentarium, clinicians are faced with two
major questions: (1) identifying the patients who benefit the
most from allo-HCT and (2) identifying the optimal time
point for referral for allo-HCT.
Who Should Be Considered for Allo-HCT for Multiple

Myeloma
In practical terms, this boils down to the identification of
patients who are at such high clinical or biologic risk that the
TRM risk of allo-HCT is balanced by the intrinsic negative
prognosis. Biologically, high-risk myeloma is defined by the
presence of chromosomal abnormalities t(4:14), t(14:16),
+1q21, 17p by fluorescence in situ hybridization (FISH); 13q
deletion by karyotyping or high-risk gene expression profiling (GEP).45,46 Based on the recently developed R-ISS
staging system that incorporates disease burden and biology, the highest risk subgroup (stage III) had a predicted
5-year OS and PFS of 40% and 24%, respectively.47 Similarly,
relapse within 18 months of ASCT is an extremely negative
prognostic marker.48 Effective treatment that establishes
long-term disease control in these high-risk patient populations is an unmet need, and allo-HCT becomes a consideration. Both prospective and retrospective studies have
explored the role of allo-HCT in patients with high-risk
multiple myeloma with variable outcomes. The EBMT
NMAM 2000 study reported a 21% PFS at the 8-year followup in patients with the higher risk deletion 13 (by FISH) who
underwent allo-HCT versus 5% in the tandem ASCT group.42
Knop et al showed that in the highest risk subgroup with
del(17p) and del(13q) abnormality,49 median PFS (p = .002)
and OS (p = .011) were superior compared with tandem
ASCT. In younger eligible patients with well-defined high-risk
features, it is reasonable to consider allo-HCT, especially
within a clinical trial and early in the clinical course (front-line
or first relapse). The mortality and morbidity in general has
steadily declined in recent years, making the short-term risks
lower than in published studies from a decade or more
ago.37 In contrast, after repeated relapses, allo-HCT itself is
associated with very poor survival, making this modality a
very poor late salvage option.50 In the United States, overall
allogeneic transplant activity has declined for patients with
multiple myeloma and more transplants, unfortunately, are
now suboptimally performed later in the disease course.
Timing of Allo-HCT for Relapsed Multiple Myeloma

Prospective data that demonstrated a benefit for allo-HCT
are all derived from the up-front setting after an initial ASCT.
Understandably in the modern era, patients and physicians
are still hesitant regarding a treatment with high immediate
TRM risk. In patients whose disease relapsed after an ASCT,
the benefit of allo-HCT was found to be highest when this
modality was used earlier in the disease course and when
used as a strategy for consolidation of remission induced
by salvage therapy.51,52 A donor versus no donor analysis
considered 169 consecutive patients who had relapsed after
an ASCT and underwent HLA-typing immediately after relapse. The 2-year PFS was higher in the allo-HCT group (with
donors) at 42% compared with 18% in the no donor group
(p = .0001) with similar OS, although TRM was 22% versus
1% in the allo-HCT and ASCT groups, respectively.53 In another
prospective phase II multicenter EBMT trial, 49 patients who
had relapsed after a previous ASCT received allo-HCT from
related or unrelated donors with an overall response rate of

90%, including a CR rate of 40%. Cumulative incidence of
1-year TRM was 25% and was significantly lower in transplants
from fully HLA-matched donors compared with mismatched
donors (10% vs. 53%, p = .001). After a median follow-up of
43 months, the 5-year PFS and OS were 20% and 26%, respectively, and were greater in patients who demonstrated
post-transplant CR.54 In light of the above data, it is reasonable to consider allo-HCT in high-risk patients who have
relapsed and who demonstrated a deep remission to salvage
regimens prior to allo-HCT.
Conditioning Regimen Intensity

The optimal intensity of conditioning regimens for allo-HCT is
still controversial for patients with multiple myeloma. Fully
myeloablative regimens have been largely abandoned,
whereas nonmyeloablative stem cell transplant regimens
(e.g., total body irradiation of 2 Gy) without anti-multiple
myeloma activity have been associated with lower TRM risk
but increased relapse.37 The pendulum of regimen intensity
has now swung back to reduced-intensity conditioning with
regimens that incorporate intermediate doses of active antimultiple myeloma therapy. The most popular approach in
the United States is a combination of fludarabine and
melphalan at a dose of 140 mg/m2 (CIBMTR data). The
addition of proteasome inhibitors to conditioning is being
explored as an additional strategy to increase the antineoplastic potential without incurring additional toxicity.
IMMUNE-BASED THERAPIES BEYOND ALLO-HCT

A variety of novel post-ASCT or allo-HCT immunotherapeutic
strategies are now being explored for patients with multiple
myeloma (Table 7). These include cellular approaches such
as myeloma-specific T cells (via T-cell expansion), marrow
infiltrating lymphocytes and redirected T cells with chimeric
antigen receptors (CAR T), and tumor-based vaccines to
induce myeloma-specific immunity in the context of enhanced antigen presentation. HCT provides an ideal platform
for additional immune-based therapies. The recovery phase
from ASCT (or other lymphodepleting therapy) represents
a favorable platform for adoptive cellular therapy. The
homeostatic lymphocyte proliferation following lymphopenia is a context in which immune checkpoint blockers may
also be able to reverse multiple myeloma-associated T-cell
exhaustion. Additionally, lymphopenia resulting from ASCT
eliminates tolerogenic antigen-presenting cells and induces
cytokine release that generates a more favorable environment for adoptive T-cell therapy. Indirect evidence suggests
that the immune system can contribute to the clinical
benefits of ASCT. For example, patients with early lymphoid
recovery after ASCT have superior long-term outcomes.55
Donor Lymphocyte Infusion

Donor lymphocyte infusions have been used upon relapse
after allo-HCT and even to preempt relapse, but exacerbation of GVHD is a major risk. Disease control is generally superior in patients in whom GVHD develops.38
217
KRISHNAN ET AL
Donor lymphocyte infusions have been combined with

IMIDs or bortezomib.56 Preemptive donor lymphocyte infusions at a defined time period has been used to enhance
reconstitution of donor T cells and antitumor immunity.57
Emergence of WT1-specific cytotoxic T lymphocytes (WT1CTL) has been correlated with better PFS after allo-HCT.
Infusion of donor-derived T cells directed against specific
myeloma antigens such as WT1 or cancer testis antigens58,59
is another promising area of adoptive T-cell therapy using
the allo-HCT platform.
Myeloma Infiltrating T Lymphocytes

Marrow lymphocytes in patients with multiple myeloma are
enriched for T cells with myeloma-specific antigen specificity.
Noonan et al60 described adoptive transfer of such autologous marrow-derived, ex vivo activated and expanded T cells
on day 3 after ASCT.60 They demonstrated measurable
myeloma-specific activity for the ex vivo expanded product
and persistence of myeloma-specific immunity even at 1 year.
along with lenalidomide.66 Haploidentical allo-HCT followed

by planned NK-cell infusion attempts to use donor-recipient
KIR ligand mismatch and NK-cell reactivity to facilitate longterm remission (NCT02100891).

Immune responses against multiple myelomaspecific antigens are minimally protective, although detectable in
patients. Inhibitors of PD-1 and PD-L1 interaction are being studied as a means of breaking down multiple myeloma immune tolerance. In patients with multiple myeloma,
PD-1 expression was upregulated on T cells concomitant
with increased PD-L1 expression on plasma cells.67,68 The
antiPD-1 agent nivolumab as monotherapy was unimpressive,69 but promising response rates were observed in
combination with lenalidomide, even for patients whose
disease was refractory.70 The antiPD-1 agent pembrolizumab is being studied in the lymphocyte recovery
phase after ASCT (NCT02331368) in combination with
lenalidomide.
CAR T-Cell Therapy

CAR T-cell therapy involves transducing activated T cells with
genes encoding T-cell receptors specific to the antigen of
interest. Although multiple myeloma is not a classic CD19positive malignancy, deep sustained response to anti-CD19
CAR T cells in conjunction with second salvage ASCT for
patients with relapsed/refractory multiple myeloma was
recently reported.61 Several promising antigenic targets
have been identified for the development of anti-multiple
myeloma CARs (40) such as Bcell maturation antigen
(BCMA), CD138, kappa light chains, and CS-1. Notably, allogeneic
CD19-directed CAR T cells (derived from donor lymphocytes)
have induced remissions without induction of GVHD in patients who relapsed after allo-HCT.61,62 Thus, the allo-HCT
or ASCT platform could be adapted to subsequent CAR T
technology.
Natural KillerCell Therapy

Natural killer cells have innate cytotoxicity against multiple
myeloma cells, while multiple myeloma exhibits specific
immune-evasive strategies to circumvent and attenuate NKcell function.63 Modulation of NK activity using anti-KIR Ab
IPH2101 (monoclonal antibody against inhibitory KIR on NK
cells) is being explored as a means to establish multiple
myelomaspecific immunity.64 Autologous-, allogeneic-, and
cord-derived NK cells have been found to be safe and efficacious for multiple myeloma.65 In a phase I study, up to
1 3 108 NK cells/kg freshly expanded cord blood NK cells
were found to be safe when given before high-dose melphan
Vaccines
The vaccine approach holds great promise for patients with
multiple myeloma. Patients who received a patient-specific
dendritic cell/myeloma fusion vaccine demonstrated the
expansion of multiple myelomaspecific T cells, as well as
upgrading of response in a subgroup of patients.71 This
concept has been expanded to a intergroup randomized
phase II trial (BMT CTN 1401) with ASCT followed by lenalidomide maintenance with or without vaccination using the
dendritic cell/myeloma fusion vaccine.
The direct manipulation of T cells by increasing T-cell
number, as well as engineering the T cells for augmented
anti-multiple myeloma affinity was studied in a phase I/II
trial. In this study, autologous T cells were transduced with a
lentiviral vector that encoded the affinity-enhanced NY-ESO
T-cell receptor. Patients received an infusion of these T cells
after ASCT, which was shown to lead to long-term engraftment, infiltration of marrow, and trafficking to other
tumor sites.
CONCLUSION
The role of transplant-based approaches as a key step in
inducing long-term remissions for patients with multiple
myeloma continues to evolve. The next decade of studies will
likely establish personalized post-transplant maintenance
strategies designed to achieve the trifecta of MRD negativity,
restoration of multiple myelomaspecific immunity, and
freedom from multiple myeloma clonal evolution.
References
1. Kumar SK, Rajkumar SV, Dispenzieri A, et al. Improved survival in
multiple myeloma and the impact of novel therapies. Blood. 2008;111:
2516-2520.
218
2. Krishnan A, Pasquini MC, Logan B, et al; Blood Marrow Transplant

Clinical Trials Network (BMT CTN). Autologous haemopoietic stem-cell
transplantation followed by allogeneic or autologous haemopoietic
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
stem-cell transplantation in patients with multiple myeloma (BMT CTN

0102): a phase 3 biological assignment trial. Lancet Oncol. 2011;12:
1195-1203.
Gahrton G, Tura S, Ljungman P, et al; European Group for Bone Marrow
Transplantation. Allogeneic bone marrow transplantation in multiple
myeloma. N Engl J Med. 1991;325:1267-1273.
Tricot G, Vesole DH, Jagannath S, et al. Graft-versus-myeloma effect:
proof of principle. Blood. 1996;87:1196-1198.
Harousseau JL, Attal M, Avet-Loiseau H. The role of complete response
in multiple myeloma. Blood. 2009;114:3139-3146.
San-Miguel JF, Mateos MV. Can multiple myeloma become a curable
disease? Haematologica. 2011;1246-1248.
Zangari M, van Rhee F, Anaissie E, et al. Eight-year median survival in
multiple myeloma after total therapy 2: roles of thalidomide and
consolidation chemotherapy in the context of total therapy 1. Br J
Haematol. 2008;141:433-444.
Vesole DH, Barlogie B, Jagannath S, et al. High-dose therapy for refractory multiple myeloma: improved prognosis with better supportive
care and double transplants. Blood. 1994;84:950-956.
Barlogie B, Jagannath S, Desikan KR, et al. Total therapy with tandem
transplants for newly diagnosed multiple myeloma. Blood.1999;93:
55-65.
Barlogie B, Tricot G, Rasmussen E, et al. Total therapy 2 without thalidomide in comparison with total therapy 1: role of intensified induction and posttransplantation consolidation therapies. Blood. 2006;
107:2633-2638.
Attal M, Lauwers-Cances V, Marit G, et al; IFM Investigators. Lenalidomide maintenance after stem-cell transplantation for multiple myeloma. N Engl J Med. 2012;366:1782-1791.
McCarthy PL, Owzar K, Hofmeister CC, et al. Lenalidomide after stemcell transplantation for multiple myeloma. N Engl J Med. 2012;366:
1770-1781.
Uy GL, Goyal SD, Fisher NM, et al. Bortezomib administered pre-autoSCT and as maintenance therapy post transplant for multiple myeloma:
a single institution phase II study. Bone Marrow Transplant. 2009;43:
793-800.
Mellqvist UH, Gimsing P, Hjertner O, et al; Nordic Myeloma Study
Group. Bortezomib consolidation after autologous stem cell transplantation in multiple myeloma: a Nordic Myeloma Study Group randomized phase 3 trial. Blood. 2013;121:4647-4654.
Cavo M, Pantani L, Petrucci MT, et al; GIMEMA (Gruppo Italiano
Malattie Ematologiche dellAdulto) Italian Myeloma Network.
Bortezomib-thalidomide-dexamethasone is superior to thalidomidedexamethasone as consolidation therapy after autologous hematopoietic stem cell transplantation in patients with newly diagnosed
multiple myeloma. Blood. 2012;120:9-19.
Roussel M, Lauwers-Cances V, Robillard N, et al. Front-line transplantation program with lenalidomide, bortezomib, and dexamethasone
combination as induction and consolidation followed by lenalidomide
maintenance in patients with multiple myeloma: a phase II study by
the Intergroupe Francophone du Mye lome. J Clin Oncol. 2014;32:
2712-2717.
Leleu X, Fouquet G, Hebraud B, et al; Intergroupe Francophone du
Myelome (IFM). Consolidation with VTd significantly improves the
complete remission rate and time to progression following VTd induction and single autologous stem cell transplantation in multiple
myeloma. Leukemia. 2013;27:2242-2244.
Nooka AK, Kaufman JL, Muppidi S, et al. Consolidation and maintenance
therapy with lenalidomide, bortezomib and dexamethasone (RVD) in
high-risk myeloma patients. Leukemia. 2014;28:690-693.
Attal M, Lauwers-Cances V, Hulin C, et al. Autologous transplantation
for multiple myeloma in the era of new drugs: a phase III study of the
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
Intergroupe Francophone Du Myelome (IFM/DFCI 2009 Trial). Blood.

2015;126:391-391.
Sonneveld P, Asselbergs E, Zweegman S, et al. Phase 2 study of carfilzomib,
thalidomide, and dexamethasone as induction/consolidation therapy for
newly diagnosed multiple myeloma. Blood. 2015;125:449-456.
Jakubowiak A, Griffith K, Jasielec J, et al. Post-transplant carfilzomib
(Kyprolis), lenalidomide (Revlimid), and dexamethasone (KRd) consolidation in newly diagnosed multiple myeloma (NDMM): efficacy and tolerability of extended treatment. Haematologica. 2015;100:1-800.
Ladetto M, Pagliano G, Ferrero S, et al. Major tumor shrinking and
persistent molecular remissions after consolidation with bortezomib,
thalidomide, and dexamethasone in patients with autografted myeloma. J Clin Oncol. 2010;28:2077-2084.
Bjorkstrand B, Svensson H, Goldschmidt H, et al. Alpha-interferon
maintenance treatment is associated with improved survival after
high-dose treatment and autologous stem cell transplantation in patients with multiple myeloma: a retrospective registry study from the
European Group for Blood and Marrow Transplantation (EBMT). Bone
Marrow Transplant. 2001;27:511-515.
Attal M, Harousseau JL, Leyvraz S, et al; Inter-Groupe Francophone du
Myelome (IFM). Maintenance therapy with thalidomide improves
survival in patients with multiple myeloma. Blood. 2006;108:3289-3294.
Morgan GJGW, Gregory WM, Davies FE, et al; National Cancer Research
Institute Haematological Oncology Clinical Studies Group. The role of
maintenance thalidomide therapy in multiple myeloma: MRC Myeloma
IX results and meta-analysis. Blood. 2012;119:7-15.
Holstein SA, Owzar K, Richardson PG, et al. Updated analysis of CALGB/
ECOG/BMT CTN 100104: lenalidomide maintenance therapy after
single autologous stem cell transplant (ASCT) for multiple myeloma.
J Clin Oncol. 2015;33: (suppl; abstr 8523).
Palumbo A, Cavallo F, Gay F, et al. Autologous transplantation and maintenance therapy in multiple myeloma. N Engl J Med. 2014;371:895-905.
Avet-Loiseau H, Corre J, Lauwers-Cances V, et al. Evaluation of minimal
residual disease (MRD) By next generation sequencing (NGS) is highly
predictive of progression free survival in the IFM/DFCI 2009 trial. Blood.
2015;126:191-191.
Sonneveld P, Schmidt-Wolf IG, van der Holt B, et al. Bortezomib induction and maintenance treatment in patients with newly diagnosed
multiple myeloma: results of the randomized phase III HOVON-65/
GMMG-HD4 trial. J Clin Oncol. 2012;30:2946-2955.
Neben K, Lokhorst HM, Jauch A, et al. Administration of bortezomib
before and after autologous stem cell transplantation improves outcome in multiple myeloma patients with deletion 17p. Blood. 2012;119:
940-948.
Rosinol L OA, Teruel A. Maintenance therapy after stem cell transplantation for multiple myeloma with bortezomib/thalidomide vs
thalidomide vs alfa 2 b interferon: final results of a phase 3 Pethema/
GEM randomized trial. Blood. 2012;120: (abstr 334).
Shah JJ, Feng L, Weber D, et al. Phase II study of the combination of
ixazomib with lenalidomide as maintenance therapy following autologous stem cell transplant in patients with multiple myeloma. Blood.
2015;126:3155-3155.
Lokhorst HM, Plesner T, Laubach JP, et al. Targeting CD38 with daratumumab monotherapy in multiple myeloma. N Engl J Med. 2015;373:
1207-1219.
Barlogie B, Kyle RA, Anderson KC, et al. Standard chemotherapy
compared with high-dose chemoradiotherapy for multiple myeloma:
final results of phase III US Intergroup Trial S9321. J Clin Oncol. 2006;24:
929-936.
Giralt S, Aleman A, Anagnostopoulos A, et al. Fludarabine/melphalan
conditioning for allogeneic transplantation in patients with multiple
myeloma. Bone Marrow Transplant. 2002;30:367-373.
219
KRISHNAN ET AL
36. Kroger N, Sayer HG, Schwerdtfeger R, et al. Unrelated stem cell

transplantation in multiple myeloma after a reduced-intensity conditioning with pretransplantation antithymocyte globulin is highly effective with low transplantation-related mortality. Blood. 2002;100:
3919-3924.
37. Kumar S, Zhang MJ, Li P, et al. Trends in allogeneic stem cell transplantation for multiple myeloma: a CIBMTR analysis. Blood. 2011;118:
1979-1988.
38. Lokhorst HM, Wu K, Verdonck LF, et al. The occurrence of graft-versus-host
disease is the major predictive factor for response to donor lymphocyte
infusions in multiple myeloma. Blood. 2004;103:4362-4364.
39. Bruno B, Rotta M, Patriarca F, et al. A comparison of allografting with
autografting for newly diagnosed myeloma. N Engl J Med. 2007;356:
1110-1120.
40. Bjorkstrand B, Iacobelli S, Hegenbart U, et al. Tandem autologous/
reduced-intensity conditioning allogeneic stem-cell transplantation
versus autologous transplantation in myeloma: long-term follow-up.
J Clin Oncol. 2011;29:3016-3022.
41. Giaccone L, Storer B, Patriarca F, et al. Long-term follow-up of a
comparison of nonmyeloablative allografting with autografting for
newly diagnosed myeloma. Blood. 2011;117:6721-6727.
42. Gahrton G, Iacobelli S, Bjorkstrand B, et al; EBMT Chronic Malignancies
Working Party Plasma Cell Disorders Subcommittee. Autologous/
reduced-intensity allogeneic stem cell transplantation vs autologous
transplantation in multiple myeloma: long-term results of the EBMTNMAM2000 study. Blood. 2013;121:5055-5063.
43. Kneppers E, van der Holt B, Kersten MJ, et al. Lenalidomide maintenance after nonmyeloablative allogeneic stem cell transplantation in
multiple myeloma is not feasible: results of the HOVON 76 Trial. Blood.
2011;118:2413-2419.
44. Alsina M, Becker PS, Zhong X, et al. Lenalidomide maintenance for highrisk multiple myeloma after allogeneic hematopoietic cell transplantation. Biol Blood Marrow Transplant. 2014;20:1183-1189.
45. Chng WJ, Dispenzieri A, Chim CS, et al; International Myeloma Working
Group. IMWG consensus on risk stratification in multiple myeloma.
Leukemia. 2014;28:269-277.
46. Mikhael JR, Dingli D, Roy V, et al; Mayo Clinic. Management of newly
diagnosed symptomatic multiple myeloma: updated Mayo Stratification of Myeloma and Risk-Adapted Therapy (mSMART) consensus
guidelines 2013. Mayo Clin Proc. 2013;88:360-376.
47. Palumbo A, Avet-Loiseau H, Oliva S, et al. Revised International Staging
System for Multiple Myeloma: a report from International Myeloma
Working Group. J Clin Oncol. 2015;33:2863-2869.
48. Kumar S, Mahmood ST, Lacy MQ, et al. Impact of early relapse after
auto-SCT for multiple myeloma. Bone Marrow Transplant. 2008;42:
413-420.
49. Knop S LP, Hebart H, et al. Autologous followed by allogeneic versus
tandem-autologous stem cell transplant in newly diagnosed FISHdel13q myeloma. Blood. 2014;124:43.
50. Freytes CO, Vesole DH, LeRademacher J, et al. Second transplants for
multiple myeloma relapsing after a previous autotransplant-reducedintensity allogeneic vs autologous transplantation. Bone Marrow
Transplant. 2014;49:416-421.
51. de Lavallade H, El-Cheikh J, Faucher C, et al. Reduced-intensity conditioning allogeneic SCT as salvage treatment for relapsed multiple
myeloma. Bone Marrow Transplant. 2008;41:953-960.
52. Michallet M, Sobh M, El-Cheikh J, et al. Evolving strategies with
immunomodulating drugs and tandem autologous/allogeneic hematopoietic stem cell transplantation in first line high risk multiple myeloma patients. Exp Hematol. 2013;41:1008-1015.
53. Patriarca F, Einsele H, Spina F, et al. Allogeneic stem cell transplantation
in multiple myeloma relapsed after autograft: a multicenter retrospective
220
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
study based on donor availability. Biol Blood Marrow Transplant. 2012;18:

617-626.
Kroger N, Schwerdtfeger R, Kiehl M, et al. Autologous stem cell
transplantation followed by a dose-reduced allograft induces high
complete remission rate in multiple myeloma. Blood. 2002;100:
755-760.
Porrata LF, Gertz MA, Inwards DJ, et al. Early lymphocyte recovery
predicts superior survival after autologous hematopoietic stem cell
transplantation in multiple myeloma or non-Hodgkin lymphoma. Blood.
2001;98:579-585.
Kroger N, Badbaran A, Lioznov M, et al. Post-transplant immunotherapy
with donor-lymphocyte infusion and novel agents to upgrade partial
into complete and molecular remission in allografted patients with
multiple myeloma. Exp Hematol. 2009;37:791-798.
Bellucci R, Alyea EP, Weller E, et al. Immunologic effects of prophylactic
donor lymphocyte infusion after allogeneic marrow transplantation for
de Carvalho F, Alves VL, Braga WM, et al. MAGE-C1/CT7 and MAGE-C2/
CT10 are frequently expressed in multiple myeloma and can be explored in combined immunotherapy for this malignancy. Cancer
Immunol Immunother. 2013;62:191-195.
Tyler EM, Jungbluth AA, OReilly RJ, et al. WT1-specific T-cell responses
in high-risk multiple myeloma patients undergoing allogeneic T celldepleted hematopoietic stem cell transplantation and donor lymphocyte
infusions. Blood. 2013;121:308-317.
Noonan KA, Huff CA, Davis J, et al. Adoptive transfer of activated
marrow-infiltrating lymphocytes induces measurable antitumor immunity in the bone marrow in multiple myeloma. Sci Transl Med. 2015;
7:288ra78.
Garfall AL, Stadtmauer EA, June CH. Chimeric antigen receptor T cells in
myeloma. N Engl J Med. 2016;374:194.
Garfall AL, Fraietta JA, Maus MV. Immunotherapy with chimeric antigen
receptors for multiple myeloma. Discov Med. 2014;17:37-46.
Childs RW, Carlsten M. Therapeutic approaches to enhance natural
killer cell cytotoxicity against cancer: the force awakens. Nat Rev Drug
Discov. 2015;14:487-498.
Benson DM Jr, Hofmeister CC, Padmanabhan S, et al. A phase 1 trial of
the anti-KIR antibody IPH2101 in patients with relapsed/refractory
Szmania S, Lapteva N, Garg T, et al. Ex vivo-expanded natural killer cells
demonstrate robust proliferation in vivo in high-risk relapsed multiple
myeloma patients. J Immunother. 2015;38:24-36.
Shah N, Li L, Kaur I, et al. Infusion of ex vivo expanded allogeneic cord
blood-derived natural killer cells in combination with autologous stem
cell transplantation for multiple myeloma: results of a phase I study.
Blood. 2015;126:929-929.
Benson DM Jr, Bakan CE, Mishra A, et al. The PD-1/PD-L1 axis modulates
the natural killer cell versus multiple myeloma effect: a therapeutic
target for CT-011, a novel monoclonal anti-PD-1 antibody. Blood. 2010;
116:2286-2294.
Rosenblatt J, Glotzbecker B, Mills H, et al. PD-1 blockade by CT-011, antiPD-1 antibody, enhances ex vivo T-cell responses to autologous dendritic cell/myeloma fusion vaccine. J Immunother. 2011;34:409-418.
Lesokhin AM, Ansell SM, Armand P, et al. Preliminary results of a phase I
study of nivolumab(BMS-936558) in patients with relapsed or refractory lymphoid malignancies. Paper presented at: 56th ASH Annual
Meeting and Exposition; December 2014; San Francisco, CA.
San Miguel J, Mateos M-V, Shah JJ, et al. Pembrolizumab in combination
with lenalidomide and low-dose dexamethasone for relapsed/refractory
multiple myeloma (RRMM): Keynote-023. Blood. 2015;126:505-505.
Rosenblatt J, Avivi I, Vasir B, et al. Vaccination with dendritic cell/tumor
fusions following autologous stem cell transplant induces immunologic
and clinical responses in multiple myeloma patients. Clin Cancer Res.

2013;19:3640-3648.
72. Cavo M, Pantani L, Petrucci MT, et al; GIMEMA (Gruppo Italiano Malattie
Ematologiche dellAdulto) Italian Myeloma Network. Bortezomibthalidomide-dexamethasone is superior to thalidomide-dexamethasone
as consolidation therapy after autologous hematopoietic stem cell
transplantation in patients with newly diagnosed multiple myeloma.
Blood. 2012;120:9-19.
73. Roussel M, Lauwers-Cances V, Robillard N, et al. Front-line transplantation program with lenalidomide, bortezomib, and dexamethasone
combination as induction and consolidation followed by lenalidomide
maintenance in patients with multiple myeloma: a phase II study by
the Intergroupe Francophone du Myelome. J Clin Oncol. 2014;32:
2712-2717.
74. Barlogie B, Pineda-Roman M, van Rhee F, et al. Thalidomide arm of Total
Therapy 2 improves complete remission duration and survival in
myeloma patients with metaphase cytogenetic abnormalities. Blood.

2008;112:3115-3121.
75. Spencer A PH, Roberts AW, et al. Consolidation therapy with low dose
thalidomide and prednisolone prolong survival of multiple myeloma
patients undergoing a single autologous stem cell transplantation
procedure. J Clin Oncol. 2009;27:6.
76. Lokhorst HM, van der Holt B, Zweegman S, et al. A randomized phase
3 study on the effect of thalidomide combined with adriamycin,
dexamethasone and high dose melphalan, followed by thalidomide
maintenance in patients with multiple myeloma. Blood. 2010;115:
1113-1120.
77. Stewart AK, Trudel S, Bahlis NJ, et al. A randomized phase 3 trial of
thalidomide and prednisone as maintenance therapy after ASCT in
patients with MM with a quality-of-life assessment: the National Cancer
Institute of Canada Clinicals Trials Group Myeloma 10 Trial. Blood. 2013;
121:1517-1523.
221

DYSCRASIA
Defining and Redefining Myeloma

CHAIR
S. Vincent Rajkumar, MD
Mayo Clinic
Rochester, MN
SPEAKERS
Evangelos Terpos, MD, PhD
National and Kapodistrian University of Athens
Athens, Greece
Irene M. Ghobrial, MD
Boston, MA
GHOBRIAL ET AL
Future Directions in the Evaluation and Treatment of

Precursor Plasma Cell Disorders
Salomon Manier, MD, Karma Z. Salem, MD, David Liu, MD, and Irene M. Ghobrial, MD
OVERVIEW
Multiple myeloma (MM) is an incurable disease that progresses from a premalignant stage termed monoclonal gammopathy
of undetermined significance (MGUS) and an intermediate stage of smoldering multiple myeloma (SMM). Recent major
advances in therapy with more effective and less toxic treatments have brought reconsideration of early therapeutic
intervention in management of SMM, with the goal of reducing progression of the disease before the occurrence of endorgan damage to MM and improving survival. Key to this effort is accurate identification of patients at high risk of progression who would truly benefit from early intervention. In this review, we discuss the current definitions, risk factors, risk
stratification, prognosis, and management of MGUS and SMM, as well as new emerging therapeutic options under active
investigation.
ultiple myeloma evolves through a spectrum of disease from a premalignant stage of MGUS to an intermediate stage of SMM and finally presents with symptoms
and signs of end-organ damage that lead to the diagnosis of
MM.1 Recent studies have indicated that almost all cases of
MM are preceded by the precursor state of MGUS or SMM.2,3
Over the years, the diagnosis of MM required evidence of
end-organ damage attributable to the neoplastic clone of
plasma cells, the so-called CRAB criteria identified by hypercalcemia, renal failure, anemia, and osteolytic bone
lesions.4 This definition was intended to be conservative to
avoid unnecessary and toxic administration of chemotherapy
to patients with asymptomatic disease. However, with major
advances in therapy, and identification of potential biomarkers that can distinguish MM from premalignant stages, it
became necessary to revise the disease definition of MM.5-7
In 2014, the International Myeloma Working Group (IMWG)
updated the diagnostic criteria for MM to add three markers
that can identify additional patients not yet meeting CRAB
criteria who should nevertheless be treated. These were
termed myeloma-defining events (MDEs) and constituted
the presence of clonal bone marrow (BM) plasma cells
greater than or equal to 60%, serum free light chain (FLC)
ratio greater than or equal to 100, provided the involved
FLC level is higher than or equal to 100 mg/L, or more than
one focal lesion on MRI or CT and PET-CT.8
Importantly, the revised definition excluded patients
previously considered to have SMM with ultra-high risk
of progression (80% within 2 years) who are now classified
as MM based on the updated diagnostic criteria. However,

it should be noted that the revised diagnostic criteria for
MM upstages only a small proportion of patients with
SMM. Thus, SMM as an entity continue to be relevant and
important. Therefore, there is a need to better define the
clinical and molecular characteristics of patients with
MGUS and SMM that predict progression to MM. Here,
we review the diagnostic criteria and risks of progression for MGUS and SMM and delineate potential therapeutic options that can be used for patients with high-risk
SMM.
DEFINITION, INCIDENCE, AND RISK OF

PROGRESSION OF MGUS
MGUS is a premalignant precursor of MM.9,10 It is defined
by a serum M-protein concentration lower than 3 g/dL and/or
an abnormal FLC ratio (, 0.26 or . 1.65), less than 10%
clonal plasma cells in the BM, and absence of MDEs or
evidence of Waldenstrom macroglobulinemia or amyloidosis (Table 1).8 When first clinically recognized, MGUS has
likely been present in an undetected state for a median
duration of more than 10 years.11
MGUS is present in over 3% to 4% of the population older
than age 5010 and increases with age to 8.9% in people older
than age 85.1 Therefore, one of the most important risk
factors for MGUS is age.1,12 In addition to an accumulation of
cases, the age-related increase in prevalence of MGUS is
related to a true increase in incidence with age.
From the Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA.
Corresponding author: Irene M. Ghobrial, MD, Dana-Farber Cancer Institute, 450 Brookline Ave., Boston, MA 02115; email: Irene_ghobrial@dfci.harvard.edu.
e400
DIAGNOSIS AND MANAGEMENT OF MGUS AND SMOLDERING MYELOMA
Apart from age, there are several factors that are associated with a higher risk of developing MGUS, including
African American race and a familial history of myeloma.
Other factors such as the presence of autoimmune disorders
are less well delineated. Persons of African and African
American descent have a threefold increased prevalence
even after adjusting for socioeconomic and other risk factors, suggesting a genetic predisposition among this population.13 An increased prevalence of MGUS in Africans
relative to Caucasians has also been reported in Ghana.14,15
Two studies have shown that the risk of MM and MGUS
is increased threefold in relatives of individuals with
MGUS.16,17 In addition, an increased risk of developing nonHodgkin lymphoma and chronic lymphoid leukemia was also
observed in this group. Collectively, these data are consistent with an inherited twofold to fourfold genetic susceptibility to MM.18 The genetic basis of inherited MM
susceptibility is incompletely understood. However, recent
genome-wide association studies (GWAS) identified seven
loci contributing to inherited genetic susceptibility to
MM.18-21 These seven loci were also associated with an
increased risk of developing MGUS.22 Another GWAS study
in the Nordic region identified a novel MM risk locus at
ELL2.23
There are three subtypes of MGUS: nonimmunoglobulin
M (IgM) MGUS, IgM MGUS, and light-chain MGUS. Non-IgM
MGUS carries a risk of progression to MM or solitary
plasmacytoma, whereas IgM MGUS is associated with a risk
of progression to Waldenstrom macroglobulinemia. Lightchain MGUS is a newly discovered entity that is associated
with a risk of progression to the light-chain type of MM. All
forms of MGUS can progress to Ig light-chain amyloidosis.
MGUS is associated with a lifelong risk of transformation
to MM or a related malignancy at a rate of 1% per year.9 In
some patients, however, the risk may be as high as 58% in
20 years. The size and subtype of the M protein at diagnosis of
MGUS and an abnormal serum FLC ratio are well-established
prognostic factors for progression.24 As such, MGUS can be
risk stratified using three simple variables: the serum FLC
KEY POINTS
Multiple myeloma is always preceded by precursor

conditions including MGUS and smoldering myeloma.
The rate of progression of MGUS is at 1% per year, but
the rate of progression of SMM is 10% per year, with a
high rate of progression of 50% at 2 years for those with
high-risk features.
Patients with high-risk SMM should be offered clinical
trials to prevent progression or carefully observed to
avoid end-organ damage.
Patients with MGUS and SMM have different risk
stratifications and rates of progression.
Some patients may benefit from early therapeutic
interventions, specifically in high-risk SMM.
TABLE 1. IMWG Diagnostic Criteria for MGUS, SMM,

and MDEs
IMWG Criteria, 2014 Version8
MGUS
Serum M protein , 3 g/dL and/or abnormal FLC ratio

(, 0.26 or . 1.65) with increased level of appropriate
involved light chain
Clonal BM plasma cells , 10%
Absence of MDEs or amyloidosis
SMM
Serum M protein . 3 g/dL

AND/OR clonal BM plasma cells . 10% and , 60%
AND/OR urinary monoclonal protein . 500 mg per 24
hours
Absence of MDEs or amyloidosis
MDEs
Evidence of end-organ damage (CRAB criteria):

Hypercalcemia: serum Ca2+ . 0.25 mmol/L (. 1 mg/dL)
above upper limit of normal or . 2.75 mmol/L
(. 11 mg/dL)
Renal insufficiency: serum creatinine . 173 mmol/L
(. 2 mg/dL)
Anemia: hemoglobin value of . 2 g/dL below the lower
limit of normal or a hemoglobin value , 10 g/dL
Bone lesions: one or more lytic lesion(s) on skeletal
radiography, CT, or PET-CT
Clonal BM plasma cell percentage $ 60%
Involved/uninvolved serum FLC ratio $ 100
More than one focal lesions on MRI
Abbreviations: IMWG, International Myeloma Working Group; MGUS,

monoclonalgammopathy of undetermined significance; SMM, smoldering multiple
myeloma; MM, multiple myeloma; MDEs, myeloma-defining events; FLC, serum free
light chain; BM, bone marrow.
ratio, the size of the M protein, and the type of M protein

(Table 2). Patients with serum M protein , 1.5 g/dL, IgG
class, and normal FLC ratio are considered low-risk MGUS
and have only a 2% lifetime risk of progression. In general,
patients with MGUS must be reassessed in 6 months, and
if stable yearly thereafter. A number of lifestyle and environmental risk factors have been proposed to increase the
risk of progression from MGUS to MM, including obesity,
immune dysfunction, and agricultural, chemical, and radiation exposure. Of these, obesity is probably the most consistently reported association.25-27 Studies are ongoing to
determine the mechanism of progression of MGUS and to
explore preventive strategies.
In addition to the risk of progression, MGUS can also be
associated with several other clinical problems including
sensorimotor peripheral neuropathy (MGUS neuropathy),
membranoproliferative glomerulonephritis, lichen myxedematosus (papular mucinosis, scleromyxedema), pyoderma gangrenosum, or necrobiotic xanthogranuloma.
DEFINITION, INCIDENCE, AND RISK OF

PROGRESSION OF SMM
SMM is a heterogeneous disease entity that includes patients with higher disease burden than in MGUS but remain
asymptomatic.28 The term SMM was first described by Kyle
e401
GHOBRIAL ET AL
TABLE 2. Risk of Progression of Monoclonal Gammopathy of Undetermined Significance to Myeloma or Related

Disorders
Relative Risk of
Progression
Risk Group
Low Risk: Serum M Protein
< 1.5 g/dL, IgG Subtype,
Normal Free Light Chain Ratio
(0.261.65)
Low-Intermediate Risk: Any One

Factor Abnormal
Cumulative Absolute Risk of

Progression at 20 Years (%)*
Cumulative Absolute Risk of

Progression at 20 Years Accounting
for Death as a Competing Risk (%)**
5.4
21
10
High-Intermediate Risk: Any Two

Factors Abnormal
10.1
37
18
High Risk: All Three Factors

Abnormal
20.8
58
27
Adapted from Rajkumar et al56 with permission from the publisher.

*Estimates in this column represent the risk of progression assuming that patients do not die of other causes during this period.
**Estimates in this column represent the risk of progression calculated by using a model that accounts for the fact that patients can die of unrelated causes during this time.
Abbreviation: IgG, immunoglobulin G.
and Greipp in 198029 and was followed by many other

descriptions terming it indolent MM,30 or Durie Salmon
stage I.31 In 2003, the IMWG defined SMM as having a serum
M-protein level higher than or equal to 3 g/L and/or greater
than or equal to 10% monoclonal plasma cells in the BM
(Table 3).1,4 Although the incidence and prevalence of SMM
in the population is not well defined, it has been estimated to
represent approximately 8% to 20% of patients within the
MM spectrum.28
The update to the disease definition of MM automatically

resulted in revised diagnostic criteria for SMM as it excludes
patients with ultra-highrisk SMM who are now considered
as having overt MM (Table 1).8 Despite that, it remains a
major clinical dilemma with an overall risk of progression of
10% per year for the first 5 years, 3% per year for the next
5 years, and 1% per year for the last 10 years. This suggests
that the current definition of SMM is highly biologically
and clinically heterogeneous.32 Indeed, SMM represents a
TABLE 3. Risk Stratification of Patients with Smoldering Myeloma Based on Mayo Clinic and Spanish Criteria
Model
Risk Factors
No. of Risk Factors
5-Year Progression (%)
Relative Risk
Mayo Clinic Model
M protein $ 3 g/dL
25
$ 10% BM plasma cells
51
2.0
FLC ratio , 0.125 or . 8
76
3.0
Total
51
NA
Spanish (PETHEMA) Model
Proposed New Criteria for

High-Risk SMM
$ 95% aPC
Immunoparesis
46
11.5
72
18
Total
46
NA
$ 10% BM plasma cells
Serum M protein $ 30 g/L
With one of the criteria in the

right column
IgA SMM
Immunoparesis with reduction of two uninvolved Ig isotypes

Serum involved/uninvolved FLC ratio $ 8 (but , 100)
Progressive increase in M-protein level (evolving type of SMM)
BM clonal plasma cells 50% to 60%
aPC immunophenotype ($ 95% of BM plasma cells are clonal)
and reduction of one or more uninvolved Ig isotypes
t(4;14) or del 17p or 1q gain
Increased circulating plasma cells
MRI with diffuse abnormalities or one focal lesion
PET-CT with focal lesion with increased uptake without underlying osteolytic bone
destruction
Abbreviations: BM, bone marrow; FLC, serum free light chain; NA, not applicable, aPC, abnormal plasma cells; IgA, immunoglobulin A; SMM, smoldering multiple myeloma.
e402
clinical entity where a subset of patients have a very indolent

course of disease that mimics an MGUS-like state, whereas
others have a more aggressive course of disease that has
been described as early myeloma or CRAB-negative
myeloma. Unfortunately, there are currently no molecular
factors to differentiate these two clinically and biologically
distinct groups of patients. For this reason, additional studies
are required to identify markers of progression within these
patients.
There are three subtypes of SMM: IgA, IgG, and light chain.
The median time to progression (TTP) for IgA and IgG SMM
is 27 and 75 months, respectively.32,33 Light-chain SMM
has a cumulative probability of progression to active MM
or light-chain amyloidosis of 27.8% at 5 years, 44.6% at
10 years, and 56.5% at 15 years.34 The current factors associated with risk of progression are mainly based on the
level of tumor burden in these patients assessed by the
degree of tumor involvement in the BM and the quantification of monoclonal protein in the peripheral blood.
The two most widely used risk stratification methods
are the Mayo Clinic32 and the PETHEMA Spanish group
classifications.35 The Mayo Clinic criteria are primarily based
on the levels of serum protein markers (serum protein
electrophoresis and FLC assay) and the percentage of plasma
cells in the BM.32,33 The risk stratification of the PETHEMA
Study Group, on the other hand, focused on the use of
multiparameter flow cytometry of the BM to quantify the
ratio of abnormal, neoplastic plasma cells to normal plasma
cells and reduction of uninvolved Igs.35 Interestingly, a headto-head comparison between the PETHEMA and the Mayo
Clinic risk models showed notable discordance reflected in
many patients being classified as high risk in one model and
low risk in the other model.36 Other risk factors that have
been examined include the role of IgA (vs. IgG) isotype, the
presence of proteinuria, circulating plasma cells, a high
proliferative rate of BM plasma cells, and abnormal MRI
findings.12,28,37
SMM can also be subclassified based on underlying cytogenetic abnormalities.38,39 Patients with t(4;14), 1q gain,
and/or del17p are considered high-risk SMM (median TTP of
approximately 24 months). Patients with hyperdiploidy are
considered intermediate-risk (median TTP of approximately
34 months). Other cytogenetic abnormalities including t(11;14)
are considered standard risk (median TTP, 54 months).
Patients with SMM who have no evidence of cytogenetic
abnormalities on fluorescence in situ hybridization (FISH)
studies are considered low risk (median TTP, 101 months).
An international workshop assembled to review cytogenetic studies to evaluate whether MGUS and SMM cases
have the same detectable anomalies that are often found in
MM.40 Point mutations in NRAS and KRAS, MYC upregulation,41
and gain or loss of chromosome 1q or 1p seem to correlate
with disease progression from MGUS and SMM.42 A progressive increase in the incidence of copy number abnormalities from MGUS to SMM and to MM has also been
recently observed.43 Although MM has more copy number abnormalities than its precursor states, MGUS is as
genetically aberrant as MM and does not appear to be

associated with a particular chromosomal imbalance but
rather with an expansion of altered clones that are already
present in MGUS.43
A study of transcriptional profiling using gene expression
profiling has identified signatures that can identify patients
with MGUS versus SMM versus MM.44 However, this study
has major limitations given that the percentage of plasma
cells present in MGUS cases is low and the samples are
inherently contaminated with normal nontransformed plasma
cells.
Based on all these factors, a new classification for high-risk
SMM has been proposed by Rajkumar et al to identify patients at high risk of progression to MM (25% per year) as
defined by the criteria listed in Table 3.32,33,35,38,39,45-49
Although these biomarkers were defined before the revisions to the diagnostic criteria, because the proportion of
patients with SMM affected and upstaged as a result of the
new criteria are small, the effect on the estimates for
progression would likely be minimal. Identification of highrisk SMM is of particular importance because these patients
are at considerable risk of end-organ damage and are
candidates for clinical trials and/or intervention.7 In contrast, patients with SMM without high-risk factors likely
have a risk of progression of 5% per year or less.
MANAGEMENT OF MGUS AND SMM

Patients with MGUS are not offered therapeutic options to
date and close observation is indicated in these patients. A
recent study demonstrated that prior knowledge of MGUS
and close monitoring had significantly better overall survival
(median survival, 2.8 years) than patients with MM without
prior knowledge of MGUS (median survival, 2.1 years).50 In
fact, low M-protein concentration (, 0.5 g/dL) at MGUS
diagnosis was associated with poorer MM survival, presumably because those patients were not closely followed
for disease progression.50
All patients suspected to have SMM need an MRI of the
spine and pelvis (or ideally whole-body MRI) or whole-body
CT or PET-CT.1,8,51 BM examination with FISH studies and
multiparametric flow cytometry are also required. The
standard of care for SMM remains observation with reevaluations of the patients every 34 months.1,8,49 In lowrisk patients, follow-up can be reduced to once every
6 months after the first 5 years. Imaging studies must be
repeated if changes in clinical features or M protein occur.
For patients at high risk, follow-up should continue indefinitely, and should include periodic imaging studies to
rule out asymptomatic progression. Patients with SMM can
be initiated on therapy without waiting for CRAB features to
appear if follow-up testing shows the development of other
MDE or early detection of MM bone disease on the basis of
advanced imaging studies. Patients with a baseline MRI
showing diffuse infiltration, solitary focal lesion, or equivocal
lesions, need follow-up examinations in 36 months to rule
out progression.8
e403
GHOBRIAL ET AL
Investigators attempted to assess whether early therapeutic intervention can lead to substantial improvement in
survival and response.28 There are two major goals in early
therapeutic intervention: the first is prevention of progression, and the second is definitive therapy to achieve
complete remission with the hope that all subclones will be
eradicated at this early disease state and cure can be
achieved.37,52
The major barrier to early intervention has been defining
the group of patients who would truly benefit from this early
treatment as they would have otherwise progressed to
symptomatic disease. Indeed, if SMM is a heterogeneous
mix of patients with early myeloma and MGUS-like myeloma, then identification of those with early MM will allow
for intervention only in those patients who truly warrant
therapy.
Bisphophonates
Bisphosphonates have shown promise in reducing the risk of
skeletal-related events (SREs) in SMM, but have not been
shown to delay progression to MM or prolong survival. In
a randomized trial of pamidronate (once monthly for
12 months) versus observation, the incidence of SRE was
39% versus 73%, respectively (p = .009).53 In another trial of
zoledronic acid (monthly for 12 months) versus observation,
the incidence of SRE was 56% versus 78%, respectively
(p = .041).54 The reduction in SREs is an important endpoint,
and based on these two trials, consideration must be given
to bisphosphonates. We recommend once-yearly bisphosphonate in patients at low risk, and once every 34 months in select
patients with high-risk SMM.
Lenalidomide
The most critical study of therapy in SMM that has reignited
interest in therapeutic intervention in this patient population came from the PETHEMA group using lenalidomide
and dexamethasone compared with observation. Mateos
et al7 reported on 119 patients with high-risk SMM who
received either observation or lenalidomide and dexamethasone in an open-label randomized trial. Patients
treated with lenalidomide and dexamethasone had a superior 3-year survival without progression to symptomatic
disease (progression-free survival; 77% vs. 30%; p , .001)
and a superior 3-year overall survival (94% vs. 80%; p = .03)

from the time of registration. However, this study was
criticized by many groups because of how asymptomatic
biochemical progression was handled in both arms, the short
overall survival of the abstention group and the use of
salvage therapy in the abstention group. Because of these
concerns, additional studies are needed before implementing therapeutic interventions as a standard of care in
Investigational Therapies
More intensive treatment approaches similar to MM are
also being investigated in high-risk SMM. These include
triplet therapy with carfilzomib, lenalidomide, and dexamethasone, as well as combining combination therapy,
autologous stem cell transplantation, and maintenance.55
Immune-based approaches using elotuzumab or daratumumab or vaccine therapies are also being investigated in
On the basis of available data, we recommend that patients with high-risk SMM be offered clinical trials testing
early intervention or observed closely. However, given the
high risk of progression, select patients with high-risk SMM
with multiple risk factors or evidence of biologic progression
(rising M-protein level) can be considered for MM therapy.49
There are no specific factors to make this determination, and
clinical judgment is needed.
FUTURE DIRECTIONS
Further research is needed to identify factors involved in
progression of MGUS and SMM to MM. We must identify
additional reliable biomarkers of malignancy. Finally, we
need a portfolio of clinical trials designed to determine
whether early intervention can improve survival or provide a
path to prevent or cure MM. The question remains whether
these efforts will lead to a cure in myeloma or potentially to
an early screening and intervention modality that will completely eradicate the progression to symptomatic disease,
making myeloma a preventable disease. Only well-designed
clinical trials will be able to determine which interventions
are most effective, what populations should be targeted,
and when interventions should take place.
References
1. Kyle RA, Durie BG, Rajkumar SV, et al; International Myeloma Working
Group. Monoclonal gammopathy of undetermined significance (MGUS)
and smoldering (asymptomatic) multiple myeloma: IMWG consensus
perspectives risk factors for progression and guidelines for monitoring
and management. Leukemia. 2010;24:1121-1127.
2. Landgren O, Kyle RA, Pfeiffer RM, et al. Monoclonal gammopathy of
undetermined significance (MGUS) consistently precedes multiple
myeloma: a prospective study. Blood. 2009;113:5412-5417.
e404
3. Weiss BM, Abadie J, Verma P, et al. A monoclonal gammopathy precedes multiple myeloma in most patients. Blood. 2009;113:5418-5422.
4. International Myeloma Working Group. Criteria for the classification of
monoclonal gammopathies, multiple myeloma and related disorders:
a report of the International Myeloma Working Group. Br J Haematol.
2003;121:749-757.
5. Rajkumar SV, Larson D, Kyle RA. Diagnosis of smoldering multiple
6. Larsen JT, Kumar SK, Dispenzieri A, et al. Serum free light chain ratio as a
biomarker for high-risk smoldering multiple myeloma. Leukemia. 2013;
27:941-946.
7. Mateos M-V, Hernandez MT, Giraldo P, et al. Lenalidomide plus
dexamethasone for high-risk smoldering multiple myeloma. N Engl J
Med. 2013;369:438-447.
8. Rajkumar SV, Dimopoulos MA, Palumbo A, et al. International Myeloma
Working Group updated criteria for the diagnosis of multiple myeloma.
9. Kyle RA, Therneau TM, Rajkumar SV, et al. A long-term study of
prognosis in monoclonal gammopathy of undetermined significance.
N Engl J Med. 2002;346:564-569.
10. Kyle RA, Therneau TM, Rajkumar SV, et al. Prevalence of monoclonal
gammopathy of undetermined significance. N Engl J Med. 2006;354:
1362-1369.
11. Therneau TM, Kyle RA, Melton LJ III, et al. Incidence of monoclonal
gammopathy of undetermined significance and estimation of duration
before first clinical recognition. Mayo Clin Proc. 2012;87:1071-1079.
12. Agarwal A, Ghobrial IM. Monoclonal gammopathy of undetermined
significance and smoldering multiple myeloma: a review of the current
understanding of epidemiology, biology, risk stratification, and management of myeloma precursor disease. Clin Cancer Res. 2013;19:985994.
13. Greenberg AJ, Vachon CM, Rajkumar SV. Disparities in the prevalence,
pathogenesis and progression of monoclonal gammopathy of undetermined significance and multiple myeloma between blacks and
whites. Leukemia. 2012;26:609-614.
14. Buadi F, Hsing AW, Katzmann JA, et al. High prevalence of polyclonal
hypergamma-globulinemia in adult males in Ghana, Africa. Am J Hematol. 2011;86:554-558.
15. Landgren O, Katzmann JA, Hsing AW, et al. Prevalence of monoclonal
gammopathy of undetermined significance among men in Ghana. Mayo
Clin Proc. 2007;82:1468-1473.
16. Landgren O, Kristinsson SY, Goldin LR, et al. Risk of plasma cell and
lymphoproliferative disorders among 14621 first-degree relatives of
4458 patients with monoclonal gammopathy of undetermined significance in Sweden. Blood. 2009;114:791-795.
17. Vachon CM, Kyle RA, Therneau TM, et al. Increased risk of monoclonal
gammopathy in first-degree relatives of patients with multiple myeloma or monoclonal gammopathy of undetermined significance. Blood.
2009;114:785-790.
18. Morgan GJ, Johnson DC, Weinhold N, et al. Inherited genetic susceptibility to multiple myeloma. Leukemia. 2014;28:518-524.
19. Chubb D, Weinhold N, Broderick P, et al. Common variation at 3q26.2,
6p21.33, 17p11.2 and 22q13.1 influences multiple myeloma risk. Nat
Genet. 2013;45:1221-1225.
20. Weinhold N, Johnson DC, Chubb D, et al. The CCND1 c.870G.A
polymorphism is a risk factor for t(11;14)(q13;q32) multiple myeloma.
Nat Genet. 2013;45:522-525.
21. Broderick P, Chubb D, Johnson DC, et al. Common variation at 3p22.1
and 7p15.3 influences multiple myeloma risk. Nat Genet. 2012;44:
58-61.
22. Weinhold N, Johnson DC, Rawstron AC, et al. Inherited genetic susceptibility to monoclonal gammopathy of unknown significance. Blood.
2014;123:2513-2517, quiz 2593.
23. Swaminathan B, Thorleifsson G, Joud M, et al. Variants in ELL2 influencing immunoglobulin levels associate with multiple myeloma. Nat
Commun. 2015;6:7213.
24. Rajkumar SV, Kyle RA, Therneau TM, et al. Serum free light chain ratio is
an independent risk factor for progression in monoclonal gammopathy
of undetermined significance. Blood. 2005;106:812-817.
25. Thompson MA, Kyle RA, Melton LJ III, et al. Effect of statins, smoking and
obesity on progression of monoclonal gammopathy of undetermined
significance: a case-control study. Haematologica. 2004;89:626-628.
26. Kyle RA, Rajkumar SV. Epidemiology of the plasma-cell disorders. Best
Pract Res Clin Haematol. 2007;20:637-664.
27. Landgren O, Rajkumar SV, Pfeiffer RM, et al. Obesity is associated with
an increased risk of monoclonal gammopathy of undetermined significance among black and white women. Blood. 2010;116:1056-1059.
28. Dispenzieri A, Stewart AK, Chanan-Khan A, et al. Smoldering multiple
myeloma requiring treatment: time for a new definition? Blood. 2013;
122:4172-4181.
29. Kyle RA, Greipp PR. Smoldering multiple myeloma. N Engl J Med. 1980;
302:1347-1349.
30. Alexanian R. Localized and indolent myeloma. Blood. 1980;56:521-525.
31. Durie BG, Salmon SE. A clinical staging system for multiple myeloma.
Correlation of measured myeloma cell mass with presenting clinical
features, response to treatment, and survival. Cancer. 1975;36:
842-854.
32. Kyle RA, Remstein ED, Therneau TM, et al. Clinical course and prognosis
of smoldering (asymptomatic) multiple myeloma. N Engl J Med. 2007;
356:2582-2590.
33. Dispenzieri A, Kyle RA, Katzmann JA, et al. Immunoglobulin free light
chain ratio is an independent risk factor for progression of smoldering
(asymptomatic) multiple myeloma. Blood. 2008;111:785-789.
34. Kyle RA, Larson DR, Therneau TM, et al. Clinical course of light-chain
smouldering multiple myeloma (idiopathic Bence Jones proteinuria):
a retrospective cohort study. Lancet Haematol. 2014;111:e28-e36.
35. Perez-Persona E, Vidriales MB, Mateo G, et al. New criteria to identify
risk of progression in monoclonal gammopathy of uncertain significance
and smoldering multiple myeloma based on multiparameter flow
cytometry analysis of bone marrow plasma cells. Blood. 2007;110:
2586-2592.
36. Cherry BM, Korde N, Kwok M, et al. Modeling progression risk for
smoldering multiple myeloma: results from a prospective clinical study.
Leuk Lymphoma. 2013;54:2215-2218.
37. Landgren O. Monoclonal gammopathy of undetermined significance
and smoldering multiple myeloma: biological insights and early
treatment strategies. Hematology Am Soc Hematol Educ Program.
2013;2013:478-487.
38. Rajkumar SV, Gupta V, Fonseca R, et al. Impact of primary molecular
cytogenetic abnormalities and risk of progression in smoldering multiple myeloma. Leukemia. 2013;27:1738-1744.
39. Neben K, Jauch A, Hielscher T, et al. Progression in smoldering myeloma
is independently determined by the chromosomal abnormalities del
(17p), t(4;14), gain 1q, hyperdiploidy, and tumor load. J Clin Oncol. 2013;
31:4325-4332.
40. Fonseca R, Barlogie B, Bataille R, et al. Genetics and cytogenetics of
multiple myeloma: a workshop report. Cancer Res. 2004;64:1546-1558.
41. Chng WJ, Huang GF, Chung TH, et al. Clinical and biological implications
of MYC activation: a common difference between MGUS and newly
diagnosed multiple myeloma. Leukemia. 2011;25:1026-1035.
42. Chiecchio L, Dagrada GP, Protheroe RK, et al; UK Myeloma Forum. Loss
of 1p and rearrangement of MYC are associated with progression of
smouldering myeloma to myeloma: sequential analysis of a single case.
Haematologica. 2009;94:1024-1028.
43. Lopez-Corral
L, Sarasquete ME, Beà S, et al. SNP-based mapping arrays

reveal high genomic complexity in monoclonal gammopathies, from
MGUS to myeloma status. Leukemia. 2012;26:2521-2529.
44. Zhan F, Barlogie B, Arzoumanian V, et al. Gene-expression signature of
benign monoclonal gammopathy evident in multiple myeloma is linked
to good prognosis. Blood. 2007;109:1692-1700.
e405
GHOBRIAL ET AL
~ol L, Blade J, Esteve J, et al. Smoldering multiple myeloma: natural

45. Rosin
history and recognition of an evolving type. Br J Haematol. 2003;123:
631-636.
46. Dhodapkar MV, Sexton R, Waheed S, et al. Clinical, genomic, and
imaging predictors of myeloma progression from asymptomatic
monoclonal gammopathies (SWOG S0120). Blood. 2014;123:78-85.
47. Bianchi G, Kyle RA, Larson DR, et al. High levels of peripheral blood
circulating plasma cells as a specific risk factor for progression of
smoldering multiple myeloma. Leukemia. 2013;27:680-685.
48. Hillengass J, Fechtner K, Weber MA, et al. Prognostic significance of
focal lesions in whole-body magnetic resonance imaging in patients
with asymptomatic multiple myeloma. J Clin Oncol. 2010;28:1606-1610.
49. Rajkumar SV, Landgren O, Mateos MV. Smoldering multiple myeloma.
Blood. 2015;125:3069-3075.
50. Sigurdardottir EE, Turesson I, Lund SH, et al. The role of diagnosis and
clinical follow-up of monoclonal gammopathy of undetermined significance on survival in multiple myeloma. JAMA Oncol. 2015;1:168-174.
51. Blade J, Dimopoulos M, Rosin~ol L, et al. Smoldering (asymptomatic)
multiple myeloma: current diagnostic criteria, new predictors of outcome, and follow-up recommendations. J Clin Oncol. 2010;28:690-697.
e406
52. Ghobrial IM, Landgren O. How I treat smoldering multiple myeloma.

Blood. 2014;124:3380-3388.
53. DArena G, Gobbi PG, Broglia C, et al; Gimema (Gruppo Italiano Malattie
Ematologiche DellAdulto); Multiple Myeloma Working Party; Gisl
(Gruppo Italiano Studio Linfomi) Cooperative Group. Pamidronate
versus observation in asymptomatic myeloma: final results with longterm follow-up of a randomized study. Leuk Lymphoma. 2011;52:
771-775.
54. Musto P, Petrucci MT, Bringhen S, et al; GIMEMA (Italian Group for
Adult Hematologic Diseases)/Multiple Myeloma Working Party and the
Italian Myeloma Network. A multicenter, randomized clinical trial
comparing zoledronic acid versus observation in patients with
asymptomatic myeloma. Cancer. 2008;113:1588-1595.
55. Zingone A, et al. Phase II clinical and correlative study of carfilzomib,
lenalidomide, and dexamethasone followed by lenalidomide extended
dosing (CRD-R) induces high rates of MRD negativity in newly diagnosed
multiple myeloma (MM) patients. Blood. 2013;122:538-538.
56. Rajkumar SV, Kyle RA, Therneau TM, et al. Serum free light chain ratio is
an independent risk factor for progression in monoclonal gammopathy
of undetermined significance (MGUS). Blood. 2005;106:812-817.
IMAGING FOR MULTIPLE MYELOMA
The Role of Imaging in the Treatment of Patients With

Multiple Myeloma in 2016
Evangelos Terpos, MD, PhD, Meletios A. Dimopoulos, MD, and Lia A. Moulopoulos, MD
OVERVIEW
The novel criteria for the diagnosis of symptomatic multiple myeloma have revealed the value of modern imaging for the
management of patients with myeloma. Whole-body low-dose CT (LDCT) has increased sensitivity over conventional radiography for the detection of osteolytic lesions, and several myeloma organizations and institutions have suggested that
whole-body LDCT should replace conventional radiography for the work-up of patients with myeloma. MRI is the best
imaging method for the depiction of marrow infiltration by myeloma cells. Whole-body MRI (or at least MRI of the spine and
pelvis if whole-body MRI is not available) should be performed for all patients with smoldering multiple myeloma with no
lytic lesions to look for occult disease, which may justify treatment. In addition, MRI accurately illustrates the presence of
plasmacytomas, spinal cord, and/or nerve compression for surgical intervention or radiation therapy; it is also recommended for the work-up of solitary bone plasmacytoma, and it may distinguish malignant from benign fractures (which is
very important in cases of patients in biochemical remission with no other signs of progression). Diffusion weighted imaging
(DWI) seems to improve MRI diagnosis in patients with myeloma. PET/CT is a functional imaging technique, more sensitive
than conventional radiography for the detection of lytic lesions, which probably allows better definition of complete
response and minimal residual disease compared with all other imaging methods. PET/CT has shown the best results in the
follow-up of patients with myeloma and has an independent prognostic value both at diagnosis and following treatment.
PET/CT can also be used for the work-up of solitary bone plasmacytoma and nonsecretory myeloma.
ytic bone disease is a major feature of multiple myeloma:

70% to 80% of patients have osteolytic lesions at diagnosis,
and up to 90% develop lytic lesions during the course of their
disease.1 The novel criteria for the diagnosis of symptomatic
multiple myeloma have revealed the value of modern imaging
for the management of patients with multiple myeloma, as
they include (1) the presence at least one lytic lesion detected
not only by conventional radiography but also by CT, wholebody LDCT, or PET/CT, and (2) the presence of one or more
focal bone marrow lesions on MRI studies.2 Furthermore, novel
imaging techniques, such as MRI and PET/CT, provide prognostic information and have been recently proven of value, for
the better definition of response to anti-myeloma therapy.
IS CONVENTIONAL RADIOGRAPHY THE GOLD

STANDARD FOR THE DIAGNOSIS OF BONE
DISEASE IN PATIENTS WITH MYELOMA IN 2016?
The Role of Whole-Body LDCT
Conventional radiography has been widely used for the
identification of osteolytic lesions both at diagnosis and
during the course of the disease. The skeletal survey (whole

body x-rays) at diagnosis should include plain radiographs of
the whole skeleton, as previously described.3 Osteolyses
have the typical appearance of punched-out lesions with
the absence of reactive sclerosis (Fig. 1) and are more
common in the vertebrae, ribs, skull, and pelvis.4 Although
the whole-body x-ray was the standard of care for many
years, it has several limitations: (1) for a lytic lesion to become apparent, more than 30% loss of trabecular bone must
occur; (2) it is difficult to assess certain areas, such as the
pelvis and the spine; (3) there are limitations to the detection of lytic lesion response to anti-myeloma therapy
because of delayed evidence of healing; (4) specificity is
reduced for the differential diagnosis of myeloma-related
fracture and benign fracture (very important, particularly in
cases of new vertebral compression fractures in the absence
of other criteria of relapse); (5) it is dependent on the observer (there is very low reproducibility among centers;
higher number of osteolytic lesions detected in academic
centers compared with nonacademic centers); and (6)
From the School of Medicine, National and Kapodistrian University of Athens, Alexandra General Hospital, Athens, Greece; School of Medicine, National and Kapodistrian University of
Athens, Areteion Hospital, Athens, Greece.
Corresponding author: Evangelos Terpos, MD, PhD, Department of Clinical Therapeutics, University of Athens School of Medicine, Alexandra General Hospital, 80 Vas. Sofias Ave.,
11528, Athens, Greece; email: eterpos@med.uoa.gr.
e407
TERPOS, DIMOPOULOS, AND MOULOPOULOS
studies are long and often not tolerable for patients in severe
pain.3,4 Thus, the development of novel imaging methods
has led to the substitution of whole-body x-ray by more
advanced techniques, such as the whole-body LDCT in many
European centers or by PET/CT in the United States.
Whole-body LDCT was introduced to allow the detection of
osteolytic lesions in the whole skeleton with high accuracy, no
need for contrast agents and low radiation dose compared
with standard CT (two- to threefold lower radiation dose
versus conventional CT).5,6 In several studies, whole-body LDCT
was found to be superior to whole-body x-ray for the detection
of osteolytic lesions (Table 1).5,7-11 In one of the largest studies
staging patients with myeloma, 61% of patients with normal
whole-body x-ray results had more than one osteolytic lesions
on whole-body LDCT.10 According to the new criteria for
symptomatic myeloma, these patients should receive therapy.
In the same study, the total number of lesions detected by
whole-body LDCT was 968 compared with 248 for whole-body
x-ray (p , .001), with the only limitation of this study being its
retrospective origin.10 In a more recent prospective study,
which included 52 patients with myeloma at diagnosis, wholebody LDCT revealed osteolysis in 12 patients (23%) with
negative whole-body x-ray, and it proved to be more sensitive
than whole-body x-ray mainly in the axial skeleton (p , .001).
Whole-body LDCT was superior in detecting lesions in patients
with osteopenia and osteoporosis.11
In total, whole-body LDCT advantages over whole-body
x-ray include (1) superior diagnostic sensitivity for depiction of osteolytic lesions, especially in areas where the
whole-body x-ray detection rate is low (i.e., pelvis and spine);
(2) superiority in estimating fracture risk and bone instability;
(3) duration of the examination, which is 5 minutes or less, an
important issue for patients in extreme pain; (4) production of
higher-quality 3D high resolution images for planning biopsies
and therapeutic interventions; and (5) demonstration of
unsuspected manifestations of myeloma or other disease,
especially in the lungs and kidneys (33% in the study by Wolf
KEY POINTS
e408
Whole-body LDCT is superior to conventional

radiography for the detection of osteolytic lesions, and it
is suggested to replace it in the work-up of patients with
myeloma.
MRI is the best imaging method for the depiction of
marrow infiltration by myeloma cells.
Whole-body MRI (or at least MRI of the spine and pelvis
if whole-body MRI is not available) should be performed
for all patients with smoldering multiple myeloma with
no lytic lesions to look for occult disease, which may
justify treatment.
PET/CT allows better definition of complete response
and minimal residual disease.
PET/CT has an independent prognostic value both at
diagnosis and after treatment.
et al).5-11 Major disadvantages of whole-body LDCT include

increased length of time required for radiologists to report
their findings, lack of availability in several centers,2,5 and lack
of specificity for the differential diagnosis between malignant
and osteoporotic fractures, despite improvements during
recent years.12 Furthermore, although exposure to radiation
is much lower compared with standard CT, it continues to be
higher than whole-body x-ray: mean dose of whole-body
LDCT is approximately 3.6 and 2.8 mSv for females and
males, respectively, compared with 1.2 mSv for whole-body
x-ray.13 Nevertheless, the higher diagnostic accuracy of
whole-body LDCT and patient comfort, which is particularly
important for older patients, renders the dose/quality ratio
favorable for whole-body LDCT. For these reasons, the European Myeloma Network has suggested that whole-body
LDCT replace conventional radiography as the standard imaging technique for evaluation of bone disease in multiple
myeloma, where available.14 Figure 2 shows images from a
whole-body LDCT study of one of our patients with myeloma.
The Value of MRI in Myeloma: Advantages and

Limitations
MRI detection of bone involvement in myeloma: comparison
with conventional radiography. MRI has been established as a valuable technique for the diagnosis of bone
involvement in multiple myeloma. MRI detects bone marrow infiltration by myeloma cells; whole-body x-ray and CT
detect osseous destruction. Conventional MRI protocols for
multiple myeloma include T1-weighted, T2-weighted with
fat suppression, in- and opposed-phase imaging, and
contrast-enhanced T1-weighted sequences. Five MRI patterns of marrow involvement have been recognized in
multiple myeloma: (1) a focal pattern that consists of localized areas of myeloma cell infiltration 5 mm or greater in
diameter (present in 30%50% of patients with multiple
myeloma); (2) a diffuse pattern characterized by almost
complete replacement of normal marrow by myeloma cells
(present in 25%40% of patients with multiple myeloma);
(3) a combined diffuse and focal pattern (present in 10% of
patients with multiple myeloma); (4) a normal bone marrow
pattern (present in 15%25% of patients with multiple
myeloma); and (5) a variegated or salt and pepper pattern
with innumerable small bone marrow focal lesions (present
in 1%5% of patients with multiple myeloma; Fig. 3).3,4,15
Several studies have shown that MRI is more sensitive than
whole-body x-ray for the detection of bone involvement in
multiple myeloma.16-19 The largest included 611 patients
who were treated with total therapy protocols in Arkansas.
MRI detected focal lesions in 74% of the patients, while
whole-body x-ray detected osteolytic lesions in 56% of them.
MRI was superior to whole-body x-ray particularly in
detecting bone involvement.16 A meta-analysis, which
compared the detection rate of different imaging techniques, confirmed the superiority of MRI over whole-body
x-ray, mainly in the axial skeleton.20
Because of its high sensitivity in revealing bone marrow
involvement, MRI is now used for the discrimination
FIGURE 1. (A) Characteristic Punched-Out Lesions in the Skull and (B) Large Osteolytic Lesions in the Left
Humerus
between asymptomatic/smoldering and symptomatic multiple myeloma. Several studies have shown that approximately 40% to 50% of patients with normal whole-body x-ray
had abnormal findings on MRI examinations.17-19 Two
studies showed that patients with smoldering multiple
myeloma who had more than one focal lesion on MRI had a
median time to progression (TTP) to symptomatic disease
of 13 to 15 months and a 2-year probability of progression of
approximately 70%.21,22 In both studies, the presence of
more than one focal lesion on MRI was an independent
adverse prognostic factor for progression to symptomatic
disease. Based on these studies, the International Myeloma
Working Group (IMWG) included the use of whole-body MRI
(or MRI of the spine and pelvis if whole-body MRI is not
available) in the work-up for smoldering multiple myeloma.2
In this point, it is important to note the differences between

the detection rates of whole-body MRI, which is not available everywhere, and the detection rate of spine and pelvis
MRI. In a study of 100 patients with multiple myeloma and
monoclonal gammopathy of undetermined significance who
underwent whole-body MRI, 10% of patients presented with
focal lesions in the peripheral skeleton only; this is the
number of patients who will not be diagnosed with lesions if
an MRI of the spine and pelvis is performed, instead of
whole-body MRI.23
MRI detection of bone involvement in myeloma:
comparison with whole-body LDCT and PET/CT. In two
small studies, with 41 and 46 patients with newly diagnosed
multiple myeloma, respectively, whole-body MRI was able
e409
TABLE 1. Major Studies Comparing Whole-Body LDCT and Conventional Radiography

Study
Study Design
No. of Patients
Reference Test
Key Findings
Gleeson et al9
39
WBXR, WB-MRI, bone marrow

biopsy
WBLDCT . WBXR, superior detection rate, comparable

with WB-MRI
Kropil et al8
29
WBXR
WBLDCT . WBXR, ratio of detection of lytic lesion

WBLDCT/WBXR = 2.06 (4.60 for the spine; 7 for the
skull; 2.54 for the pelvis and 2.07 for the thoracic cage)
Princewill et al10
51
WBXR
WBLDCT . WBXR; ratio of detection of lytic lesion

WBLDCT/WBXR = 3.9 (74 for the thoracic cage; 6.67 for
the pelvis and the fat bones; 4.92 for the spine; 2.31 for
the long bone and 1.09 for the skull); 61% of the
patients upstaged with WBLDCT
Wolf et al11
52
WBXR
WBLDCT . WBXR, particularly in the axial skeleton. 63%

patients upstaged with WBLDCT, 23% patients negative at WBXR and positive at WBLDCT
Abbreviations: P, prospective; WBRX; whole-body x-ray (conventional skeletal survey); WB-MRI, whole-body MRI; WBLDCT, whole-body low-dose CT; R, retrospective.
to detect a higher number of patients with bone marrow

involvement compared with whole-body LDCT or PET/CT.
The superiority of MRI over PET/CT was mainly because of
the higher sensitivity of MRI for the detection of diffuse
disease.24,25 However, in the largest comparison to date of

303 patients with myeloma, there was no superiority of MRI
over PET/CT for the detection of focal lesions; though, a
diffuse MRI pattern of bone marrow involvement was not
FIGURE 2. Whole-Body LDCT Images Showing Lytic Lesions in the Skull, Spine, and Pelvis, Bone Marrow Deposits in
the Femoral Cavity, and Vertebral Body Fractures
e410
FIGURE 3. MRI Patterns of Marrow Involvement in Myeloma
(A) Focal pattern: On T1-weighted images, focal lesions are darker than yellow marrow and slightly hypointense or isointense to intervertebral disc and muscle. (B) Diffuse
pattern: On T1-weighted images, the normal bone marrow signal is completely replaced by the abnormal process, and the intervertebral discs appear brighter or isointense
to the diseased marrow. (C) Variegated pattern: On T1-weighted images, the bone marrow is very inhomogeneous with innumerable small lesions.
included in this comparison.26 Finally, in a retrospective

study of 210 MRI and 210 PET/CT studies of patients with
multiple myeloma, MRI achieved better results than PET/CT
for the detection of myeloma at diagnosis and at progression; PET/CT, however, detected findings of response to antimyeloma therapy earlier than MRI.27
Prognostic significance of MRI. MRI patterns have been
compared with myeloma features at presentation. More
specifically, a diffuse MRI pattern correlated with high-risk
cytogenetics, advanced disease stage (ISS-3), and increased
angiogenesis.28-30 Patients with diffuse MRI pattern experienced poorer overall survival (OS) compared with patients with focal or normal patterns, both in the era of
conventional chemotherapy or in the era of novel antimyeloma agents.29-32 More importantly, in a study that
included 228 symptomatic patients who received first-line
therapy with bortezomib- or immunomodulatory drugbased
regimens, the combination of diffuse MRI pattern and highrisk cytogenetics and ISS-3, identified a group of patients
who experienced poor outcome: a median OS of 21 months
and a 35% probability for 3-year OS.29 Similarly, in a recent
study of 161 patients with multiple myeloma who underwent an autologous transplantation (ASCT), the combination of a high-risk MRI score (defined as the presence of
moderate to severe diffuse infiltration, or the presence of
more than 25 focal lesions on whole-body MRI or more than
seven focal lesions on axial MRI) and high-risk cytogenetics
identified a group of patients with median progression-free
survival (PFS) and OS of 23 and 56 months, respectively.33
The cut-off value of seven focal lesions on MRI was also used
by the Arkansas group to identify patients with inferior OS
(3-year OS probability: 73% vs. 86% for those with zero to
seven focal lesions).34 Despite the prognostic significance of
MRI, no treatment has been suggested to date for patients
with high-risk MRI features.
e411
The role of MRI in the follow-up of patients with

myeloma. Although there are several studies evaluating the
role of imaging at diagnosis, there are limited data for the
value of imaging during treatment and follow-up of patients
with myeloma. The IMWG suggests that there is no need to
repeat the skeletal survey in a patient who is responding to
treatment unless he develops bone symptoms and that
other imaging studies (CT, MRI, and PET/CT) to detect soft
tissue masses arising from bone lesions or extramedullary
disease may be indicated according to clinical circumstances.35 We believe that these recommendations must
be changed, as the physician cannot detect asymptomatic
increases in the dimensions of pre-existing lytic lesions or the
presence of new lytic lesions or new plasmacytomas (all are
criteria for defining disease progression36) without appropriate imaging.
Is there any role for MRI in this setting? And if yes, how
often must we repeat MRI after initial evaluation? Resolution of previously detected focal lesions occurred in approximately 60% of patients who were treated with total
therapy protocols in Arkansas. When the disease progressed, after the achievement of complete response (CR),
whole-body MRI revealed new focal lesions or enlargement
of previously detected focal lesions in 26% and 28% of
patients, respectively, while 15% of patients demonstrated
both of these features at their MRI evaluation.16 In general,
MRI findings of response to therapy (CR or partial response
[PR]) include resolution or decrease of the extent of focal
lesions or diffuse patterns in conventional chemotherapy or
novel anti-myeloma agent era.37-39 In a small study with 33
patients who underwent an ASCT, MRI had 79% accuracy
and 86% specificity but only 64% sensitivity for the detection
of remission because of false-positive results of nonviable
lesions.39 Thus, for better evaluation of remission, functional
imaging techniques (e.g., PET/CT or possibly DWI MRI) seem
to be more appropriate and will be discussed in the following
sections.
Regarding patients with smoldering multiple myeloma
with a negative MRI study at diagnosis, the timing of followup with MRI and the significance of any findings on these
studies have not yet been established. In a single such study,
the Heidelberg group performed a second MRI 3 to 6 months
after the first in 63 patients with smoldering multiple myeloma. Approximately one-half of the patients showed
progression on MRI, while 40% developed symptomatic
disease, according to CRAB criteria (i.e., calcium, renal
failure, anemia, and bone disease).40 More data are needed
before recommending time intervals between the first and
subsequent MRIs for patients with smoldering multiple
myeloma.
Advantages and limitations of conventional MRI. The
major advantage of MRI over whole-body LDCT or conventional CT is the discrimination between myelomatous
and normal marrow. This is extremely helpful in the diagnosis of the malignant or benign nature of a vertebral
fracture (Fig. 4). Several times, clinicians face the condition
e412
of a patient with myeloma in remission with a new vertebral

fracture and no other evidence of progression. MRI can offer
important information and is able to differentiate myeloma
from osteoporotic fractures in more than 90% of the cases.41
Furthermore, extramedullary plasmacytomas, present in up
to 20% of patients during the course of their disease, are
easily seen on whole-body MRI.42 MRI is also recommended
for the work-up of patients with a solitary plasmacytoma of
the bone. MRI may reveal unsuspected bone lesions in
approximately one-third of such patients43; in these patients, systemic treatment must be applied instead of radiation therapy, which is the treatment of choice for solitary
bone plasmacytoma.44 MRI accurately localizes spinal cord
and/or nerve root compression for surgical intervention or
radiation therapy.2,4,15 Finally, avascular necrosis of the
femoral head and cardiac amyloidosis may also be evaluated
with MRI.45,46
Limitations of MRI include prolonged acquisition time,
availability issues, problems with claustrophobic patients
and with patients carrying metal devices in their body, and
high cost.15
Novel MRI techniques. DWI MRI, which derives its contrast
mainly from differences in the diffusivity of water molecules in the tissue environment, is a novel and promising
MRI technique. DWI MRI uses the calculation of apparent
diffusion coefficient values to better evaluate myeloma
burden and MRI infiltration patterns.47,48 Apparent diffusion coefficient values are higher in patients with multiple
myeloma at diagnosis, compared with patients in remission
20 weeks after initiation of treatment.49 DWI MRI was
found superior to whole-body x-ray for the detection of
bone involvement in 20 patients with relapsed/refractory
FIGURE 4. MRI Differentiating a Vertebral Fracture
Left: From an underlying tumor. Right: From an osteoporotic fracture without

signs of myeloma cell infiltration.
multiple myeloma in all areas of the skeleton except the

skull, where both examinations had equal sensitivity.50 In
another small study with 24 patients with myeloma (both
treated and untreated), DWI MRI was found more sensitive
than 18F-fluorodeoxyglucose (FDG)-PET in detecting myeloma lesions.51 In a recent study, 17 patients were evaluated using DWI MRI and FDG-PET/CT, and the findings
were compared with bone marrow biopsy data. In all
studied regions, whole-body DWI scores were higher
compared with FDG-PET/CT. DWI MRI was particularly
accurate in diagnosing diffuse disease (diffuse disease was
observed in 37% of regions imaged on whole-body DWI
scans compared with only 7% on FDG-PET/CT); both
techniques were equally sensitive in the detection of focal
lesions.52 Preliminary reports suggest that DWI MRI may be
used for the better definition of response to therapy, but
this has to be confirmed in larger studies and in comparison
with PET/CT results.48,53
Dynamic contrast-enhanced (DCE) MRI evaluates the
distribution of an intravenous contrast agent both inside and
outside the blood vessels over time. DCE MRI provides data
for the microcirculation of a specific area. In multiple myeloma, DCE MRI derives parameters correlated with microvascular density in the bone marrow both before and
after therapy,54-56 as well as with adverse disease features
and OS.57 Finally, the combination of DWI and DCE MRI
produced a score with 76% agreement with the IMWG
criteria of response to therapy in 27 patients with myeloma
under treatment.58 More data are needed to evaluate such
scores in larger studies.
PET/MRI represents a novel imaging modality with attractive potential clinical applications. In multiple myeloma,
there is only one prospective study, which compared

PET/MRI with PET/CT in 30 patients with myeloma with both
techniques performed sequentially. There was high correlation between the two techniques, regarding the number of
active lesions and average standardized uptake value
(SUV).59 Further studies with PET/MRI will reveal if there is
any value of this technique for patients with multiple
myeloma.
PET/CT and the Treatment of Patients With Myeloma

PET/CT detection of bone involvement in myeloma. FDGPET/CT is a functional imaging method that combines
demonstration of hypermetabolic activity in intramedullary
and extramedullary sites (PET) with evidence of osteolysis
(CT). Several studies have shown that PET/CT is more
sensitive compared with whole-body x-ray for the detection of osteolytic lesions in multiple myeloma.25,60-62
This has been confirmed by the largest meta-analysis in
the field.20 The higher detection rate of PET/CT over
whole-body x-ray for the presence of osteolytic lesions is
especially important for patients with smoldering multiple myeloma. In one study with 120 patients with smoldering multiple myeloma based on the previous IMWG
criteria,25 16% of patients with normal whole-body x-ray
had positive PET/CT results. The median TTP for patients
with positive PET/CT results was 1.1 years compared with
4.5 for patients with negative PET/CT results, while the
probability of progression at 2 years for patients with
positive PET/CT results was 58%.63 The largest study in the
field involved 188 patients with suspected smoldering
multiple myeloma examined with PET/CT. PET/CT was
positive in 39% of patients. The probability of progression
TABLE 2. Summary of the Novel IMPeTUs Criteria for the PET/CT Standardization for Multiple Myeloma
Lesion Type
Site
Diffuse
Bone marrow*
Number of Lesions (x)
Focal
Skull
x =1 (no lesions)
Spine
x = 2 (1-3 lesions)
Extraspinal
x = 3 (4-10 lesions)
Grading
Deauville five-point scale
x = 4 (. 10 lesions)
Lytic
x = 1 (no lesions)
x = 2 (1-3 lesions)
x = 3 (4-10 lesions)
x = 4 (. 10 lesions
Fracture
At least one
Paramedullary
At least one
Extramedullary
At least one
Nodal**/extranodal
*A if hypermetabolism in limbs and ribs.

**For nodal disease: C, cervical; SC, supraclavicular; M, mediastinal; Ax, axillary; Rp, retroperitoneal; Mes, mesenteric; In, inguinal.
For extranodal disease: Li, liver; Mus, muscle; Spl, spleen; Sk, skin; Oth, other.
Adapted by Nanni et al76 with permission.
Deauville five-point scale:
1 5 No uptake at all
2 # Mediastinal blood pool uptake (SUVmax)
3 . Mediastinal blood pool uptake, # liver uptake
4 . Liver uptake more than 10%
5 . Liver uptake (twice)
e413
to symptomatic multiple myeloma within 2 years was 75%

for patients with a positive PET/CT under observation
compared with only 30% for patients with a negative
PET/CT. This probability was higher if hypermetabolic
activity was combined with underlying osteolysis (2-year
progression rate: 87%). The median TTP was 21 months
compared with 60 months for patients with a positive
PET/CT and those with a negative PET/CT, respectively.64 The
results of these two studies support the integration of
changes in imaging requirements in the new IMWG diagnostic criteria for multiple myeloma; detection of osteolytic
lesions by PET/CT is a criterion for symptomatic multiple
myeloma.2
Compared with MRI, as mentioned previously, PET/CT
performs equally well in detecting focal lesions, but MRI
is better at detecting diffuse disease.24,25,62
Value of PET/CT for better definition of CR to anti-myeloma

therapy. Data obtained from PET/CT of 40 patients with
multiple myeloma, including average SUV and FDG kinetic
parameters K1, influx and fractal dimension correlated
significantly with percentage of bone marrow infiltration by
plasma cells on trephine biopsies (PC %).65 Furthermore,
PET/CT efficiently detected extramedullary disease in patients both at diagnosis and at relapse.66 Consequently,
PET/CT was tested for better definition of CR in 282 patients
with multiple myeloma. It was performed at diagnosis and
every 12 to 18 months afterward. At diagnosis, 42% of
patients with multiple myeloma had more than three focal
lesions; in 50% of these patients, SUVmax was more than 4.2.
After treatment, PET/CT was negative for 70% of patients,
while 53% of patients achieved CR according to IMWG
criteria. Approximately 30% of patients at CR had a positive
TABLE 3. Advantages and Limitations of Different Imaging Techniques

Advantages
WBXR
Limitations
Cost
Poor sensitivity/low detection rate
Availability
Findings present only after advanced bone damage
Historical use/validation
Patient discomfort with repositioning, multiple films; long image

acquisition time
No evaluation of bone marrow; no differentiation between
malignant and benign fractures
Lack of accurate visualization of specific areas, such as the pelvis
and the spine
Difficulties in the assessment of lytic lesions response to antimyeloma therapy
Observer dependency
WBLDCT
Increased sensitivity and specificity for the detection of lytic

lesions
Unclear prognostic significance of lesion number
3D structural information for CT-guided biopsy, surgery, and

radiotherapy planning
Relatively higher radiation exposure and more expensive than

WBXR
Can depict EMD, bone marrow involvement, and lytic lesions

Rapid acquisition time
Inexpensive compared with MRI and PET
Comfortable for patients
PET/CT
MRI
Functional method
High cost
Assesses disease activity before and after treatment; better

definition of CR and MRD
Limited availability
Depicts extramedullary disease
False-positive diagnoses because of infection/inflammation
Prognostic significance pre- and post-treatment
Lack of standardization
Novel radioisotopes may offer additional disease-relevant

information
Poor spatial resolution
Is able to exclude smoldering/asymptomatic myeloma
High cost
No radiation exposure
Long acquisition time, claustrophobia
Assesses bone marrow both diffuse infiltration and focal

bone marrow lesions
May exclude patients with indwelling metal objects; contrast

contraindicated in severe renal insufficiency
Superior for detection of spinal cord compression and softtissue masses

Number of focal lesions has prognostic significance
Detects extramedullary disease
3D structural information for CT-guided biopsy, surgery, and
radiotherapy planning
Abbreviations: WBXR, whole-body x-ray; WBLDCT, whole-body low-dose CT; EMD, extramedullary disease; CR, complete response; MRD, minimal residual disease.
e414
PET/CT. More importantly, PET/CT negativity was an independent predictor for prolonged PFS and OS in patients
with a CR. In addition, for patients with a CR, median PFS was
50 months for patients with a positive PET/CT and 90 months
for patients with a negative PET/CT.67 PET/CT, therefore,
provides more accurate definition of CR, and it has been
suggested that it should be incorporated to CR criteria.68
Prognostic significance of PET/CT. Several studies have
confirmed the value of PET/CT as an independent factor for
survival of patients with multiple myeloma both at diagnosis and after treatment.69-74 In 192 newly diagnosed
patients who underwent ASCT, the presence of extramedullary disease and SUVmax greater than 4.2 on PET/CT
performed at diagnosis and the persistence of FDG uptake
post-ASCT were independent variables adversely affecting
PFS.69 In the largest study in the field, 429 patients who
were treated with total therapy protocols in Arkansas were
evaluated with both MRI and PET/CT at diagnosis and seven
days post-ASCT. From the imaging variables, in the multivariate
analysis, only the detection of more than two osteolytic lesions
by whole-body x-ray at diagnosis and the detection of more
than three focal lesions by PET/CT 7 days post-ASCT were
independent prognostic factors for inferior OS. Limitation of
this study was the exclusion of the diffuse MRI pattern from the
analysis.34 Despite this limitation, studies reported to date
support the role of PET/CT after therapy, deeming it the best
imaging technique for the follow-up of patients with myeloma.
Indeed, in a recent study, 134 patients who were eligible for
treatment with ASCT were randomly assigned to eight cycles of
bortezomib-lenalidomide-dexamethasone (VRD) followed by
1-year maintenance with lenalidomide, or three cycles of VRD
followed by ASCT plus two cycles of VRD consolidation and
1-year lenalidomide maintenance. PET/CT and whole-body
MRI were performed after induction and before maintenance. Both techniques were positive at diagnosis in more than
90% of patients. After induction therapy and before maintenance, more patients continued to have positive MRI than
PET/CT (93% vs. 55% and 83% vs. 21%, respectively), possibly
because of earlier reduction of activity of PET/CT lesions. Both
after induction and before maintenance, normalization of
PET/CT and not of MRI could predict for PFS, while only
normalization of PET/CT before maintenance could predict for

OS (30-month OS rate: 70% in patients with a positive PET/CT
vs. 94.6% in patients with negative PET/CT; p = .01).75
At this point, it is crucial to mention that one of the major
limitations of PET/CT is the lack of standardization and the
controversies regarding SUV level of positivity. Recently, an
Italian panel of experts introduced novel criteria for the interpretation of PET/CT images; these are summarized in
Table 2.76 Large multicenter studies with prospective evaluation of these new criteria will reveal their clinical effect.
Other PET/CT indications and limitations. PET/CT may be
used for the work-up of patients with solitary bone plasmacytoma at diagnosis.77 However, it is not clear whether
PET/CT or MRI is more suitable in this setting because
restaging PET/CT after radiotherapy has a number of falsepositive findings.78 PET/CT also has a role for patients with
nonsecretory or oligosecretory myeloma for the detection of
active lesions in the body.79 Major limitations of PET/CT include high cost, lack of availability in many centers and
countries, and false-positive results because of inflammation
from other underlying pathology.
CONCLUSION
Whole-body LDCT seems to be most suitable for the detection
of osteolytic bone disease in patients with multiple myeloma
at diagnosis, replacing whole-body x-ray. Whole-body MRI (or
at least MRI of the spine and pelvis if whole-body MRI is not
available) should be performed for all patients with smoldering multiple myeloma with no lytic lesions to look for
occult disease, which may justify treatment. PET/CT seems to
be inferior to MRI regarding the detection of marrow involvement in multiple myeloma, but it is probably the best
technique for optimal definition of CR and follow-up of patients with myeloma. Table 3 lists the advantages and limitations of the most important imaging modalities. More
studies with novel imaging MRI techniques and the broader
use of PET/CT will define the role of these techniques in
myeloma at diagnosis and follow-up, with the particular aim
to better define CR or minimal residual disease and to select
the optimal window for their performance.
References
1. Terpos E, Berenson J, Raje N, et al. Management of bone disease in
multiple myeloma. Expert Rev Hematol. 2014;7:113-125.
3. Dimopoulos M, Terpos E, Comenzo RL, et al. International myeloma
working group consensus statement and guidelines regarding the
current role of imaging techniques in the diagnosis and monitoring of
multiple myeloma. Leukemia. 2009;23:1545-1556.
4. Terpos E, Moulopoulos LA, Dimopoulos MA. Advances in imaging and the
management of myeloma bone disease. J Clin Oncol. 2011;29:1907-1915.
5. Pianko MJ, Terpos E, Roodman GD, et al. Whole-body low-dose

computed tomography and advanced imaging techniques for multiple myeloma bone disease. Clin Cancer Res. 2014;20:5888-5897.
6. Ippolito D, Besostri V, Bonaffini PA, et al. Diagnostic value of wholebody low-dose computed tomography (WBLDCT) in bone lesions
detection in patients with multiple myeloma (MM). Eur J Radiol. 2013;82:
2322-2327.
7. Horger M, Claussen CD, Bross-Bach U, et al. Whole-body low-dose
multidetector row-CT in the diagnosis of multiple myeloma: an
alternative to conventional radiography. Eur J Radiol. 2005;54:
289-297.
e415
8. Kropil P, Fenk R, Fritz LB, et al. Comparison of whole-body 64-slice

multidetector computed tomography and conventional radiography
in staging of multiple myeloma. Eur Radiol. 2008;18:51-58.
9. Gleeson TG, Moriarty J, Shortt CP, et al. Accuracy of whole-body lowdose multidetector CT (WBLDCT) versus skeletal survey in the detection
of myelomatous lesions, and correlation of disease distribution with
whole-body MRI (WBMRI). Skeletal Radiol. 2009;38:225-236.
10. Princewill K, Kyere S, Awan O, et al. Multiple myeloma lesion detection
with whole body CT versus radiographic skeletal survey. Cancer Invest.
2013;31:206-211.
11. Wolf MB, Murray F, Kilk K, et al. Sensitivity of whole-body CT and MRI
versus projection radiography in the detection of osteolyses in patients
with monoclonal plasma cell disease. Eur J Radiol. 2014;83:1222-1230.
12. Cretti F, Perugini G. Patient dose evaluation for the whole-body lowdose multidetector CT (WBLDMDCT) skeleton study in multiple myeloma (MM). Radiol Med (Torino). 2016;121:93-105.
13. Borggrefe J, Giravent S, Campbell G, et al. Association of osteolytic
lesions, bone mineral loss and trabecular sclerosis with prevalent
vertebral fractures in patients with multiple myeloma. Eur J Radiol.
2015;84:2269-2274.
14. Terpos E, Kleber M, Engelhardt M, et al. European Myeloma Network
guidelines for the management of multiple myeloma-related complications. Haematologica. 2015;100:1254-1266.
15. Dimopoulos MA, Hillengass J, Usmani S, et al. Role of magnetic resonance imaging in the management of patients with multiple myeloma:
a consensus statement. J Clin Oncol. 2015;33:657-664.
16. Walker R, Barlogie B, Haessler J, et al. Magnetic resonance imaging in
multiple myeloma: diagnostic and clinical implications. J Clin Oncol.
2007;25:1121-1128.
17. Moulopoulos LA, Varma DG, Dimopoulos MA, et al. Multiple myeloma:
spinal MR imaging in patients with untreated newly diagnosed disease.
Radiology. 1992;185:833-840.
18. Moulopoulos LA, Dimopoulos MA, Smith TL, et al. Prognostic significance of magnetic resonance imaging in patients with asymptomatic
multiple myeloma. J Clin Oncol. 1995;13:251-256.
19. Kastritis E, Terpos E, Moulopoulos L, et al. Extensive bone marrow
infiltration and abnormal free light chain ratio identifies patients with
asymptomatic myeloma at high risk for progression to symptomatic
disease. Leukemia. 2013;27:947-953.
20. Regelink JC, Minnema MC, Terpos E, et al. Comparison of modern and
conventional imaging techniques in establishing multiple myelomarelated bone disease: a systematic review. Br J Haematol. 2013;162:
50-61 .
22. Kastritis E, Moulopoulos LA, Terpos E, et al. The prognostic importance
of the presence of more than one focal lesion in spine MRI of patients
with asymptomatic (smoldering) multiple myeloma. Leukemia. 2014;
28:2402-2403.
23. Bauerle T, Hillengass J, Fechtner K, et al. Multiple myeloma and
monoclonal gammopathy of undetermined significance: importance of
whole-body versus spinal MR imaging. Radiology. 2009;252:477-485.
24. Baur-Melnyk A, Buhmann S, Becker C, et al. Whole-body MRI versus
whole-body MDCT for staging of multiple myeloma. AJR Am J Roentgenol. 2008;190:1097-1104.
25. Zamagni E, Nanni C, Patriarca F, et al. A prospective comparison of
18
F-fluorodeoxyglucose positron emission tomography-computed tomography, magnetic resonance imaging and whole-body planar radiographs in
the assessment of bone disease in newly diagnosed multiple myeloma.
Haematologica. 2007;92:50-55.
e416
26. Waheed S, Mitchell A, Usmani S, et al. Standard and novel imaging

methods for multiple myeloma: correlates with prognostic laboratory
variables including gene expression profiling data. Haematologica.
2013;98:71-78.
27. Spinnato P, Bazzocchi A, Brioli A, et al. Contrast enhanced MRI and
18
F-FDG PET-CT in the assessment of multiple myeloma: a comparison
of results in different phases of the disease. Eur J Radiol. 2012;81:
4013-4018.
28. Moulopoulos LA, Dimopoulos MA, Christoulas D, et al. Diffuse MRI
marrow pattern correlates with increased angiogenesis, advanced
disease features and poor prognosis in newly diagnosed myeloma
treated with novel agents. Leukemia. 2010;24:1206-1212.
29. Moulopoulos LA, Dimopoulos MA, Kastritis E, et al. Diffuse pattern of
bone marrow involvement on magnetic resonance imaging is associated with high risk cytogenetics and poor outcome in newly diagnosed,
symptomatic patients with multiple myeloma: a single center experience on 228 patients. Am J Hematol. 2012;87:861-864.
30. Mai EK, Hielscher T, Kloth JK, et al. Association between magnetic
resonance imaging patterns and baseline disease features in multiple
myeloma: analyzing surrogates of tumour mass and biology. Eur Radiol.
Epub 2016 Jan 15.
31. Moulopoulos LA, Gika D, Anagnostopoulos A, et al. Prognostic significance
of magnetic resonance imaging of bone marrow in previously untreated
patients with multiple myeloma. Ann Oncol. 2005;16:1824-1828.
32. Song MK, Chung JS, Lee JJ, et al. Magnetic resonance imaging pattern of
bone marrow involvement as a new predictive parameter of disease
progression in newly diagnosed patients with multiple myeloma eligible for
autologous stem cell transplantation. Br J Haematol. 2014;165:777-785.
33. Mai EK, Hielscher T, Kloth JK, et al. A magnetic resonance imaging-based
prognostic scoring system to predict outcome in transplant-eligible
patients with multiple myeloma. Haematologica. 2015;100:818-825.
34. Usmani SZ, Mitchell A, Waheed S, et al. Prognostic implications of serial
18-fluoro-deoxyglucose emission tomography in multiple myeloma
treated with total therapy 3. Blood. 2013;121:1819-1823.
35. Dimopoulos M, Kyle R, Fermand JP, et al. Consensus recommendations
for standard investigative workup: report of the International Myeloma
Workshop Consensus Panel 3. Blood. 2011;117:4701-4705.
36. Kyle RA, Rajkumar SV. Criteria for diagnosis, staging, risk stratification
and response assessment of multiple myeloma. Leukemia. 2009;23:3-9.
37. Moulopoulos LA, Dimopoulos MA, Alexanian R, et al. Multiple myeloma:
MR patterns of response to treatment. Radiology. 1994;193:441-446.
38. Hillengass J, Ayyaz S, Kilk K, et al. Changes in magnetic resonance
imaging before and after autologous stem cell transplantation correlate
with response and survival in multiple myeloma. Haematologica. 2012;
97:1757-1760.
39. Bannas P, Hentschel HB, Bley TA, et al. Diagnostic performance of
whole-body MRI for the detection of persistent or relapsing disease in
multiple myeloma after stem cell transplantation. Eur Radiol. 2012;22:
2007-2012.
40. Merz M, Hielscher T, Wagner B, et al. Predictive value of longitudinal
whole-body magnetic resonance imaging in patients with smoldering
multiple myeloma. Leukemia. 2014;28:1902-1908.
41. Baur A, Stabler A, Bruning R, et al. Diffusion-weighted MR imaging of
bone marrow: differentiation of benign versus pathologic compression
fractures. Radiology. 1998;207:349-356.
42. Varettoni M, Corso A, Pica G, et al. Incidence, presenting features and
outcome of extramedullary disease in multiple myeloma: a longitudinal
study on 1003 consecutive patients. Ann Oncol. 2010;21:325-330.
43. Moulopoulos LA, Dimopoulos MA, Weber D, et al. Magnetic resonance
imaging in the staging of solitary plasmacytoma of bone. J Clin Oncol.
1993;11:1311-1315.
44. Dimopoulos MA, Moulopoulos LA, Maniatis A, et al. Solitary plasmacytoma of bone and asymptomatic multiple myeloma. Blood. 2000;96:
2037-2044.
45. Lafforgue P, Dahan E, Chagnaud C, et al. Early-stage avascular necrosis
of the femoral head: MR imaging for prognosis in 31 cases with at least 2
years of follow-up. Radiology. 1993;187:199-204.
46. Syed IS, Glockner JF, Feng D, et al. Role of cardiac magnetic resonance
imaging in the detection of cardiac amyloidosis. JACC Cardiovasc Imaging. 2010;3:155-164.
47. Nonomura Y, Yasumoto M, Yoshimura R, et al. Relationship between
bone marrow cellularity and apparent diffusion coefficient. J Magn
Reson Imaging. 2001;13:757-760.
48. Terpos E, Koutoulidis V, Fontara S, et al. Diffusion-weighted imaging
improves accuracy in the diagnosis of MRI patterns of marrow involvement in newly diagnosed myeloma: results of a prospective study
in 99 patients. Blood. 2015;126 (suppl; abstr 4178).
49. Messiou C, Giles S, Collins DJ, et al. Assessing response of myeloma
bone disease with diffusion-weighted MRI. Br J Radiol. 2012;85:
e1198-e1203.
50. Giles SL, deSouza NM, Collins DJ, et al. Assessing myeloma bone disease
with whole-body diffusion-weighted imaging: comparison with x-ray
skeletal survey by region and relationship with laboratory estimates of
disease burden. Clin Radiol. 2015;70:614-621.
51. Sachpekidis C, Mosebach J, Freitag MT, et al. Application of (18)F-FDG
PET and diffusion weighted imaging (DWI) in multiple myeloma:
comparison of functional imaging modalities. Am J Nucl Med Mol
Imaging. 2015;5:479-492.
52. Pawlyn C, Fowkes L, Otero S, et al. Whole-body diffusion-weighted MRI:
a new gold standard for assessing disease burden in patients with
multiple myeloma? Leukemia. Epub 2015 Dec 9.
53. Horger M, Weisel K, Horger W, et al. Whole-body diffusion-weighted
MRI with apparent diffusion coefficient mapping for early response
monitoring in multiple myeloma: preliminary results. AJR Am J
Roentgenol. 2011;196:W790-W795.
54. Hillengass J, Zechmann C, Bauerle T, et al. Dynamic contrast-enhanced
magnetic resonance imaging identifies a subgroup of patients with
asymptomatic monoclonal plasma cell disease and pathologic microcirculation. Clin Cancer Res. 2009;15:3118-3125.
55. Huang SY, Chen BB, Lu HY, et al. Correlation among DCE-MRI measurements of bone marrow angiogenesis, microvessel density, and
extramedullary disease in patients with multiple myeloma. Am J Hematol. 2012;87:837-839.
56. Merz M, Ritsch J, Kunz C, et al. Dynamic contrast-enhanced magnetic
resonance imaging for assessment of antiangiogenic treatment effects
in multiple myeloma. Clin Cancer Res. 2015;21:106-112.
57. Merz M, Moehler TM, Ritsch J, et al. Prognostic significance of
increased bone marrow microcirculation in newly diagnosed multiple myeloma: results of a prospective DCE-MRI study. Eur Radiol.
Epub 2015 Jul 29.
58. Dutoit JC, Claus E, Offner F, et al. Combined evaluation of conventional
MRI, dynamic contrast-enhanced MRI and diffusion weighted imaging
for response evaluation of patients with multiple myeloma. Eur J Radiol.
2016;85:373-382.
59. Sachpekidis C, Hillengass J, Goldschmidt H, et al. Comparison of (18)
F-FDG PET/CT and PET/MRI in patients with multiple myeloma. Am J
Nucl Med Mol Imaging. 2015;5:469-478.
60. Bredella MA, Steinbach L, Caputo G, et al. Value of FDG PET in the
assessment of patients with multiple myeloma. AJR Am J Roentgenol.
2005;184:1199-1204.
61. Lutje S, de Rooy JW, Croockewit S, et al. Role of radiography, MRI and
FDG-PET/CT in diagnosing, staging and therapeutical evaluation of
patients with multiple myeloma. Ann Hematol. 2009;88:1161-1168.
62. Breyer RJ III, Mulligan ME, Smith SE, et al. Comparison of imaging with
FDG PET/CT with other imaging modalities in myeloma. Skeletal Radiol.
2006;35:632-640.
63. Zamagni E, Nanni C, Gay F, et al. 18F-FDG PET/CT focal, but not
osteolytic, lesions predict the progression of smoldering myeloma to
active disease. Leukemia. 2015;30:417-422.
64. Siontis B, Kumar S, Dispenzieri A, et al. Positron emission tomographycomputed tomography in the diagnostic evaluation of smoldering
multiple myeloma: identification of patients needing therapy. Blood
Cancer J. 2015;5:e364.
65. Sachpekidis C, Mai EK, Goldschmidt H, et al. (18)F-FDG dynamic PET/CT
in patients with multiple myeloma: patterns of tracer uptake and
correlation with bone marrow plasma cell infiltration rate. Clin Nucl
Med. 2015;40:e300-e307.
66. Tirumani SH, Sakellis C, Jacene H, et al. Role of FDG-PET/CT in extramedullary multiple myeloma: correlation of FDG-PET/CT findings with
clinical outcome. Clin Nucl Med. 2016;41:e7-e13.
67. Zamagni E, Nanni C, Mancuso K, et al. PET/CT improves the definition of
complete response and allows to detect otherwise unidentifiable skeletal
progression in multiple myeloma. Clin Cancer Res. 2015;21:4384-4390.
68. Paiva B, van Dongen JJ, Orfao A. New criteria for response assessment:
role of minimal residual disease in multiple myeloma. Blood. 2015;125:
3059-3068.
69. Zamagni E, Patriarca F, Nanni C, et al. Prognostic relevance of 18-F FDG
PET/CT in newly diagnosed multiple myeloma patients treated with upfront autologous transplantation. Blood. 2011;118:5989-5995.
70. Patriarca F, Carobolante F, Zamagni E, et al. The role of positron
emission tomography with 18F-fluorodeoxyglucose integrated with
computed tomography in the evaluation of patients with multiple
myeloma undergoing allogeneic stem cell transplantation. Biol Blood
Marrow Transplant. 2015;21:1068-1073.
71. Fonti R, Pace L, Cerchione C, et al. 18F-FDG PET/CT, 99mTc-MIBI, and MRI
in the prediction of outcome of patients with multiple myeloma:
a comparative study. Clin Nucl Med. 2015;40:303-308.
72. Bartel TB, Haessler J, Brown TL, et al. F18-fluorodeoxyglucose positron emission tomography in the context of other imaging techniques and prognostic factors in multiple myeloma. Blood. 2009;114:
2068-2076.
73. Lapa C, Luckerath K, Malzahn U, et al. 18 FDG-PET/CT for prognostic
stratification of patients with multiple myeloma relapse after stem cell
transplantation. Oncotarget. 2014;5:7381-7391.
74. Cascini GL, Falcone C, Console D, et al. Whole-body MRI and PET/CT in
multiple myeloma patients during staging and after treatment: personal
experience in a longitudinal study. Radiol Med (Torino). 2013;118:
930-948.
75. Moreau P, Attal M, Karlin L, et al. Prospective evaluation of MRI and PETCT at diagnosis and before maintenance therapy in symptomatic patients with multiple myeloma included in the IFM/DFCI 2009 trial.
Blood. 2015;126 (suppl; abstr 395).
76. Nanni C, Zamagni E, Versari A, et al. Image interpretation criteria for FDG
PET/CT in multiple myeloma: a new proposal from an Italian expert
panel. IMPeTUs (Italian Myeloma criteria for PET USe). Eur J Nucl Med
Mol Imaging. 2015;43:414-421.
77. Fouquet G, Guidez S, Herbaux C, et al. Impact of initial FDG-PET/CT and
serum-free light chain on transformation of conventionally defined
solitary plasmacytoma to multiple myeloma. Clin Cancer Res. 2014;20:
3254-3260.
78. Alongi P, Zanoni L, Incerti E, et al. 18F-FDG PET/CT for early postradiotherapy assessment in solitary bone plasmacytomas. Clin Nucl
Med. 2015;40:e399-e404.
79. Lonial S, Kaufman JL. Non-secretory myeloma: a clinicians guide.
Oncology (Williston Park). 2013;27:924-928, 930.
e417
S. VINCENT RAJKUMAR
Updated Diagnostic Criteria and Staging System for Multiple

Myeloma
S. Vincent Rajkumar, MD
OVERVIEW
There has been remarkable progress made in the diagnosis and treatment of multiple myeloma (MM). The median survival of
the disease has doubled as a result of several new active drugs. These advances have necessitated a revision of the disease
definition and staging of MM. Until recently, MM was defined by the presence of end-organ damage, specifically hypercalcemia, renal failure, anemia, and bone lesions (CRAB features) that can be attributed to the clonal process. In 2014, the
International Myeloma Working Group (IMWG) updated the diagnostic criteria for MM to add three specific biomarkers that
can be used to diagnose the disease in patients who did not have CRAB features: clonal bone marrow plasma cells greater
than or equal to 60%, serum free light chain (FLC) ratio greater than or equal to 100 provided involved FLC level is 100 mg/L or
higher, or more than one focal lesion on MRI. In addition, the definition was revised to allow CT and PET-CT to diagnose MM
bone disease. These changes enable early diagnosis and allow the initiation of effective therapy to prevent the development
of end-organ damage for patients who are at the highest risk. A new staging system has been developed that incorporates
high-risk cytogenetic abnormalities in addition to standard laboratory markers of prognosis.
ultiple myeloma evolves from a clinically silent premalignant stage termed monoclonal gammopathy of
undetermined significance (MGUS).1,2 MGUS is a classic
premalignant condition with a low risk of malignant conversion, but the risk of progression persists indefinitely. A
small subset of patients has an intermediate clinical phenotype between MGUS and MM, and they are referred to
as having smoldering multiple myeloma (SMM).3 MGUS
is associated with a risk of progression to MM or related
malignancy at a rate of approximately 1% per year, whereas
SMM has a much higher risk of progression of approximately 10% per year.4,5 MGUS and SMM are typically
asymptomatic and are typically diagnosed incidentally
when a monoclonal (M) protein is detected during laboratory work-up of patients who have a wide spectrum of
clinical conditions.
Over the years the diagnosis of MM required evidence
of end-organ damage attributable to the neoplastic clone
of plasma cells: hypercalcemia, renal failure, anemia, and
osteolytic bone lesions, commonly referred to as CRAB
features.6 This definition was conservative and intended to
prevent patients with MGUS and SMM from receiving unnecessary and toxic chemotherapy. With major advances in
therapy and identification of biomarkers that can distinguish
MM from premalignant phases, it became necessary to
revise the disease definition of MM.7-9
REVISED DIAGNOSTIC CRITERIA FOR MULTIPLE

MYELOMA
In 2014, the IMWG revised the diagnostic criteria for MM.10
The revised diagnostic criteria for MM allow the use of
specific biomarkers to define the disease in addition to the
established CRAB features. They also allow the use of
modern imaging tools to diagnose MM bone disease and
clarify several other diagnostic requirements. Table 1 provides the revised IMWG criteria for diagnosis of MM and
related plasma cell disorders.10
Myeloma-Defining Events
The diagnosis of MM requires 10% or more clonal plasma
cells on bone marrow examination or a biopsy-proven
plasmacytoma plus the presence of one or more
myeloma-defining events.10 Myeloma-defining events include the presence of one or more CRAB features, or one
or more biomarkers of malignancy. The three biomarkers
included in the definition of MM are associated with an
approximately 80% risk of progression to symptomatic endorgan damage in two or more independent studies. They are
clonal bone marrow plasma cells greater than or equal to
60%, serum FLC ratio of 100 or higher, provided involved FLC
level is 100 mg/L or higher, or more than one focal lesion
on MRI.
From the Division of Hematology, Mayo Clinic, Rochester, MN.

Corresponding author: S. Vincent Rajkumar, MD, Division of Hematology, Mayo Clinic, 200 First St. SW, Rochester, MN 55905; email: rajkumar.vincent@mayo.edu.
e418
NEW DIAGNOSTIC AND STAGING CRITERIA IN MYELOMA
Extreme bone marrow clonal plasmacytosis. Clonal bone

marrow plasma cell involvement of greater than or equal to
60% is extremely unusual in the absence of CRAB features.
In a Mayo Clinic study of SMM, only six of 276 patients (2%)
had clonal bone marrow plasma cells greater than or equal
to 60%. In that study, patients with this level of marrow
involvement progressed to symptomatic malignancy rapidly,
with a median progression-free survival of 7.7 months.7 In a
separate cohort of 651 patients with SMM, only 21 (3.2%)
had clonal bone marrow plasma cells greater than or equal
to 60%, and 95% of these patients progressed to having MM
within 2 years.7 A study of 96 patients with SMM from the
Greek Myeloma Group also found a markedly high risk of
progression in this subgroup of patients, with a median time
to progression (TTP) of 15 months.11 Similar results were
reported by the University of Pennsylvania; six of 121 patients (5%) with SMM had greater than or equal to 60% bone
marrow involvement, and all progressed to having MM in
less than 2 years.12
Marked elevation of serum involved/uninvolved FLC
ratio. In SMM, an abnormal FLC ratio is associated with a
higher risk of progression to MM.13 As the ratio increases,
so does the risk of progression. In a Mayo Clinic study of
586 patients with SMM, a markedly abnormal involved/
uninvolved FLC ratio ($ 100) was seen in 90 (15%) patients.8 The risk of progression to symptomatic end-organ
damage within the first 2 years with an FLC ratio of 100 or
higher was 72%; the risk of progression to MM or amyloid
light-chain amyloidosis in 2 years was 79%. A similar finding
was reported by Kastritis et al, who studied 96 patients with
SMM. In their study, 7% of patients with SMM had an
involved/uninvolved FLC ratio of 100 or higher, and almost
all progressed to symptomatic disease within 18 months.11
In a third study, at the University of Pennsylvania, patients
KEY POINTS
The diagnostic criteria for multiple myeloma and related

disorders have been updated by the International
Myeloma Working Group.
Patients with 60% or more clonal plasma cell
involvement of the marrow, serum free light chain ratio
of 100 or higher (provided involved free light chain
level 100 mg/L), and/or greater than one focal lesion
on MRI are defined as MM even in the absence of endorgan damage.
Low-dose whole-body CT, MRI, and FDG-PET and FDGPET with PET-CT are more sensitive in detecting
myeloma and must be considered in diagnosis and
monitoring.
In terms of renal involvement, only light chain cast
nephropathy is considered a myeloma-defining event.
The Revised International Staging System has been
developed for myeloma that incorportates high-risk
cytogenetic abnormalities.
with SMM with an involved/uninvolved FLC ratio of 100 or

higher had a 64% risk of progression within 2 years.12 A
markedly abnormal FLC ratio is thus an accurate predictor of
patients who have a high risk of progression to end-organ
damage within a short period of time. To reduce the possibility of error, in addition to the FLC ratio being 100 or
higher, the IMWG also added a requirement for a minimal
involved FLC level of at least 100 mg/L to be considered as
a myeloma-defining event.10
MRI with more than one focal lesion. MRI imaging study of
the whole-body spine/pelvis can reveal focal or diffuse
changes in SMM. In a study by Hillengass et al, 23 of 149
(15%) patients with SMM had more than one focal lesion
seen on whole-body MRI.14 The median TTP in these patients
was 13 months, and the progression rate at 2 years was 70%.
These results were confirmed by Kastritis et al, who found
more than one focal lesion on spinal MRI in nine of 65
patients (14%) with SMM.15 The median TTP was 15 months
and 69% progressed to MM within 2 years. To increase
predictive value, the IMWG added a requirement that focal
lesions must be at least 5 mm or more in size and recommended follow-up examinations in 3 to 6 months for patients who had a solitary focal lesion, equivocal findings, or
diffuse infiltration.10
Assessment of myeloma bone disease. There have been
many advances to imaging methods used to detect bone and
extramedullary disease in MM. These include low-dose
whole-body CT, MRI, and (18)Ffluorodeoxyglucose PET
(FDG-PET) and FDG-PET with PET-CT.16-22 The old IMWG
criteria for the diagnosis of MM primarily relied on conventional radiographs to detect bone disease.6 A systematic
review compared MRI, FDG-PET, PET CT, and whole-body CT
to conventional whole-body skeletal radiography in MM.23
Newer imaging techniques had greater sensitivity compared
with radiographic bone survey for detection of MM bone
lesions, with as high as 80% more lesions detected by the
newer imaging techniques. CT and MRI performed equally
with respect to sensitivity.
Based on these studies, the revised IMWG criteria for MM
permit the use of CT, low-dose whole-body CT, PET-CT, and
whole-body CT to diagnose lytic bone disease in MM. One or
more sites of osteolytic bone destruction of at least 5 mm or
more in size is required. Increased uptake on PET-CT alone is
not adequate; there must be evidence of actual osteolytic
bone destruction on the CT portion of the examination. A
biopsy of one of the bone lesions should be considered if
there is any doubt about the diagnosis. These changes will
enable early and accurate diagnosis of MM. In addition to
these changes, the IMWG clarified that the presence of
osteoporosis, vertebral compression fractures, or bone
densitometric changes in the absence of lytic lesions is not
sufficient evidence of myeloma bone disease.
Because the accurate diagnosis of MM is critical, at least
one advanced imaging examination (PET-CT, low-dose
whole-body CT, or MRI of the whole body or spine) is
e419
S. VINCENT RAJKUMAR
TABLE 1. International Myeloma Working Group Diagnostic Criteria for Multiple Myeloma and Related Plasma
Cell Disorders
Disorder
Disease Definition
Non-IgM MGUS
All three criteria must be met:

Serum monoclonal protein (non-IgM type) , 3 g/dL
Clonal bone marrow plasma cells , 10%*
Absence of end-organ damage such as CRAB features that can be attributed to the plasma cell proliferative disorder
Smoldering MM
Both criteria must be met:

Serum monoclonal protein (IgG or IgA) $ 3 gm/dL, or urinary monoclonal protein $ 500 mg per 24 h and/or
clonal bone marrow plasma cells 10%60%
Absence of myeloma-defining events or amyloidosis
MM
Both criteria must be met:

Clonal bone marrow plasma cells $ 10% or biopsy-proven bony or extramedullary plasmacytoma
Any one or more of the following myeloma defining events:
Evidence of end organ damage that can be attributed to the underlying plasma cell proliferative disorder,
specifically:
Hypercalcemia: serum calcium . 0.25 mmol/L (. 1 mg/dL) higher than the upper limit of normal or
. 2.75 mmol/L (. 11 mg/dL)
Renal insufficiency: creatinine clearance , 40 mL/min or serum creatinine . 177 mmol/L (. 2 mg/dL)
Anemia: hemoglobin value of . 2 g/dL below the lower limit of normal, or a hemoglobin value
, 10 g/dL
Bone lesions: one or more osteolytic lesions on skeletal radiography, CT, or PET-CT
Clonal bone marrow plasma cell percentage $ 60%
Involved: uninvolved serum FLC ratio $ 100 (involved FLC level must be $ 100 mg/L)
. 1 focal lesion on MRI studies (at least 5 mm in size)
IgM MGUS
All three criteria must be met:

Serum IgM monoclonal protein , 3 gm/dL
Bone marrow lymphoplasmacytic infiltration , 10%
No evidence of anemia, constitutional symptoms, hyperviscosity, lymphadenopathy, or hepatosplenomegaly that can be attributed to the underlying lymphoproliferative disorder.
Light-Chain MGUS
All criteria must be met:

Abnormal FLC ratio (, 0.26 or . 1.65)
Increased level of the appropriate involved light chain (increased kappa FLC in patients with ratio . 1.65 and
increased lambda FLC in patients with ratio , 0.26)
No immunoglobulin heavy-chain expression on immunofixation
Absence of end-organ damage that can be attributed to the plasma cell proliferative disorder
Clonal bone marrow plasma cells , 10%
Urinary monoclonal protein , 500 mg/24 h
Solitary Plasmacytoma
All four criteria must be met:

Biopsy-proven solitary lesion of bone or soft tissue with evidence of clonal plasma cells
Normal bone marrow with no evidence of clonal plasma cells
Normal skeletal survey and MRI (or CT) of spine and pelvis (except for the primary solitary lesion)
Absence of end-organ damage such as CRAB features that can be attributed to a lympho-plasma cell
proliferative disorder
Solitary Plasmacytoma With Minimal

Marrow Involvement**
All four criteria must be met:

Biopsy-proven solitary lesion of bone or soft tissue with evidence of clonal plasma cells
Clonal bone marrow plasma cells , 10%
Normal skeletal survey and MRI (or CT) of spine and pelvis (except for the primary solitary lesion)
Absence of end-organ damage such as CRAB features that can be attributed to a lympho-plasma cell
proliferative disorder
Abbreviations: MGUS, monoclonal gammopathy of undertermined significance; CRAB features, hypercalcemia, renal insufficiency, anemia, and bone lesions; MM, multiple myeloma; FLC, free light chain;
SMM, smoldering multiple myeloma.
*A bone marrow examination can be deferred for patients with low-risk MGUS (IgG type, M protein , 15 gm/L, normal FLC ratio) in whom there are no clinical features concerning for myeloma.
**
Solitary plasmacytoma with 10% or more clonal plasma cells is considered as MM.
Reproduced from Rajkumar et al.10
e420
recommended for patients before concluding that a patient

has SMM or solitary plasmacytoma.10,24 The choice between
various imaging methods can vary depending on the clinical
situation and availability.
Other clarifications to MM diagnostic criteria. Several
clarifications to the diagnostic criteria for MM were made.
Importantly, the IMWG clarified that in terms of renal disease,
only suspected or proven light-chain cast nephropathy is
considered a myeloma-defining event. Other renal disorders
associated with M proteins, such as light-chain deposition
disease, membranoproliferative glomerulonephritis, and
amyloid light-chain amyloidosis, are considered unique diseases and not MM. A renal biopsy to clarify the underlying
cause of renal failure is recommended for patients with
suspected cast nephropathy, especially if the serum involved
FLC levels are less than 500 mg/L.25 An estimated glomerular
filtration rate less than 40 mL/min is preferred to the serum
creatinine concentration for purposes of fulfilling the CRAB
criteria.
Solitary plasmacytoma. Solitary plasmacytoma is an earlystage plasma cell malignancy that is in between MGUS/SMM
and MM along the spectrum of plasma cell disorders. It is
defined by the presence of a single biopsy-proven plasmacytoma (bony or extramedullary) and a normal bone
marrow examination (Table 1). Treatment consists of radiation therapy at 40 Gy to 50 Gy to the involved site.
Patients with an apparent solitary plasmacytoma who
have limited (, 10%) clonal marrow involvement are considered to have solitary plasmacytoma with minimal marrow
involvement. These patients are also treated similar to
patients with solitary plasmacytoma. The risk of recurrence
or progression to myeloma within 3 years is approximately
10% for patients with solitary plasmacytoma versus 20% to
60% for patients with solitary plasmacytoma and minimal
marrow involvement.
REVISED INTERNATIONAL STAGING SYSTEM FOR

MYELOMA
There is major variation in survival of MM depending on
host factors, tumor burden (stage), biology (cytogenetic
abnormalities), and response to therapy.26 Tumor burden
in MM has traditionally been assessed using the DurieSalmon Staging (DSS) 27 and the International Staging
System (ISS).28,29 Both these staging systems have some
limitations. The DSS primarily classified patients based on
tumor burden and had problems with reproducibility because some judgment is needed in interpretation of MM
bone disease. The ISS is more reproducible but includes a
host factor determinant, namely serum albumin. Thus,
outcome using ISS can be disproportionately affected by
TABLE 2. Cytogenetic Abnormalities on Clinical Course and Prognosis in Multiple Myeloma

Clinical Setting in Which Abnormality Is Detected
Cytogenetic Abnormality
Smoldering Multiple Myeloma
Multiple Myeloma
Trisomies
Intermediate-risk of progression, median TTP of 3 years
Good prognosis, standard-risk MM, median

OS 7-10 years
Most have myeloma bone disease at diagnosis
Excellent response to lenalidomide-based therapy
t(11;14) (q13;q32)
Standard-risk of progression, median TTP of 5 years

OS 7-10 years
t(6;14) (p21;q32)

OS 7-10 years
t(4;14) (p16;q32)
High-risk of progression, median TTP of 2 years
Intermediate-risk MM, median OS 5 years

Needs bortezomib-based initial therapy, early
ASCT (if eligible), followed by bortezomibbased consolidation/maintenance
t(14;16) (q32;q23)
High-risk MM, median OS 3 years

Associated with high levels of FLC and 25% present
with acute renal failure as initial MDE
t(14;20) (q32;q11)
Gain(1q21)
Intermediate-risk MM, median OS 5 years
Del(17p)
Trisomies Plus Any One of the IgH

Translocations
May ameliorate adverse prognosis conferred by

high risk IgH translocations, and del 17p
Isolated Monosomy 13, or Isolated

Monosomy 14
Effect on prognosis is not clear
Normal
Low-risk of progression, median TTP of 7-10 years
Good prognosis, probably reflecting low tumor

burden, median OS . 7-10 years
Abbreviations: TTP, time to progression; MM, multiple myeloma; OS, overall survival; ASCT, autologous stem cell transplantation; MDE, myeloma-defining event.
Reproduced from Rajan and Rajkumar.32
e421
S. VINCENT RAJKUMAR
TABLE 3. Revised International Staging System for Myeloma

Stage
Frequency (% of Patients)
5-Year Survival Rate (%)
Stage I
28
82
62
62
10
40
ISS stage I (serum albumin > 3.5, serum beta-2-microglobulin < 3.5) and
No high-risk cytogenetics
Normal LDH
Stage II
Neither stage I or III
Stage III
ISS stage III (serum beta-2-microglobulin > 5.5) and
High-risk cytogenetics [t(4;14), t(14;16), or del(17p)] or elevated LDH
Abbreviations: LDH, lactate dehydrogenase; ISS, International Staging System.
Derived from Palumbo et al.33
factors that are not disease-specific. Neither staging system

considers disease biologya key determinant of overall
survival in the disease.
Disease biology in MM is best reflected based on the
molecular subtype of the disease and the presence or absence of specific cytogenetic abnormalities.30,31 For example, abnormalities such as t(4;14), t(14;16), t(14;20), gain
(1q), del(1p), and del(17p) influence disease course, response to therapy, and prognosis in MM (Table 2).32 The
Revised International Staging System (RISS) combines elements of tumor burden (ISS) and disease biology (presence
of high-risk cytogenetic abnormalities or elevated lactate
dehydrogenase level) to create a unified prognostic index
that helps in clinical care as well as in comparison of clinical
trial data (Table 3).33 The RISS was developed based on a
study of 4,445 patients with newly diagnosed MM from
11 international trials. This study showed that the 5-year
survival rate of patients with stage I, II, and III RISS was 82%,
62%, and 40%, respectively. The RISS will be of importance in
the clinic in terms of counseling patients regarding prognosis, as well as in clinical trials to compare outcomes across
clinical trials.
FUTURE DIRECTIONS
The updated diagnostic criteria for MM represent a paradigm shift in our approach to the disease. These changes will
allow us to intervene before end-organ damage in selected
patients who are at imminent risk of symptomatic progression. The revised staging system for MM allows us to
better predict outcome and tailor treatments accordingly.
Advances in imaging have not only enabled early and accurate diagnosis, but also provide ways of monitoring response to therapy. We must identify additional reliable
biomarkers of malignancy. We must also develop a portfolio
of clinical trials designed to determine whether early intervention can improve survival or provide a path to
cure MM.
References
1. Landgren O, Kyle RA, Pfeiffer RM, et al. Monoclonal gammopathy of
undetermined significance (MGUS) consistently precedes multiple
myeloma: a prospective study. Blood. 2009;113:5412-5417.
2. Weiss BM, Abadie J, Verma P, et al. A monoclonal gammopathy precedes multiple myeloma in most patients. Blood. 2009;113:5418-5422.
3. Rajkumar SV, Merlini G, San Miguel JF. Haematological cancer: Redefining myeloma. Nat Rev Clin Oncol. 2012;9:494-496.
4. Kyle RA, Remstein ED, Therneau TM, et al. Clinical course and prognosis
of smoldering (asymptomatic) multiple myeloma. N Engl J Med. 2007;
356:2582-2590.
5. Kyle RA, Therneau TM, Rajkumar SV, et al. A long-term study of
prognosis in monoclonal gammopathy of undetermined significance.
N Engl J Med. 2002;346:564-569.
6. International Myeloma Working Group. Criteria for the classification of
monoclonal gammopathies, multiple myeloma and related disorders:
a report of the International Myeloma Working Group. Br J Haematol.
2003;121:749-757.
7. Rajkumar SV, Larson D, Kyle RA. Diagnosis of smoldering multiple
e422
8. Larsen JT, Kumar SK, Dispenzieri A, et al. Serum free light chain ratio as a
biomarker for high-risk smoldering multiple myeloma. Leukemia. 2013;
27:941-946.
9. Mateos M-V, Hernandez M-T, Giraldo P, et al. Lenalidomide plus
dexamethasone for high-risk smoldering multiple myeloma. N Engl J
Med. 2013;369:438-447.
11. Kastritis E, Terpos E, Moulopoulos L, et al. Extensive bone marrow
infiltration and abnormal free light chain ratio identifies patients with
asymptomatic myeloma at high risk for progression to symptomatic
disease. Leukemia. 2013;27:947-953.
12. Waxman AJ, Mick R, Garfall AL, et al. Modeling the risk of progression in
smoldering multiple myeloma. J Clin Oncol. 2014;32:A8607.
13. Dispenzieri A, Kyle RA, Katzmann JA, et al. Immunoglobulin free
light chain ratio is an independent risk factor for progression of
smoldering (asymptomatic) multiple myeloma. Blood. 2008;111:
785-789.

15. Kastritis E, Moulopoulos LA, Terpos E, et al. The prognostic importance
of the presence of more than one focal lesion in spine MRI of patients
with asymptomatic (smoldering) multiple myeloma. Leukemia. 2014;
28:2402-2403.
16. Dimopoulos M, Terpos E, Comenzo RL, et al; IMWG. International
myeloma working group consensus statement and guidelines regarding
the current role of imaging techniques in the diagnosis and monitoring
of multiple Myeloma. Leukemia. 2009;23:1545-1556.
17. Hillengass J, Landgren O. Challenges and opportunities of novel imaging
techniques in monoclonal plasma cell disorders: imaging early myeloma. Leuk Lymphoma. 2013;54:1355-1363.
18. Zamagni E, Cavo M. The role of imaging techniques in the management
of multiple myeloma. Br J Haematol. 2012;159:499-513.
19. Zamagni E, Nanni C, Patriarca F, et al. A prospective comparison of 18Ffluorodeoxyglucose positron emission tomography-computed tomography, magnetic resonance imaging and whole-body planar radiographs
in the assessment of bone disease in newly diagnosed multiple myeloma. Haematologica. 2007;92:50-55.
20. Walker R, Barlogie B, Haessler J, et al. Magnetic resonance imaging in
multiple myeloma: diagnostic and clinical implications. J Clin Oncol.
2007;25:1121-1128.
21. Bartel TB, Haessler J, Brown TL, et al. F18-fluorodeoxyglucose positron
emission tomography in the context of other imaging techniques and
prognostic factors in multiple myeloma. Blood. 2009;114:2068-2076.
22. Waheed S, Mitchell A, Usmani S, et al. Standard and novel imaging
methods for multiple myeloma: correlates with prognostic laboratory
variables including gene expression profiling data. Haematologica.
2013;98:71-78.
23. Regelink JC, Minnema MC, Terpos E, et al. Comparison of modern and
conventional imaging techniques in establishing multiple myeloma-
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
related bone disease: a systematic review. Br J Haematol. 2013;162:

50-61.
Dimopoulos MA, Hillengass J, Usmani S, et al. Role of magnetic resonance imaging in the management of patients with multiple myeloma:
a consensus statement. J Clin Oncol. 2015;33:657-664.
Hutchison CA, Batuman V, Behrens J, et al; International Kidney and
Monoclonal Gammopathy Research Group. The pathogenesis and diagnosis of acute kidney injury in multiple myeloma. Nat Rev Nephrol.
2012;8:43-51.
Russell SJ, Rajkumar SV. Multiple myeloma and the road to personalised
medicine. Lancet Oncol. 2011;12:617-619.
Durie BG, Salmon SE. A clinical staging system for multiple myeloma.
Correlation of measured myeloma cell mass with presenting clinical
features, response to treatment, and survival. Cancer. 1975;36:
842-854.
Greipp PR, San Miguel J, Durie BG, et al. International staging system for
multiple myeloma. J Clin Oncol. 2005;23:3412-3420.
Hari PN, Zhang MJ, Roy V, et al. Is the International Staging System
superior to the Durie-Salmon staging system? A comparison in multiple
myeloma patients undergoing autologous transplant. Leukemia. 2009;
23:1528-1534.
Kumar SK, Mikhael JR, Buadi FK, et al. Management of newly diagnosed
symptomatic multiple myeloma: updated Mayo Stratification of Myeloma and Risk-Adapted Therapy (mSMART) consensus guidelines.
Mayo Clin Proc. 2009;84:1095-1110.
Kumar S, Fonseca R, Ketterling RP, et al. Trisomies in multiple myeloma:
impact on survival in patients with high-risk cytogenetics. Blood. 2012;
119:2100-2105.
Rajan AM, Rajkumar SV. Interpretation of cytogenetic results in multiple
myeloma for clinical practice. Blood Cancer J. 2015;5:e365.
Palumbo A, Avet-Loiseau H, Oliva S, et al. Revised International Staging
System for multiple myeloma: a report from International Myeloma
e423

DYSCRASIA
Multiple Myeloma: Are We Ready

for Personalized Therapy?
CHAIR
Nikhil C. Munshi, MD
Boston, MA
SPEAKERS
Herve Avet-Loiseau, MD, PhD
University Hospital
Nantes, France
Sagar Lonial, MD, FACP
Winship Cancer Institute of Emory University
Atlanta, GA
MINIMAL RESIDUAL DISEASE BY NEXT-GENERATION SEQUENCING
Minimal Residual Disease by Next-Generation Sequencing:

Pros and Cons
Herve Avet-Loiseau, MD, PhD
OVERVIEW
The wealth of data recently generated highlights that minimal residual disease (MRD)negative status can be achieved in a
large proportion of patients. These studies, in addition to a meta-analysis, clearly suggest significant improvement in both
event-free survival (EFS) and overall survival (OS) among those patients achieving MRDnegative status, especially with
sensitivity of one cell in 1 million bone marrow cells. There is an evolving consensus that achieving MRDnegative status
should become the ultimate goal of therapeutic intervention. Further future efforts should now be directed at determining
how MRD status can be used to guide and personalize further therapy including type of consolidation and maintenance
therapy.
he treatment landscape for multiple myeloma (MM) has

been radically transformed during the past decade by the
introduction of several new drugs with different mechanisms of action, which in turn has led to the improved
survival of patients with MM.1 Diagnostic and response
criteria in MM have used a combination of tumor burden
and functional consequences of tumor expansion (such as
CRAB features), with recent revisions incorporating thresholds for the serum-free light chain (sFLC) ratio and bone
marrow plasma cell percentage as well as using bone lesions
on sensitive imaging as predictors of incipient progression to
MM and, therefore, as acceptable triggers for initiation of
therapy.2 Response evaluation in MM has traditionally been
based on assessing serum and/or urine monoclonal protein
concentrations via protein electrophoresis as a surrogate for
tumor burden with immunofixation, allowing for detection
of trace amounts of paraprotein. Thus, the most recent iteration of the response criteria was developed a decade ago
by the International Myeloma Working Group (IMWG; IMWG
Consensus Criteria for Response Assessment) and was based
on serum and urine protein electrophoresis, serum and urine
immunofixation, sFLC, and bone marrow plasma cell quantitation plus clonality assessment.3
The consensus criteria were uniformly incorporated into
clinical trials, allowing better comparison of different drugs,
drug combinations, and treatment strategies, and the revisions over the years have allowed it to stay ahead of the
advances in treatment. With the recent introduction of more
effective multidrug combinations, especially when used in
the context of allogeneic stem cell transplantation (ASCT),
post-transplant consolidation, and prolonged maintenance

therapy, nearly all patients achieve a treatment response,
with over 50% of these patients reaching complete remission (CR) in some of the more recently reported studies
and up to 80% in recent phase II trials.4-9 Frustratingly, the
vast majority of patients experience relapse despite achieving such deep responses, reflecting persistent disease that
cannot be detected using the currently recommended disease
evaluation techniques. Consequently, new methods are urgently required to detect and quantitate MRD beyond the
level of detection of the current clinical response criteria, and
there is a need to revise the current definition of disease
response for it to evolve in parallel to the changing treatment
paradigm.
DEPTH OF RESPONSE AND LONG-TERM

OUTCOME
The relationship between depth of response and long-term
outcomes has been one of the most debated topics in MM.
The relationship between CR and progression-free survival
(PFS) and/or time to progression (TTP) has been more
consistent than the relationship between CR and OS. This
putative incongruity is a frequent phenomenon in cancer
therapy and probably due to multiple factors, including
interactions between (1) disease biology, (2) different treatment strategies after CR has been reached, and (3) the true
depth of response beyond the conventional (and low-sensitive)
approaches defining CR after different therapies. In this
context, several studies using newer and more sensitive
From the Laboratory for Genomics in Myeloma, Institut Universitaire du Cancer and University Hospital, Centre de Recherche en Cancerologie de Toulouse, Toulouse, France.
Corresponding author: Herve Avet-Loiseau, MD, PhD, Unite de Genomique du Myelome, IUC-T Oncopole, 1 Avenue Irene Joliot-Curie, 31059, Toulouse, France;
email: avet-loiseau.h@chu-toulouse.fr.
e425
HERVE AVET-LOISEAU
techniques have been able to demonstrate the persistence

of MRD not detected by the current CR evaluation methodologies in a significant fraction of patients. The level of
persistent MRD, undetected by conventional methods, is
likely to be one of the most important features contributing
to the link between the depth of response and long-term
outcomes. Independent of the method used to define MRD
(cell-based, molecular-based, or imaging-based), prior studies
consistently show that among patients who have achieved
CR, MRDpositive cases consistently have an inferior PFS
than patients with MRDnegative disease.10-16 Given the
substantial proportion of patients achieving CR with the current therapies, it is time to go beyond the CR criteria and
define MRD assessment in MM.16
DETECTION OF MINIMAL RESIDUAL DISEASE IN

BONE MARROW
Bone marrow examination has been the cornerstone
of disease assessment in the absence of a measurable
monoclonal protein in serum or urine, whether this represents nonsecretory disease or complete response to therapy. In recent years, increasingly sensitive assays have
been adopted for the evaluation of bone marrow aspirates, including multiparameter flow cytometry (MFC),
allele-specific oligonucleotide polymerase chain reaction
(ASO-PCR), and next-generation sequencing (NGS) of immunoglobulin gene sequences, in an effort to increase the
sensitivity of detection of MM cells. Such methods allow
examination of several hundred thousand to millions of
bone marrow cells (or their corresponding amount of DNA)
per assay and can provide a quantitative assessment of any
residual tumor cells in the bone marrow in a relatively short
timeframe.
Multiparameter Flow Cytometric Methods for

Minimal Residual Disease Detection
First-generation MFC methods. Although MFC-based assessment of bone marrow has been conducted in MM for a
number of years, it is only recently that the technology has
gained wide acceptance in the routine work-up of patients
with MM. The most commonly used surface markers used
KEY POINTS
e426
Traditional complete response is not adequate for

predicting long-term outcome or acting as an endpoint
for clinical studies.
Newer technologies allow detection of myeloma cells in
bone marrow with great sensitivity.
All studies confirm that patients with MRDnegative
status have improved EFS and OS.
MRD status should become a standard in clinical studies
and be considered for intervention decision making in
the future.
for discriminating and categorizing normal and MM plasma

cells include CD138, CD38, CD45, CD56, CD19, and cytoplasmic kappa and lambda immunoglobulin light chains.
Additional markers, many of which are aberrantly expressed
on MM plasma cells, have also been found to be of value and
include CD27, CD28, CD81, and CD117. However, given the
heterogeneity of expression of these markers and differences in both the number of events studied and in the
analytical strategies used, there has been significant confusion in the field and inconsistent clinical interpretation of
results from different studies.20 Recently, attempts have
been made to develop consensus guidelines to standardize
the flow-based assessment of disease in MM and other
related plasma cell disorders.17,18
Several studies have demonstrated the utility of MFC in
the detection of MRD in the bone marrow. The Spanish
Myeloma Group (PETHEMA/GEM) used four-color flow
cytometry to study MRD among 295 patients with newly
diagnosed MM receiving uniform treatment including ASCT,
and they demonstrated that MRD was one of the most
important predictors of outcome.10 Minimal residual disease negativity at day 100 after ASCT correlated with improved PFS and OS; furthermore, the impact of MRD
negativity was equally relevant among patients who had
achieved a conventional CR. Similarly, Rawstron et al evaluated the role of six-color MFC in the assessment of MRD at
various stages of therapy among patients with newly diagnosed MM who enrolled on the MRC IX clinical trial.13
Among patients undergoing ASCT, absence of MRD at day
100 was associated with substantially improved PFS, irrespective of cytogenetics or achievement of CR. Of note, in all
these studies, three- to six-color MFC approaches with a
sensitivity of one in 10,000 myeloma cells were used.
Thus, the recent advances in the flow technology allowing
interrogation of several million cells have significantly
improved the sensitivity of the assay, particularly when
combined with simultaneous usage of eight or more colors/
markers for an increased specificity. Recent consensus indicates that such approaches are optimally suited for MRD
testing in the settings of multiple myeloma.20
Next-generation flow. Recent attempts at standardization
and automated read outs makes MFC an attractive test for
sensitive, routine detection of MRD in the bone marrow
compartment.21,22 However, to have uniform MFC-based
MRD response criteria, it is mandatory to have consensus in
the way MRD is evaluated; accordingly, concerted effort has
been put into effect to standardize the flow-based approaches and remove the subjectivity introduced by individual interpretations by defining the reagent characteristics,
defining the acquisition and plasma cell identification parameters, and introducing novel common data analysis tools.
The current EuroFlow next-generation flow (NGF) method
for MRD detection in MM relies on two 8-color combinations
that combine surface antigens for the identification of
phenotypically aberrant clonal plasma cells, as well as
intracytoplasmic kappa and lambda light chain expression to
confirm their clonality. This two-tube NGF approach has now

been extensively validated (over 1,000 MRD samples). It is
very robust and improves reliability, consistency, and sensitivity because of the acquisition of a greater number of cells.
One of the most attractive features of the eight-color
method is its balance between effectiveness (sensitivity plus
specificity) and wide availability because eight-color instruments are commonly used in many hospitals. To improve
efficiency in the laboratory and to reduce costs, it is anticipated that alternate single-tube 10- and 14-color methods will
be suggested by some centers. The single-tube approach will
undergo detailed cross-validation with the reference NGF
method for standardization. To fulfill criteria for NGF, documentation of cross-validation with reference NGF, ongoing
quality control assessment, routine assessment of more than
5 million cells to estimate MRD, and a sensitivity of one in
100,000 or higher is necessary.
Molecular Methods for Minimal Residual Disease

Detection
Allele-specific oligonucleotide polymerase chain reaction. The
other methodology that has been studied extensively in the
past and for which a wealth of data exist is ASO-qPCR, including a few reports in which it has been compared head-tohead with MFC assays. Despite MRD evaluation by ASO-qPCR
being a sensitive and specific approach, it is applicable in a
lower proportion of patients with MM and is more timeconsuming when compared with MFC. This technique is now
supplanted by NGS methods.
Next-generation sequencing. Recently, there has been significant interest in NGS for detection of MM MRD in the
bone marrow, and data supporting its use continue to
emerge. Most of the published data have been generated
using the LymphoSIGHT platform (Sequenta/Adaptive Inc.),
which utilizes consensus primers for the amplification and
sequencing of immunoglobulin gene segments. Specifically,
genomic DNA is amplified using locus-specific primers
designed for IGH-VDJH or IGH-DJH or IGK. Once amplified, the
immunoglobulin gene DNA is sequenced and the frequencies
of the different clonotypes in the sample are determined.
Patients with detectable MM clones (. 5%) at baseline can
then be studied at subsequent time points to determine the
presence and quantity of that particular clone using sequencing approaches. Ladetto et al compared IGH gene
based MRD detection by ASO-qPCR and NGS to assess
whether NGS could overcome some of the limitations of ASOqPCR and further increase its sensitivity and specificity.23
Clonotypes identified by NGS and ASO-qPCR were identical or
greater than 97% homologous in 96% of cases, and both tools
appeared to have a sensitivity of approximately one in
100,000, but NGS has the added advantage of not requiring
patient-specific primers. Recent studies show that NGS can
achieve a sensitivity of one in 1,000,000.
Next-generation sequencing has been compared with
first-generation four-color MFC in one Spanish study.24 Bone
marrow samples from baseline and from the time of very

good partial response or CR from patients enrolled in the
GEM trials were studied by NGS to identify a tumor clonotype at baseline and then re-evaluated for the presence of
the same clonotype in the subsequent sample. A dominant
MM clone could be identified at baseline in 91% of the
patients. Of the 121 patients with an identifiable clonotype
at baseline, 110 had follow-up samples, of which 73%
remained positive by sequencing at MRD levels of one in
100,000 MM cells. Among patients who achieved a very
good partial response, an MRD-negative status (less than
one tumor cell in 100,000 cells) bestowed a better PFS and
OS. Among the group of patients with a CR, a higher proportion of cases had MRD negativity that also associated
with improved PFS. Korde et al also used NGS in 43 patients
with MM treated with carfilzomib, lenalidomide, dexamethasone, and they observed a 12-month PFS for MRDnegative versus MRD-positive patients of 100% and 79%,
respectively (p , .001).24 Results from the IFM-DFCI trial
that randomly selected 700 patients to receive either eight
cycles of VRD (arm A) or three cycles of VRD, ASCT, and
then two consolidation VRD cycles (arm B).25 All patients
then received lenalidomide maintenance for 12 months.
A total of 246 patients were evaluated by NGS before
maintenance and 178 after maintenance. Among the
patients in CR, the 4-year PFS was 83% and 30% for patients
who were MRD negative and MRD positive, respectively,
postmaintenance.
Comparison of Techniques for Minimal Residual

Disease Detection in Bone Marrow
As described above, various techniques have been studied
for detection of MRD. Each of these techniques (based on
the plasma cell phenotype and/or plasma cell genotype) has
advantages and disadvantages that must be taken into
consideration (Table 1). The ideal MRD test should fulfill
several relevant characteristics: (1) high applicability (useful among most patients), (2) highly sensitive and specific,
(3) excellent feasibility (results can be obtained for most
patients), (4) easily accessible, (5) requirement for a limited
sample that can be transported with ease, (6) reproducibility, and (7) proven clinical value.
While none of these tests fully satisfies all these ideal
characteristics at the current time, NGS and NGF fulfill most
of them and can be translated into an advanced platform
that can be uniformly applied across institutions and countries. In terms of sensitivity, the different methodologies
have been reported to have variable levels of sensitivity.
Both NGF and NGS have been reported to have the ability to
detect one in 100,0001,000,000 MM cells in more recent
studies. We strongly encourage the inclusion of both methodologies in prospective trials, to the extent possible, to
collect data that would allow us to better understand the
advantages and disadvantages of the individual approaches
as well as the sensitivity of detection required in various
clinical settings.
e427
HERVE AVET-LOISEAU
Table 1. Comparison of Different Bone Marrow Minimal Residual Disease Techniques

Characteristic
ASO-qPCR
MFC
VDJ Sequencing
Applicability
60%70%
Nearly 100%
$ 90%
Need for Baseline

Sample
Yes, requires production of

patient-specific probes
Not required, abnormal plasma

cells can be identified in any sample
by their distinct immunophenotypic
pattern vs. normal plasma cells
Baseline samples required for

identification of the dominant
clonotype; alternatively, a stored
sample from time-point with
detectable disease can be used
to define baseline status
Sample Requirements
, 1 million cells sufficient
Recent, sensitive methodology

requires over 5 million cells
to be analyzed
, 1 million cells sufficient
Sample Processing
Can be delayed
Needs assessment within 2448

hours; requires a fresh sample
Can be delayed; can use both fresh

and stored samples
Sample Quality
Control
Not possible. Additional

studies required
Immediate with global bone

marrow cell analysis
Not possible. Additional studies

required.
Sensitivity
$ 1 in 100,000
$ 1 in 100,000
$ 1 in 100,000
In frequent clones may not

be evaluable
Turnaround and
Complexity
Labor intensive, requires

development of patient-specific
primers/probes, may take several
days
Can be done in few hours, automated

software available
Typically days for turnaround,

requires intense bioinformatics
support. Use of local laboratories
may speed up turnaround.
Clonal Consideration
Detects only single clone or

some of the related clones
Considers all clones with similar

phenotype but evolving clone with
change in phenotype may not be
evaluable
Can take into account all minor

clones with infrequent occurrence
Abbreviations: ASO-qPCR, allele-specific oligonucleotide polymerase chain reaction; MFC, multiparameter flow cytometry.
DETECTION OF BONE AND EXTRAMEDULLARY

DISEASE
The current approaches for the detection and measurement
of tumor burden after therapy relies on bone marrow assessment. However, bone marrow involvement in MM can
be heterogeneous, thus increasing the likelihood of a falsenegative assessment. More importantly, it does not allow
detection of the disease outside the bone marrow. The recognition of extramedullary disease is increasingly encountered in the clinic as a result of more sensitive imaging studies
and extended survival of patients with MM.
18
F-fluorodeoxyglucose PET (FDG-PET) is a powerful tool
used to assess tumor metabolic activity and the effect of
therapy on the tumor cell metabolism. Multiple studies have
demonstrated a prognostic capacity related to the detection
of PET-positive lesions in patients with MM at diagnosis and
at time of relapse.26-31 In addition to the metabolic assessment, the low-dose CT that is typically done for localization along with FDG-PET allows for a sensitive screen for
the detection of MM-associated bone disease. In an initial
study from the University of Arkansas, complete FDG suppression in the focal lesions before first transplantation was
associated with substantially improved survival outcomes.
In a subsequent study in the context of Total Therapy (TT) 3,
it was shown that persistent FDG avidity 7 days after initiation of therapy was associated with poorer survival outcomes and was independent of other prognostic factors. In
an Italian study, PET-CT was performed at diagnosis, after
thalidomide-dexamethasone induction therapy, and after
e428
double ASCT among 192 patients with newly diagnosed MM.

Persistence of a standard uptake value (SUV) of greater than
4.2 after induction therapy predicted for early relapse, and
the 4-year PFS and OS were superior for those patients
with a negative PET-CT at day 100 post-ASCT. Recently, the
Italian group presented updated results from their study of
282 patients with newly diagnosed MM who had PET imaging
at baseline. After treatment, PET negativity was achieved in
70% of patients, whereas conventionally defined CR was
achieved in 53%. Among the patients in CR, 29% still had
positive PET, translating to an inferior PFS (44 vs. 84 months)
and OS (5-year estimate of 70% vs. 90%). In this study,
persistence of SUVmax greater than 4.2 was the single factor
independently associated with skeletal progression in the
absence of conventional measures of disease progression.
Similar data have been recently reported by the French group.
In the IFM-DFCI trial, which randomized high-dose melphalan,
PET-CT positivity premaintenance was predictive of a shorter
PFS and OS.
INCORPORATION IN CLINICAL PRACTICE

Having established that achieving MRD-negative status
provides significantly superior survival outcome and that we
can reliably measure MRD-negative status with a sensitivity
of one cell in 1 million, the next important question in
myeloma is how MRD status will inform our therapeutic
decision algorithm, to both provide the most effective
therapy and develop strategies to individualize therapy. The
questions to be answered are whether outcome is similar
between patients achieving early MRD negativity versus

those who achieve late MRD-negative status following
multiple interventions. This may help determine the need
for consolidation therapy for a given patient. The next
question is whether patients with MRD-negative status can
get similar benefit with less intense (one drug vs. two drugs)
or shorter duration of maintenance. The last questions will
be whether the same principles apply to older individuals
and whether MRD negativity should also be an endpoint for
patients with relapsed disease.
CONCLUSION
In the era of novel agents and combination therapies
achieving very high conventional CR rates, MRD status is
begging to play an important role in predicting superior

outcome. Minimal residual disease may serve as a biomarker
to inform therapy and as a surrogate for OS. Specifically,
achieving MRD-negative status may become a goal of future
studies using induction, transplant, consolidation, and/or
maintenance therapies. With available technologies and
ease of MRD measurement, it is now time to carry out
additional large prospective studies to define the clinical
significance of MRD and its impact on patient outcome in
myeloma. In our quest toward personalizing therapy for
patients, as in chronic myeloid leukemia, it may now be
possible to both assess and monitor MRD using standardized
assays and decide both intensity and length of therapy for
individual patients to improve patient outcome.
References
1. Kumar SK, Dispenzieri A, Lacy MQ, et al. Continued improvement in
survival in multiple myeloma: changes in early mortality and outcomes
in older patients. Leukemia. 2013;28:122-128.
3. Durie BG, Harousseau JL, Miguel JS, et al; International Myeloma
Working Group. International uniform response criteria for multiple
4. Attal M, Lauwers-Cances V, Marit G, et al; IFM Investigators. Lenalidomide maintenance after stem-cell transplantation for multiple myeloma. N Engl J Med. 2012;366:1782-1791.
5. McCarthy PL, Owzar K, Hofmeister CC, et al. Lenalidomide after stem-cell
transplantation for multiple myeloma. N Engl J Med. 2012;366:1770-1781.
6. Jakubowiak AJ, Dytfeld D, Griffith KA, et al. A phase 1/2 study of
carfilzomib in combination with lenalidomide and low-dose dexamethasone as a frontline treatment for multiple myeloma. Blood.
2012;120:1801-1809.
7. Kumar S, Flinn I, Richardson PG, et al. Randomized, multicenter, phase 2
study (EVOLUTION) of combinations of bortezomib, dexamethasone,
cyclophosphamide, and lenalidomide in previously untreated multiple
myeloma. Blood. 2012;119:4375-4382.
8. Cavo M, Tacchetti P, Patriarca F, et al; GIMEMA Italian Myeloma
Network. Bortezomib with thalidomide plus dexamethasone compared
with thalidomide plus dexamethasone as induction therapy before, and
consolidation therapy after, double autologous stem-cell transplantation in newly diagnosed multiple myeloma: a randomised phase 3
study. Lancet. 2010;376:2075-2085.
9. Harousseau JL, Attal M, Avet-Loiseau H. The role of complete response
in multiple myeloma. Blood. 2009;114:3139-3146.
~ol de MM)/
10. Paiva B, Vidriales MB, Cervero J, et al; GEM (Grupo Espan
PETHEMA Cooperative Study Groups. Multiparameter flow cytometric
remission is the most relevant prognostic factor for multiple myeloma
patients who undergo autologous stem cell transplantation. Blood.
2008;112:4017-4023.
11. Paiva B, Martinez-Lopez J, Vidriales MB, et al. Comparison of immunofixation, serum free light chain, and immunophenotyping for response evaluation and prognostication in multiple myeloma. J Clin
Oncol. 2011;29:1627-1633.
~ol L, et al; PETHEMA/GEM Cooperative
12. Paiva B, Gutierrez NC, Rosin
Study Groups. High-risk cytogenetics and persistent minimal residual
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
disease by multiparameter flow cytometry predict unsustained complete response after autologous stem cell transplantation in multiple
myeloma. Blood. 2012;119:687-691.
Rawstron AC, Child JA, de Tute RM, et al. Minimal residual disease
assessed by multiparameter flow cytometry in multiple myeloma:
impact on outcome in the Medical Research Council Myeloma IX Study.
J Clin Oncol. 2013;31:2540-2547.
Puig N, Sarasquete ME, Balanzategui A, et al. Critical evaluation of ASO
RQ-PCR for minimal residual disease evaluation in multiple myeloma.
A comparative analysis with flow cytometry. Leukemia. 2013;28:391-397.
Martinez-Lopez J, Lahuerta JJ, Pepin F, et al. Prognostic value of deep
sequencing method for minimal residual disease detection in multiple
myeloma. Blood. 2014;123:3073-3079.
Paiva B, Puig N, Garcia-Sanz R, et al. Grupo Espanol de Mieloma
/Programa para el Estudio de la Terapeutica en Hemopatias Malignas
cooperative study g. Is this the time to introduce minimal residual
disease in multiple myeloma clinical practice? Clin Cancer Res. 2015;21:
2001-2008.
van Dongen JJ, Lhermitte L, Bottcher S, et al; EuroFlow Consortium
(EU-FP6, LSHB-CT-2006-018708). EuroFlow antibody panels for standardized n-dimensional flow cytometric immunophenotyping of normal,
reactive and malignant leukocytes. Leukemia. 2012;26:1908-1975.
Kalina T, Flores-Montero J, van der Velden VH, et al; EuroFlow Consortium (EU-FP6, LSHB-CT-2006-018708). EuroFlow standardization of
flow cytometer instrument settings and immunophenotyping protocols. Leukemia. 2012;26:1986-2010.
Flanders A, Stetler-Stevenson M, Landgren O. Minimal residual disease
testing in multiple myeloma by flow cytometry: major heterogeneity.
Blood. 2013;122:1088-1089.
Stetler-Stevenson M, Paiva B, Stoolman L, et al. Consensus guidelines
for myeloma minimal residual disease sample staining and data acquisition. Cytometry B Clin Cytom. Epub 2015 Jul 6.
van Dongen JJ, van der Velden VH, Bruggemann M, et al. Minimal residual
disease diagnostics in acute lymphoblastic leukemia: need for sensitive,
fast, and standardized technologies. Blood. 2015;125:3996-4009.
Paiva B, van Dongen JJ, Orfao A. New criteria for response assessment:
role of minimal residual disease in multiple myeloma. Blood. 2015;125:
3059-3068.
Ladetto M, Bruggemann M, Monitillo L, et al. Next-generation sequencing and real-time quantitative PCR for minimal residual disease
detection in B-cell disorders. Leukemia. 2013;28:1299-1307.
e429
HERVE AVET-LOISEAU
24. Korde N, Roschewski M, Zingone A, et al. Treatment with carfilzomiblenalidomide-dexamethasone with lenalidomide extension in patients
with smoldering or newly diagnosed multiple myeloma. JAMA Oncol.
2015;1:746-754.
25. Avet-Loiseau H, Corre L, Lauwers-Cances V, et al. Evaluation of minimal
residual disease (MRD) by next generation sequencing (NGS) is highly
predictive of PFS in the IFM/DFCI 2009 trial. Blood. 2015;126:191.
26. Zamagni E, Patriarca F, Nanni C, et al. Prognostic relevance of 18-F FDG
PET/CT in newly diagnosed multiple myeloma patients treated with upfront autologous transplantation. Blood. 2011;118:5989-5995.
27. Usmani SZ, Mitchell A, Waheed S, et al. Prognostic implications of serial
18-fluoro-deoxyglucose emission tomography in multiple myeloma
treated with total therapy 3. Blood. 2013;121:1819-1823.
e430
28. Bartel TB, Haessler J, Brown TL, et al. F18-fluorodeoxyglucose positron

emission tomography in the context of other imaging techniques and
prognostic factors in multiple myeloma. Blood. 2009;114:2068-2076.
29. van Lammeren-Venema D, Regelink JC, Riphagen II, et al. F-fluorodeoxyglucose positron emission tomography in assessment of myelomarelated bone disease: a systematic review. Cancer. 2012;118:1971-1981.
30. Elliott BM, Peti S, Osman K, et al. Combining FDG-PET/CT with laboratory data yields superior results for prediction of relapse in multiple
myeloma. Eur J Haematol. 2011;86:289-298.
31. Zamagni E, Nanni C, Mancuso K, et al. PET/CT improves the definition of
complete response and allows to detect otherwise unidentifiable
skeletal progression in multiple myeloma. Clin Cancer Res. 2015;21:
4384-4390.
NEW TARGETS AND AGENTS IN HIGH-RISK MULTIPLE MYELOMA
New Targets and New Agents in High-Risk Multiple Myeloma

Ajay K. Nooka, MD, FACP, and Sagar Lonial, MD
OVERVIEW
Advances in the treatment of multiple myeloma have resulted in dramatic improvements in outcomes for patients. The
newly emerging profiling of mutations emerging as a consequence of large prospective sequencing studies such as the
CoMMpass Study or other efforts from European investigators are not further helping to define the place and role for
personalized medicine in myeloma. While mutations such as NRAS, KRAS, and BRAF do occur in myeloma, it is not clear that
targeting them as a single drug strategy will result in meaningful responses or durations of response. Personalized medicine
in multiple myeloma at this time likely entails the use of risk-based approaches for maintenance therapy, the use of current
biology-based treatments such as proteasome inhibitors, and immunomodulatory agents, with an eye towards the use of
mutation-specific treatments in the setting of minimal residual disease or in concert with biology-based treatments overall.
he improvement in outcomes of patients with high-risk

myeloma in recent years can be attributed to a confluence of numerous factors. First, improved stratification of
patients with myeloma by their risk status has allowed us to
better define the high-risk myeloma subset. Improved
stratification enabled us to specifically focus on developing
treatment algorithms aimed at delivering uninterrupted
intensive antimyeloma therapies, with resultant improvement in survival outcomes. Second, the availability of novel
therapies with proven clinical efficacy among patients with
myeloma, especially among high-risk patients, led to unprecedented survival improvements. More importantly, vast
clinical experience with these newer agents has not only
enabled us to better manage the toxicities of new agents,
but has also enabled us to consciously withdraw from the
excessive use of cytotoxic agents, such as alkylating agents,
among these patients with a highly genomically unstable
disease. This has further led to minimizing regimen-related
toxicity and decreased the incidence of alkylator-related
secondary myelodysplastic syndromes. In this article, we
describe the clinical experience of effective antimyeloma
agents among high-risk patients and provide the rationale
for incorporating these most effective combination regimens in the treatment of patients with high-risk myeloma.
CURRENTLY APPROVED AGENTS AND THEIR

ACTIVITY IN HIGH-RISK MYELOMA
Successfully treating high-risk patients requires two important steps. First, among all patients with myeloma, nowhere is the importance of achieving a complete response
(CR) more important than in the high-risk cohort. Attaining

CR using agents with novel mechanisms of action is critical to
success. Second, the treatment approach used must be well
tolerated and not induce further genomic changes leading to
drug resistance. This is important because long-term duration of therapy is a critical part of this strategic approach
for high-risk patients. Although the high-risk myeloma
spectrum has been defined mostly by cytogenetics, a few
groups have identified high-risk patients based on the disease presentation as extramedullary myelomathat is,
plasma cell leukemiaand also by gene expression profiling. We have incorporated these high-risk attributes in the
review of individual drugs
Immunomodulatory Drugs
Thalidomide. The use of thalidomide is currently limited
because of the availability of next-generation agents with
greater efficacy and safety. In the context of high-risk patents, thalidomide usage has not been shown to improve
prognosis. The MRC-IX trial evaluated its role among patients with adverse immunofluorescence with fluorescence
in situ hybridization (iFISH; defined as gain(1q), t(4;14),
t(14;16), t(14;20), del(17p), and del(1p32)) demonstrated no
progression-free survival (PFS) benefit and, in fact, worse
overall survival (OS).1 The poor survival among the high-risk
patients receiving thalidomide maintenance was attributed
to the clonal advantage of the high-risk patients, the selective pressure of thalidomide accounting for acquired drug
resistance, and the subsequent poor survival after relapse.
Conflicting data were seen from the other groups that have
From the Winship Cancer Institute, Emory University School of Medicine, Atlanta, GA.
Corresponding author: Sagar Lonial, MD, Emory University, 1365 Clifton Rd., Building C, Room 3003, Atlanta, GA 30322; email: sloni01@emory.edu.
e431
NOOKA AND LONIAL
used metaphase cytogenetic abnormalities to define high

risk. The OS of the thalidomide arm of the Total Therapy 2
(TT2) regimen was superior among patients exhibiting cytogenetic abnormalities, but when examined in the context
of gene expression profiling data, the benefit of thalidomide
was limited to those patients with low-risk disease defined
by gene expression profiling.2 Several other studies have
repeatedly demonstrated the inability to induce responses
with the use of thalidomide among patients with high-risk
myeloma presenting with extramedullary myeloma.3,4 In
current practice, thalidomide has a nominal role in the
treatment of patients with high-risk myeloma.
poor prognosis conferred by t(4;14). Although additional

studies have confirmed that lenalidomide/dexamethasone
is a suboptimal regimen among patients with del(17)p, the
positive impact of this regimen on patients with t(4;14)
has never been recapitulated in any other trial to date.8,9
In the maintenance setting, Attal at al presented their
findings comparing fixed-duration lenalidomide maintenance placebo post-transplant.10 They reported that the
presence of t(4;14) or del(17)p are independent compared with risk factors for poor survival on the multivariate analysis, but they did not comment on the impact
of maintenance therapy on these high-risk patients.10
Lenalidomide. The use of lenalidomide and dexamethasone as induction therapy among high-risk patients did not
yield superior results. Among 40 patients (16%) exhibiting
t(4;14), t(14;16), t(14;20), or del(17)p on FISH analysis, the
median PFS was 1.4 years (compared with standard-risk
median PFS of 3.4 years; p = .0002). Not surprisingly, the
5-year OS was 47% versus 77% (p = .0001), favoring the
standard-risk patients.5 Clearly, lenalidomide/dexamethasone
was deemed as a suboptimal induction regimen, and a shift
toward triplet induction regimen has been made by all
parties, at least for the high-risk subgroup. The results from
the FIRST trial also did not demonstrate any survival benefit
for transplant-ineligible patients exhibiting high-risk features treated with lenalidomide/dexamethasone continuously (Table 1).6 These results suggest that a modified
combination regimen such as lenalidomide, bortezomib, and
dexamethasone (RVD)-lite might be an appropriate induction regimen for high-risk transplant-ineligible patients, though this has yet to be demonstrated. Among
patients with relapsed or refractory myeloma, Reece et al
evaluated the impact of cytogenetics, using lenalidomide/
dexamethasone in relapsed myeloma.7 Consistent with
other experiences, patients with del(17)p experienced
a worse outcome, with a median OS of 4.7 months. In
contrast, patients with t(4;14) experienced a median OS
similar to patients without any cytogenetic abnormalities,
suggesting lenalidomide/dexamethasone may abrogate the
Pomalidomide. The influence of cytogenetics in patients

with relapsed or refractory myeloma treated with pomalidomide and low-dose dexamethasone have been evaluated
by Dimopoulos and colleagues as a subgroup analysis from
the registration phase III MM003 trial. The results demonstrate quite a variance between the outcomes of patients with del(17p) and t(4;14). The median PFS rates for
standard risk versus del(17p) versus t(4;14) were 4.2, 4.6,
and 2.8 months, respectively; the median OS rates were
14.0, 12.6, and 7.5 months, respectively, suggesting that
the pomalidomide/low-dose dexamethasone regimen may
overcome the poor prognosis of patients with del(17p) but
not t(4;14).11 Similar results were demonstrated in a dedicated high-risk trial from the Intergroupe Francophone du
Myelome; patients with del(17)p and t(4:14) treated with
pomalidomide/low-dose dexamethasone achieved median
time to progression (TTP) for del(17p) and t(4;14) of 7.3
versus 2.8 months, and the 8-month OS was 41% versus
12.4%, respectively.12 This regimen could serve as an interesting backbone upon which to further build combinations that would benefit patients with del(17)p and t(4;14).
Based on this concept, IFM-2014-01 is evaluating the regimen of pomalidomide/low-dose dexamethasone with the
addition of ixazomib exclusively among patients with relapsed or refractory high-risk myeloma who have received
more than two lines of therapy with an enrollment goal of
50 patients.
Proteasome Inhibitors
KEY POINTS
e432
Identification of risk status should be performed on all

patients at diagnosis.
Immune-based treatments may provide unique ways by
which to target high-risk myeloma.
Combination therapy with currently available novel
agents may help mitigate the outcomes of high-risk
disease.
Cellular immune-based treatments are currently in
evolution with rapidly expanding targets.
Specific mutation-driven treatments are in
development and may offer benefit as part of
a combination approach with biology-based treatments.
Bortezomib. The UAMS group as part of the Total Therapy 3

(TT3) experience incorporated bortezomib in the induction
treatment of myeloma. Compared with TT2, a meaningful
benefit was apparent among younger patients with high-risk
myeloma defined by GEP70. The 2-year event-free survival
rates (TT3 vs. TT2: 68% vs. 30%) and OS rates (TT3 vs. TT2:
75% vs. 50%) strongly favored inclusion of bortezomib in
induction therapy for patients with high-risk myeloma.13
Additionally, in the earlier relapsed trials with bortezomib,
Jagannath et al evaluated the utility of bortezomib among
high-risk cohorts (defined as del13 by FISH and/or metaphase cytogenetics). This analysis suggested that bortezomib may be able to deliver responses uniformly, irrespective
of the risk status.14 Among patients with newly diagnosed
VTD SCT 3 2
VTD (236) vs.
TD SCT 3
2 TD (238)
Cavo
et al,16
2010/III
RVD SCT RVD

(45)
N/A
45 (100)
620 (81.3) 142 (18.7) NR
2 (5)
19 (42)
10.6%
38.8%
96%
34.7%
N/A
30.2%
47.3%
46%
68%
49.3%
66%
NR
NR
NR
96%
10.6%
34.7%
30.2%
NR
NR
NR
NR
17.5%
8.4%
24 mo
30 mo
8% (5 y PFS)
16%
28 mo
37 (3 y PFS)
69
t(4;14) PFS
17.5%
8.4%
24 mo
30 mo
5% (5 y PFS)
22%
14 mo
37 (3 y PFS)
69
del(17)p PFS
71%
77.1%
72% (4 y OS)
65%
64% (5 y OS)
72%
79% (4 y OS)
NR
Std Risk OS
39.6%
40.7%
27% (4 y OS)
45%
33% (5 y OS)
52%
63% (4 y OS)
NR
t(4;14) OS
N/A
32 mo
28 mo
N/A
93% (3 y OS)
NR
Median
Followup
39.6%
40.7%
27% (4 y OS)
45%
18% (5 y OS)
65%
94% (3 y OS) 26
NR
37
94
50% (4 y OS) 24
NR
del(17)p OS
24.9% (3 y PFS) 14.6% (3 y PFS) 14.6% (3 y PFS) 64.8% (3 y OS) 46.8% (3 y OS) 46.8% (3 y
OS)
21.2%
31.1%
36 mo
32 mo
24% (5 y PFS)
27%
36 mo
63 (3 y PFS)
74
del(17)p
VGPR Std Risk PFS
Abbreviations: VGPR, very good partial response; PFS, progression-free survival; OS, overall survival; mo, months; VTD, bortezomib, thalidomide, and dexamethasone; SCT, stem cell transplant; NR, not reported; TD, thalidomide and
dexamethasone; VD, bortezomib and dexamethasone; y, years; V, bortezomib; T, thalidomide; VMP, bortezomib/melphalan/prednisone; Rd, lenalidomide and dexamethasone; MPT, melphalan/thalidomide/prednisone; RVD, lenalidomide,
bortezomib, and dexamethasone.
*Cytogenetics available for 762 of 1,623 patients in the FIRST trial.
Nooka
et al,19
2013
Avet-Loiseau Rd continuous
et al,22
(248) vs. Rd18
(261) vs. MPT
2015/III*
(253)
NR
63%
13
19
66
VMP Rd
(sequential)
76
(118) vs. VMPRd
alternating (115)
Mateos
et al,59
2015/III
NR
56%
54 (11)
NR
50/295 (17%) 39/312 (13%) 76%
106 (21)
17 (8)
15 (7)
Std Risk t(4;14)

VGPR VGPR
NR
NR
del(17)p,
No. Patients
(%)
PAD SCT 3
2 V (413) vs.
VADSCT 32T
(4141)
NR
57 (26)
NR
NR
53 (24)
NR
High Risk,
t(4;14),
No.
No. Patients
Patients
(%)
(%)
Sonneveld
et al,18
2015/III
Avet-Loiseau VDSCT (507)

et al,15
2010/III
Regimen (No.
Patients)
Study, Year/
Phase
Std Risk,
No.
Patients
(%)
TABLE 1. Outcomes Among Patients With High-Risk Myeloma From Newly Diagnosed Trials
e433
NOOKA AND LONIAL
myeloma, several trials evaluated the impact of bortezomib

on high-risk patients. From the IFM trial that compared
bortezomib and dexamethasone to the chemotherapy
combination regimen vincristine, doxorubicin, and dexamethasone as induction therapy before high-dose therapy
and autologous transplant, differential outcomes were
observed for t(4;14) and del(17)p. Patients with t(4;14)
appeared to have significantly improved outcomes with
bortezomib/dexamethasone as induction therapy compared
with those with del(17)p (Table 1).15 Another induction trial
evaluated the impact of bortezomib, thalidomide, and
dexamethasone (VTD) versus thalidomide and dexamethasone on outcomes of high-risk cohorts including patients
with newly diagnosed myeloma. Cavo and colleagues
demonstrated that VTD with bortezomib as part of consolidation could eliminate the negative impact of t(4;14),
although this was not the case with the thalidomide/
dexamethasone arm.16 A subsequent trial, HOVON/GMMG,
also evaluated the impact of bortezomib as part of induction
and as maintenance.17 As illustrated in Table 1, the use of
bortezomib as maintenance therapy was able to reduce the
negative impact of high-risk cytogenetics on outcomes.
Prolonged bortezomib treatment mostly abrogated the
del(17p) effect on PFS and OS but had limited benefit among
patients with t(4;14).18 Our group evaluated the use of the
combination of RVD at modified doses as a post-transplant
maintenance strategy among patients with high-risk myeloma.19 The use of RVD maintenance after transplant significantly improved PFS and OS for high-risk patients, greater
than what has been reported with single-agent maintenance
with either lenalidomide or bortezomib.
Carfilzomib. The efficacy of carfilzomib as a single agent was
evaluated among patients with high-risk myeloma in the PX171-003-A1 study. Patients with t(4;14) had the highest
overall response rate (ORR) of 38.9%. Those with del(17)p
had the lowest ORR of 16.7%. Among high-risk patients, the
median PFS and OS trended higher among the patients with
t(4;14). Among patients with t(4;14) only, the OS was
15.8 months, and, among all high-risk patients,20 the OS was
shorter compared with the standard-risk patients (9 vs.
23 months; Table 2). In the ENDEAVOR trial, a phase III
study that compared two proteasome inhibitors,21 carfilzomib with dexamethasone was superior to bortezomib/
dexamethasone among all patients with myeloma irrespective
of baseline cytogenetic risk status. However, the median
PFS for high-risk patients was lower compared with the
standard-risk patients. Carfilzomib/dexamethasone improved the outcome of high-risk patients but did not
overcome the adverse impact of high-risk cytogenetics. A
similar experience from the ASPIRE trial that compared
carfilzomib, lenalidomide, and dexamethasone (KRD) with
lenalidomide/dexamethasone in a phase III trial demonstrated KRD exerting higher-quality responses both
among the high-risk and standard-risk patients. The
overall responses among the high-risk versus standardrisk patients with KRD was 79.2% versus 91.2% and for
e434
lenalidomide/dexamethasone was 59.6% versus 73.5%,

respectively. The median PFS among patients who received
KRD with standard risk versus t(4;14) versus del(17)p was
29.6, 23.1, and 24.5 months, and for Rd was 19.5, 16.7, and
11.1 months, respectively (Table 2). Though there are
several limitations to the data set including unavailability of
cytogenetics for more than half the patients, these results
suggest the proteasome inhibitor/immunomodulatory drug
combination can overcome the poor prognosis conferred
by these high-risk cytogenetics.22 In another recent trial
evaluating carfilzomib in combination with pomalidomide/
low-dose dexamethasone (KPD) for relapsed/refractory myeloma, high ORRs were seen with this combination, even
among patients with high-risk genetics (Table 2).23
Ixazomib. Ixazomib is the most recently approved oral
proteasome inhibitor that has been evaluated in the
TOURMALINE-MM1 trial. This phase III study randomly selected 722 patients to receive lenalidomide/dexamethasone
or to receive ixazomib in combination with lenalidomide/
dexamethasone (IRD). The combination of IRD was shown
to be superior in delivering high or at least very good
partial response (VGPR) rates among both high-risk and
standard-risk patients. At least VGPR with IRD for standard
risk versus t(4;14) versus del(17)p was 51%, 53%, and 39%,
and for lenalidomide/dexamethasone was 44%, 28%, and
21%, respectively. The median PFS with IRD for standard
risk versus t(4;14) versus del(17)p was 20.6, 18.5, and
21.4 months, and for lenalidomide/dexamethasone was
15.6, 12, and 9 months, respectively (Table 2).24 These results also suggest the same observations seen with KRD,
a proteasome inhibitor/immunomodulatory drug combination is the most optimal regimen to deliver the best results
for patients with high-risk cytogenetics.
IMMUNE-BASED APPROACHES
Immunotherapeutic approaches as antimyeloma therapies
aimed at targeting the malignant plasma cell have been
explored for some time with limited success. Though identifying the ideal target was a major challenge, the current
clinical activity using the monoclonal antibodies targeting
SLAMF7 and CD38 demonstrate much promise. Acceptable
safety in combination with other antimyeloma agents allow
for their utility in treating myeloma, even among the patients heavily pretreated for myeloma. Furthermore, drug
development using novel immune-based antimyeloma approaches such as checkpoint inhibitors and cellular therapies
are highly promising.
Monoclonal Antibodies
Elotuzumab. Elotuzumab is a recently approved monoclonal antibody targeting the surface protein SLAMF7. Its U.S.
Food and Drug Administration approval in combination with
lenalidomide and dexamethasone for the treatment of
patients with multiple myeloma who have received one to
three prior lines of therapies was based on the results
IRd (360) vs. 199 (55)

Rd (362)
216 (60)
Kd (464) vs. 284 (61)

Vd (465)
291 (63)
Rd (130)
Rd (207)
Chng
et al,21
2015/III
Reece
et al,7
2009/II
Avet-Loiseau
et al15, 2010
K (266)
KPd (32)
Jakubowiak
et al,20
2013/II
Shah et al,60
2015/I
15 (10)
10 (31)
3 (10)
62 (27.1) 18 (8)
44 (15)
41 (38)
22 (46)
99 (44)
50 (100)
14 (5)
27.4%
56.7%
44%
51%
45.3
75.5
5 (15)
30 (13)
23 (15)
44 (15)
32 (64)
16%
8.4
1.4
6.8
N/A
NR
12 (9.2) NR
NR
NR
NR
NR
21 (40)
18 (38)
NR
NR
2.3
3%
NR
NR
27.5%
44.3%
28%
53%
27%
60.5%
NR
t(4;14)
VGPR
NR
6.8
9%
NR
NR
27.5%
44.3%
15%
39%
27%
60.5%
NR
del(17)p
VGPR
10.2
NR
15.6
20.6
19.5
29.6
7.2
1.9
NR
4.6
2.3
4.2
2.2 (TTP)
6.0
8.8
9.7
21.4
11.1
24.1
14.85
18.46
del(17)p
PFS
15.1
22.7
NR
NR
NR
NR
NR
4.5
1.9
2.8
60% (12 mo)
3.5
1.1
4.6
20.6
7.7
12.6
NR
19
14
N/A
NR
NR
NR
NR
NR
OS
Std Risk
(mo) OS
41% (8 mo TTP) 12.4% (8 mo TTP) 12
5.5
8 (TTP)
6.0
8.8
12.0
18.5
16.7
23.1
15.74
20.34
Std Risk t(4;14)

PFS
PFS
7.3
N/A
(TTP)
9.6
7.1
(TTP)
9.4
18.7
14.7
20.6
17.6
26.3
14.9
19.4
PFS
(mo)
NR
11.8
4.9
7.5
NR
9.4
NR
NR
NR
NR
NR
t(4;14)
OS
N/A
15.4
10
19.7
N/A
23
32
24.5
Median
Follow-up
(mo)
80% (12 mo) 26
7.0
7.7
12.6
NR
4.7
NR
NR
NR
NR
del(17)p
OS
Abbreviations: VGPR, very good partial response; PFS, progression-free survival; OS, overall survival; mo, months; ERd, elotuzumab in combination with lenalidomide and dexamethasone; Rd, lenalidomide and dexamethasone; NR, not reported;
KRd, carfilzomib, lenalidomide, and dexamethasone; IRd, ixazomib in combination with lenalidomide and dexamethasone; Kd, carfilzomib and dexamethasone; Vd, bortezomib and dexamethasone; TTP, time to progression; Pd, pomalidomide and
low-dose dexamethasone; HiDex, high-dose dexamethasone; K, carfilzomib; KPd, carfilzomib in combination with pomalidomide and low-dose dexamethasone.
Note: unknown cytogenetic status: KRD, 201 (51%) vs. Rd, 174 (44%) from ASPIRE trial; unknown cytogenetic status: IRd, 86 (24%) vs. Rd, 84 (23%) from TOURMALINE-MM1 trial; unknown cytogenetic status: KD, (18%) vs. Vd, (13%) from
TOURMALINE-MM1 trial; unknown cytogenetic status for 47 patients in PX-171-003-A1 study. Available for 229 of 266 patients; MM003 cytogenetics available for 225 (75%) in Pd and 107 (70%) patients.
167 (72.9)
Pd (302) vs. 126 (56)

HiDex
66 (62)
(153)
Dimopoulos
et al,11
2015/III
N/A
Pd (50)
Leleu
et al,12
2015/III
7 (6)
28 (21.5)
NR
NR
97 (21)
NR
113 (24)
NR
62 (17)
31 (60)
33 (69)
104 (32)
75 (21)
52 (13)
Moreau
et al,24
2015/III
48 (12)
170 (43)
KRd (396)
vs.
Rd (396)
Avet-Loiseau
et al,22
2015/III
102 (32)
31 (10)
del(17)p,
No.
Patients Std Risk
(%)
VGPR
30 (9)
High Risk,
t(4;14),
No.
No. Patients
Patients
(%)
(%)
147 (37)
ERd (321)
vs.
Rd (325)
Std Risk,
No.
Patients
(%)
Lonial
et al,25 2015
Study, Year/
Phase
Regimen
(No.
Patients)
TABLE 2. Outcomes Among Patients With High-Risk Myeloma From Relapsed/Refractory Trials
e435
NOOKA AND LONIAL
meeting the primary endpoint of PFS and ORR from the

ELOQUENT-2 trial. In this trial, 646 patients were randomly selected to receive elotuzumab in combination with
lenalidomide and dexamethasone (ERD) or lenalidomide/
dexamethasone among patients sensitive to lenalidomide.
The ORRs were 79% and median PFS of 19.4 versus
14.9 months (p = .0014) favored the ERD arm.25 The 3-year
extended PFS was recently presented at the 2015 American
Society of Hematology (ASH) Annual Meeting and demonstrated
that more than one-quarter of the patients receiving ERD
continue to be free from progression, suggesting long-term
benefits with immune-based approaches. The prespecified
interim analysis for OS indicated a strong trend favoring ERD
(43.7 vs. 39.6 months; p = .0257).
Data regarding the efficacy of elotuzumab among patients at
high risk are very encouraging. Approximately 32% of patients
had the cytogenetic abnormality of del(17)p and 9% of patients
had t(4;14), and the outcomes for these patients have been
surprisingly comparable with those of the patients at standard
risk. The PFS benefit of the combination of ERD relative to
lenalidomide/dexamethasone was evident among both of
these subsets; hazard ratios (HR) for del(17)p and t(4;14) were
0.65 (95% CI, 0.450.94) and 0.53 (95% CI, 0.290.95), respectively. At a median follow-up of 24 months, the PFS for
standard risk versus del(17)p versus t(4;14) patients were
similar at 19.4, 18.46, and 20.34 months, respectively.
Based on a phase I trial showing the tolerability of elotuzumab in combination with bortezomib and dexamethasone (EVD) among heavily pretreated patients,26 another
randomized phase II trial was designed that randomly selected 152 bortezomib-sensitive patients with myeloma who
had received one to three prior lines of therapy to receive
EVD or bortezomib/dexamethasone. At a median followup of 27.3 months, a PFS benefit of close to 3 months
was observed for the EVD arm (EVD vs. bortezomib/
dexamethasone: 9.7 vs. 6.9 months, respectively; p = .018).
Similarly, the OS trended in favor of EVD (2-year OS, EVD vs.
bortezomib/dexamethasone: 73% vs. 66%).27
Multiple combination trials of elotuzumab with other
antimyeloma agents are ongoing. Study of the safety and
efficacy of elotuzumab in combination with RVD (ERVD)
among transplant-eligible patients is currently ongoing.
Preliminary data suggest that this as a safe combination.
Another trial (S1211) of RVD with or without elotuzumab
specifically focused on accruing newly diagnosed patients
with high-risk multiple myeloma is underway with an expected accrual of 122 patients. GMMG-HD6 in a randomized
phase III trial is evaluating the same combination (ERVD)
versus RVD among newly diagnosed transplant-eligible
patients. The study also has a second set of patients randomly selected to receive RVD after receiving stem cell
transplant (SCT) versus ERVD consolidation and lenalidomide versus lenalidomide with elotuzumab maintenance,
with the primary endpoint of PFS. Another similarly designed
trial from the DSMM XVII evaluating the combination (KRD
vs. KRD with elotuzumab) also has a second trial of randomly
selected patients after receiving SCT and KRD versus KRD
e436
with elotuzumab consolidation to receive lenalidomide

versus lenalidomide with elotuzumab maintenance for the
primary endpoint of CR post-induction and 3-year PFS
post = first randomization. In the relapsed setting, trials of
elotuzumab in combination with pomalidomide and lowdose dexamethasone (EPD; accrual goal: 60 patients) and
combination nivolumab to EPD (accrual goal: 40 patients)
are underway. More importantly, the safety and tolerability
of elotuzumab in combination with other monoclonal antibodies such as lirilumab or urelumab are also ongoing.
Daratumumab. Daratumumab, a humanized IgG1 antiCD38 antibody, is the most active monoclonal antibody in
myeloma to date. It was recently approved in the United
States at the active dose of 16 mg/kg as a single agent
resulting in response rates of 29%.28 The combined results of
GEN501 part 2 and the SIRIUS trials comprising 148 patients
were presented at ASH 2015. At a median follow-up of
14.8 months, 50% of the responders were progression free
at 1 year. Among most patients, responses deepened with
continued treatment and were tolerated well. The 1-year OS
rate among this heavily pretreated patient population was
69% (95% CI, 60.4%75.6%) and the median OS was
19.9 months, demonstrating its efficacy as a single agent.29
More importantly, the minimal adverse event profile and the
established safety and tolerability of this antibody (48% of
patients had infusion reactions, mostly # grade 2) led to the
design of multiple combination trials.
Daratumumab in combination with lenalidomide/
dexamethasone, among patients with early-relapse multiple
myeloma who have received a median of two lines of therapy,
demonstrated response rates of 81%, and close to two-thirds
of patients who enrolled in the trial appeared to have
achieved at least a VGPR (63%).30 In this trial, the toxicity
profile was acceptable and more likely related to the lenalidomide than the daratumumab. Although greater than or
equal to grade 3 or worse toxicities were seen in 88% of
patients, most of these were at least grade 3 neutropenia
(78%) that have resolved with supportive growth factor
therapies. In another recent study presented by Chari et al,
the combination of daratumumab with pomalidomide/lowdose dexamethasone among patients who have received four
lines of therapy resulted in response rates of 71% and at least
VGPR rates seen among 43% of patients. More than half (53%)
of patients were progression free at 1-year among this heavily
pretreated population.31 Daratumumab certainly has a good
safety and efficacy profile and can be combined with a
number of antimyeloma therapies. The question of whether
we can rechallenge with a subsequent daratumumab-based
combination therapy among those patients who are deemed
refractory to daratumumab single-agent therapy remains to
be answered. Data of responses among the high-risk patients
are encouraging. At least a 20% ORR among patients with
heavily pretreated, relapsed, refractory, high-risk multiple
myeloma and those with extramedullary myeloma as a single
agent is of great interest. More than one quarter of the
patients in the SIRIUS trial had either del(17)p (17%) or t(4;14)
(10%), and the responses were not limited to standard-risk

patients alone.28
Several daratumumab trials are in progress. The HOVON/
IFM trial is evaluating the combination of VTD with daratumumab in the induction, consolidation, and daratumumab
maintenance for the primary endpoint of stringent complete
remission, PFS, and OS. From the Alliance groups, the
concept of evaluating lenalidomide/dexamethasone versus
daratumumab/lenalidomide/dexamethasone in transplantineligible patients and RVD versus daratumumab with RVD
are in the concept phase. In the maintenance setting,
daratumumab/lenalidomide versus lenalidomide posttransplant until progression or unacceptable toxicity is
under evaluation (communication with Dr. McCarthy).
Several other phase III trials of daratumumab in combination
with lenalidomide/dexamethasone versus lenalidomide/
dexamethasone or in combination with bortezomib/
dexamethasone versus bortezomib/dexamethasone have
completed accrual and are awaiting results. Several other
phase III trials in transplant-ineligible patients (daratumumab
with or without bortezomib/melphalan/prednisone) are
ongoing. Given the longer infusion times and the higher
rate of infusion reactions seen (although majority of these
are # grade 2), daratumumab is currently being evaluated as a
subcutaneous formulation with the addition of recombinant
human hyaluronidase (rHuPH20) among patients with multiple
myeloma.
A second CD38 monoclonal antibody, SAR650984 (isatuximab), in combination with lenalidomide/dexamethasone
demonstrated an ORR of 65% and VGPR rates of 32% among
patients with multiple myeloma who were heavily pretreated (six
prior lines of therapy).32 Patients experienced a median PFS of
6.2 months reported at a median follow-up of 6 months. Another
phase IB study of isatuximab in combination with pomalidomide
and dexamethasone is currently enrolling patients.
and, at the same time, limiting the toxicities seen with an allogeneic transplant. The main goal of cellular therapies is to
induce a robust antimyeloma immune response that can have a
potential effect on disease control in the long term. The utility of
chimeric antigen receptor T cells (CAR-T cells), natural killer cells,
and regulatory T cells is encouraging but preliminary (Table 3).
Cellular Therapies
Therapeutic Vaccinations
The idea behind using cellular immunotherapies is to attain

similar benefits to those observed after allogeneic transplant
Dendritic cell vaccines. The utility of dendritic cell vaccines

is currently being explored in myeloma. Vaccines aimed at
CAR-T cells. Engineering autologous T cells to express antigenspecific CAR-T cells can be a potential immune-based
therapeutic approach for B-cell malignancies. The utility of
CAR-T cells has also been recently investigated in myeloma
management. One of the means by which myeloma cells evade
immunosurveillance is by enabling mutations in the major histocompatibility complex (MHC). T cells bind to the T-cell receptor
with the MHC on the tumor cell. However, in the presence of
MHC mutations, normal T cells cannot bind to MHC T-cell receptors. CAR-T cells have been genetically modified to have a
different extracellular antigen-binding site that allows CAR-T
cells to identify malignant cells without the use of the MHC.33 A
case of sustained CR to infusion of cytotoxic T lymphocyte (CTL)
019 cells in conjunction with autologous transplantation in a
patient with refractory myeloma was recently reported by
Stadtmauer and colleagues.34 They were able to demonstrate
that the clinically relevant target of CTL019 in this case may have
been non-neoplastic CD19+ cells, which have been implicated in
immune evasion and resistance to therapy in solid tumors.
However, one can argue the observed response could be attributed to be a response from alkylating therapies alone based
on the normal reconstitution of CD19+ B cells and loss of detectable CTL019. Nevertheless, the fact that six of the 10 patients
treated remained progression free is hypothesis-generating in
that this is not just an effect of melphalan alone. Additional
CAR-T cell studies are in progress, and early clinical results are
promising, though other targets such as B-cell maturation antigen may prove to be more suitable for myeloma.35
TABLE 3. Other Cellular Therapies

Cellular Therapy Identification
Rationale
Clinical Activity
NK Cells
NK cells are members of the cellular immune Given the depression of NK activity in
refractory myeloma, and the idea of
system that play a key role in defense
enhancing NK efficacy to improve
against viruses and tumors. In humans,
antimyeloma activity has led to ongoing
NK cells are identified as CD32CD56
lymphocytes but may differ in their
trials exploring the utility of NK cellbased
immunoregulatory role based on their
therapy in myeloma.
further expression.61
Tregs
Tregs are CD4 + T-cell population

(CD4+CD25+FoxP3+ T cells), comprise
5%10% of peripheral CD4 T cells, and
are responsible for the control of
autoimmune phenomena.
They are believed to strongly inhibit

Among 10 patients undergoing ASCT, Treg
antitumor immune responses among
depletion for patients with myeloma was
patients with myeloma. It is unclear if
deemed to be safe, and further studies to
Tregs play a role in protection from
determine the role of Treg depletion are
malignancy, but Treg depletion leads to
ongoing.62
induction of tumor rejection in murine
models. Treg depletion may enhance the
function of tumor antigenspecific T cells.
Abbreviations: NK, natural killer; Tregs, regulatory T cells; ASCT, autologous stem cell transplant.
e437
NOOKA AND LONIAL
developing myeloma-specific immunity have become a

major area of focus. Currently, several groups are working on
either whole-cell or antigen-specific dendritic cell vaccines
to enhance innate immunity to obtain the antitumor effect.
Currently, these are early approaches and in phase I. A
dendritic cellmyeloma cell fusion vaccine study among
patients with multiple myeloma supported by the Bone
Marrow Transplant Clinical Trials Network as a posttransplant maneuver to gain disease control is underway.
Peptide vaccines. PVX-410 is composed of a tetrapeptide
from three unique regions of myeloma-associated antigens
(XBP1, CD138, and CS1). The goal is to induce immunity
against myeloma cells by selectively stimulating tumorassociated antigen-specific CTLs. The safety and tolerability of PVX-410, alone and in combination with lenalidomide,
show considerable promise in treatment of patients with
smoldering myeloma.36 Future studies in combination with
MEDI4736, a PD-L1 antibody and lenalidomide PVX-410, will
be initiated shortly. Immucin, an anti-MUC1 signal peptide
vaccine, induced a robust CD8+- and CD4+-specific T-cell
response in all 15 patients following autologous SCT and
demonstrated a marked antitumor humoral response.37
Checkpoint Blockade
PD-1, or its ligand, PD-L1, interactions may indirectly
modulate the response to tumor antigens through T-cell/
antigen-presenting cell interactions. PD-1 engagement may
represent one way by which tumors evade immunosurveillance.38 The PD-1/PD-L1 pathway is shown by multiple groups to promote progression of myeloma indirectly
by undermining immune control of the malignancy.39 PD-1
blockade has been pursued as a promising therapeutic
strategy to reverse immune tolerance and enhance T-cell
effector function in several tumor types. The broad expression of PD-1 and its ligands in the microenvironment of
myeloma and the preclinical data reveal an important role of
the PD-1 pathway in immune evasion by myeloma cells, and
there may be immunotherapeutic potential of PD-1/PD-L
inhibition in myeloma. Several PD-1 antibodies such as
nivolumab, pidiluzumab (CT-011), and PD-L1 antibodies
(MEDI4736) are under investigation. Pembrolizumab, a
humanized IgG4 antiPD-1 monoclonal antibody designed
to block interaction of PD-1 with PD-L1 and PD-L2, was
evaluated in combination with immunomodulatory drugs
(lenalidomide/dexamethasone40 and pomalidomide/lowdose dexamethasone41) with an intent to enhance tumor
suppression. The ORR with lenalidomide/dexamethasone
was 76% ($ VGPR 24%)40 and the ORR with pomalidomide/
low-dose dexamethasone was 60% ($ VGPR 19%), 41
showing impressive activity with a tolerable safety profile.
Forty-two percent of the patients in the pembrolizumab and
pomalidomide/low-dose dexamethasone studies exhibited
high-risk features (del 17p, t(4:14) and/or t(14:16)). Among
these high-risk patients, a 50% ORR was observed.41 Several
other combination trials with PD-1 antibodies and PD-L1
antibodies are under investigation
e438
TARGETED THERAPIES
Boise and colleagues have published an interesting model for
the two sides of myeloma therapy, the Tao of myeloma,
where Yin and Yang represented the normal plasma cell and
malignant plasma cell biology, respectively.42 Our current
agents have focused on the biology side of the Tao, but among
high-risk patients, focusing on the mutational profile and the
abnormal expression may be equally important.43
MDM2 Inhibitors Targeting p53 Deletion or Mutation

p53, a tumor suppressor gene located on the short arm of
chromosome 17, is implicated in normal cellular proliferation, differentiation, and apoptosis. Loss of p53 is associated with poor survival among patients with myeloma
(del(17)p).44 Another protein, the murine double-minute 2
(MDM2), is a negative regulator of p53 and plays a role in
proliferation and survival of myeloma cells by suppressing
the p53 function. MDM2 inhibitors to block the MDM2-p53
protein-protein interaction have potential as antimyeloma
therapeutic strategies for patients with myeloma with
del(17)p; however, this hypothesis presumes there is a
normal and functional copy of p53 in the cell.45 There are
emerging data suggesting an increased frequency of p53
mutations among with del(17)p upon greater exposure
to treatment. This finding may limit the utility of MDM2
inhibitors. Currently, DS-3032b, an oral MDM2 inhibitor, is in initial trials for relapsed refractory myeloma
(NCT02579824).
BRAF Inhibitors
Prevalence of patients with myeloma with actionable BRAF
mutations (V600E) is much higher than what was originally
thought. Approximately 5% of patients harbor either a clone
or a subclone of BRAF V600E.46 It has been postulated that
myeloma clones harboring BRAF V600E might have a survival
advantage and likely present as aggressive, extramedullary
disease.47,48 Though recent analysis evaluating BRAF V600E
mutations in early-stage myeloma suggests no relation to
prognosis, this could represent the differential prognosis of
BRAF subclones at various phases of myeloma progression.46
There have been several reports of patients with BRAFmutated, relapsed, refractory myeloma who were treated
successfully with the BRAF inhibitor vemurafenib.47
MEK Inhibitors
The deregulation of the RAS/RAF/MEK/extracellular signal
regulated kinase (ERK; RAS/mitogen-activated protein
kinases) signal transduction pathway leads to abnormal
cellular proliferation, impaired apoptosis, enhanced angiogenesis, metastasis, and the development of drug resistance.49 Activating RAS mutations, which have a reported
incidence varying between 32% and 50% in multiple myeloma, are also thought to deregulate this pathway.50 The
presence of RAS mutations have previously been shown to
confer clinical prognosis.50,51 Mutation-specific inhibitor use
in myeloma so far is in the early phases. Several factors that
are not understood fully limit the therapeutic application of

the KRAS/NRAS inhibitors in myeloma. Variability of the
dominant clone relative to size of the clonal population
carrying the KRAS/NRAS mutations point to the presence of
intraclonal heterogeneity and also the potential difficulties
in the use of targeted treatment strategies.43
processing and the regulation of gene transcription. BRD as a

potential therapeutic target has resulted in the development
of BRD and BRD and extraterminal protein inhibitors.56 The
BRD inhibitor CPI203 exerted much preclinical activity in
both the bortezomib- and melphalan-resistant lines and
patient samples.57 Specific BRD4 degraders show promising
activity against preclinical models of multiple myeloma.58
FGFR3 Inhibitors
Simultaneous overexpression of two oncogenes, FGFR3
(fibroblast growth factor receptor 3) and MMSET (multiple
myeloma SET domain), represent a subset of patients with
myeloma with t(4;14) translocation, often indicating poor
prognosis.52 Though preclinical experience suggested antimyeloma activity in FGFR3-expressing cell lines,53 experience with small-molecule inhibitors has not made much
progress. It could represent that 30% of t(4;14) patients do
not express FGFR3.52 Further drug development with FGFR3
monoclonal antibodies54 and MMSET inhibitors55 are currently under evaluation.
BRD4 Inhibitors
Bromodomain (BRD) and extraterminal proteins play an
important role in cellular proliferation, cell cycle progression, and chromatin compaction. They use BRD modules
for lysine acetylation, a pivotal mechanism in chromatin
CONCLUSION
It is an exciting time in myeloma research, where future
progress is being built on the current success. The successful incorporation of immunomodulatory agents and
proteasome inhibitor combinations in high-risk multiple
myeloma management has improved the survival among
this group of patients. With monoclonal antibodies targeting
CD38, SLAMF7, and PD-1/PD-L1 pathways available in
clinical practice and, more importantly, the preliminary
results showing superior efficacy among high-risk patients,
future strategies of treating high-risk patients should incorporate these agents in the absence of target-specific
antimyeloma therapy. Identifying more relevant biologic
targets/actionable mutations and developing drugs for
these targets will be the next step to enhance deeper responses, resulting in durable remissions among the high-risk
patients with multiple myeloma.
References
1. Morgan GJ, Gregory WM, Davies FE, et al; National Cancer Research
Institute Haematological Oncology Clinical Studies Group. The role of
maintenance thalidomide therapy in multiple myeloma: MRC Myeloma
IX results and meta-analysis. Blood. 2012;119:7-15.
2. Barlogie B, Pineda-Roman M, van Rhee F, et al. Thalidomide arm of total
therapy 2 improves complete remission duration and survival in myeloma patients with metaphase cytogenetic abnormalities. Blood.
2008;112:3115-3121.
3. Wu P, Davies FE, Boyd K, et al. The impact of extramedullary disease at
presentation on the outcome of myeloma. Leuk Lymphoma. 2009;50:
230-235.
~ol L, Cibeira MT, Blade J, et al. Extramedullary multiple myeloma
4. Rosin
escapes the effect of thalidomide. Haematologica. 2004;89:832-836.
5. Srivastava G, Rana V, Lacy MQ, et al. Long-term outcome with lenalidomide and dexamethasone therapy for newly diagnosed multiple
6. Avet-Loiseau H, Hulin C, Benboubker L, et al. Impact of cytogenetics on
outcomes of transplant-ineligible patients with newly diagnosed
multiple myeloma treated with continuous lenalidomide plus low-dose
dexamethasone in the first (MM-020) trial. Blood. 2015;126:730.
7. Reece D, Song KW, Fu T, et al. Influence of cytogenetics in patients with
relapsed or refractory multiple myeloma treated with lenalidomide plus
dexamethasone: adverse effect of deletion 17p13. Blood. 2009;114:
522-525.
8. Avet-Loiseau H, Soulier J, Fermand JP, et al; IFM and MAG groups.
Impact of high-risk cytogenetics and prior therapy on outcomes in
patients with advanced relapsed or refractory multiple myeloma
treated with lenalidomide plus dexamethasone. Leukemia. 2010;24:
623-628.
9. Kapoor P, Kumar S, Fonseca R, et al. Impact of risk stratification on

outcome among patients with multiple myeloma receiving initial
therapy with lenalidomide and dexamethasone. Blood. 2009;114:
518-521.
10. Attal M, Lauwers-Cances V, Marit G, et al. Lenalidomide maintenance
after stem-cell transplantation for multiple myeloma: follow-up analysis of the IFM 2005-02 trial. Blood. 2013;122:406.
11. Dimopoulos MA, Weisel KC, Song KW, et al. Cytogenetics and long-term
survival of patients with refractory or relapsed and refractory multiple
myeloma treated with pomalidomide and low-dose dexamethasone.
Haematologica. 2015;100:1327-1333.
12. Leleu X, Karlin L, Macro M, et al; Intergroupe Francophone du Mye lome
(IFM). Pomalidomide plus low-dose dexamethasone in multiple myeloma with deletion 17p and/or translocation (4;14): IFM 2010-02 trial
results. Blood. 2015;125:1411-1417.
13. Barlogie B, Anaissie E, van Rhee F, et al. Incorporating bortezomib into
upfront treatment for multiple myeloma: early results of total therapy
3. Br J Haematol. 2007;138:176-185.
14. Jagannath S, Richardson PG, Sonneveld P, et al. Bortezomib appears to
overcome the poor prognosis conferred by chromosome 13 deletion in
phase 2 and 3 trials. Leukemia. 2007;21:151-157.
15. Avet-Loiseau H, Leleu X, Roussel M, et al. Bortezomib plus dexamethasone induction improves outcome of patients with t(4;14) myeloma but not outcome of patients with del(17p). J Clin Oncol. 2010;28:
4630-4634.
16. Cavo M, Tacchetti P, Patriarca F, et al; GIMEMA Italian Myeloma
Network. Bortezomib with thalidomide plus dexamethasone compared
with thalidomide plus dexamethasone as induction therapy before, and
consolidation therapy after, double autologous stem-cell transplantation
e439
NOOKA AND LONIAL
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
in newly diagnosed multiple myeloma: a randomised phase 3 study.

Lancet. 2010;376:2075-2085.
Sonneveld P, Schmidt-Wolf IG, van der Holt B, et al. Bortezomib induction and maintenance treatment in patients with newly diagnosed
multiple myeloma: results of the randomized phase III HOVON-65/
GMMG-HD4 trial. J Clin Oncol. 2012;30:2946-2955.
Sonneveld P, Salwender H-J, Van Der Holt B, et al. Bortezomib induction
and maintenance in patients with newly diagnosed multiple myeloma:
long-term follow-up of the HOVON-65/GMMG-HD4 trial. Blood. 2015;
126:27.
Nooka AK, Kaufman JL, Muppidi S, et al. Consolidation and maintenance
therapy with lenalidomide, bortezomib and dexamethasone (RVD) in
high-risk myeloma patients. Leukemia. 2014;28:690-693.
Jakubowiak AJ, Siegel DS, Martin T, et al. Treatment outcomes in patients with relapsed and refractory multiple myeloma and high-risk
cytogenetics receiving single-agent carfilzomib in the PX-171-003-A1
study. Leukemia. 2013;27:2351-2356.
Chng W-J, Goldschmidt H, Dimopoulos MA, et al. Efficacy and safety of
carfilzomib and dexamethasone vs bortezomib and dexamethasone in
patients with relapsed multiple myeloma based on cytogenetic risk
status: subgroup analysis from the phase 3 study endeavor
(NCT01568866). Blood. 2015;126:30.
Avet-Loiseau H, Fonseca R, Siegel D, et al. Efficacy and safety of carfilzomib, lenalidomide, and dexamethasone vs lenalidomide and
dexamethasone in patients with relapsed multiple myeloma based on
cytogenetic risk status: subgroup analysis from the phase 3 study aspire
(NCT01080391). Blood. 2015;126:731.
Shah J, Stadtmauer E, Abonour R, et al. Phase I/II dose expansion of a
multi-center trial of carfilzomib and pomalidomide with dexamethasone (Car-Pom-d) in patients with relapsed/refractory multiple myeloma. Paper presented at: 2013 American Society of Hematology
Annual Meeting; December 2013; New Orleans, LA.
Moreau P, Masszi T, Grzasko N, et al. Ixazomib, an investigational oral
proteasome inhibitor (PI), in combination with lenalidomide and
dexamethasone (IRd), significantly extends progression-free survival
(PFS) for patients (Pts) with relapsed and/or refractory multiple myeloma (RRMM): the phase 3 tourmaline-MM1 study (NCT01564537).
Blood. 2015;126:727.
Lonial S, Dimopoulos M, Palumbo A, et al; ELOQUENT-2 Investigators.
Elotuzumab therapy for relapsed or refractory multiple myeloma.
N Engl J Med. 2015;373:621-631.
Jakubowiak AJ, Benson DM, Bensinger W, et al. Phase I trial of anti-CS1
monoclonal antibody elotuzumab in combination with bortezomib in
the treatment of relapsed/refractory multiple myeloma. J Clin Oncol.
2012;30:1960-1965.
Palumbo A, Offidani M, Pegourie B, et al. Elotuzumab plus bortezomib
and dexamethasone versus bortezomib and dexamethasone in patients
with relapsed/refractory multiple myeloma: 2-year follow-up. Blood.
2015;126:510.
Lonial S, Weiss BM, Usmani SZ, et al. Daratumumab monotherapy in
patients with treatment-refractory multiple myeloma (SIRIUS): an
open-label, randomised, phase 2 trial. The Lancet.
Usmani S, Weiss B, Bahlis NJ, et al. Clinical efficacy of daratumumab
monotherapy in patients with heavily pretreated relapsed or refractory
multiple myeloma. Blood. 2015;126:29.
Plesner T, Arkenau H-T, Gimsing P, et al. Daratumumab in combination
with lenalidomide and dexamethasone in patients with relapsed or
relapsed and refractory multiple myeloma: updated results of a phase
1/2 Study (GEN503). Blood. 2015;126:507.
Chari A, Lonial S, Suvannasankha A, et al. Open-label, multicenter, phase
1b study of daratumumab in combination with pomalidomide and
dexamethasone in patients with at least 2 lines of prior therapy and
e440
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
relapsed or relapsed and refractory multiple myeloma. Blood. 2015;

126:508.
Martin TG, Baz R, Benson DM, et al. A phase Ib dose escalation trial of
SAR650984 (Anti-CD-38 mAb) in combination with lenalidomide and
dexamethasone in relapsed/refractory multiple myeloma. Blood. 2014;
124:83.
Ayed AO, Chang LJ, Moreb JS. Immunotherapy for multiple myeloma:
current status and future directions. Crit Rev Oncol Hematol. 2015;96:
399-412.
Garfall AL, Maus MV, Hwang W-T, et al. Chimeric antigen receptor
T cells against CD19 for multiple myeloma. N Engl J Med. 2015;373:
1040-1047.
Chekmasova AA, Horton HM, Garrett TE, et al. A novel and highly potent
CAR T cell drug product for treatment of BCMA-expressing hematological malignances. Blood. 2015;126:3094.
Wang M, Nooka AK, Yee AJ, et al. Initial results of a phase 1/2a, dose
escalation study of PVX-410 multi-peptide cancer vaccine in patients
with smoldering multiple myeloma (SMM). Blood. 2014;124:4737.
Carmon L, Avivi I, Riva K, et al. Phase I/II trial of immucin a pan HLA, antiMUC1 signal peptide vaccine, in multiple myeloma patients with
residual disease or progression following autologous stem cell transplantation. Blood. 2013;122:1943.
Brahmer JR, Tykodi SS, Chow LQM, et al. Safety and activity of anti-PDL1 antibody in patients with advanced cancer. N Engl J Med. 2012;366:
2455-2465.
Iwai Y, Ishida M, Tanaka Y, et al. Involvement of PD-L1 on tumor cells in
the escape from host immune system and tumor immunotherapy by
PD-L1 blockade. Proc Natl Acad Sci USA. 2002;99:12293-12297.
San Miguel J, Mateos M-V, Shah JJ, et al. Pembrolizumab in combination with lenalidomide and low-dose dexamethasone for relapsed/
refractory multiple myeloma (RRMM): keynote-023. Blood. 2015;
126:505.
Badros AZ, Kocoglu MH, Ma N, et al. A phase II Study of anti PD-1
antibody pembrolizumab, pomalidomide and dexamethasone in patients with relapsed/refractory multiple myeloma (RRMM). Blood.
2015;126:506.
Boise LH, Kaufman JL, Bahlis NJ, et al. The tao of myeloma. Blood. 2014;
124:1873-1879.
Lohr JG, Stojanov P, Carter SL, et al; Multiple Myeloma Research
Consortium. Widespread genetic heterogeneity in multiple myeloma:
implications for targeted therapy. Cancer Cell. 2014;25:91-101.
Chng WJ, Dispenzieri A, Chim CS, et al; International Myeloma Working
Leukemia. 2014;28:269-277.
Zhang B, Golding BT, Hardcastle IR. Small-molecule MDM2-p53 inhibitors: recent advances. Future Med Chem. 2015;7:631-645.
Rustad EH, Dai HY, Hov H, et al. BRAF V600E mutation in early-stage
multiple myeloma: good response to broad acting drugs and no relation
to prognosis. Blood Cancer J. 2015;5:e299.
Andrulis M, Lehners N, Capper D, et al. Targeting the BRAF V600E
mutation in multiple myeloma. Cancer Discov. 2013;3:862-869.
Bohn OL, Hsu K, Hyman DM, et al. BRAF V600E mutation and clonal
evolution in a patient with relapsed refractory myeloma with plasmablastic differentiation. Clin Lymphoma Myeloma Leuk. 2014;14:
e65-e68.
McCubrey JA, Steelman LS, Chappell WH, et al. Roles of the Raf/MEK/
ERK pathway in cell growth, malignant transformation and drug resistance. Biochim Biophys Acta. 2007;1773:1263-1284.
Liu P, Leong T, Quam L, et al. Activating mutations of N- and K-RAS in
multiple myeloma show different clinical associations: analysis of the
Eastern Cooperative Oncology Group phase III trial. Blood. 1996;88:
2699-2706.
51. Mulligan G, Lichter DI, Di Bacco A, et al. Mutation of NRAS but not KRAS
significantly reduces myeloma sensitivity to single-agent bortezomib
therapy. Blood. 2014;123:632-639.
52. Xie Z, Chng WJ. MMSET: role and therapeutic opportunities in multiple
myeloma. Biomed Res Intl. 2014;2014:636514.
53. Trudel S, Li ZH, Wei E, et al. CHIR-258, a novel, multitargeted tyrosine
kinase inhibitor for the potential treatment of t(4;14) multiple myeloma. Blood. 2005;105:2941-2948.
54. Trudel S, Stewart AK, Rom E, et al. The inhibitory anti-FGFR3 antibody,
PRO-001, is cytotoxic to t(4;14) multiple myeloma cells. Blood. 2006;
107:4039-4046.
55. Martinez-Garcia E, Popovic R, Min DJ, et al. The MMSET histone
methyl transferase switches global histone methylation and alters
gene expression in t(4;14) multiple myeloma cells. Blood. 2011;117:
211-220.
56. Chaidos A, Caputo V, Karadimitris A. Inhibition of bromodomain and
extra-terminal proteins (BET) as a potential therapeutic approach in
haematological malignancies: emerging preclinical and clinical evidence. Ther Adv Hematol. 2015;6:128-141.
57. Siegel MB, Liu SQ, Davare MA, et al. Small molecule inhibitor screen
identifies synergistic activity of the bromodomain inhibitor CPI203 and
bortezomib in drug resistant myeloma. Oncotarget. 2015;6:18921-18932.
58. Zhang X, Lu J, Qian Y, et al. Proteolytic targeting chimeric molecules
(PROTACs) specific for bromodomain-containing protein (BRD) 4 are
active against pre-clinical models of multiple myeloma. Blood. 2015;
126:917.
59. Mateos M-V, Martnez-Lopez J, Hernandez M-T, et al. Sequential versus
alternating administration of VMP and Rd in elderly patients with newly
diagnosed MM. Blood. 2016;127(4):420-425.
60. Shah JJ, Stadtmauer EA, Abonour R, et al. Carfilzomib, pomalidomide,
and dexamethasone for relapsed or refractory myeloma. Blood. 2015;
126:2284-2290.
61. Dosani T, Carlsten M, Maric I, et al. The cellular immune system in
myelomagenesis: NK cells and T cells in the development of myeloma
and their uses in immunotherapies. Blood Cancer J. 2015;5:e306.
62. Keudell Gv, Rosenbaum CA, Zimmerman TM, et al. Pilot study of
regulatory T cell depletion in the setting of autologous stem cell
transplantation for multiple myeloma. Blood. 2013;122:4607.
e441
SZALAT AND MUNSHI
Next-Generation Sequencing Informing Therapeutic Decisions

and Personalized Approaches
Raphael Szalat, MD, and Nikhil C. Munshi, MD
OVERVIEW
Multiple myeloma is a heterogeneous disease featured by different molecular subtypes. In the last decade, new therapeutics
including second- and third-generation proteasome inhibitors and immunomodulatory agents, monoclonal antibodies,
and other novel targeted agents have completely transformed the outcome of the disease. The task ahead is to develop
strategies to identify effective combinations and sequences of agents that can exploit the genetic make-up of myeloma cells
to improve efficacy. Moreover, a subgroup of high-risk patients who experience early disease relapse and shorter survival
also requires early identification and specific intervention. Next-generation sequencing (NGS) technologies now allow us to
accomplish some of these goals. As described here, besides improving our understanding of the disease, it is beginning to
influence our clinical decisions and therapeutic choices. In this article, we describe the current state-of-the-art role of NGS in
myeloma from identifying high-risk disease, to drug selection, and, ultimately, to guide personalized therapy.
ultiple myeloma is a clonal plasma cell disease that

evolves as a multistep process classically characterized
by a premalignant phase (monoclonal gammopathy of undetermined significance), a nonsymptomatic phase (smoldering multiple myeloma), and, eventually, a symptomatic
disease that will be subjected to remissions and relapses
over time. Multiple myeloma is biologically highly heterogeneous, defined by two groupshyperdiploid and nonhyperdiploid multiple myelomaaccording to karyotype
studies, and featured by 710 molecular subgroups based on
gene expression studies.1-3
The therapeutic landscape of multiple myeloma has been
transformed by the advent of new and very efficient drugs
including second and third generations of proteasome inhibitors as well as immunomodulatory agents and monoclonal antibodies. Such therapies have transformed patient
outcomes and quality of life.4-9 However, the majority of
patients experience relapse, and, for a subgroup of patients,
the prognosis remains poor with early relapse and short
survival. Risk stratification is crucial to identify these patients
and to develop and adapt treatment regimens accordingly.
The actual recommendations to assess multiple myeloma
risk at diagnosis are based on the combination of two biologic criteria determined by the International Staging
System (ISS; beta-2 microglobulin and albumin levels) and
three cytogenetic abnormalities (17p13 deletion, t(4;14),
and t(14;16)), usually identified by fluorescence in situ hybridization.10,11 These criteria allow identification of patients
who experience shorter survival at diagnosis. A classification

combining these characteristics has also been proposed.10
To date, there are no validated criteria that take into account
the chemosensitivity of the disease or status and changes at
relapse. Furthermore, these criteria do not take into account the heterogeneity existing within patients to define
high or low risk. As an example, based on cytogenetic
classification, a combination of high-risk features such as
17p13 deletion may coexist with trisomies of some of the
chromosomes, a good-risk feature. A recent study by the
Intergroupe Francophone du Myelome has now reported
how the combination of these features change the outcome.
If patients have trisomies of chromosome 3 and 5 along with
other poor risk cytogenetics, the overall outcome may improve; however, the presence of chromosome 21 trisomy
may further adversely affect the outcome.12,13 Conversely,
some patients not identified by the current stratification as
high-risk patients have an unpredicted short survival. With
the availability of additional novel therapeutic options, new
markers are now more important than ever in predicting not
only patient risk but also treatment selection and outcome
to provide the best available therapy in multiple myeloma.
Over last 15 years, various high-throughput technologies
have become available. DNA-based studies include wholegenome and whole-exome sequencing, array comparative
genomic hybridization, and high-density single nucleotide
polymorphism arrays. These techniques allow identification
of recurrent mutations and affected pathways, mutational
From the Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA; VA Boston Healthcare System, Boston, MA.
Corresponding author: Nikhil C. Munshi, MD, Dana-Farber Cancer Institute, 450 Brookline Ave., Boston, MA 02215; email: nikhil_munshi@dfci.harvard.edu.
e442
NEXT-GENERATION SEQUENCING
signatures, features of clonality and clonal evolution, and

copy number alterations. At the RNA level, gene expression
profiling of the malignant plasma cells has been widely
evaluated and reported as a promising method for prognostication in multiple myeloma.14-16 RNA sequencing is
now progressively replacing microarray by providing wider
information related to protein-coding RNA (gene expression, splicing and isoforms expression, fusions, mutations)
and to noncoding RNA (micro-RNA, long noncoding RNA,
small nucleolar RNA, small nuclear RNA, transfer RNA, ribosomal RNA, and extracellular RNA). The clinical significance of noncoding RNA, which constitutes a large volume
of RNA, is still being investigated. These and other technological advances including chromatin studies have been
explored and validated for their ability to risk stratify and
develop personalized effective therapy. In this article, we
describe the most recent and important findings generated
by NGS (Table 1) and how it may contribute to better
identify high-risk patients and to guide therapy in multiple
myeloma.
WHOLE-GENOME AND WHOLE-EXOME

SEQUENCING
Three independent large studies (total of 733 patients) have
evaluated the mutational profile of multiple myeloma by
whole-exome and/or whole-genome sequencing.17-19 These
studies have led to the discovery of the recurrent mutational
landscape in multiple myeloma. Although no universal driver
mutation has been discovered, KRAS, NRAS, FAM46C, DIS3,
BRAF, and TP53 have been identified as the most recurrent mutations. NRAS, KRAS, and BRAF, which are part of the
mitogen-activated protein kinase (MAPK) pathway, and
TP53 are also recurrently mutated in other malignancies.20
Along with the MAPK pathway, genes involved in the nuclear
factor (NF)-kappa B pathway are recurrently mutated in
up to 17% of multiple myeloma cases, suggesting that
KEY POINTS
Whole-genome and whole-exome sequencing have

identified recurrent mutations in multiple myeloma
including mutations associated with poor outcome
(ATM, ATR, TP53, and CCND1) and druggable target
mutations (BRAF, MAPK pathway, FGFR3).
Specific mutations, mutational load, copy number
abnormalities, and gene expression signatures are
reliable markers for prognostication in multiple
myeloma.
NGS data can be used to accurately evaluate response to
treatment characterizing minimal residual disease.
The molecular and genomic characteristics of a patient
continue to evolve and thus may require repeated
reassessment over disease course
New markers generated by NGS can be incorporated in
the clinical practice to elaborate a personalized therapy.
mutations in the MAPK and NF-kappa B pathway are most

probably drivers of myelomagenesis.19
Although most of the recurrent mutations do not appear
to correlate with survival, a global increased number of
mutations correlate with a higher risk of relapse and death.17
In one study, some less-recurrent mutations involving DNA
repair and apoptotic pathways have been reported to be
associated with poor survival. Indeed TP53, ATM, ATR, and
CCND1 mutations are significantly associated with poor
outcome.19 As described below, mutational information can
be used for various aspects of patient care in myeloma.
Besides prognostication, mutational profiling is beginning
to guide therapy and aid in development of personalized
medicine. The increasing number of targeted or nontargeted
therapies that are now available make this strategy particularly attractive. Thus, in a cohort of 133 patients with
relapsing myeloma, the presence of NRAS mutations evaluated by targeted sequencing was significantly associated
with poor response to bortezomib.21 Conversely, IRF4
mutations, if present, are associated with better outcome
with immunomodulatory agent therapy.19 MEK inhibitors
for MAPK pathway, ATR/ATM inhibitors, and CCND1 inhibitors are some examples of adapted strategies that
can now be considered for patients with these specific
mutations.
Study of NGS has confirmed that, at the time of diagnosis,
patients have a number of coexistent clones. Some mutations, especially those occurring early in the tumor development, are present in all cells and are called clonal
mutations. However, some mutations occur later and are
present in subset of cells; these are known as subclonal
mutations. The clonal and subclonal content changes and
evolves over time. In multiple myeloma, no clear data are
currently available to precisely understand the impact of
clonal composition on survival.17 When two or more samples
from the same patients are studied in myeloma, four distinct
patterns of clonal evolution are observed (linear, branching,
no change, and differential clonal response).17 This ability to
monitor the clonal evolution over time is also an important
source of information that can be exploited in clinic. Moreover, the study of types of mutations have now allowed
us to identify two signatures ascribed to mechanisms of
mutational processes.22 One of the two signatures belong to
the APOBEC family and is observed more often in relapsed
samples.17 These signatures should lead to new therapeutic
strategies that may reduce the genomic instability.
Targeted sequencing strategies focusing on a relatively
small number of genes are under evaluation.23,24 This has
included studies of frequently mutated groups of genes or in
one specific instance using deep sequencing of a single gene,
for example, IgH, to study minimal residual disease.25
RNA PROFILE
Gene expression profiling using microarrays has identified
different multiple myeloma subtypes.2,3,26-28 It has allowed
improved understanding of the myeloma biology, however,
it has not yet helped to select specific therapeutic agents
e443
SZALAT AND MUNSHI
TABLE 1. Genomic Methodologies and Their Practical Clinical Uses

Possible Results
Clinical Application
Comments
Mutations; clonality and clonal

evolution; chromosome
rearrangements
Identify mutations for prognostication This method provides extensive useful data but
and druggable targets; identify clonal
requires expertise to interpret data
content; Clonal evolution
DNA-Based Studies
Whole-Genome or
Whole-Exome
Sequencing
Targeted Sequencing Identifies specific mutations;

Identify specific mutations impacting
minimal residual disease evaluaoutcome and with potential for
tion (IgH deep sequencing)
therapeutic targeting; response
assessment for MRD
Most practical clinically applicable method; MRD

evaluation is becoming an important endpoint in
therapy and also for prognostication; useful for
personalized therapy in future
aCGH
SNP Array
Evaluation of copy number abnormalities; identification of amplifications and deletions; identify

specific SNPs
Correlate copy number changes to

Limited utilization as same information can be
prognosis; correlate specific SNPs to
obtained by WGS/WES
disease occurrence by GWAS studies
Methylation Array
Regulation of gene expression;

evaluation of promoter
methylation
May inform utilization of methylation

targeting agents
Currently not used to inform therapy; useful in

research; future potential for personalized
therapy as more agents targeting
methylation are developed
ChIP Sequencing
Histone and other epigenomic

DNA-binding modifications
impacting gene expression
Select epigenome targeting agents
Will allow selection of agents in myeloma with

specific epigenomic changes, however,
currently remains investigational; useful in
research to understand biology
RNA-Based Studies
Microarray-Based
Studies
Gene expression profile; micro-RNA GEP-based signatures allow risk

expression profile
stratification
Widely and commercially available; ease of utilization of the data; being phased-out by RNA
sequencing
RNA Sequencing
Gene expression profile; microIdentify high-risk disease; may inform

RNA; long noncoding RNA; alterselection of agents in future
nate splicing isoforms; mutations
This method provides broader information and is

more reliable for clinical trials; requires
expertise to interpret data
Abbreviations: aCGH, array comparative genomic hybridization; ChIP, chromatin immunoprecipitation; GEP, gene-expression profiling; GWAS, genome-wide association study; MRD,
minimal residual disease; SNP, single nucleotide polymorphism; WGS, whole-genome sequencing; WES, whole-exome sequencing
that could be predicted to be effective. Using gene expression profiling, various gene expression signatures have
been developed that could identify patients with high- and
low-risk disease. Three different signatures have been
described; a 70-gene signature by the Arkansas group, a
15-gene signature by the Intergroupe Francophone du
Myelome, and a 92-gene signature by the HOVON
group.14-16 All three signatures identify around 20% to 25%
of patients who are high risk with less than 2-year survival.
However, each of these signatures includes distinct genes
with almost no overlap between the signatures, making its
practical and universal use complicated. A recent study has
proposed a combined unique signature that may be used in
clinical practice in the future.29
Importantly, RNA sequencing has now replaced microarray to study gene expression. This method allows study of
coding and noncoding RNA, identifying not only gene expression levels, but also differential splicing and isoform
expression, mutational profiling, and gene fusions. RNA sequencing data from a large cohort of uniformly treated
patients have also identified that alternative splicing and
small noncoding RNA expression patterns are specific in
multiple myeloma compared with normal plasma cells. Interestingly, within the same level of expression of a given
gene, only some isoforms are associated with poor outcome,30 suggesting that post-translational regulation, such
as splicing, plays an important role in multiple myeloma
e444
biology and that any new gene expression studies will require integration of these data in the future.31,32
NONCODING RNA
RNA sequencing data also provide information on long
noncoding RNA, which can significantly affect cellular biology. Large RNA sequencing data have identified a unique
long noncoding RNA pattern in multiple myeloma and an
independent, significant correlation with poor survival.33
Similarly, micro-RNAs are noncoding RNAs that affect
transcriptome function. It has been reported as a marker
for prognosis in multiple myeloma in several studies.34-37
In addition to prognostication, micro-RNA are being investigated as therapeutic targets with potential for development of small molecules that affect micro-RNA
function.38-42
EPIGENOMIC MODIFICATIONS
DNA methylation and histone modifications that include
acetylation and methylation lead to modification of gene
expression and function. Histone methylation has been
studied, mostly in the t(4;14) of multiple myeloma cells that
overexpress histone methyl transferase and multiple myeloma SET domain (MMSET, NSD2, or WHSC1)regulating
gene expression43-45 and influence multiple myeloma cell
behavior.46,47 Although the impact of epigenomic changes
on outcome has not been studied in detail, it has already

provided various therapeutic targets including inhibitors of
histone deacetylase (HDACs) or enhancer of zeste homolog 2
(EZH2) inhibitors.48 Several studies have also demonstrated
that DNA methylation is differentially regulated in multiple myeloma cells compared with normal plasma cells,
with a pattern of global hypomethylation and focal
hypermethylation.49-51 However, therapies targeting DNA
methylation have not been effective in multiple myeloma.
CLINICAL APPLICATION OF GENOMIC DATA

Prognostic Marker
Next-generation sequencing data enlighten the biology of
myeloma and provide information on disease behavior to
inform prognosis and, particularly, to define low- and
high-risk disease. To date, several validated markers can be
integrated and combined with ISS and cytogenetic features
to accurately identify low- and high-risk multiple myeloma.
Array comparative genomic hybridization and high-density
single nucleotide polymorphism arrays have been particularly useful to study large cohorts of patients. Major
findings have been amp(1q23.3), amp(5q31.3), and
del(12p13.31), which carry poor prognosis. Now, the newer
gene and micro-RNA expression signatures, and specific
mutations and clonal shifts, constitute the basis of new
markers of prognosis. An important challenge is to
generate a unique tool that would combine all of these
validated data. Ideally, such a tool would use gene and
micro-RNA expression signatures, copy number alterations, chromosome rearrangements, and relevant mutations for their prognostic value and possibly to be
potentially targeted by specific therapies. Additional data
concerning noncoding RNA, the roles of clonality and clonal
evolution, the roles of specific mutations, and epigenomic
and post-translational regulation are also under evaluation
and should provide even more reliable markers in the near
future (Table 2).
The feasibility of using a single or limited number of

techniques to get the most from this information is under
development. As an example, DNA and RNA sequencing can
provide most of the information.52 Although currently used
in a limited number of institutions for research purposes
only, routine DNA sequencingand in some cases, targeted
sequencingis performed along with either array-based
gene expression profiling or RNA sequencing. As an integrated approach of analysis is developed and validated, a
more robust prognostic model will be available for clinical
application. Moreover, the NGS data have revealed that a
substantial number of new mutations evolve at the time of
relapse and suggest the need to re-examine genomic profiles
on more than one occasion during the course of the disease.
Inform Therapy
The ability to characterize the prognosis of patients with
multiple myeloma by NGS can be used to rationally design a
personalized therapeutic plan at diagnosis and at relapse.
First, risk stratification allows for identification of high-risk
patients who may benefit from more intense therapeutic
interventions. Currently, the main focus of intensification for
these patients is in consolidation and maintenance therapies. We suggest that patients identified as high risk at diagnosis should be treated with the most efficient and
available therapy as an induction regimen, including a
combination of immunomodulatory agents, proteasome
inhibitors, dexamethasone, and possibly a monoclonal antibody. The impact of autotransplant among high-risk patients may not be of the same extent as among patients
with low-risk disease; however, it remains an important consolidation therapy. This phase must be followed for high-risk
patients by consolidation with a more intense regimen and a
two-drug maintenance regimen.53
Attempts at identifying effective therapy using gene expression profiling have not been very successful. 54 The
TABLE 2. New Markers Generated by Next-Generation Sequencing Under Evaluation to Predict High-Risk Multiple
Myeloma
Marker
Method
Potential Impact
Mutational Signature WES/WGS
Distinct mutational processes may contribute differentially to tumor progression
Clonal and Subclonal WES/WGS

Evolution
Clonal content may have prognostic significance
Lnc-RNA Signature
RNA-seq
Lnc-RNA are involved in MM biology and may provide potential therapeutic targets
RNA Splicing
RNA-seq
Differentially expressed isoforms constitute a marker of prognosis
May provide future therapeutic targets

May predict treatment sensitivity
May require special consideration to judge response to therapy
Lnc-RNA signature may have prognostic significance
Spliced isoforms have differential biologic activity making them attractive therapeutic targets with increased
specificity
Epigenomic Profile
Microarray-ChIP Highly regulated region may correlate with patient outcome

Some epigenomic modifications have already been targeted in myeloma (panobinostat, HDAC inhibitor)
Abbreviations: ChIP, chromatin immunoprecipitation; lnc-RNA, long noncoding RNA; MM, multiple myeloma; RNA-seq, RNA sequencing; WGS, whole-genome sequencing; WES,
whole-exome sequencing.
e445
SZALAT AND MUNSHI
results suggest that a simple gene expression level is unable

to identify effective agents that can achieve complete
remission. The use of DNA sequencing seems particularly
promising to identify specific mutations that correlate
with a clinical response to therapy or those that are
druggable targets. A strategy that takes into account
mutations can be integrated into the patient assessment
plan. For example, the presence of NRAS mutations at
relapse is associated with low response to bortezomib,21
whereas the presence of IRF4 mutations confers higher
sensitivity to immunomodulatory agents.18 Currently, no
single agent that specifically targets a mutation is approved
for routine clinical use in multiple myeloma. However,
smaller studies have evaluated the efficacy of such targeted
agents. The presence of an activating BRAF mutation can be
specifically targeted using a BRAF inhibitor, and efficacy can
also be monitored using assays devised to detect the
presence of BRAF-mutated myeloma cells.55 In a single case
report, vemurafenib, which exclusively inhibits the V600Emutated BRAF, was effective for a patient with the BRAF
V600E mutation.56 Similarly, in a RAS-mutated, MAPK
pathwayactivated, relapsed, refractory myeloma, the
MEK inhibitor trametinib has been shown to achieve deep
response. A larger study of trametinib has also shown very
encouraging efficacy.57 Numbers of such druggable mutations are being recognized, each in a small subset of
patients, with the potential to add a targeted agent for
these patients. The large NCI MATCH trial, which is evaluating targeted strategies for a number of malignancies, is
expanding to include myeloma. In this distinctive trial
concept, patients tumor cells are initially sequenced to
identify unique mutations and for those druggable mutations, a specific inhibitor will be used to evaluate efficacy
across various malignancies. NCI MATCH and other similar
studies will usher in an era where therapy will be tailored to
molecular and genomic characteristics rather than just
phenotypic or pathologic diagnosis.
Mutations in SF3B1, FGFR3, ATM/ATR, IDH1/2, and CCND1
as well as RAS/RAF, NF-kappa B, and mTOR pathwayrelated
genes have been reported in myeloma. These mutations can
be targeted by appropriate inhibitors. To identify these
mutations, a targeted sequencing approach is being pursued.52,58 The study of the epigenome has also provided
useful information to select a specific therapy that would

target DNA methylation or histone methylation or acetylation (MMSET expression abnormality, HDAC expression
changes). The development of new histone deacetylase
inhibitors and new histone methylase inhibitors is particularly promising.9 However, some mutations may not have
any biologic impact because of low or no expression of the
mutated gene. We have observed that only a little over a
quarter of mutated genes are expressed, and the global
expression of mutated genes is lower than the expression of
nonmutated genes, with approximately only 25% expression. This low level of expression may be due to the level of
expression of the nonmutated allele or by the presence of
only a subclonal population featured by the mutation. Thus,
theoretically, the presence of mutation does not necessarily
mean that it has an important role and that we should
consider it as therapeutic target. This precision is important
for the interpretation of data in ongoing clinical studies.59
Mutational profiling ideally should be coupled with RNA
sequencing studies.
Important obstacles to these therapies must be considered as we develop personalized approaches (Table 3). The
major limitations may include a dynamic disease process
with a constantly evolving genome, the coexistence of
subclonal populations with only a subset of subclones that
harbor a mutation, and lack of expressed mutant alleles
making the presence of a mutation not a clinically important
target and rendering a specific targeted treatment inefficient. Moreover, other downstream activating mutations
in the same pathway may coexist and would make a targeted
therapy ineffective as well. In one study, a number of patients had coexistent KRAS and BRAF mutations.17 For these
patients, BRAF-directed therapy will not be effective. Despite these potential limitations, NGS is a promising tool
for targeted therapy or to identify a specific drug that
could be combined with current therapies including immunomodulatory agents, proteasome inhibitors, and monoclonal antibodies.
Response Assessment
Deep IgH sequencing has been used to evaluate minimal
residual disease in multiple myeloma. Next-generation sequencing has been recently reported to be especially useful
TABLE 3. Obstacles to Successful Therapeutic Application of Next-Generation Sequencing Data

Genomic Change
Clinical Impact
Possible Interventions to Overcome Obstacles
Subclonal Nature of Mutation
Limited clinical response despite effective

targeted therapy
Evaluate subclonal response
Lack or Limited Expression of the

Mutant Allele
Minimal to no effect of targeted agent
Must evaluate both mutation and expression
Presence of Mutation Downstream

of Targeted Mutation
Transient or no response
Prefer targeting downstream mutation from the beginning
Evolution of New Driver Mutation
Transient or no response
Resequence genome if response plateau or progression
Target multiple subclones, combination therapy

Select alternate target/s
Resequence genome if response plateau or progression
Direct therapy at new target or use broader multitarget agent
e446
to detect one cell among 1,000,000 in bone marrow, and the

absence of myeloma with such precision predicts superior
survival.25 This aspect of minimal residual disease is discussed in accompanying article. In regards to the evaluation
of agents targeting specific mutations, it is necessary to
consider clonal content; what proportion of myeloma cells
carry such a mutation? It is well described that some of the
driver mutations are subclonal and only present in small
number of cells. As we use specific inhibitors, this therapy
will potentially only impact the proportion of cells that carry
such mutations. Thus, the ability to detect a response at the
clonal level will be an important response criterion in the
future. For example, if 30% of cells contain a BRAF mutation
and BRAF-directed therapy is able to eliminate all of these
mutated cells, it will only remove 30% of the cells with

overall response, well below partial response criteria.
However, if we measure BRAF-mutated cells, we can
conclude a complete clonal response. Sensitive laboratory
methods are being developed to detect such responses to
consider their practical applications.
CONCLUSION
Next-generation sequencing data identify new markers
that accurately characterize prognosis, patient outcome,
and response to treatment allowing elaboration of personalized therapy. Incorporation of these markers into the
clinical practice should be evaluated and be made widely
available.
References
1. Morgan GJ, Walker BA, Davies FE. The genetic architecture of multiple
myeloma. Nat Rev Cancer. 2012;12:335-348.
2. Zhan F, Huang Y, Colla S, et al. The molecular classification of multiple
myeloma. Blood. 2006;108:2020-2028.
3. Broyl A, Hose D, Lokhorst H, et al. Gene expression profiling for molecular classification of multiple myeloma in newly diagnosed patients.
Blood. 2010;116:2543-2553.
4. Lokhorst HM, Plesner T, Laubach JP, et al. Targeting CD38 with daratumumab monotherapy in multiple myeloma. N Engl J Med. 2015;373:
1207-1219.
5. Stewart AK, Rajkumar SV, Dimopoulos MA, et al; ASPIRE Investigators.
Carfilzomib, lenalidomide, and dexamethasone for relapsed multiple
6. Attal M, Lauwers-Cances V, Marit G, et al; IFM Investigators. Lenalidomide maintenance after stem-cell transplantation for multiple myeloma. N Engl J Med. 2012;366:1782-1791.
7. Benboubker L, Dimopoulos MA, Dispenzieri A, et al; FIRST Trial Team.
Lenalidomide and dexamethasone in transplant-ineligible patients with
8. Lonial S, Dimopoulos M, Palumbo A, et al; ELOQUENT-2 Investigators.
Elotuzumab therapy for relapsed or refractory multiple myeloma.
N Engl J Med. 2015;373:621-631.
9. Mimura N, Hideshima T, Anderson KC. Novel therapeutic strategies for
multiple myeloma. Exp Hematol. 2015;43:732-741.
10. Chng WJ, Dispenzieri A, Chim CS, et al; International Myeloma Working
Leukemia. 2014;28:269-277.
11. Palumbo A, Avet-Loiseau H, Oliva S, et al. Revised International Staging
System for multiple myeloma: a report from International Myeloma
12. Moreau P, Attal M, Garban F, et al; SAKK; IFM Group. Heterogeneity
of t(4;14) in multiple myeloma. Long-term follow-up of 100 cases
treated with tandem transplantation in IFM99 trials. Leukemia. 2007;
21:2020-2024.
13. Kumar S, Fonseca R, Ketterling RP, et al. Trisomies in multiple myeloma:
impact on survival in patients with high-risk cytogenetics. Blood. 2012;
119:2100-2105.
14. Decaux O, Lode L, Magrangeas F, et al; Intergroupe Francophone du
Myelome. Prediction of survival in multiple myeloma based on gene
expression profiles reveals cell cycle and chromosomal instability signatures in high-risk patients and hyperdiploid signatures in low-risk
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
patients: a study of the Intergroupe Francophone du Myelome. J Clin

Oncol. 2008;26:4798-4805.
Kuiper R, Broyl A, de Knegt Y, et al. A gene expression signature for highrisk multiple myeloma. Leukemia. 2012;26:2406-2413.
Shaughnessy JD Jr, Zhan F, Burington BE, et al. A validated gene expression
model of high-risk multiple myeloma is defined by deregulated expression
of genes mapping to chromosome 1. Blood. 2007;109:2276-2284.
Bolli N, Avet-Loiseau H, Wedge DC, et al. Heterogeneity of genomic
evolution and mutational profiles in multiple myeloma. Nat Commun.
2014;5:2997.
Lohr JG, Stojanov P, Carter SL, et al; Multiple Myeloma Research
Consortium. Widespread genetic heterogeneity in multiple myeloma:
implications for targeted therapy. Cancer Cell. 2014;25:91-101.
Walker BA, Boyle EM, Wardell CP, et al. Mutational spectrum, copy
number changes, and outcome: results of a sequencing study of
patients with newly diagnosed myeloma. J Clin Oncol. 2015;33:
3911-3920.
Mulligan G, Lichter DI, Di Bacco A, et al. Mutation of NRAS but not KRAS
significantly reduces myeloma sensitivity to single-agent bortezomib
therapy. Blood. 2014;123:632-639.
Alexandrov LB, Nik-Zainal S, Wedge DC, et al; Australian Pancreatic
Cancer Genome Initiative; ICGC Breast Cancer Consortium; ICGC
MMML-Seq Consortium; ICGC PedBrain. Signatures of mutational
processes in human cancer. Nature. 2013;500:415-421.
Kortum KM, Langer C, Monge J, et al. Longitudinal analysis of 25
sequential sample-pairs using a custom multiple myeloma mutation
sequencing panel (M(3)P). Ann Hematol. 2015;94:1205-1211.
Kortum KM, Langer C, Monge J, et al. Targeted sequencing using a 47
gene multiple myeloma mutation panel (M(3) P) in -17p high risk
disease. Br J Haematol. 2015;168:507-510.
Avet-Loiseau H, Corre J, Lauwers-Cances V, et al. Evaluation of minimal
residual disease (MRD) by next generation sequencing (ngs) is highly
predictive of progression free survival in the IFM/DFCI 2009 trial. Blood.
2015;126:191.
Davies FE, Dring AM, Li C, et al. Insights into the multistep transformation of MGUS to myeloma using microarray expression analysis.
Blood. 2003;102:4504-4511.
Zhan F, Hardin J, Kordsmeier B, et al. Global gene expression profiling
of multiple myeloma, monoclonal gammopathy of undetermined
e447
SZALAT AND MUNSHI
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
significance, and normal bone marrow plasma cells. Blood. 2002;99:

1745-1757.
Mulligan G, Mitsiades C, Bryant B, et al. Gene expression profiling and
correlation with outcome in clinical trials of the proteasome inhibitor
bortezomib. Blood. 2007;109:3177-3188.
Chng WJ, Chung TH, Kumar S, et al. Gene signature combinations
improve prognostic stratification of multiple myeloma patients. Leukemia. Epub 2015 Dec 16.
Rashid N, Minvielle S, Magrangeas F, et al. Alternative splicing is a
frequent event and impacts clinical outcome in myeloma: a large RNAseq data analysis of newly-diagnosed myeloma patients. Presented at:
56th ASH Annual Meeting and Exposition; December 2014; San Francisco, CA; abstract 638.
Fulciniti M, Samur MK, Rashid NU, et al. Functional and clinical impact of
splicing factor dysregulation in multiple myeloma. Blood. 2015;126:726.
Cleynen A, Szalat R, Samur MK, et al. The fusion gene landscape in
multiple myeloma, with clinical impact. Blood. 2015;126:835.
Samur MK, Gulla A, Cleynen A, et al. Differentially expressed and
prognostically significant lincrnas may impact immune system and
tumor progression in multiple myeloma (MM). Blood. 2015;126:2989.
Seckinger A, Meiner T, Moreaux J, et al. miRNAs in multiple myeloma
a survival relevant complex regulator of gene expression. Oncotarget.
2015;6:39165-39183.
Wu P, Agnelli L, Walker BA, et al. Improved risk stratification in myeloma
using a microRNA-based classifier. Br J Haematol. 2013;162:348-359.
Zhou Y, Chen L, Barlogie B, et al. High-risk myeloma is associated with
global elevation of miRNAs and overexpression of EIF2C2/AGO2. Proc
Natl Acad Sci USA. 2010;107:7904-7909.
Lionetti M, Biasiolo M, Agnelli L, et al. Identification of microRNA expression patterns and definition of a microRNA/mRNA regulatory
network in distinct molecular groups of multiple myeloma. Blood. 2009;
114:e20-e26.
Gatt ME, Zhao JJ, Ebert MS, et al. MicroRNAs 15a/16-1 function as tumor
suppressor genes in multiple myeloma. Blood. Epub 2010 Oct 20.
Roccaro AM, Sacco A, Thompson B, et al. MicroRNAs 15a and 16
regulate tumor proliferation in multiple myeloma. Blood. 2009;113:
6669-6680.
Fulciniti M, Amodio N, Bandi RL, et al. miR-23b/SP1/c-myc forms a feedforward loop supporting multiple myeloma cell growth. Blood Cancer J.
2016;6:e380.
Min DJ, Ezponda T, Kim MK, et al. MMSET stimulates myeloma cell
growth through microRNA-mediated modulation of c-MYC. Leukemia.
2013;27:686-694.
Rio-Machin A, Ferreira BI, Henry T, et al. Downregulation of specific
miRNAs in hyperdiploid multiple myeloma mimics the oncogenic effect
of IgH translocations occurring in the non-hyperdiploid subtype. Leukemia. 2013;27:925-931.
e448
43. Pei H, Zhang L, Luo K, et al. MMSET regulates histone H4K20 methylation
and 53BP1 accumulation at DNA damage sites. Nature. 2011;470:
124-128.
44. Martinez-Garcia E, Popovic R, Min DJ, et al. The MMSET histone methyl
transferase switches global histone methylation and alters gene expression in t(4;14) multiple myeloma cells. Blood. 2011;117:211-220.
45. Kuo AJ, Cheung P, Chen K, et al. NSD2 links dimethylation of histone H3
at lysine 36 to oncogenic programming. Mol Cell. 2011;44:609-620.
46. Popovic R, Martinez-Garcia E, Giannopoulou EG, et al. Histone methyltransferase MMSET/NSD2 alters EZH2 binding and reprograms the
myeloma epigenome through global and focal changes in H3K36 and
H3K27 methylation. PLoS Genet. 2014;10:e1004566.
47. Marango J, Shimoyama M, Nishio H, et al. The MMSET protein is a
histone methyltransferase with characteristics of a transcriptional
corepressor. Blood. 2008;111:3145-3154.
48. Afifi S, Michael A, Azimi M, et al. Role of histone deacetylase inhibitors
in relapsed refractory multiple myeloma: a focus on vorinostat and
panobinostat. Pharmacotherapy. 2015;35:1173-1188.
49. Kaiser MF, Johnson DC, Wu P, et al. Global methylation analysis
identifies prognostically important epigenetically inactivated tumor
suppressor genes in multiple myeloma. Blood. 2013;122:219-226.
50. Walker BA, Wardell CP, Chiecchio L, et al. Aberrant global methylation
patterns affect the molecular pathogenesis and prognosis of multiple
myeloma. Blood. 2011;117:553-562.
51. Agirre X, Castellano G, Pascual M, et al. Whole-epigenome analysis in
multiple myeloma reveals DNA hypermethylation of B cell-specific
enhancers. Genome Res. 2015;25:478-487.
52. Bolli N, Avet-Loiseau H, Gimondi S, et al. Genomic landscape and its
prognostic implications in multiple myeloma using a targeted sequencing approach. Blood. 2015;126:370.
53. Lonial S, Boise LH, Kaufman J. How I treat high-risk myeloma. Blood.
2015;126:1536-1543.
54. Amin SB, Yip WK, Minvielle S, et al. Gene expression profile alone is
inadequate in predicting complete response in multiple myeloma.
Leukemia. 2014;28:2229-2234.
55. Raab MS, Lehners N, Xu J, et al. Spatially divergent clonal evolution in
multiple myeloma: overcoming resistance to BRAF inhibition. Blood.
Epub 2016 Feb 16.
56. Andrulis M, Lehners N, Capper D, et al. Targeting the BRAF V600E
mutation in multiple myeloma. Cancer Discov. 2013;3:862-869.
57. Heuck C, Jethava Y, Khan RZ, et al. Targeted MEK inhibition in patients
with previously treated multiple myeloma. Blood. 2014;124:4775.
58. Heuck C, Chavan SS, Stein CK, et al. Comprehensive genomic profiling of
multiple myeloma in the course of clinical care identifies targetable and
prognostically significant genomic alterations. Blood. 2015;126:369.
59. Rashid NU, Sperling AS, Bolli N, et al. Differential and limited expression
of mutant alleles in multiple myeloma. Blood. 2014;124:3110-3117.
LUNG CANCER
Immunotherapy: Beyond
AntiPD-1 and AntiPD-L1
Therapies
CHAIR
Scott J. Antonia, MD, PhD
Tampa, FL
SPEAKERS
Edmund Moon, MD
Hospital of the University of Pennsylvania
Philadelphia, PA
Johan F. Vansteenkiste, MD, PhD
University Hospital KU Leuven
Leuven, Belgium
ANTONIA, VANSTEENKISTE, AND MOON
Immunotherapy: Beyond AntiPD-1 and AntiPD-L1

Therapies
Scott J. Antonia, MD, PhD, Johan F. Vansteenkiste, MD, PhD, and Edmund Moon, MD
OVERVIEW
Advanced-stage nonsmall cell lung cancer (NSCLC) and small cell lung cancer are cancers in which chemotherapy produces a
survival benefit, although it is small. We now know that antiPD-1/PD-L1 has substantial clinical activity in both of these
diseases, with an overall response rate (ORR) of 15%20%. These responses are frequently rapid and durable, increase
median overall survival (OS) compared with chemotherapy, and produce long-term survivors. Despite these very significant
results, many patients do not benefit from antiPD-1/PD-L1. This is because of the potential for malignancies to co-opt
myriad immunosuppressive mechanisms other than aberrant expression of PD-L1. Conceptually, these can be divided into
three categories. First, for some patients there is likely a failure to generate sufficient functional tumor antigen-specific
T cells. Second, for others, tumor antigenspecific T cells may be generated but fail to enter into the tumor parenchyma.
Finally, there are a large number of immunosuppressive mechanisms that have the potential to be operational within the
tumor microenvironment: surface membrane immune checkpoint proteins PD-1, CTLA-4, LAG3, TIM3, BTLA, and adenosine
A2AR; soluble factors and metabolic alterations interleukin (IL)-10, transforming growth factor (TGF)-b, adenosine, IDO, and
arginase; and inhibitory cells, cancer-associated fibroblasts (CAFs), regulatory T cells, myeloid-derived suppressor cells
(MDSCs), and tumor-associated macrophages. In this article, we discuss three strategies to generate more tumor-reactive
T cells for patients: antiCTLA-4, therapeutic tumor vaccination, and adoptive cellular therapy, with T cells redirected to
tumor antigens using T-cell receptor (TCR) or chimeric antigen receptor (CAR) gene modification. We also review some of the
various strategies in development to thwart tumor microenvironment immunosuppressive mechanisms. Strategies to drive
more T cells into tumors remain a significant challenge.
STRATEGIES TO INCREASE THE NUMBER OF

TUMOR-REACTIVE T CELLS
AntiCTLA-4
CTLA-4 is an immune checkpoint protein. Unlike PD-1,
CTLA-4 for patients with cancer suppresses T cells within
the lymphoid compartment rather than within the tumor
microenvironment by limiting T-cell proliferation, and thus
preventing expansion of antitumor T-cell responses. Evidence
for this comes from a study demonstrating that among patients with melanoma who were treated with ipilimumab,
there was the appearance of tumor antigenspecific T cells
with specificities in the blood of clinically responsive patients
not present in pretreatment and not appearing in the blood
of clinically nonresponsive patients.1
AntiPD-1/PD-L1 in combination with antiCTLA-4. Anti
CTLA-4 (ipilimumab) has not been tested as a single agent
for patients with NSCLC, but has some activity in combination
with chemotherapy.2,3 In combination with antiPD-1/PD-L1,
the CTLA-4expanded T cells are prevented from being

functionally inhibited when they enter into the tumor microenvironment. We recently reported the results of two
separate trials with these monoclonal antibodies, one with
nivolumab (antiPD-1) and ipilimumab (antiCTLA-4) and the
other with MEDI4736 (antiPD-L1) and tremelimumab
(antiCTLA-4). The initial doses and schedules of the nivolumab/
ipilimumab combination were toxic and produced a response
rate of 15%; however, more tolerable doses have now been
discovered.4 The durvalumab/tremelimumab combination
was associated with manageable toxicity and produced responses for patients who were PD-L1 biomarkernegative to a
similar degree as antiPD-1 monotherapy for patients who
were PD-L1 positive.5
Lung Cancer Vaccines

Melanoma-associated antigen 3 cancer vaccination and
MAGRIT. We start this section with the MAGRIT trial, as this
is the largest vaccination study, in a population that has long
From the Department of Thoracic Oncology Moffitt Cancer Center, Tampa, FL; Respiratory Oncology Unit, University Hospital KU Leuven, Leuven, Belgium; Hospital of the University of
Pennsylvania, Philadelphia, PA.
Corresponding author: Scott J. Antonia, MD, PhD, Moffitt Cancer Center, 12902 Magnolia Dr., MRC 3-East, Tampa, FL 33612; email: scott.antonia@moffitt.org.
e450
BEYOND ANTIPD-1 AND ANTIPD-L1 THERAPIES
been considered to be optimal candidates: patients with

minimal residual disease after definitive therapy. This is also
the largest therapeutic study overall in NSCLC, and the
program nicely shows how vaccines were developed in
NSCLC (Table 1).
The melanoma-associated antigen 3 (MAGE-A3) protein is
totally tumor-specific and present in approximately 35% of
early-stage NSCLC.6 In a mechanism-of-action study, it was
shown that MAGE-3 protein with adjuvant AS02B successfully
induced specific antibodies, CD8+, and CD4+ T cells (whereas
MAGE-3 protein alone did not).7 In the hypothesis-generating
phase II, double-blind, randomized, placebo-controlled trial,
182 patients with completely resected MAGE-A3positive stage
IB-II NSCLC received the MAGE-A3 cancer vaccine (recombinant
MAGE-A3 protein combined with an immunostimulant) or
placebo.8 No significant toxicity was observed, and a
potential predictive biomarker was suggested.9
The ensuing large phase III study, MAGRIT (MAGE-A3 as
Adjuvant NonSmall Cell Lung Cancer Immunotherapy) was
reported at the 2014 European Society for Medical Oncology
Congress (Table 2). Patients who were MAGE-A3 positive with
completely resected stage IB-II-IIIA NSCLC and adjuvant chemotherapy, as clinically indicated, were randomly assigned 2:1
to receive MAGE-A3 vaccine or placebo. Almost 14,000 surgical patients were screened; 4,210 patients were MAGE-A3
positive (33%) and 2,312 patients were randomly assigned.
The median disease-free survival (primary endpoint) was slightly
better with the MAGE-A3 cancer vaccine (60.5 vs. 57.9 months),
but the difference was unfortunately not significant (hazard
ratio [HR] 1.02; 95% CI, 0.891.18; p = .74). No subgroups with
potential benefit could be identified. The safety of this approach
as adjuvant therapy in NSCLC was confirmed, as there were
higher rates of mild injection-site reactions and flu-like symptoms than with placebo, but grade 3 and 4 treatment-related
symptoms and treatment-related signigicant adverse events
were infrequent. On the basis of this disappointing result,
further development of the MAGE-A3 vaccine in NSCLC has
been abandoned.
KEY POINTS
AntiPD-1/PD-L1 have clinical activity in lung cancer.

The majority of patients treated with single agent
antiPD-1/PD-L1 do not benefit.
Combination therapy with multiple immunotherapeutics
will be necessary to improve clinical efficacy.
Strategies are being developed to increase the number
of tumor antigen reactive T cells, to drive more of these
into the tumor parenchyma, and to block a variety of
immunosuppressive mechanisms in the tumor
microenvironment.
Given the heterogeneity regarding the specific ways that
individual tumors evade immune-mediated rejection,
relevant biomarkers will be necessary to guide therapy.
TABLE 1. Steps in the Clinical Development of Cancer

Vaccines
Phase
Objective
Questions
Preclinical
Expression
Broadly expressed?
Conserved in metastatic cells?
Specificity
Tumor specific or not?
Mechanism of
Action
Immunogenic
Effective humoral and cellular

response?
Phase I-II
Clinical effects
Activity?
Tolerability?
Predictive biomarker?
Phase III
Patient benefit
Outcome (survival)?
TolerabilityQuality of life effects?
Predictive biomarker confirmed?
Other recent phase III trials. The mucin MUC1 expression is

altered, mainly by aberrant glycosylation, in many cancer
types, including NSCLC. The tandem repeat MUC1-peptide
in a liposomal formulation is the BLP-25, or tecemotide,
vaccine. START (Stimulating Targeted Antigenic Responses
to NSCLC Trial) was a phase III, double-blind, randomized, placebo-controlled trial comparing maintenance
therapy with tecemotide (829 patients) or placebo (410
patients) for patients with unresectable stage III NSCLC
who did not progress after sequential or concurrent
chemoradiotherapy.10 The primary endpointoverall
survivalwas not significantly different between the vaccine and placebo group (25.6 vs. 22.3 months). However,
preplanned subgroup analysis showed that the patients
treated with concurrent chemoradiotherapy (829 patients)
had a 10.2-month improvement in OS (30.8 vs. 20.6 months;
adjusted HR 0.78; p = .016). The consequential trial was
START 2, a similar large randomized, placebo-controlled trial
for patients who completed concurrent chemoradiotherapy
for unresectable stage III NSCLC (NCT02049151). However,
this trial and further development of tecemotide was
abandoned after disappointing results of a smaller trial
among Japanese patients with stage III NSCLC and concurrent chemoradiotherapy.
TG4010 is a vaccine based on a recombinant viral vector
(attenuated strain of vaccinia virus) expressing both the
tumor-associated antigen MUC1 and IL-2. This vaccine is
explored in the phase IIB/III TIME trial (NCT01383148). This
double-blind, placebo-controlled trial evaluates standard
first-line chemotherapy with or without TG4010 for patients
with MUC1-positive stage IV NSCLC. In the phase IIB part,
the predictive value of activated natural killer cells (triplepositive activated lymphocytes [TrPAL], CD16+ CD56+ CD69+)
was evaluated on the basis of a progression-free survival (PFS)
endpoint.11 If patients with TrPAL lower than normal values
had a Bayesian probability of greater than 95% so that the
HR and PFS were less than 1, and if those with TrPAL values
greater than normal had a probability of greater than 80%
so that the HR for PFS was greater than 1, then the TrPAL
biomarker would be considered validated. The former was
e451
TABLE 2. Recent Randomized Trials of Cancer Vaccines

No.
Patients Study Population
Compound
Trial Design
MAGE-A3
Phase III RCT

(MAGRIT)5
2,270
Completely resected
MAGE-A3positive
stage IB-IIIA NSCLC
MAGE-A3 vaccine vs. Disease-free

Prospective validation
placebo
survival: HR 1.02; 95%
of predictive gene
CI, 0.891.18; p = .738
signature not
feasible
Tecemotide
Phase III RCT

(START)6
1,513
Stage III NSCLC after

chemoradiation
therapy
LBLP-25 vaccine vs.

placebo
TG4010
Phase II B RCT
222
(TIME;
NCT01383148)
Untreated patients with

stage IV MUC1positive NSCLC
Chemotherapy with TrPAL biomarker

TG4010 vaccine vs.
validation:
chemotherapy
achieved in part
plus placebo
Phase III RCT

800
(TIME;
NCT01383148)
Untreated patients with Chemotherapy plus

OS; phase III part
stage IV nonsquamous
TG4010 vaccine vs.
ongoing
MUC1-positive NSCLC
chemotherapy
plus placebo
Safety, response rate,

duration of response
Advanced NSCLC in
disease control after
standard therapy
Subgroup analysis;
OS for patients
with radiotherapy:
HR 0.61; 95% CI,
0.380.96; p = .032
Belagen-pumatucel-L Phase III RCT

(STOP)8
700
Treatment Arms
Primary Endpoint
OS: HR 0.88; 95% CI,

0.751.03; p = .123
Belagen pumatucel-L OS: HR 0.94; 95% CI,

vs. placebo
0.731.20; p = .594
Other Endpoints
Subgroup analysis;
concurrent
chemoradiation:
OS: HR 0.78; 95% CI,
0.640.95; p = .016
PFS: HR 0.74; 95% CI,
0.550.98
Abbreviations: MAGE-A3, melanoma-associated antigen 3; HR, hazard ratio; NSCLC, nonsmall cell lung cancer; OS, overall survival; LBLP, stimuvax; RCT, randomized, placebocontrolled trial; PFS, progression-free survival.
the case; the latter was not. Overall, median PFS was
5.9 months in the TG4010 group and 5.1 months in the
placebo group (HR 0.74; 95% CI, 0.550.98; p = .019). On
the basis of these findings, the phase III part of the trial
continues.
Belagenpumatucel-L is a vaccine based on a mixture of
allogeneic tumor cells with TGF-b2 antisense blockade as
adjuvant. It was studied in the phase III STOP (Survival:
Tumor-free, Overall, and Progression-free) trial for patients
with stage III-IV NSCLC achieving a response or disease control after first-line therapy.12 Median OS was
20.3 months with belagenpumatucel-L versus 17.8 months
with placebo (HR 0.94; p = .594). In the (small) subgroup of
patients treated with previous radiotherapy, the OS HR was
0.61 (p = .032).
Why were early positive findings not translated into patients benefits in phase III? Looking at MAGRIT, it was most
disappointing that the very well tolerated MAGE-A3 vaccine
could not enrich our adjuvant therapies in NSCLC, especially as no real progress in this setting has been seen for
more than a decade. Indeed, the IALT trial established
cisplatin-based chemotherapy as a standard for resected
stage II and IIIA tumors13 and was initially presented at the
2003 American Society of Clinical Oncology (ASCO) Annual
Meeting. Since then, we have witnessed failures of gefitinib,14
erlotinib,15 bevacizumab, and pharmacogenetically tailored
chemotherapy.16-18
MAGRIT was started on the basis of previously documented clear responses to the MAGE-A3 cancer vaccine in
metastatic melanoma,19 a phase II randomized study with an
HR for OS similar to modern adjuvant chemotherapy with an
e452
agent with almost no grade 3/4 toxicity,8 and a predictive

gene signature from metastatic melanoma reproducible in
early-stage NSCLC.9 Differences in population (stage IB-II-IIIA
vs. IB-II), diagnostic testing (MAGE-A3 reverse transcriptase
polymerase chain reaction on formalin-fixed, paraffinembedded vs. fresh, frozen tissue), and drug composition
(adjuvants AS15 vs. AS02B) could be ruled out as causes of
the negative MAGRIT results in comparison with the initial
phase II randomized trial. Overinterpretation of the existing
data, as is the case for many compounds, may have been in
place: factors for the preclinical data were that most of the
immune responses were antibodies or CD4+ T cells, whereas
it was more difficult to demonstrate CD8+ T cells. For the
phase II randomized data, the sample size was limited, with
possible hidden imbalances across treatment arms. Another
factor was raising the bar in outcome compared with historical data in the placebo arm, thus making it more difficult
to improve. Indeed, in the adjuvant landmark study IALT,13
the 5-year OS was 40.4% with surgery alone and 44.5% with
surgery followed by adjuvant chemotherapy. In MAGRIT, the
5-year OS in the placebo arm was 58.7%.
Where should we go with cancer vaccination in
NSCLC? Several recently reported phase III trials with vaccination approaches in NSCLC did not reach their primary
endpoint (MAGRIT, START, STOP). This essentially means
that immunologic response vaccines are not translated into
clinical benefits. Immunologic responses are measured in
the circulatory compartment, typically with antibody responses and less easily with specific lymphocyte responses,
and these may not be translated sufficiently in the peritumoral stromal immune system because of the strong
immunosuppressive environment on NSCLC. Indeed, findings from recent years gave deeper understanding of the
numerous mechanisms of immune suppression induced
by cancer cells: inhibitory cells such as regulatory T-lymphocytes
and MDSCs, inhibitory molecules such as TGF-b and IDO,
and immune checkpointinhibiting pathways such as those
related to CTLA-4 and the PD-1 receptor and its ligand
(PD-L1). 20
Consequently, combination therapies of agents that reverse this immunosuppressive environment of NSCLC (such
as the monoclonal antibodies that inhibit the PD-1/PD-L1
pathway) with vaccination may be the best way forward.
Both complement each other in the loop of immune response,
vaccination with delivery and enhanced presentation of
antigens, and checkpoint inhibitors with alleviation of the
immune-suppressive tumor microenvironment. It would be
particularly interesting to deliver neo-antigens of neoepitopes, as the latter have recently been correlated with
the response to antiPD-1/PD-L1 therapies.21 Moreover,
this mutational load of neo-antigens and neo-epitopes is
particularly high in NSCLC found in smokers, especially
squamous cell lung cancer. Such mutational epitopes are
immunogenic when presented by major histocompatibility
complex (MHC)receptors and have recently been associated
with increased patient survival.22
Engineering T Cells for Lung Cancer and Thoracic

Malignancies
Efforts to redirect T cells to tumor. The more traditional
strategy of adoptive T-cell transfer involved polyclonal
expansion of tumor-infiltrating lymphocytes (TILs) from
tumor resection specimens and reinfusion of expanded
populations of naturally occurring tumor-reactive TILs back
into the patient. Although this strategy induced responses
for patients with metastatic melanoma,23-25 identification
and ex vivo culturing of TILs from most other solid tumor
types is difficult. Over the last 2 decades, this challenge has
been overcome by genetically modifying otherwise tumornonreactive T cells to bear tumor reactivity. This involves
taking bulk CD8+ and CD4+ T cells via leukapharesis,
expanding them via coculture with artificial antigenpresenting cells or microbeads coated with anti-CD3/CD28
protein, and transducing them to express either a TCR clone
or a CAR, conferring specificity to a tumor-associated antigen (TAA). Transduction can be performed using viral systems (e.g., retrovirus and lentivirus) 26,27 and nonviral
systems (e.g., Sleeping Beauty transposon28 and mRNA
electroporation29).
T-cell receptor engineering involves the modification of
T cells to express high-affinity alpha/beta TCRs bearing
known specificity and avidity for TAAs (Fig. 1A).30 One major
advantage of TCR engineering is the ability to confer T-cell
reactivity against both surface and intracellular antigens.
One potential drawback is the potential generation of additional, unintended receptor specificities via rearranged
pairing of the a and b chains with those of the endogenous
TCRs. This could translate into increased toxicity. Another

limitation is that each TCR is specific for a given peptide-MHC
complex, thus making it applicable to patients who shared
both MHC alleles and TAAs.
Chimeric antigen receptor engineering artificially combines a single-chain Fv fragment fused from the variable light
and heavy chains of a TAA-specific antibody with the CD3zsignaling chain of the TCR.31 This so-called first-generation
CAR, which confers antigen-specific tumor lytic capability,
can be further modified to express one or more additional
costimulatory domains to increase T-cell persistence and
induce proliferation upon antigen-binding. CD28, 41BB,
and ICOS are a few examples of such domains that can be
included in these second- and third-generation CARs
(Fig. 1B).32 The advantages of CARs include highly specific
and avid recognition of the targeted antigen and MHC independent signaling initiated by antigen binding. However,
in general, CARs are only able to recognize antigens
expressed or presented on the cell membrane of target cells,
limiting the number of candidate TAAs that can be targeted
using CAR technology.
Tumor antigens as targets for genetically redirected
T cells. Both TCR and CAR engineering strategies to redirect
adoptively transferred T cells require choosing an optimal
TAA for targeting. Ideally, the targeted TAA is highly
expressed by tumor cells but not by vital normal tissues (or at
least expressed at low levels by normal tissues). Otherwise,
adoptively transferred T cells would be able to induce what
has been coined off-tumor, on-target toxicity. Tumorassociated antigens that have been targeted via engineered
T cells in lung cancer and/or mesothelioma include (1)
NYESO1 (NCT01697527, NCT01967823), a cancer testis antigen found in up to 30% of lung cancers;33 (2) VEGF receptor 2
(NCT01218867), expressed in the majority of lung cancer
samples;34 (3) MAGE-A3 (NCT02111850), found in up to
40% of lung cancers;6,35 (4) mesothelin (NCT02159716,
NCT01583686, NCT02414269),36,37 a molecule involved in
cell-cell adhesion but otherwise unclear in function,
expressed in virtually all mesothelioma samples of the epithelioid subtype and up to 80% of lung cancer samples;38,39
and (5) Wilm tumor protein 1 (NCT02408016), Wilm tumor
suppressor gene 1, which encodes a transcription factor
essential in the normal development of the urogenital
system. Refer to the publications by Cheever et al40 and
Hinrichs and Restifo 41 for a more comprehensive review
of TAA targets for T-cell immunotherapy in thoracic
malignancies.
Challenges to genetically redirecting T cells to thoracic
malignancies. Engineered T-cell therapy has demonstrated
dramatic and durable responses in patients with melanoma
(TCR engineering)42 and leukemia.43 However, applying this
immunotherapeutic strategy to many other solid tumors,
such as lung cancer and mesothelioma,44,45 has been met
with many challenges, limiting the objective responses
among patients. There have been no published reports of
e453
FIGURE 1. Chimeric Antigen Receptors
Abbreviation: TAA, tumor-associated antigen.
clinical responses to TCR-engineered T cells among patients

with lung cancer, and the limited data on the clinical experience of CAR-engineered T cells for patients with mesothelioma describe partial responses in only a small
e454
proportion of patients.36 The challenges inherent to solid

tumors have been described repeatedly for many years in
naturally occurring TILs.46-48 T-cell receptorengineered
T cells,49 and even CAR-engineered T cells, despite their
built-in costimulatory signaling motifs,50 succumb to similar

changes. These challenges include, but are not limited to,
heterogeneous TAA expression and TAA shedding, short
T-cell survival/persistence, suboptimal T-cell trafficking, the
barrier of tumor-associated stroma, the presence of suppressive immune cells, upregulation of inhibitory checkpoints, the expression of regulatory genes, lack of oxygen
and cellular nutrients, and immunosuppressive soluble
factors. Moon et al50,51 recently demonstrated that both
TCR- and CAR-engineered T cells undergo profound suppression of their tumor-lytic and cytokine secretion capabilities, leading to an inability to cure human lung cancer
and human mesothelioma xenograft tumors. When these
TCR/CAR TILs were isolated and allowed to rest away from
the tumor microenvironment, they recovered a significant
degree of their antitumor effector function. Preclinical
studies such as these are being conducted in parallel with the
early-phase trials of adoptive T-cell therapy for thoracic
malignancies, testing ways to overcome the much anticipated challenges in the solid tumor microenvironment. We
describe a few salient examples.
The challenge of T-cell survival/persistence. There is a
positive correlation between survival/persistence of genetically engineered T cells and clinical response.52 There are
a few potential strategies of optimizing T-cell survival/
persistence in the context of solid tumors.
Host preparative conditioning regimens (e.g., fludarabine
and/or cyclophosphamide)53,54 can be administered prior to
adoptive transfer of genetically engineered T cells to reduce
the number of circulating T cells (i.e., lymphodepletion). This
will promote the in vivo expansion of transferred T cells by
limiting the competition for cytokines necessary for T-cell
survival/persistence (e.g., IL-7, IL-15).55 The majority of
ongoing trials in thoracic malignancies (see above) have
conditioning regimens incorporated in their protocols.
Another strategy is further genetically modifying the
TCR/CAR-bearing T cells to secrete the cytokines necessary
for their survival and persistence.56 In some cases, the same
cytokines can promote the secondary influx of other host
immune cells, leading to additional antitumor/stroma effects.57 Optimizing the combination of costimulatory domains in CARs will also augment the cytokine secretion
profiles of transferred T cells, and the costimulatory motifs
being tested in CARs are constantly evolving.58,59
The challenge of tumor stroma. Stroma in both lung cancer
and mesothelioma consists of a multicellular matrix at the
invasion front of tumor and has a very crucial role in tumor
progression.45,60 Mesothelioma tumors have especially
pronounced stromal components compared with other solid
tumors.61 There is a complex network of cross-talk between
stromal cells and tumor cells via both cell-cell interactions
and soluble factors. Wang et al62 demonstrated the ability to
augment mesothelioma tumor control by combining CAR
T cells targeting CAFs and mesothelin-directed CAR T cells.
With the well-described observation that tumor progression
is associated with hyaluronan accumulation in stroma, there

are efforts to combine hyaluronidase administration with
engineered T cells to overcome the stromal challenge.63
The challenge of inhibitory checkpoints. Checkpoint
blockade has recently received a lot of attention as a promising immunotherapeutic strategy in lung cancer.64 The most
studied checkpoint at the present time is PD-1, a negative
regulator of the T-cell immune response that signals upon
binding its ligands, PD-L1 and PD-L2, which are often highly
expressed in solid tumors and their tumor microenvironments. However, the majority of patients treated with PD-1
checkpoint blockade demonstrate no objective clinical response. One likely reason for this is that many patients tumors lack sufficient infiltration of tumor-reactive TILs to be
unmasked by checkpoint blockade. Thus, combining adoptive
transfer of engineered tumor-reactive T cells with checkpoint
blockade may be the key to successfully applying immunotherapy to cancers such as lung cancer and mesothelioma.
Preclinical studies have already demonstrated the ability to
augment TCR and CAR human T-cell control of lung cancer
and mesothelioma tumors by combining them with PD-1
blockade.51
IMMUNOSUPPRESSION IN THE TUMOR

MICROENVIRONMENT
Immune Checkpoint Proteins
There are several T-cell surface receptors called immune
checkpoint proteins that deliver negative signals to T cells
when they engage their ligands expressed on tissue- or
antigen-presenting cells.65 This is a feedback mechanism
whereby immune responses are dampened when no longer
needed. This mechanism of control of T cells can be co-opted
by tumors to escape rejection by the immune system.66,67 A
hallmark of malignancy is an inflamed tumor microenvironment,68 and inflammatory cytokines such as g-interferon
can induce immune checkpoint ligand expression on tumor
cells.69 Therefore tumor antigenspecific T cells activated in
tumor-draining lymph nodes are shut down when they enter
into the tumor microenvironment.
AntiPD-1/PD-L1 antibodies have significant antitumor
activity in lung cancer. The discovery of aberrant expression of PD-L1 in tumors leads to the development of the
therapeutic strategy of monoclonal antibodies designed to
prevent PD-1 binding to PD-L1. There are a number of
different monoclonal antibodies specific to PD-1 or PD-L1
that are in clinical development for lung cancer, all having
similar tumor response rates of 15%20%.70 This demonstrates that lung cancer is an immunotherapeutically responsive disease, and this strategy can have a major impact
on OS and, more importantly, can influence the tail of the
survival curve.
Other immune checkpoint proteins that may be operational within the tumor microenvironment. It is known
that the extracellular tumor microenvironment contains
e455
high levels of adenosine as a consequence of anaerobic

glycolysis in hypoxic regions and preferential utilization of
aerobic glycolysis for energy metabolism in nonhypoxic
regions (the Warburg effect), producing a relative excess of
AMP; and tumor cell expression of the ectonucleotidase
CD73 that catabolyzes AMP to produce adenosine.71,72 It
had been known that adenosine produced within the
hypoxic microenvironment of inflamed tissue functions to
limit the exuberance of inflammatory responses to reduce
collateral damage of normal tissue by inflammatory cells and
cytokines.73,74 This is because of a direct inhibitory effect of
adenosine on T cells that express adenosine A2ARs, a T-cell
surface immune checkpoint protein,75 and leads to the
discovery that adenosine in the tumor microenvironment
interferes with antitumor immunity,76 suggesting that antagonism of the A2AR could be an effective cancer immunotherapeutic.77 Other immune checkpoint proteins that
can be expressed on T cells include BTLA, LAG3, and TIM3.
Antibodies that block the binding of these proteins to their
ligands are being developed and tested in clinical trials.
Immune Suppressor Cells

Cancer-associated fibroblasts. Tumors secrete a variety of
growth factors and cytokines, including TGF-b, plateletderived growth factor, basic fibroblast growth factor, IL-6,
and IL-1, to recruit and activate CAFs.4-8 Fearons group
first demonstrated that CAFs contribute significantly to
immunosuppression within the tumor microenvironment.78
Subsequently, it has been discovered that they do so
through a variety of mechanisms. Recent studies demonstrated that CAFs suppressed T-cell proliferation more
efficiently than normal fibroblasts and collaborate with
tumor cells to create an immunosuppressive tumor microenvironment in head and neck squamous cell carcinoma.10
Cancer-associated fibroblasts have been shown to protect
lung cancer cells from T-cellmediated destruction. Fibroblast activation protein (FAP)-apositive CAFs were able to
protect tumor cells from tumor necrosis factormediated
and interferon gmediated necrosis of tumor cells by T cells
and were also able to sequester T cells away from neoplastic
cells by expressing C-X-C motif chemokine ligand 12. When
FAP-apositive CAFs were depleted, immunologic detection
and destruction of tumor cells was restored, and interferon-g
or tumor necrosis factor were able to cause rapid hypoxic
necrosis.4,11,12 Inhibition of CXC receptor type 4 interacting
with C-X-C motif chemokine ligand 12 of CAFs was shown to
promote T-cell accumulation in the pancreatic tumor site
and synergize with antiPD-L1.13 Cancer-associated fibroblasts are also known to induce immune suppressor cell
types such as regulatory T cells and MDSCs, and immune
suppressor cells were reduced when CAFs were inhibited.14
Cancer-associated fibroblasts have also been shown to recruit macrophages into tumors and activate M2 phenotypes
through secretion of IL-4, IL-6, and IL-8.9 All of these varied
mechanisms help to support an immunosuppressive, tumorpromoting microenvironment.
e456
Recently, two drugs have been approved for the treatment

of idiopathic pulmonary fibrosis, pirfenidone and nintedanib.
The fact that the fibroblasts in this disease resemble
CAFse.g., both express FAPled us to develop clinical
trials where these drugs will be repurposed as cancer immunotherapeutics. Because CAFs are also known to support
the malignant phenotype, these drugs may have anticancer
effects beyond relief of immunosuppression.
Myeloid-derived suppressor cells. Inflammation associated with the tumor microenvironment plays an important
role in the development and progression of lung cancer. In
the context of an inflammatory response, myeloid cells are
the primary recruited effectors. In lung cancer, these cells
are activated macrophages, granulocytes, and MDSC.79
Production of reactive oxygen and nitrogen species is one of
the major characteristics of all activated myeloid cells. The
formation of the free radical peroxynitrite (PNT) is the main
result of interaction between superoxide and nitric oxide.
Peroxynitrite can react directly with cysteine, methionine,
and tryptophan. A substantial number of studies have
demonstrated high levels of nitrotyrosine in lung cancer.
Gabrilovich et al79 demonstrated that MDSC can induce
T-cell tolerance via production of PNT and nitration/
nitrosylation of TCR and CD8 molecules on the surface of
T cells. T-cell receptors lose their ability to bind specific
peptide/MHC complexes and kill tumor cells.80
Synthetic triterpenoids work through activation of the
transcription factor Nrf2 (nuclear factor [erythroid-derived]
like 2), which induces the up-regulation of an array of antioxidant molecules, such as glutathione, thioredoxin, catalase,
and superoxide dismutase.81 We are testing one such triterprenoid as an anticancer immunotherapeutic.
CONCLUSION
The discovery that blocking signaling through the T-cell
surface immune checkpoint protein PD-1 can produce a
very significant impact on clinical outcomes of patients with
lung cancer not only gives us effective drugs that work as
single agents, but perhaps more importantly has given
us the proof-of-principle that lung cancer is certainly an
immunotherapeutically responsive disease. With the discovery of the many ways that tumors can evade rejection by
the immune system and the development of strategies to
thwart these comes real hope that we will be able to significantly improve on results obtained with antiPD-1/PD-L1
blocking antibody monotherapy. It is clear that there is
considerable heterogeneity among patients with lung cancer
with respect to which of the numerous potential immunoevasive mechanisms is operational. This means that in all
likelihood we must interfere with several immunoevasive
mechanisms for individual patients and that we must develop good biomarkers to identify which of these mechanisms is operational among individual patients to choose
the proper combination of immunotherapeutics for each
individual patient.
References
1. Kvistborg P, Philips D, Kelderman S, et al. Anti-CTLA-4 therapy broadens
the melanoma-reactive CD8+ T cell response. Sci Transl Med. 2014;6:
254ra128.
2. Lynch TJ, Bondarenko I, Luft A, et al. Ipilimumab in combination with
paclitaxel and carboplatin as first-line treatment in stage IIIB/IV nonsmall-cell lung cancer: results from a randomized, double-blind,
multicenter phase II study. J Clin Oncol. 2012;30:2046-2054.
3. Reck M, Bondarenko I, Luft A, et al. Ipilimumab in combination with
paclitaxel and carboplatin as first-line therapy in extensive-diseasesmall-cell lung cancer: results from a randomized, double-blind,
multicenter phase 2 trial. Ann Oncol. 2013;24:75-83.
4. Antonia SJ, Gettinger SN, Chow LQ, et al. Nivolumab (anti-PD-1; BMS936558, ONO-4538) and ipilimumab in first-line NSCLC: interim phase I
results. J Clin Oncol. 2014;32 (suppl; abstr 8023).
5. Antonia S, Goldberg SB, Balmanoukian A, et al. Safety and antitumour
activity of durvalumab plus tremelimumab in non-small cell lung
cancer: a multicentre, phase 1b study. Lancet Oncol. Epub 2016 Feb 5.
6. Sienel W, Varwerk C, Linder A, et al. Melanoma associated antigen
(MAGE)-A3 expression in stages I and II non-small cell lung cancer: results
of a multi-center study. Eur J Cardiothorac Surg. 2004;25:131-134.
7. Atanackovic D, Altorki NK, Stockert E, et al. Vaccine-induced CD4+ T cell
responses to MAGE-3 protein in lung cancer patients. J Immunol. 2004;
172:3289-3296.
8. Vansteenkiste J, Zielinski M, Linder A, et al. Adjuvant MAGE-A3 immunotherapy in resected non-small-cell lung cancer: phase II randomized study results. J Clin Oncol. 2013;31:2396-2403.
9. Ulloa-Montoya F, Louahed J, Dizier B, et al. Predictive gene signature in
MAGE-A3 antigen-specific cancer immunotherapy. J Clin Oncol. 2013;
31:2388-2395.
10. Butts C, Socinski MA, Mitchell PL, et al; START trial team. Tecemotide
(L-BLP25) versus placebo after chemoradiotherapy for stage III nonsmall-cell lung cancer (START): a randomised, double-blind, phase 3
trial. Lancet Oncol 2014;15:59-68.
11. Quoix EA, Forget F, Papai-Szekely Z, et al. Results of the phase IIb part of
TIME study evaluating TG4010 immunotherapy in stage IV non-small
cell lung cancer (NSCLC) patients receiving first line chemotherapy. J Clin
Oncol. 2015;33 (suppl; abstr 3034).
12. Giaccone G, Bazhenova LA, Nemunaitis J, et al. A phase III study of
belagenpumatucel-L, an allogeneic tumour cell vaccine, as maintenance
therapy for non-small cell lung cancer. Eur J Cancer. 2015;51:2321-2329.
13. Arriagada R, Bergman B, Dunant A, et al. Cisplatin-based adjuvant
chemotherapy in patients with completely resected non-small-cell lung
14. Goss GD, Lorimer I, Tsao MS, et al. A phase III randomized, double-blind,
placebo-controlled trial of the epidermal growth factor receptor inhibitor
gefitinb in completely resected stage IB-IIIA non-small cell lung cancer
(NSCLC): NCIC CTG BR.19. J Clin Oncol. 2010;28:18s(suppl; abstr LBA7005).
15. Kelly K, Altorki NK, Eberhardt WE, et al. Adjuvant erlotinib versus placebo in
patients with stage IB-IIIA non-small-cell lung cancer (RADIANT): a randomized, double-blind, phase III trial. J Clin Oncol. 2015;33:4007-4014.
16. Friboulet L, Olaussen KA, Pignon JP, et al. ERCC1 isoform expression and DNA
repair in non-small-cell lung cancer. N Engl J Med. 2013;368:1101-1110.
17. Bepler G, Williams CC, Schell MJ, et al. Molecular analysis-directed,
international, phase III trial in patients with advanced non-small-cell
lung cancer. J Clin Oncol. 2013;31:(suppl; abstr 8001).
18. Massuti B, Cobo M, Rodriguez-Paniagua JM, et al. SLCG SLCGG. Randomized phase III trial of customized adjuvant chemotherapy (CT)
according BRCA-1 expression levels in patients with node positive
resected non-small cell lung cancer (NSCLS) SCAT: a Spanish Lung
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
Cancer Group trial (Eudract:2007-000067-15; NCTgov:00478699). J Clin

Kruit WH, van Ojik HH, Brichard VG, et al. Phase 1/2 study of subcutaneous and intradermal immunization with a recombinant MAGE-3
protein in patients with detectable metastatic melanoma. Int J Cancer.
2005;117:596-604.
Berraondo P, Umansky V, Melero I. Changing the tumor microenvironment: new strategies for immunotherapy. Cancer Res. 2012;72:
5159-5164.
Rizvi N, Shepherd FA, Antonia SJ, et al. First-line monotherapy with
nivolumab (anti-PD-1; BMS-936558, ONO-4538) in advanced non-small
cell lung cancer (NSCLC): safety, efficacy, and correlation of outcomes
with PD-L1 status. Int J Radiat Oncol Biol Phys. 2014;90S31 (suppl 5).
Brown SD, Warren RL, Gibb EA, et al. Neo-antigens predicted by tumor
genome meta-analysis correlate with increased patient survival.
Genome Res. 2014;24:743-750.
Dudley ME, Wunderlich JR, Yang JC, et al. Adoptive cell transfer therapy
following non-myeloablative but lymphodepleting chemotherapy for
the treatment of patients with refractory metastatic melanoma. J Clin
Oncol. 2005;23:2346-2357.
Dudley ME, Wunderlich JR, Robbins PF, et al. Cancer regression and
autoimmunity in patients after clonal repopulation with antitumor
lymphocytes. Science. 2002;298:850-854.
Rosenberg SA, Spiess P, Lafreniere R. A new approach to the adoptive
immunotherapy of cancer with tumor-infiltrating lymphocytes. Science.
1986;233:1318-1321.
Naldini L, Blomer U, Gallay P, et al. In vivo gene delivery and stable
transduction of nondividing cells by a lentiviral vector. Science. 1996;
272:263-267.
Sharma S, Cantwell M, Kipps TJ, et al. Efficient infection of a human
T-cell line and of human primary peripheral blood leukocytes with a
pseudotyped retrovirus vector. Proc Natl Acad Sci USA. 1996;93:1184211847.
Singh H, Huls H, Kebriaei P, et al. A new approach to gene therapy using
Sleeping Beauty to genetically modify clinical-grade T cells to target
CD19. Immunol Rev. 2014;257:181-190.
Zhao Y, Zheng Z, Cohen CJ, et al. High-efficiency transfection of primary
human and mouse T lymphocytes using RNA electroporation. Mol Ther.
2006;13:151-159.
Schumacher TN. T-cell-receptor gene therapy. Nat Rev Immunol. 2002;
2:512-519.
Gross G, Waks T, Eshhar Z. Expression of immunoglobulin-T-cell receptor chimeric molecules as functional receptors with antibody-type
specificity. Proc Natl Acad Sci USA. 1989;86:10024-10028.
Eshhar Z, Waks T, Bendavid A, et al. Functional expression of chimeric
receptor genes in human T cells. J Immunol Methods. 2001;248:67-76.
Nicholaou T, Ebert L, Davis ID, et al. Directions in the immune targeting
of cancer: lessons learned from the cancer-testis Ag NY-ESO-1. Immunol
Cell Biol. 2006;84:303-317.
Bonnesen B, Pappot H, Holmstav J, et al. Vascular endothelial growth
factor A and vascular endothelial growth factor receptor 2 expression in
non-small cell lung cancer patients: relation to prognosis. Lung Cancer.
2009;66:314-318.
Gure AO, Chua R, Williamson B, et al. Cancer-testis genes are coordinately expressed and are markers of poor outcome in non-small cell
lung cancer. Clin Cancer Res. 2005;11:8055-8062.
Beatty GL, Haas AR, Maus MV, et al. Mesothelin-specific chimeric
antigen receptor mRNA-engineered T cells induce anti-tumor activity in
solid malignancies. Cancer Immunol Res. 2014 2:112-120.
e457
37. Maus MV, Haas AR, Beatty GL, et al. T cells expressing chimeric antigen
receptors can cause anaphylaxis in humans. Cancer Immunol Res. 2013;
1:26-31.
38. Ho M, Bera TK, Willingham MC, et al. Mesothelin expression in human
lung cancer. Clin Cancer Res. 2007;13:1571-1575.
~ez NG. Application of mesothelin immunostaining in tumor di39. Ordo n
agnosis. Am J Surg Pathol. 2003;27:1418-1428.
40. Cheever MA, Allison JP, Ferris AS, et al. The prioritization of cancer
antigens: a national cancer institute pilot project for the acceleration of
translational research. Clin Cancer Res. 2009;15:5323-5337.
41. Hinrichs CS, Restifo NP. Reassessing target antigens for adoptive T-cell
therapy. Nat Biotechnol. 2013;31:999-1008.
42. Robbins PF, Kassim SH, Tran TL, et al. A pilot trial using lymphocytes
genetically engineered with an NY-ESO-1-reactive T-cell receptor: longterm follow-up and correlates with response. Clin Cancer Res. 2015;21:
1019-1027.
43. Maude SL, Frey N, Shaw PA, et al. Chimeric antigen receptor T cells for
sustained remissions in leukemia. N Engl J Med. 2014;371:1507-1517.
44. Chen Z, Fillmore CM, Hammerman PS, et al. Non-small-cell lung
cancers: a heterogeneous set of diseases. Nat Rev Cancer. 2014;14:
535-546.
45. Ujiie H, Kadota K, Nitadori JI, et al. The tumoral and stromal immune
microenvironment in malignant pleural mesothelioma: a comprehensive analysis reveals prognostic immune markers. OncoImmunology.
2015;4:e1009285.
46. Frey AB. Suppression of T cell responses in the tumor microenvironment. Vaccine. 2015;33:7393-7400.
47. Monu N, Frey AB. Suppression of proximal T cell receptor signaling and
lytic function in CD8+ tumor-infiltrating T cells. Cancer Res. 2007;67:
11447-11454.
48. Baitsch L, Fuertes-Marraco SA, Legat A, et al. The three main stumbling
blocks for anticancer T cells. Trends Immunol. 2012;33:364-372.
49. Kunert A, Straetemans T, Govers C, et al. TCR-engineered T cells meet
new challenges to treat solid tumors: choice of antigen, T cell fitness,
and sensitization of tumor milieu. Front Immunol. 2013;4:363.
50. Moon EK, Wang LC, Dolfi DV, et al. Multifactorial T-cell hypofunction
that is reversible can limit the efficacy of chimeric antigen receptortransduced human T cells in solid tumors. Clin Cancer Res. 2014;20:
4262-4273.
51. Moon EK, Ranganathan R, Eruslanov E, et al. Blockade of programmed
death 1 augments the ability of human T cells engineered to target
NY-ESO-1 to control tumor growth after adoptive transfer. Clin Cancer
Res. 2016;22:436-447.
52. Rosenberg SA, Yang JC, Sherry RM, et al. Durable complete responses in
heavily pretreated patients with metastatic melanoma using T-cell
transfer immunotherapy. Clin Cancer Res. 2011;17:4550-4557.
53. Klebanoff CA, Khong HT, Antony PA, et al. Sinks, suppressors and antigen presenters: how lymphodepletion enhances T cell-mediated
tumor immunotherapy. Trends Immunol. 2005;26:111-117.
54. Wrzesinski C, Paulos CM, Kaiser A, et al. Increased intensity
lymphodepletion enhances tumor treatment efficacy of adoptively
transferred tumor-specific T cells. J Immunother. 2010;33:1-7.
55. Gattinoni L, Finkelstein SE, Klebanoff CA, et al. Removal of homeostatic cytokine sinks by lymphodepletion enhances the efficacy of
adoptively transferred tumor-specific CD8+ T cells. J Exp Med. 2005;
202:907-912.
56. Pegram HJ, Lee JC, Hayman EG, et al. Tumor-targeted T cells modified to
secrete IL-12 eradicate systemic tumors without need for prior conditioning. Blood. 2012;119:4133-4141.
57. Chmielewski M, Hombach AA, Abken H. Of CARs and TRUCKs: chimeric
antigen receptor (CAR) T cells engineered with an inducible cytokine to
modulate the tumor stroma. Immunol Rev. 2014;257:83-90.
e458
58. Guedan S, Chen X, Madar A, et al. ICOS-based chimeric antigen receptors program bipolar TH17/TH1 cells. Blood. 2014;124:1070-1080.
59. Hombach AA, Abken H. Of chimeric antigen receptors and antibodies:
OX40 and 41BB costimulation sharpen up T cell-based immunotherapy
of cancer. Immunotherapy 2013;5:677-681.
60. Bremnes RM, Donnem T, Al-Saad S, et al. The role of tumor stroma in
cancer progression and prognosis: emphasis on carcinoma-associated fibroblasts and non-small cell lung cancer. J Thorac Oncol. 2011;6:209-217.
61. Shia J, Qin J, Erlandson RA, et al. Malignant mesothelioma with a
pronounced myxoid stroma: a clinical and pathological evaluation of 19
cases. Virchows Arch. 2005;447:828-834.
62. Wang LC, Lo A, Scholler J, et al. Targeting fibroblast activation protein in
tumor stroma with chimeric antigen receptor T cells can inhibit tumor
growth and augment host immunity without severe toxicity. Cancer
Immunol Res. 2014;2:154-166.
63. Shepard HM. Breaching the castle walls: hyaluronan depletion as a
therapeutic approach to cancer therapy. Front Oncol. 2015;5:192.
64. Topalian SL, Hodi FS, Brahmer JR, et al. Safety, activity, and immune correlates of anti-PD-1 antibody in cancer. N Engl J Med. 2012;366:2443-2454.
65. Yao S, Zhu Y, Chen L. Advances in targeting cell surface signalling molecules for immune modulation. Nat Rev Drug Discov. 2013;12:130-146.
66. Nguyen LT, Ohashi PS. Clinical blockade of PD1 and LAG3potential
mechanisms of action. Nat Rev Immunol. 2015;15:45-56.
68. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation.
Cell. 2011;144:646-674.
69. Lee SJ, Jang BC, Lee SW, et al. Interferon regulatory factor-1 is prerequisite to the constitutive expression and IFN-gamma-induced
upregulation of B7-H1 (CD274). FEBS Lett. 2006;580:755-762.
70. Philips GK, Atkins M. Therapeutic uses of anti-PD-1 and anti-PD-L1
antibodies. Int Immunol. 2015;27:39-46.
71. Jin D, Fan J, Wang L, et al. CD73 on tumor cells impairs antitumor T-cell
responses: a novel mechanism of tumor-induced immune suppression.
Cancer Res. 2010;70:2245-2255.
72. Stagg J, Smyth MJ. Extracellular adenosine triphosphate and adenosine
in cancer. Oncogene 2010;29:5346-5358.
73. Di Paola R, Melani A, Esposito E, et al. Adenosine A2A receptor-selective
stimulation reduces signaling pathways involved in the development of
intestine ischemia and reperfusion injury. Shock. 2010;33:541-551.
74. Ohta A, Sitkovsky M. Role of G-protein-coupled adenosine receptors in
downregulation of inflammation and protection from tissue damage.
Nature. 2001;414:916-920.
75. Zarek PE, Huang CT, Lutz ER, et al. A2A receptor signaling promotes
peripheral tolerance by inducing T-cell anergy and the generation of
adaptive regulatory T cells. Blood. 2008;111:251-259.
76. Ohta A, Gorelik E, Prasad SJ, et al. A2A adenosine receptor protects
tumors from antitumor T cells. Proc Natl Acad Sci USA. 2006;103:
13132-13137.
77. Young A, Mittal D, Stagg J, et al. Targeting cancer-derived adenosine:
new therapeutic approaches. Cancer Discov. 2014;4:879-888.
78. Kraman M, Bambrough PJ, Arnold JN, et al. Suppression of antitumor
immunity by stromal cells expressing fibroblast activation proteinalpha. Science. 2010;330:827-830.
79. Gabrilovich DI, Ostrand-Rosenberg S, Bronte V. Coordinated regulation
of myeloid cells by tumours. Nat Rev Immunol. 2012;12:253-268.
80. Nagaraj S, Gupta K, Pisarev V, et al. Altered recognition of antigen is a
mechanism of CD8+ T cell tolerance in cancer. Nat Med. 2007;13:
828-835.
81. Liby KT, Yore MM, Sporn MB. Triterpenoids and rexinoids as multifunctional agents for the prevention and treatment of cancer. Nat Rev
Cancer. 2007;7:357-369.
LUNG CANCER
Local Therapies in the

Management of Oligometastatic
and Metastatic NonSmall Cell
Lung Cancer
CHAIR
Puneeth Iyengar, MD, PhD
The University of Texas Southwestern Medical Center
Dallas, TX
SPEAKERS
Jessica S. Donington, MD
New York University School of Medicine
New York, NY
Robert D. Suh, MD
David Geffen School of Medicine at UCLA
Los Angeles, CA
IYENGAR ET AL
Local Therapy for Limited Metastatic NonSmall Cell Lung

Cancer: What Are the Options and Is There a Benefit?
Puneeth Iyengar, MD, PhD, Steven Lau, MD, PhD, Jessica S. Donington, MD, MSCR, and
Robert D. Suh, MD
OVERVIEW
Distant metastasis is common in nonsmall cell lung cancer (NSCLC) and typically associated with poor prognosis. Aggressive
local therapy including surgery and/or radiation for limited metastatic disease from colorectal cancer and sarcoma is
associated with survival benefit and has become part of the standard of care. In this article, we review the literature and
ongoing studies concerning surgery, radiation, and radiofrequency ablation for oligometastatic NSCLC.
onsmall cell lung cancer continues to represent disproportionately the number of patients with and the
number of patient deaths from cancer. Over 224,000 new
cases are expected in 2016,1 and the overall survival (OS) for
all these patients is anticipated to approach only 15%20%.
Hence, a need to offer better strategies to combat the
disease should be a priority. For early-stage disease, surgery
continues to represent the gold standard for treatment. In
medically inoperable patients, stereotactic ablative radiotherapy (SABR), also referred to as stereotactic body radiation therapy (SBRT), has become a more integral part of our
treatment armamentarium. For those patients at high risk
for SABR in this medically inoperable setting, potentially
because of previous irradiation, radiofrequency ablation
(RFA) has begun to take on a greater role. Despite relentless
advances in surveillance, imaging, and knowledge, the
number of patients at initial presentation with surgically
resectable local disease remains soberly small at under 30%,
compounded by the realization that many with resectable
disease are rendered medically inoperable by comorbid
disease.
Over recent decades, most importantly the last, evolutionary inertia has refined local control measures on all
fronts, surgical and nonsurgical, through improved technology and its application and patient selection. Although
standard surgical lobectomy has been and is currently the
gold standard for early-stage NSCLC,2 noninvasive and
minimally invasive nonsurgical measures for local control
SABR and image-guided tumor ablation (IGTA), respectively,
have been increasingly accepted and used for local control
and cure for both medically inoperable but even medical

operable lung cancer patients.
Unfortunately, most patients with NSCLC are still diagnosed with advanced, metastatic disease. With the development of targeted therapies and immunotherapies,
survival is improving. Yet, for a significant proportion of
patients with stage IV NSCLC, systemic agents offer limited
potential for durable control of disease. Historically, with
limited metastatic colorectal cancer and sarcomas, the use
of local therapies including surgery and/or radiation have
become part of the standard of care as a means of drastically
improving OS along with local control. It therefore begs the
question, as we try to identify more and more subsets of
patients with advanced NSCLC with favorable biology, if
there is a group of patients with limited metastatic NSCLC
who could benefit from local treatments in addition to their
systemic therapy regimens. If so, what would be the optimal
treatments, when should they be delivered, and what type
of benefit is realistic? Finally, it may be that local therapy in
limited metastatic NSCLC is not beneficial independent of
the favorable biology of this cohort. This information would
also be of vital importance to generate.
This article offers the current evidence for the use of local
therapiessurgery, radiation in the form of SABR or
hypofractionated regimens, and RFAas part of the overall
management for patients with limited metastatic NSCLC. For
SABR and RFA, there is additional information regarding
their general acceptance into more mainstream treatment,
starting with evidence of their role in early-stage disease.
As a note, unless considering the yet to be validated abscopal
From the Department of Radiation Oncology, The University of Texas Southwestern Medical Center, Dallas, TX; NYU School of Medicine, New York, NY; Department of Radiological
Sciences, Thoracic Imaging and Intervention, and Diagnostic Radiology Education, David Geffen School of Medicine at UCLA, Los Angeles, CA; Department of Radiation Oncology, Harold
C. Simmons Comprehensive Cancer Center, The University of Texas Southwestern Medical Center, Dallas, TX.
Corresponding author: Puneeth Iyengar, MD, PhD, UT Southwestern Medical Center, 5801 Forest Park Rd., MC 9183, Dallas, TX 75235-9183; email: puneeth.iyengar@utsouthwestern.
edu.
e460
LOCAL THERAPY FOR LIMITED METASTATIC NSCLC
response associated with immunotherapy, there is no accepted role for the use of local therapies in the treatment of
widely metastatic NSCLC unless deemed necessary for
palliation or other improvement in quality of life.
SURGERY FOR METASTATIC NSCLC

Historically, surgery for patients with stage IV cancer was
limited to palliation, but in the 1980s, series began to appear
that reported prolonged survival following complete resection of primary tumors and oligometastatic disease in
selected patients. It is now recognized that there is a subset
of patients in whom an isolated metastasis represents the
entire disease burden, and removal can confer considerable
survival prolongation.
Surgery for metastases from numerous primary locations,
including the lung, has increased over the past decade.
Metastasectomy for primary NSCLC has been found second
in incidence only to colon cancers.3 An analysis from the
National Inpatient Sample uncovered a 5.8% average annual
percent increase in resections of NSCLC metastases between
2000 and 2011.3 This increase is attributed to several factors,
including more efficacious and better tolerated chemotherapies and the introduction of targeted agents, which
have slowed the progression of metastatic spread and altered patterns of resistance. Simultaneously, there have
been improvements in surgical techniques, with increased
use of minimally invasive approaches, making resection
better tolerated and negating long interruption from systemic treatments.
The majority of patients considered for resection of
metastatic NSCLC fall into three categories: those with an
isolated metastasis to the brain, adrenal glands, or contralateral lung. Occasional patients with an isolated metastasis to other sites are considered, but evidence for
prolonged survival following local therapy is sparse. Patients
who have previously undergone complete resection of earlystage NSCLC and subsequently present with isolated oligometastasis after thorough metastatic survey and have
KEY POINTS
Many patients with NSCLC will be diagnosed with distant

metastasis, but a subset of patients with limited
metastatic disease may benefit from aggressive local
therapies.
Evidence supporting an aggressive approach to
oligometastatic NSCLC is limited but suggestive of
reasonable safety and efficacy.
The potential risks and benefits of aggressive local
therapies for an individual patient should be openly
discussed prior to initiating treatment.
Synchronous versus metachronous oligometastasis may
carry disparate prognosis to guide management.
Ongoing studies will clarify the roles of various local
therapies in oligometastatic NSCLC.
good performance status should be considered for curative

intent resection. Similarly, patients who present de novo with a
single metastasis after full staging should be evaluated for
curative resection of both sites. An ongoing question in those
who present with synchronous disease is which treatment
should be undertaken first, the systemic or localized therapies?
Isolated Brain Metastasis

Some of the oldest series on surgical treatment of stage IV
NSCLC relate to treatment of isolated brain metastasis.4,5 Up
to one-quarter of all patients with stage IV NSCLC harbor
brain metastasis. Adenocarcinomas are associated with
higher rates of brain metastasis,6-8 but in 10% of patients
with metastatic adenocarcinoma, the brain is the only site of
involvement.9 Aggressive curative intent treatment of the
primary and metastatic site is encouraged in those with good
performance status and in whom both sites are amenable to
complete resection or ablation. Curative intent local treatments can only be considered after a thorough search for
disease at other sites. Mediastinal lymph node involvement
portends poor prognosis,6-8,10,11 and therefore, invasive mediastinal staging is recommended prior to starting treatment.
Brain MRI is recommended in addition to PET/CT because
of increased sensitivity12 Multiple brain metastasis are not
an absolute contraindication to this aggressive treatment
approach, but most recommend three or fewer lesions.9
Treatment of the brain lesion can be by resection or radiosurgery ablation. Radiosurgery has the advantage of being
able to be performed in almost any location, including the
brain stem.13-15
Five-year survival following definitive treatment of isolated brain metastasis and primary NSCLC is 15% and not
significantly impacted by synchronous or metachronous
presentation. Results from several series with more than 20
patients are outlined in Table 1.4-8,10,16,17 Prognosis is improved in patients who are younger, female, have a lower tstage, and good performance status.6-8 Administration of
adjuvant whole-brain radiation therapy (WBRT) following
resection or radiosurgery is recommended. Randomized data
Table 1. Survival Outcomes for Patients With Isolated

Brain Metastasis Following Local Therapy of Primary
Tumor and Metastasis
Synchronous
Metachronous
Survival (%)
Author
7
No.
2 years
5 years
Survival (%)
No.
2 years
5 years
103
28
11
Girard et al8
51
42
N/R
Granone et al6
30
47
14
Bonnette et al
86
14
145
29
17
Nakagawa et al5
60
10
N/R
28
11
N/R
Flannery et al17
42
34
21
33
59
13
45
N/R
16
Wronski
et al
10
Furak et al
Abbreviation: N/R, not recorded.
e461
IYENGAR ET AL
on WBRT in this setting are limited, but the sole trial

demonstrated a notable decrease in brain recurrence.18
There are also no randomized data specifically addressing
the use of adjuvant chemotherapy following complete
resection of stage IV disease, but with strong evidence supporting adjuvant chemotherapy for completely resected stage
II and III,19 it is recommended for patients with completely
resected primary and brain metastasis.9
Isolated Adrenal Metastasis

In well-selected patients with isolated adrenal metastasis
from NSCLC, survival following complete resection is 25%.20-23
Similar to those with isolated brain metastasis, mediastinal
lymph node involvement portends a poor prognosis, so invasive staging is recommended.24 Histology and laterality
appear to have no impact on survival, and adjuvant chemotherapy is recommended. Synchronous and metachronous
tumors have similar long-term survival rates, despite younger
age distribution in patients with synchronous presentation.25
Operative mortality is extremely low in reported series, and the
majority of patients die of disease progression.
There are no prospective studies comparing surgical resection to radiation therapy for the treatment of oligometastatic disease. Indeed, nonoperative approaches may be
preferred for some patients with oligometastasis because of
risks of surgical morbidity and mortality as well as comorbid
conditions common in patients with NSCLC, which may increase
such surgical risks. SABR adapts the techniques of stereotactic
radiosurgery for intracranial disease to the delivery of highly
conformal radiation to extracranial targets, and SABR is increasingly used to treat oligometastasis.30
SABR Is Associated With Excellent Local Control and

Acceptable Toxicity
The appearance of bilateral NSCLC lesions with the same

histology is a staging challenge. In the absence of other
disease, it is difficult to distinguish bilateral primary tumors
from stage IVa disease. Analysis of mutational status and
genetic clonality difference are being investigated, but is not
clinically reliable at this time.26 The clinical judgment of an
experienced multimodality team is essential,9 and the criteria described by Martini and Melamed in 1975 remains
relevant.27 As with isolated brain and adrenal metastasis, an
exhaustive search for additional metastatic disease and
invasive mediastinal staging are recommended prior to
considering resection. Parenchymal sparing resections are
typically recommended when possible in this setting.
Phase I/II studies have described the use of SABR for oligometastatic disease in the lung, liver, spine, and multiple
sites and reported local control rates of 70%90% with grade
3 or greater toxicity rates of less than 10%.31-36 For example,
Rusthoven et al31 enrolled 38 patients with 63 lung metastases on a phase I/II protocol using SABR to 4860 Gy in
three fractions. Twenty-seven (71%) patients had been
previously treated with at least one systemic regimen. At a
median follow-up of 15.4 months, only one local failure was
observed. Importantly, only three (8%) patients experienced at
least 3 toxicity. Separately, Salama et al35 enrolled 61 patients
with 113 extracranial metastases on a dose-escalation protocol
using SABR to 2460 Gy in three fractions. Twenty-three (38%)
patients had received metastasis-directed therapy prior to
enrollment. The maximum tolerated dose was not reached.
At a median imaging follow-up of 15.0 months, 72 (64%)
lesions were controlled despite initially low radiation doses
on this protocol. At least grade 3 acute toxicity was observed in
only two (3%) patients, and at least 3 late toxicity was observed
in six (10%) patients. Furthermore, when patients with oligometastatic disease treated with ablative local therapies experience distant progression, these failures tend to be limited
rather than widespread and more amenable to further ablative
therapies.37
Summary
Patient Selection in Oligometastatic NSCLC
Overall, with more data being generated regarding local

treatment of oligometastatic NSCLC, the role of surgery
will be more properly delineated with the goal of improving progression-free survival (PFS) at the least and
ultimately OS.
Outcomes of NSCLC are worse than other cancers, and

whether an oligometastatic disease state exists in NSCLC is
debated. Ashworth et al38 performed a systematic review of
49 studies including 2,176 patients with one to five metastases from NSCLC who underwent surgery or radiation.
Most (82%) patients had controlled primary disease, and
60% of studies were limited to intracranial metastasis.
Median survival was 14.8 months, median time to progression was 12 months, and median 5-year OS was 23.3%.
Control of primary disease, N stage, and disease-free interval
of at least 6 to 12 months prior to diagnosis of oligometastasis were found to be prognostic on multivariable
analysis.
Ashworth et al39 subsequently performed an individual
patient data meta-analysis on 757 patients with 1 to 5
synchronous or metachronous metastases from NSCLC
who underwent surgery or radiation. Median survival was
M1a Disease
RADIATION FOR METASTATIC NSCLC

Patients with metastatic cancer are historically considered
incurable, and their treatment is palliative in nature. However,
reports of patients long-term survival after surgical resection
of metastasis began to surface in the 1930s.28 The state of
limited metastatic disease without widespread progression
was ultimately termed oligometastasis in 1995.29 Oligometastasis is now recognized as a unique clinical entity
in which aggressive, ablative therapies can result in longterm cure.
e462
26 months, median PFS was 11 months, and 5-year OS was

29.4%. Recursive partitioning analysis stratified patients
into low-risk (metachronous metastasis; 5-year survival
47.8%), intermediate-risk (synchronous metastasis, N0;
5-year survival 36.2%), and high-risk disease (synchronous metastasis, N+; 5-year survival 12.1%).
De Ruysscher et al prospectively evaluated aggressive
treatment with curative intent in 39 patients diagnosed with
NSCLC and one to three synchronous metastases.40 Primary
disease was treated with radiation, and metastatic disease
was treated with either surgery (24%) or radiation (76%).
Twenty-seven (69.2%) patients had thoracic nodal involvement, 34 (87.2%) patients had a single metastasis, and
17 (43.9%) patients had brain metastasis. Median survival
was 13.5 months, median PFS was 12.1 months, and 2-year
survival was 23.3%. No grade 3 toxicity or higher related to
treatment of oligometastatic disease was observed.
Thus, appropriate patient selection is warranted when planning for an aggressive approach using ablative local therapies.
Combined Therapy for Oligometastatic NSCLC

Although ablative local therapies may control known sites
of disease, most patients will ultimately progress. Novel
therapeutic approaches are therefore needed. One promising approach is the combination of SABR and systemic
therapies. Iyengar et al41 prospectively tested the use of
SABR and concurrent erlotinib in 24 patients with 52 extracranial metastases from NSCLC who had progressed after
at least one systemic regimen. SABR was delivered to all
active sites of disease. Median survival and PFS were 20.4
and 14.7 months, respectively. Only three local failures of 47
evaluable lesions were observed, and 10 patients progressed
at distant sites. Others have reported an abscopal effect on
unirradiated metastases from the combination of SABR and
immunotherapies.42,43
Summary
SABR for the treatment of oligometastatic NSCLC is effective
and well tolerated, and combined modality therapy is a
promising approach to improve outcomes. Two studies, one out
of MD Anderson Cancer Center (NCT01725165) and the second
out of UT Southwestern Medical Center (NCT02045446), are
using SABR and hypofractionated radiation in randomized
studies with systemic therapy to determine if local treatment
can improve PFS in patients with up to three and six sites of
limited metastatic disease, respectively. Other larger studies
with OS as an endpoint will hopefully best answer how beneficial radiation may be in the stage IV NSCLC setting.
RADIOFREQUENCY ABLATION FOR METASTATIC

NSCLC
For those with NSCLC undergoing RFA, to date the most
studied thermal ablation technique, the medical literature
has focused on the high-risk, or medically inoperable,
population. RFA, or IGTA, can be considered for those
patients with locally recurrent cancer (oligorecurrent)
after local or even systemic therapy or with synchronous

or metachronous lung cancer, those with limited or low
volume metastases particularly after a prolonged diseasefree interval (oligometastatic), and those in need of palliative
and now even salvage therapy. Although multiple methods
for safe and effective local therapy are necessary, IGTA for
NSCLC is particularly poised for success, given it provides
the best balance of comparable survival, superior cost,
and experiential use patients with high-risk disease and
other emerging patient populations, much of which was
unfathomable a decade prior.
IGTA Efficacy
In 2007, Simon et al44 first published 5-year long-term safety
and efficacy data of RFA in 71 patients with stage I primary
lung cancer. Overall survival was reported at 78% at 1 year,
57% at 2 years, 36% at 3 years, and 27% at 4 and 5 years. The
authors re-emphasized that patients with treated cancers
3 cm and smaller versus cancers larger than 3 cm demonstrated better OS at every annual time point: 83% versus
45% at 1 year, 64% versus 25% at 2 years, 57% versus 25% at
3 years, and 47% versus 25% at 4 and 5 years. In addition,
those patients with 3 cm and smaller tumors showed significantly longer median time to progression: 45 versus
12 months. Although 27% 5-year OS appears dismal at first
glance, the study was quite remarkable in achieving this
survival in the high-risk population, estimated up to 30%
according to the Surveillance, Epidemiology, and End Results
(SEER) database in 2005,45 most therapeutically neglected
because of their medical ineligibility before this time. Since
2007, at least three additional series in patients with highrisk disease have shown comparable or better 5-year OS:
24%, 25%, and 61%,46-48 the last being quite formidable
and a testament to long-term operator experience and the
benefits of lessons learned regarding patient and lesion
selection. In addition to single-center experience, the medical
literature includes the results of two multicenter trials. In
Lancet Oncology, the RAPTURE study demonstrated 48%
OS and 73% cancer-specific survival at 2 years for primary
lung cancer population of patients with high-risk disease.49
Moreover, pulmonary function tests did not show any significant decline in forced expiratory volume (FEV), FEV percentage
predicted, forced vital capacity (FVC), or FVC percentage
predicted, in any follow-up visits through 12 months,
compared with baseline values. The slight decrease in these
values in treated patients with lung cancer was not clinically significant and was consistent with the progression of
underlying chronic pulmonary disease in this subgroup.
More recently, the results from the American College of
Surgeons Oncology Group (ACOSOG) Z4033 Trial were
published.50 The OS rate was 86.3% at 1 year and 69.8% at
2 years. The local tumor recurrencefree rate was 68.9% at
1 year and 59.8% at 2 years and worse for tumors at least 2 cm.
Not surprisingly, patients with tumors smaller than 2 cm and a
performance status of 0 or 1 achieved a statistically significant
improved survival of 83 and 78%, respectively, at 2 years.
e463
IYENGAR ET AL
Although not as mature as a technological application for

lung malignancies compared with RFA, microwave ablation
(MWA) and cryoablation, or percutaneous cryotherapy
(PCT), have more recently published 5-year results in the
treatment of early lung cancer in the medically inoperable
patient population, both having demonstrated comparable,
if not improved, OS and cancer-specific survival and local
control rates to RFA at 5 years with favorable complication
profiles.51,52 Yang et al51 reported OS rates at 1, 2, 3, and 5
years after MWA were 89%, 63%, 43%, and 16%, respectively.
Tumors that were 3.5 cm or smaller were associated with
better survival than were tumors that were larger than 3.5 cm.
The local control rates at 1, 3, and 5 years after MWA were
96%, 64%, and 48%, respectively. In patients undergoing PCT
for primary lung cancer, Moore et al52 in 2015 reported the 5year OS rate at 67.8%, the cancer-specific survival rate at 5
years at 56.6%, and the 5-year PFS rate at 87.9%.
Argued Comparability
Over the years, RFA has been unfavorably compared with
sublobar resection and SABR. However, upon closer scrutiny,
one major overlooked factor is that those patients treated
with RFA were always the sickest, the patients receiving RFA
in most, if not all, studies demonstrating statistically significant
poorer FEV in 1 second53 and older age54 compared with those
undergoing sublobar resection. Moreover, when comparing
patient demographics from the ACOSOG z4033 and z4032 and
RTOG 0236 trials, those patients enrolled in the z4033 were
significantly older and held significantly lower diffusion
capacity.55 And finally, Kwan et al56 recently argued comparable overall and lung-cancer specific survival between
RFA and sublobar resection in patients in the SEER database
when propensity scores were used to match patient subgroups.
In many circles, stereotactic body radiotherapy has been
largely hailed as the silver bullet against early-stage lung
cancer, setting new lofty standards for 5-year OS and local
recurrence ranging between 54% and 65% and 4% and 14%,
respectively.57,58 Although certainly impressive, the results
emerging from these and other similar series have not been
devoid of criticism. Deeper analyses show that not all treated
lesions were pathologically confirmed, with many radiographically diagnosed, prevalence of more indolent tumors specifically
adenocarcinoma in situ, and the treated population included both patients with low-risk disease and those with
high-risk disease. In fact, when parsing out the high-risk
subgroup from the Onishi et al study in 2007,57 5-year OS
drops from the overall reported 65% to 35%, similar to
published 5-year RFA results. Unlike RFA, moreover, SABRs
higher potential for collateral injury to healthy peritumoral
lung may be detrimental in many patients with high-risk
disease with tenuous lung reserve.
Will Less Equivalent Local Control Be the Death of

Radiofrequency Ablation?
In ascending order, local control improves when comparing
RFA, MWA, cryoablation, SABR, and surgical lobectomy;
e464
however, despite enhanced local control, the rate of distant

disease remains highly variable between studies whether
comparing like or unlike therapies. Simply and most plausibly, this variability is the direct result of already microscopic disease at the time of treatment, in some surgical
series reported as high as 16% in 1-cm apical cancers.59 In
other words, even complete, or 100%, local control will have
already microscopic locoregional and distant disease. This is
further exacerbated in the high-risk population, which may
be ineligible for formal nodal staging through mediastinoscopy, left to complete reliance on medical imaging, specifically CT-PET.
Ultimately, the rate of local and locoregional progression
may not negatively impact OS. Specifically in 2009, Lanuti
et al60 looked at locoregional recurrence in patients with
stage I lung cancer initially treated with RFA, and despite
retreatment with RFA, radiotherapy, chemoradiotherapy,
and chemotherapy, the authors found no notable difference
in 5-year overall and disease-free survival in patients receiving RFA without recurrence versus those patients with
treated recurrences.
Other Patient Populations

Helping to define a new patient population afforded lung
ablation, Kodama et al61 in 2012 showed 55.7% 5-year OS in
those patients with recurrent and second primary lung
cancers after surgical resection. Ridge et al62 in 2014 found
improved local progression after thermal ablation with
synchronous and metachronous lung cancers compared
with ablation of first primaries.
RFA has been studied in the advanced disease patient
population. Yu et al63 found RFA a useful tool to bridge from
failure of one therapy to the initiation of another therapy in
patients receiving EGFR. In some patients with advanced
disease, Gu et al64 found that patients receiving cryoablation
and gefitinib demonstrated significantly better survival at 1
year than those patients receiving gefitinib alone. Similarly,
Lee et al54 observed that even stage III and IV lung cancers
treated with RFA and chemotherapy demonstrated significantly longer median OS than those treated with only
chemotherapy. In patients with localized recurrence after
initial local or systemic treatment failure, RFA, when added
to a predetermined treatment regimen, lengthened overall
and PFS.65,66
Cost Benefit
Although surgical resection showed OS superiority at 1, 2,
and 3 years, Alexander et al67 argued that this superiority
came at a significantly increased median cost per months
lived of 1.93 times from reimbursement rates from the study
state. When compared with other treatment modalities in
the setting of oligometastatic NSCLC, cryoablation appeared
the most cost-effective, even when added to the cost of best
supportive care or systemic regimens with an adjunctive
cost-effectiveness ratio of $49,008$87,074. In their study,
cryoablation was associated with very low morbidity and
local tumor recurrence rates for all anatomic sites and

possibly increased OS.68
Although prospective cost-efficacy studies comparing
IGTA and SBRT are nonexistent, with the exception of a
single study using Markov modeling and older RFA data,69
global Medicare reimbursement in Rhode Island for SABR
was 4.25 times that of RFA at $17,000 versus $4,000.70 If
the higher cost of SABR for better local control and arguably
better OS is justified remains to be seen in the high-risk
population, particularly given the microscopic locoregional
and distant disease rates and RFAs repeatability at lower
cost.
Summary
In summary, advances in thermal energy devices and delivery have led to improved control. As local control improves, however, gains in survival will be limited for NSCLC,
given the limitations of radiographic staging and microscopic
lymph nodal and distant disease at the time of therapy
including ablation, rendering the need for adjuvant control
in some situations. Additive local control even in those with
advanced cancer appears to offer survival benefits in some
patient subsets with recurrent or advanced disease. When
critically compared with surgical resection and SABR, IGTA is
an attractive option with an acceptable threshold for local
control balanced with risk and cost without detriment to
survival.
CONCLUSION
Most would agree that favorable biology is the primary driver of
prognosis in the setting of oligometastasis, and the true impact
of local interventions on prognosis is unclear. But in an era when
local treatments carry minimal morbidity and mortality, the lack
of clarity should not translate into a denial of intervention in
well-selected patients with oligometastatic disease.
Evidence in favor of aggressive local therapies in oligometastatic NSCLC is limited to retrospective series and
several single-arm prospective series reporting reasonable
safety and efficacy. In the absence of more robust data and
comparative effectiveness studies, clinical judgment and
provider experience must guide management of patients
with limited metastatic NSCLC. Optimal management of
oligometastatic NSCLC is likely to be influenced by patient
selection. Patients with a single metastasis and good performance status might best be served by surgical resection. In
contrast, patients with limited multiple metastases and those
with inoperable disease may be treated with SABR. Similarly,
patients with accessible, small, and inoperable lesions may be
candidates for RFA. Multidisciplinary discussion may be helpful
for guiding management on a case-by-case basis. Ideally, patients would be enrolled and treated on clinical trials including
ongoing protocols using hypofractionated radiation and systemic therapy. Ultimately, rigorous studies in well-selected
patients will be necessary to determine the optimal management of oligometastatic NSCLC.
References
2016;66:7-30.
2. Churchill ED, Sweet RH, Soutter L, et al. The surgical management of
carcinoma of the lung; a study of the cases treated at the Massachusetts General Hospital from 1930 to 1950. J Thorac Surg. 1950;20:
349-365.
3. Bartlett EK, Simmons KD, Wachtel H, et al. The rise in metastasectomy
across cancer types over the past decade. Cancer. 2015;121:747-757.
4. Wronski
M, Arbit E, Burt M, et al. Survival after surgical treatment of
brain metastases from lung cancer: a follow-up study of 231 patients
treated between 1976 and 1991. J Neurosurg. 1995;83:605-616.
5. Nakagawa H, Miyawaki Y, Fujita T, et al. Surgical treatment of brain
metastases of lung cancer: retrospective analysis of 89 cases. J Neurol
Neurosurg Psychiatry. 1994;57:950-956.
6. Granone P, Margaritora S, DAndrilli A, et al. Non-small cell lung cancer
with single brain metastasis: the role of surgical treatment. Eur J
Cardiothorac Surg. 2001;20:361-366.
7. Bonnette P, Puyo P, Gabriel C, et al; Groupe Thorax. Surgical management of non-small cell lung cancer with synchronous brain metastases. Chest. 2001;119:1469-1475.
8. Girard N, Cottin V, Tronc F, et al. Chemotherapy is the cornerstone of
the combined surgical treatment of lung cancer with synchronous brain
metastases. Lung Cancer. 2006;53:51-58.
9. Kozower BD, Larner JM, Detterbeck FC, et al. Special treatment issues in
non-small cell lung cancer: Diagnosis and management of lung cancer,
3rd ed: American College of Chest Physicians evidence-based clinical
practice guidelines. Chest 2013;143:e369S-e399S.
10. Furak J, Trojan I, Szoke T, et al. Lung cancer and its operable brain
metastasis: survival rate and staging problems. Ann Thorac Surg. 2005;
79:241-247; discussion 247.
11. Modi A, Vohra HA, Weeden DF. Does surgery for primary non-small cell
lung cancer and cerebral metastasis have any impact on survival? Interact Cardiovasc Thorac Surg. 2009;8:467-473.
12. Billing PS, Miller DL, Allen MS, et al. Surgical treatment of primary lung
cancer with synchronous brain metastases. J Thorac Cardiovasc Surg.
2001;122:548-553.
13. Flannery TW, Suntharalingam M, Regine WF, et al. Long-term survival in
patients with synchronous, solitary brain metastasis from non-small-cell lung
cancer treated with radiosurgery. Int J Radiat Oncol Biol Phys. 2008;72:19-23.
14. Fuentes R, Bonfill X, Exposito J. Surgery versus radiosurgery for patients
with a solitary brain metastasis from non-small cell lung cancer.
Cochrane Database Syst Rev. 2006;1:CD004840.
15. Mariya Y, Sekizawa G, Matsuoka Y, et al. Outcome of stereotactic
radiosurgery for patients with non-small cell lung cancer metastatic to
the brain. J Radiat Res (Tokyo). 2010;51:333-342.
16. Ambrogi MC, Lucchi M, Dini P, et al. Percutaneous radiofrequency
ablation of lung tumours: results in the mid-term. Eur J Cardiothorac
Surg. 2006;30:177-183.
17. Flannery TW, Suntharalingam M, Kwok Y, et al. Gamma knife stereotactic
radiosurgery for synchronous versus metachronous solitary brain metastases from non-small cell lung cancer. Lung Cancer. 2003;42:327-333.
18. Patchell RA, Tibbs PA, Regine WF, et al. Postoperative radiotherapy in
the treatment of single metastases to the brain: a randomized trial.
JAMA. 1998;280:1485-1489.
e465
IYENGAR ET AL
19. Pignon JP, Tribodet H, Scagliotti GV, et al; LACE Collaborative Group.
Lung adjuvant cisplatin evaluation: a pooled analysis by the LACE
Collaborative Group. J Clin Oncol. 2008;26:3552-3559.
20. Raz DJ, Lanuti M, Gaissert HC, Wright CD, Mathisen DJ, Wain JC.
Outcomes of patients with isolated adrenal metastasis from non-small
cell lung carcinoma. Ann Thorac Surg. 2011;92:1788-1792; discussion
1793.
21. Lucchi M, Dini P, Ambrogi MC, et al. Metachronous adrenal masses in
resected non-small cell lung cancer patients: therapeutic implications of
laparoscopic adrenalectomy. Eur J Cardiothorac Surg. 2005;27:753-756.
22. Strong VE, DAngelica M, Tang L, et al. Laparoscopic adrenalectomy for
isolated adrenal metastasis. Ann Surg Oncol. 2007;14:3392-3400.
23. Mercier O, Fadel E, de Perrot M, et al. Surgical treatment of solitary
adrenal metastasis from non-small cell lung cancer. J Thorac Cardiovasc
Surg. 2005;130:136-140.
24. Porte H, Siat J, Guibert B, et al. Resection of adrenal metastases from
non-small cell lung cancer: a multicenter study. Ann Thorac Surg. 2001;
71:981-985.
25. Tanvetyanon T, Robinson LA, Schell MJ, et al. Outcomes of adrenalectomy for isolated synchronous versus metachronous adrenal metastases in non-small-cell lung cancer: a systematic review and pooled
26. Wu CT, Lin MW, Hsieh MS, et al. New aspects of the clinicopathology
and genetic profile of metachronous multiple lung cancers. Ann Surg.
2014;259:1018-1024.
27. Martini N, Melamed MR. Multiple primary lung cancers. J Thorac
Cardiovasc Surg. 1975;70:606-612.
28. Barney J, Churchill E. Adenocarcinoma of the kidney with metastasis to
the lung cured by nephrectomy and lobectomy. J Urol. 1939;42:
269-276.
29. Hellman S, Weichselbaum RR. Oligometastases. J Clin Oncol. 1995;13:
8-10.
30. Pan H, Simpson DR, Mell LK, et al. A survey of stereotactic body radiotherapy use in the United States. Cancer. 2011;117:4566-4572.
31. Rusthoven KE, Kavanagh BD, Burri SH, et al. Multi-institutional phase I/II
trial of stereotactic body radiation therapy for lung metastases. J Clin
Oncol. 2009;27:1579-1584.
32. Rusthoven KE, Kavanagh BD, Cardenes H, et al. Multi-institutional phase
I/II trial of stereotactic body radiation therapy for liver metastases. J Clin
Oncol. 2009;27:1572-1578.
33. Lee MT, Kim JJ, Dinniwell R, et al. Phase I study of individualized stereotactic body radiotherapy of liver metastases. J Clin Oncol. 2009;27:
1585-1591.
34. Wang XS, Rhines LD, Shiu AS, et al. Stereotactic body radiation therapy
for management of spinal metastases in patients without spinal cord
compression: a phase 1-2 trial. Lancet Oncol. 2012;13:395-402.
35. Salama JK, Hasselle MD, Chmura SJ, et al. Stereotactic body radiotherapy for multisite extracranial oligometastases: final report of a dose
escalation trial in patients with 1 to 5 sites of metastatic disease. Cancer.
2012;118:2962-2970.
36. Milano MT, Katz AW, Zhang H, et al. Oligometastases treated with
stereotactic body radiotherapy: long-term follow-up of prospective
study. Int J Radiat Oncol Biol Phys. 2012;83:878-886.
37. Milano MT, Philip A, Okunieff P. Analysis of patients with oligometastases undergoing two or more curative-intent stereotactic radiotherapy courses. Int J Radiat Oncol Biol Phys. 2009;73:832-837.
38. Ashworth A, Rodrigues G, Boldt G, et al. Is there an oligometastatic state
in non-small cell lung cancer? A systematic review of the literature. Lung
Cancer. 2013;82:197-203.
39. Ashworth AB, Senan S, Palma DA, et al. An individual patient data
metaanalysis of outcomes and prognostic factors after treatment of
e466
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
oligometastatic non-small-cell lung cancer. Clin Lung Cancer. 2014;15:

346-355.
De Ruysscher D, Wanders R, van Baardwijk A, et al. Radical treatment of
non-small-cell lung cancer patients with synchronous oligometastases:
long-term results of a prospective phase II trial. J Thorac Oncol. 2012;7:
1547-1555.
Iyengar P, Kavanagh BD, Wardak Z, et al. Phase II trial of stereotactic
body radiation therapy combined with erlotinib for patients with
limited but progressive metastatic non-small-cell lung cancer. J Clin
Oncol. 2014;32:3824-3830.
Postow MA, Callahan MK, Barker CA, et al. Immunologic correlates of
the abscopal effect in a patient with melanoma. N Engl J Med. 2012;366:
925-931.
Golden EB, Demaria S, Schiff PB, et al. An abscopal response to radiation
and ipilimumab in a patient with metastatic non-small cell lung cancer.
Cancer Immunol Res. 2013;1:365-372.
Simon CJ, Dupuy DE, DiPetrillo TA, et al. Pulmonary radiofrequency
ablation: long-term safety and efficacy in 153 patients. Radiology. 2007;
243:268-275.
Mery CM, Pappas AN, Bueno R, et al. Similar long-term survival of
elderly patients with non-small cell lung cancer treated with lobectomy
or wedge resection within the surveillance, epidemiology, and end
results database. Chest. 2005;128:237-245.
Huang L, Han Y, Zhao J, et al. Is radiofrequency thermal ablation a safe
and effective procedure in the treatment of pulmonary malignancies?
Eur J Cardiothoracic Surg. 2011;39:348-351.
Ambrogi MC, Fanucchi O, Cioni R, et al. Long-term results of radiofrequency ablation treatment of stage I non-small cell lung cancer:
a prospective intention-to-treat study. J Thorac Oncol. 2011;6:
2044-2051.
Hiraki T, Gobara H, Mimura H, et al. Percutaneous radiofrequency
ablation of clinical stage I non-small cell lung cancer. J Thorac Cardiovasc
Surg. 2011;142:24-30.
Lencioni R, Crocetti L, Cioni R, et al. Response to radiofrequency ablation
of pulmonary tumours: a prospective, intention-to-treat, multicentre
clinical trial (the RAPTURE study). Lancet Oncol. 2008;9:621-628.
Dupuy DE, Fernando HC, Hillman S, et al. Radiofrequency ablation of
stage IA non-small cell lung cancer in medically inoperable patients:
Results from the American College of Surgeons Oncology Group Z4033
(Alliance) trial. Cancer. 2015;121:3491-3498.
Yang X, Ye X, Zheng A, et al. Percutaneous microwave ablation of stage I
medically inoperable non-small cell lung cancer: clinical evaluation of 47
cases. J Surg Oncol. 2014;110:758-763.
Moore W, Talati R, Bhattacharji P, et al. Five-year survival after cryoablation of stage I non-small cell lung cancer in medically inoperable
patients. J Vasc Interv Radiol. 2015;26:312-319.
Kim SR, Han HJ, Park SJ, et al. Comparison between surgery and
radiofrequency ablation for stage I non-small cell lung cancer. Eur J
Radiol. 2012;81:395-399.
Lee H, Jin GY, Han YM, et al. Comparison of survival rate in primary nonsmall-cell lung cancer among elderly patients treated with radiofrequency ablation, surgery, or chemotherapy. Cardiovasc Intervent
Radiol. 2012;35:343-350.
Crabtree T, Puri V, Timmerman R, et al. Treatment of stage I lung cancer
in high-risk and inoperable patients: comparison of prospective clinical
trials using stereotactic body radiotherapy (RTOG 0236), sublobar resection (ACOSOG Z4032), and radiofrequency ablation (ACOSOG
Z4033). J Thorac Cardiovasc Surg. 2013;145:692-699.
Kwan SW, Mortell KE, Talenfeld AD, Brunner MC. Thermal ablation
matches sublobar resection outcomes in older patients with early-stage
non-small cell lung cancer. J Vasc Interv Radiol. 2014;25:1-9.e1.
57. Onishi H, Shirato H, Nagata Y, et al. Hypofractionated stereotactic

radiotherapy (HypoFXSRT) for stage I non-small cell lung cancer:
updated results of 257 patients in a Japanese multi-institutional study.
J Thorac Oncol. 2007;2:S94-S100.
58. Uematsu M, Fukui T, Tahara K, et al. Long-term results of computed
tomography guided hypofractionated stereotactic radiotherapy for
stage I non-small cell lung cancers. Int J Radiat Oncol Biol Phys. 2008;72:
S37.
59. Ketchedjian A, Daly BD, Fernando HC, et al. Location as an important
predictor of lymph node involvement for pulmonary adenocarcinoma.
J Thorac Cardiovasc Surg. 2006;132:544-548.
60. Lanuti M, Sharma A, Digumarthy SR, et al. Radiofrequency ablation for
treatment of medically inoperable stage I non-small cell lung cancer.
J Thorac Cardiovasc Surg. 2009;137:160-166.
61. Kodama H, Yamakado K, Takaki H, et al. Lung radiofrequency ablation
for the treatment of unresectable recurrent non-small-cell lung cancer
after surgical intervention. Cardiovasc Intervent Radiol. 2012;35:
563-569.
62. Ridge CA, Silk M, Petre EN, et al. Radiofrequency ablation of T1 lung
carcinoma: comparison of outcomes for first primary, metachronous,
and synchronous lung tumors. J Vasc Interv Radiol. 2014;25:989-996.
63. Yu HA, Sima CS, Huang J, et al. Local therapy with continued EGFR
tyrosine kinase inhibitor therapy as a treatment strategy in EGFRmutant advanced lung cancers that have developed acquired resistance to EGFR tyrosine kinase inhibitors. J Thorac Oncol. 2013;8:
346-351.
64. Gu XY, Jiang Z, Fang W. Cryoablation combined with molecular target

therapy improves the curative effect in patients with advanced nonsmall cell lung cancer. J Int Med Res. 2011;39:1736-1743.
65. Schoellnast H, Deodhar A, Hsu M, et al. Recurrent non-small cell lung
cancer: evaluation of CT-guided radiofrequency ablation as salvage
therapy. Acta Radiol. 2012;53:893-899.
66. Li X, Zhao M, Wang J, et al. Percutaneous CT-guided radiofrequency
ablation as supplemental therapy after systemic chemotherapy for
selected advanced non-small cell lung cancers. AJR Am J Roentgenol.
2013;201:1362-1367.
67. Alexander ES, Machan JT, Ng T, et al. Cost and effectiveness of radiofrequency ablation versus limited surgical resection for stage I nonsmall-cell lung cancer in elderly patients: is less more? J Vasc Interv
Radiol. 2013;24:476-482.
68. Bang HJ, Littrup PJ, Currier BP, et al. Percutaneous cryoablation of
metastatic lesions from non-small-cell lung carcinoma: initial survival,
local control, and cost observations. J Vasc Interv Radiol. 2012;23:
761-769.
69. Sher DJ, Wee JO, Punglia RS. Cost-effectiveness analysis of stereotactic
body radiotherapy and radiofrequency ablation for medically inoperable, early-stage non-small cell lung cancer. Int J Radiat Oncol Biol
Phys. 2011;81:e767-e774.
70. Dupuy DE. Treatment of medically inoperable non-small-cell lung
cancer with stereotactic body radiation therapy versus image-guided
tumor ablation: can interventional radiology compete? J Vasc Interv
Radiol. 2013;24:1139-1145.
e467
LUNG CANCER
Lung Cancer Screening and

Prevention
CHAIR
Denise R. Aberle, MD
David Geffen School of Medicine at UCLA
Los Angeles, CA
SPEAKERS
Abbie Begnaud, MD
University of Minnesota
Minneapolis, MN
Graham W. Warren, MD, PhD
Medical University of South Carolina
Charleston, SC
BEGNAUD, HALL, AND ALLEN
Lung Cancer Screening With Low-Dose CT: Implementation

Amid Changing Public Policy at One Health Care System
Abbie Begnaud, MD, Thomas Hall, MBA, and Tadashi Allen, MD
OVERVIEW
Screening for lung cancer with low-dose CT has evolved rapidly in recent years since the National Lung Screening Trial (NLST)
results. Subsequent professional and governmental organization guidelines have shaped policy and reimbursement for the
service. Increasingly available guidance describes eligible patients and components necessary for a high-quality lung cancer
screening program; however, practical instruction and implementation experience is not widely reported. We launched a
lung cancer screening program in the face of reimbursement and guideline uncertainties at a large academic health center.
We report our experience with implementation, including challenges and proposed solutions. Initially, we saw less referrals
than expected for screening, and many patients referred for screening did not clearly meet eligibility guidelines. We
educated primary care providers and implemented system tools to encourage referral of eligible patients. Moreover, in
response to the Centers for Medicare & Medicaid Services (CMS) final coverage determination, we report our programmatic
adaptation to meet these requirements. In addition to the components common to all quality programs, individual health
delivery systems will face unique barriers related to patient population, available resources, and referral patterns.
ince the publication of the landmark NLST in 2011, many

health care systems prepared to implement a lung cancer
screening program. The NLST was the first randomized
controlled trial to show a mortality benefit when screening
high-risk populations for lung cancer with low-dose CT of the
chest.1 At the time of NLST publication, the most updated
guideline from the U.S. Preventive Services Task Force
(USPSTF) cited insufficient evidence to recommend for or
against screening for lung cancer.2 Similarly, the American
College of Chest Physicians recommended screening only
within the confines of a clinical trial.3
The University of Minnesota Health is an academic medical
practice in the center of a large metropolitan area. Our health
system affiliate, Fairview Health Systems, is one of several that
provide care throughout the state. One advantage of this
system is that any patient treated in the hospitals or clinics has a
universal electronic health record (EHR, Epic). However, different physician practice groups provide services within the
system. Imaging services are contracted to both University of
Minnesota physicians and a large community radiology practice.
MATERIALS AND METHODS

Determination of Need and Assessment of Resources
In early 2013, we gathered a team, consisting of representatives from business development, thoracic oncology,
radiology, thoracic surgery, and pulmonology, to design the

program. Initial research included an assessment of available
clinical resources, a market analysis, and a cost analysis to
start the program. According to our estimates, approximately 94,634 people at high risk for lung cancer, based on
smoking history, live in the metropolitan area. Only one
other screening program was open in the area at the time of
our program planning and launch.
A team of clinicians with lung nodule management expertise is an integral component of a lung cancer screening
program.4 Our multidisciplinary thoracic oncology group
was established in 2011 and includes board-certified thoracic surgeons, radiation oncologists, thoracic and interventional radiologists, interventional pulmonologists, and
medical oncologistsall specializing in the care of patients
with lung cancer. An experienced, specialized team contributes to optimizing outcomes of lung cancer screening by
avoiding unnecessary diagnostic procedures and achieving
the best possible outcomes for lung resection surgery.5 The
multidisciplinary team reviews indeterminate lung nodules
in a weekly lung nodule conference with diagnostic thoracic
radiology to recommend a patient-specific plan for diagnosis
and/or disease management. Minimally invasive diagnostic
testing is available at our institution, including percutaneous
lung biopsy, video-assisted thoracic surgery, and linear
From the Department of Medicine, University of Minnesota, Minneapolis, MN; Fairview Health System, University of Minnesota Cancer Care, Minneapolis, MN; Department of Radiology,
University of Minnesota, Minneapolis, MN.
Corresponding author: Abbie Begnaud, MD, 420 East Delaware St., MMC 27, Minneapolis, MN 55455; email: abegnaud@umn.edu.
e468
LUNG CANCER SCREENING WITH LOW-DOSE CT
probe endobronchial ultrasound and guided bronchoscopy

techniques (such as electromagnetic navigational bronchoscopy and radial endobronchial ultrasound). The team
uses advanced diagnostic technologies with individualized
attention to lesion characteristics, patient preferences, and
medical comorbidities.
Program Planning
Establishing a workflow for ordering examinations, interpreting images, relaying results, and ensuring follow-up
was a crucial phase of planning. The follow-up is driven
by the guidelines by the National Comprehensive Cancer
Network, 6 American Association of Thoracic Surgery, 7
and American College of Radiology (ACR).8 Implementation
of the program required partnership with EHR builders,
billing and patient financial services, and communications
specialists.
Although self-referral for screening was initially considered, ultimately we required an order from a credentialed
provider for screening. Subsequently released CMS requirements mandate an order and ordering provider, but
health systems might elect to offer screening to patients
with private insurance or those who would pay for services
out of pocket. The anticipation of need for result follow-up
influenced our decision to require all patients to have an
ordering provider for lung cancer screening.
A unique EHR order was created specifically for lung cancer
screening. The name of the order and associated search key
words included CT, lung cancer screening, and lowdose CT. Initially the order had few mandatory responses
related to eligibility criteria: patients symptoms suggestive
of lung cancer, age, and smoking history. When this order is
signed, providers are also prompted to include in the patients printed after-visit summary the Lung Cancer
Screening, Frequently Asked Questions and Resources to
Help You Quit Smoking. Subsequent CMS requirements
changed the order to include the required elements for all
patients.
KEY POINTS
Executing a new preventive health care service requires

education of and acceptance from providers and
patients.
Implementing program for screening for a lung cancer
requires multidisciplinary collaboration.
Ensuring regulatory compliance and best practice in lung
cancer screening requires system tools and ongoing
oversight.
Despite education, providers may inappropriately refer
patients for lung cancer screening.
Electronic health record tools can be used to
supplement provider and patient education about and
compliance with lung cancer screening.
Despite the limited screening interval and upper age limit

of 75 reported in the NLST, we predicted benefit from
continued annual screening. Hence, we added lung cancer
screening to the health maintenance activity where other
preventive health measures are tracked, including cancer
screenings and vaccinations. The American Association for
Thoracic Surgery also recommended annual screening for
people age 5579.7 Ultimately, the USPFTF and CMS released recommendations for both a broader age range and
screening extending beyond three annual examinations. We
educated primary care provider groups about screening
eligibility and the process we developed.
Early Implementation
After creating a new chest CT lung cancer screening order in
the EHR, establishing a result workflow with radiology, and
educating providers, our program launched on December
12, 2013. Because most insurance companies were not yet
covering lung cancer screening examinations, we offered the
service at $150, which was a fair market price at the time.
Several insurance companies began covering screening
throughout 2014, well before the updated USPSTF guideline.
At the time of screening, patients submit detailed information about their risk factors for lung cancer via a
questionnaire (developed with collaboration from academic
epidemiologists and clinical psychologists, subsequently
adapted to match CMS registry data requirements). Imaging
technologists enter the information into the EHR and share
with the interpreting radiologist. The questionnaire assesses
each patients risk profile including smoking history, radon
and particulate exposures, family history of lung cancer, and
personal medical history.
We implemented a structured reporting system for results
called U-Lung-RADS, which was modeled after Bi-RADS and
was very similar to the system ultimately released by ACR in
April 2014 (Table 1).9 In addition to a numeric score based on
the likelihood of cancer, additional modifiers indicate the
presence of incidental or suspicious findings. We implemented the ACR Lung-RADS 1.0 immediately after it was
released.
Our results follow-up team is comprised of a call center,
certified nurse specialists, and a physician. The team
monitors semiweekly and monthly reports for screening
examination results. All patients who were screened
receive a letter describing the results of the screening CT and
the next steps for follow-up in plain language. The follow-up
team also helps facilitate orders for follow-up scans and
appointments and verifies that they are performed. The
team also notifies the primary care physician of any incidental findings. Incidental findings unrelated to lung
cancer were found in 7.5% of patients in the NLST,10
therefore a mechanism for management is imperative.
Patients who have Lung-RADs 1 or 2 (no suspicious findings)
and meet eligibility for screening have the lung cancer
screening topic added to their health maintenance modifier.
The health maintenance activity is visible to all providers,
e469
TABLE 1. Lung-RADS Version 1.0 Assessment Categories

Category Descriptor
Incomplete
Negative
Primary
Category
Category Descriptor
No nodules and definitely benign nodules
Management
Additional lung cancer screening CT images and/or

comparison with prior chest CT examinations is
needed
Continue annual screening with LDCT in 12 months
Benign Appearance or Nodules with a very low likelihood of becoming a

Behavior
clinically active cancer due to size or lack of growth
Probably Benign
Probably benign finding(s): short term follow-up

suggested, includes nodules with a low likelihood
of becoming a clinically active cancer
Suspicious
Findings for which additional diagnostic testing and/or 4A

tissue sampling is recommended
4B
Significant, Other
Prior Lung Cancer
6-month LDCT
3-month LDCT; PET/CT may be used when there is a

$ 8-mm solid component
Chest CT with or without contrast, PET/CT and/or tissue
sampling depending on the probability of malignancy
and comorbidities; PET/CT may be used when there is
a $ 8-mm solid component
Release date: April 28, 2014.

Abbreviation: LDCT, low-dose CT.
and, when overdue, an alert appears in the patients chart.

For any positive findings (Lung-RADS 3 or higher), the certified nurse specialist or physician is notified immediately so
that the patient can be discussed in multidisciplinary team
conference to establish a plan for follow-up or diagnostic
procedure. A unified EHR allows multidisciplinary chart and
image review as requested by ordering providers for externally performed screening examinations with abnormal
findings.
Overcoming Barriers
Improving examination ordering rates. The number of
orders for screening has been less than expected relative to
the estimated eligible population. Several identifiable barriers are likely responsible. First, adoption of new recommendations usually grows over time.11 Adoption is almost
certainly limited by provider awareness of screening availability and eligibility, as well as the time required to incorporate yet another preventive care service into patient
interactions. Secondly, uncertainty regarding insurance
coverage may discourage health care providers from
ordering a screening. As we expand to new regions in the
state, we offer educational sessions to primary care practices
served by those centers. Furthermore, systemwide educational events have been held in the form of grand rounds at
multiple sites and lunch webinars for primary care providers.
An effective method of improving adoption of guidelines
for lung cancer screening may be to use the EHRs best
practice alert (BPA) functionality. A pop-up alert prompts a
health care provider about a clinical topic, including eligibility for preventive care activities. Activating an alert for
lung cancer screening CT orders for all of our patients
who meet the criteria would probably improve ordering
rates and awareness among our ordering physicians.
e470
However, implementing alerts like these must balance patient care and health care provider information overload and
alert desensitization.12 Moreover, in order for a BPA to be
useful, patients must have a very detailed smoking history
recorded in the EHR.
Improving examination completion rates. Even as examination orders increased, many went unscheduled and were
not performed. We believe this low completion rate to be
primarily due to reimbursement uncertainty and out-ofpocket payment for the examination. Because several national private insurance companies added lung cancer
screening coverage prior to the USPSTF recommendation,
we eliminated the self-pay approach and began reviewing
orders for insurance eligibility. If an eligible patients insurance did not cover screening, they could receive the
screening examination under a local grant for this purpose.
Thus, we were able to remove a cash price barrier and offer
the procedure to all eligible patients as an insurance-neutral
screening examination. We notified providers about the
grant and elimination of the self-pay option using an internal
newsletter. We also sent letters that explained the benefits
of screening and offered examination with insurance or
grant coverage to all patients with orders that had not been
completed. Finally, our imaging centers do not proactively
schedule ordered examinations; rather, patients are asked
to initiate the scheduling process. Proactive scheduling of
ordered examinations by radiology would likely improve
completion rates as well.
Improving result reporting. Streamlining result reporting
also was an initial challenge. We believe it is important to
deliver results to patients in a way that minimizes anxiety
about abnormal findings. We envisioned an experienced
certified nurse specialist to deliver abnormal results along

with a treatment recommendation to the patient. However,
early examination results follow-up by the ordering provider
was generally not in accordance with national guidelines.
When this happened, we contacted the providers to notify
them of the service we intended to provide and revised
any misinformed recommendations. As providers learned to
trust that we would handle results follow-up, the process
became more streamlined. Primary care physicians are informed of all patient communication and recommendations.
As a service to patients and providers who scheduled at a
non-university-affiliated imaging center in the health system, we added an opt-in question to the imaging order in
September 2015. Ordering providers who opt-in transfer
nodule follow-up to the centralized results team.
Smoking Cessation Program

The effect of screening for lung cancer on smoking rates is
not fully known13 but thought to be most favorable when
smoking cessation assistance is offered with screening.14
Moreover, smoking cessation is an important part of preventive health and cost-effective lung cancer prevention.15
We use a variety of tools to help patients quit smoking. A
pharmacist provides pharmacologic assistance in medication therapy management for smoking cessation, and
smoking cessation tools are incorporated in the lung cancer
screening ordering process in the EHR. Furthermore, our
institution is a site for a National Institutes of Healthfunded
smoking cessation trial, with participation offered to
smokers who are being screened. Therefore, we consolidated available institution resources within a website to
provide centralized access to smoking cessation assistance.
We incorporated smoking cessation resources in our lung
cancer screening FAQs, which was given to patients at the
time of order placement. Although many barriers to ideal
provision of smoking cessation assistance exist,16,17 it is
unequivocally recommended to accompany lung cancer
screening.4
Adapting to New Guidelines and CMS Requirements

Professional society guidelines. In late 2014, the American
College of Chest Physicians and the American Thoracic Society jointly published a statement that outlined the key
components of a high-quality lung cancer screening program
(Table 2).4 Guidelines are also available for lung cancer
screening from the ACR, Society of Thoracic Radiology,8 and
the American Association of Thoracic Surgeons.7 Our program had already been built with these best practices in
mind and required no modifications to meet the guidelines.
CMS Decision Memo. The CMS Decision Memo was released in November 201418 and incorporated many of the
elements of a high-quality screening program previously
recommended (Table 3). In addition, a credentialed provider
must conduct a shared decision-making visit with the patient
when ordering the initial screening examination, document
TABLE 2. Recommended Components of High-Quality

Lung Cancer Screening Program4
Policy in place to offer lung cancer screening only to patients who meet
criteria established by the U.S. Preventive Services Task Force. At
least 90% of patients screening should match this policy.
Annual screening until the age of 80, a person has not smoked for
15 years, or becomes unwilling or unable to undergo curative lung
resection surgery.
Technical specifications of CT conform to ACR-STR standards, and data
are collected to ensure mean radiation dose is in compliance with
those.
Threshold of nodule size that is considered positive, and data are
collected describing the number and size of nodules detected.
Structured reporting system, such as Lung-RADS, is used, and data are
collected on compliance with structured reporting.
Designated clinicians with expertise in lung nodule management
(diagnostic radiology, interventional radiology, pulmonary
medicine, thoracic surgery, medical oncology, radiation oncology)
use accepted lung nodule management strategy.
Integrated smoking cessation program and data collection related to
smoking cessation interventions offered.
eligibility and discussion of risks and benefits of screening,

and offer smoking cessation if applicable. The screening
order itself should also include eligibility criteria.
To render an interpretation that will be reimbursed by
CMS, radiologists must be board certified, experienced in
chest CT, and participate in continuing medical education. Imaging centers eligible for reimbursement of lung
cancer screening must have experience with low-dose CT
lung cancer screening and be accredited as an advanced
diagnostic imaging center. In addition, effective radiation
dose of reimbursable examinations must be less than 1.5
mSv, and data must be collected and submitted to a CMSapproved registry for each screening examination. Although an extensive list of data elements was initially
proposed, the final Decision Memo included 10 data
elements pertaining to the ordering provider, interpreting radiologist, patient smoking history, and CT
scanner.
Adapting EHR tools. To facilitate compliance with CMS
requirements, we modified the EHR tools. In addition to
making modifications to the lung cancer screening imaging
order, we created a SmartSet, which is a bundle of orders
and documentation related to a topic. The SmartSet guides
clinicians on patient eligibility, required billing codes, suggested wording for the shared decision-making visit, orders
for smoking cessation tools, and the imaging order. When
signed, the SmartSet outputs patient instructions with
screening FAQs and scheduling information. A BPA was
created to direct the ordering provider from the stand-alone
order to the SmartSet, as many providers had become accustomed to using the order laone.
RESULTS
Examination Ordering and Completion Rates
Systemwide, 902 unique patients received orders for
screening using low-dose CT. However, only 63% of these
e471
TABLE 3. CMS Lung Cancer Screening Eligibility Criteria19

Type
Criteria
Beneficiary Eligibility Criteria
1. Age 5577
2. Asymptomatic (no signs or symptoms of lung cancer)
3. Tobacco smoking history of at least 30 pack-years (one pack-year 5 smoking one pack per day for one year; 1 pack 5
20 cigarettes)
4. Current smoker or one who has quit smoking within the last 15 years
5. Receives a written order for LDCT lung cancer screening that meets the following criteria:
a. For the initial LDCT lung cancer screening service: a beneficiary must receive a written order for LDCT lung cancer
screening during a lung cancer screening counseling and shared decision-making visit, furnished by a physician (as
defined in Section 1861(r)(1) of the Social Security Act) or qualified nonphysician practitioner (meaning a physician
assistant, nurse practitioner, or clinical nurse specialist as defined in 1861(aa)(5) of the Social Security Act). A lung
cancer screening counseling and shared decision-making visit includes the following elements (and is appropriately
documented in the beneficiarys medical records):
i. Determination of beneficiary eligibility including age, absence of signs or symptoms of lung cancer, a specific
calculation of cigarette smoking pack-years; and if a former smoker, the number of years since quitting;
ii. Shared decision-making, including the use of one or more decision aids, to include benefits and harms of
screening, follow-up diagnostic testing, overdiagnosis, false-positive rate, and total radiation exposure;
iii. Counseling on the importance of adherence to annual lung cancer LDCT screening, impact of comorbidities, and
ability or willingness to undergo diagnosis and treatment;
iv. Counseling on the importance of maintaining cigarette smoking abstinence if former smoker; or the importance
of smoking cessation if current smoker and, if appropriate, furnishing of information about tobacco cessation
interventions; and
v. If appropriate, the furnishing of a written order for lung cancer screening with LDCT.
b. For subsequent LDCT lung cancer screenings: the beneficiary must receive a written order for LDCT lung cancer
screening, which may be furnished during any appropriate visit with a physician (as defined in Section 1861(r)(1) of the
Social Security Act) or qualified nonphysician practitioner (meaning a physician assistant, nurse practitioner, or clinical
nurse specialist as defined in Section 1861(aa)(5) of the Social Security Act). If a physician or qualified nonphysician
practitioner elects to provide a lung cancer screening counseling and shared decision-making visit for subsequent lung
cancer screenings with LDCT, the visit must meet the criteria described above for a counseling and shared decisionmaking visit.
c. Written orders for both initial and subsequent LDCT lung cancer screenings must contain the following information,
which must also be appropriately documented in the beneficiarys medical records:
i. Beneficiary date of birth;
ii. Actual pack-year smoking history (number);
iii. Current smoking status, and for former smokers, the number of years since quitting smoking;
iv. Statement that the beneficiary is asymptomatic (no signs or symptoms of lung cancer); and
National Provider Identifier (NPI) of the ordering practitioner.
Reading Radiologist Eligibility

Criteria
1.
2.
3.
4.
Radiology Imaging Facility

Eligibility Criteria
1. Performs LDCT with volumetric CT dose index (CTDIvol) of # 3.0 mGy (milligray) for standard size patients (defined to
be 57 and approximately 155 pounds) with appropriate reductions in CTDIvol for smaller patients and appropriate
increases in CTDIvol for larger patients
2. Utilizes a standardized lung nodule identification, classification, and reporting system
3. Makes available smoking cessation interventions for current smokers and collects and submits data to a CMSapproved registry for each LDCT lung cancer screening performed.
Board certification or board eligibility with the American Board of Radiology or equivalent organization
Documented training in diagnostic radiology and radiation safety
Involvement in the supervision and interpretation of at least 300 chest CT acquisitions in the past 3 years
Documented participation in continuing medical education in accordance with current American College of Radiology
standards
5. Furnish lung cancer screening with LDCT in a radiology imaging facility that meets the radiology imaging facility
eligibility criteria below
examinations have been completed (Fig. 1). Over time, both

examinations orders and completion rates increased. Since
the function has been offered, 81% of examinations completed opted in for centralized result management. Using
the Lung-RADS interpretation strategy, 70% of patients had a
positive finding (Lung-RADS 2 or higher). However, only 17%
of patients had a finding that required follow-up sooner than
annually (Fig. 2).
Patient Eligibility
EHR record of eligibility was measured by completed
smoking history tool. Of the 572 screening examinations
completed, less than 20% were eligible by smoking history
e472
(Table 4). About one-third of patients had inadequate

smoking history as recorded in the EHR history tool to
determine eligibility. Many of these patients did report
smoking for at least 30 pack-years on the patient
questionnaire administered prior to low-dose CT. Another nearly one-third of patients were not eligible for
screening based on age or smoking history, as recorded.
Among patients who had smoking intensity (packs per
day and smoking years) recorded, implementation of a
BPA clarifying eligibility criteria had no impact on the
proportion of patients referred for screening who met
eligibility criteria (30 or more pack years). Similarly, the
proportion of screening referrals for patients whose
recorded quit smoking date was more than 15 years ago
FIGURE 1. Lung Cancer Screening Order Completion

by Quarter
TABLE 4. Patient Characteristics Contributing to

Eligibility for Lung Cancer Screening as Recorded in
Electronic Health Records
Patient Characteristics
Number of Patients (572, %)
Current Smoker
268 (47%)
Former Smoker
282 (49%)
No Recorded Quit Date
31 (5%)
Pack-year < 30
117 (20%)
Inadequate Entry to Calculate
did not substantially change after implementation of

the BPA.
Counseling and Shared Decision-Making Visit

Nearly all examinations (95%) were ordered during an office
visit. Yet many of those did not have documentation of
counseling and a shared decision-making visit. Shareddecision making documentation has improved over time
(Fig. 3), but, in January 2016, it still only occurred in half of
patients with completed screening examinations. Prior to
implementation of a BPA indicating CMS requirements, 87%
of patients did not have documentation of a shared decisionmaking visit. Since the BPA function, compliance with shared
decisions-making documentation in patients with Medicare
or Medicaid has improved to 46%.
DISCUSSION
Lung cancer screening is a relatively new preventive health
care service (secondary prevention). As such, developing and
implementing new processes can be guided by national
leaders but must also be customized to meet the needs of
local and regional health care systems. We began implementing
FIGURE 2. Lung-RADS Results for Completed Orders
48 (282 patients, 17%)
Quit > 15 years
206 (36%)
Age-inappropriate
21 (4%)
Apparently Eligible
123 (22%)
a screening program before current regulatory requirements

were released.
Approximately one-third of the orders were for patients
whose eligibility for screening could not be confirmed.
Eligibility was unconfirmed due to insufficient documentation of smoking history, documented smoking intensity
less than 30 pack-years, or quit more than 15 years prior to
examination. Our health system has had some success but
continues to optimize EHR tools to facilitate compliance
with recommended eligibility. Most patients did report
smoking history on prescreening questionnaires that was
adequate for lung cancer screening eligibility. We believe
that infrequent and inaccurate recording of smoking history by clinicians to be responsible. Furthermore, the EHR
tool for recording smoking history is inadequate to fully
capture the variability in smoking intensity over time that
describes many smokers habits. A tool that more specifically captures light smoking, heavy smoking, and quit
periods can help record an accurate account of smoke
exposure. In addition to improving EHR tools and utilization, confirmation of eligibility prior to screening using lowdose CT will likely be needed to reach perfect compliance
with eligibility requirements.
Early orders had a low completion rate, which we believe
to be due to the absence of insurance coverage and the selfpay cost of the examination ($150). After the CMS final
Decision Memo, examination orders quadrupled, supporting
our notion that uncertainty about reimbursement drove
initial low uptake. Feedback from providers suggests that
anxiety prevents patients from scheduling examinations
despite having the information to do so. Furthermore, in
most clinics, patients are asked to schedule examinations
themselves rather than having examinations scheduled
for them. Completion rates continue to improve, despite
having a patient-initiated scheduling system.
Our findings suggest that reimbursement-driven enforcement of patient eligibility as required by Medicare is
necessary to apply screening to the high-risk groups who are
most likely to benefit. We also concur with the need for
continued monitoring of screening outcomes through
national registries to further refine screening eligibility and
utility.
e473
FIGURE 3. Lung Cancer Screening Orders Completed in 2015 and Accompanied by Shared Decision Making
Documentation
Abbreviation: SDM, shared decision making.
This article demonstrates our experience implementing

a lung cancer screening program outside of a federally
supported clinical trial. Future trends will be greatly shaped
by outcomes reported from national registries and
by experience with pervasive screening in a climate of
reimbursement for eligible patients. As diagnostic radiology

technology improves and adjunctive methods of early detection of lung cancer become available, the riskbenefit
ratio for screening for lung cancer will continue to evolve
and, ideally, improve.
References
1. Aberle DR, Adams AM, Berg CD, et al. Reduced lung-cancer mortality
with low-dose computed tomographic screening. N Engl J Med. 2011;
365:395-409.
2. U.S. Preventive Services Task Force. Lung cancer screening: recommendation statement. Ann Intern Med. 2004;140:738-739.
3. Bach PB, Silvestri GA, Hanger M, et al. Screening for lung cancer: ACCP
evidence-based clinical practice guidelines (2nd edition). Chest. 2007;
132:69S-77S.
4. Mazzone P, Powell CA, Arenberg D, et al. Components necessary for high
quality lung cancer screening: American College of Chest Physicians and
American Thoracic Society Policy Statement. Chest. 2015;147:295-303.
5. Farjah F, Flum DR, Varghese TK Jr, et al. Surgeon specialty and long-term
survival after pulmonary resection for lung cancer. Ann Thorac Surg.
2009;87:995-1004, discussion 1005-1006.
6. National Comprehensive Cancer Network. Lung Cancer Screening. https://
www.nccn.org/store/login/login.aspx?ReturnURL=http://www.nccn.org/
professionals/physician_gls/pdf/lung_screening.pdf. Accessed February
13, 2016.
7. Jaklitsch MT, Jacobson FL, Austin JH, et al. The American Association for
Thoracic Surgery guidelines for lung cancer screening using low-dose
computed tomography scans for lung cancer survivors and other highrisk groups. J Thorac Cardiovasc Surg. 2012;144:33-38.
8. Kazerooni EA, Austin JH, Black WC, et al. ACR-STR practice parameter for
the performance and reporting of lung cancer screening thoracic
e474
9.
10.
11.
12.
13.
14.
15.
computed tomography (CT): 2014 (Resolution 4). J Thorac Imaging.

2014;29:310-316.
American College of Radiology. Lung-RADS Version 1.0 Assessment Categories Release date: April 28, 2014. https://www.acr.org/
~/media/ACR/Documents/PDF/QualitySafety/Resources/LungRADS/
AssessmentCategories.pdf. Accessed February 13, 2016.
Church TR, Black WC, Aberle DR, et al. Results of initial low-dose
computed tomographic screening for lung cancer. N Engl J Med.
2013;368:1980-1991.
Francke AL, Smit MC, de Veer AJ, et al. Factors influencing the
implementation of clinical guidelines for health care professionals:
a systematic meta-review. BMC Med Inform Decis Mak. 2008;8:38.
Ash JS, Berg M, Coiera E. Some unintended consequences of information
technology in health care: the nature of patient care information systemrelated errors. J Am Med Inform Assoc. 2004;11:104-112.
Tanoue LT, Tanner NT, Gould MK, et al. Lung cancer screening. Am J
Respir Crit Care Med. 2015;191:19-33.
Park ER, Gareen IF, Japuntich S, et al. Primary care provider-delivered
smoking cessation interventions and smoking cessation among participants in the National Lung Screening Trial. JAMA Intern Med. 2015;
175:1509-1516.
Villanti AC, Jiang Y, Abrams DB, et al. A cost-utility analysis of lung cancer
screening and the additional benefits of incorporating smoking cessation interventions. PLoS One. 2013;8:e71379.
16. Manley MW, Griffin T, Foldes SS, et al. The role of health plans in
tobacco control. Annu Rev Public Health. 2003;24:247-266.
17. Panaitescu C, Moffat MA, Williams S, et al. Barriers to the provision
of smoking cessation assistance: a qualitative study among Romanian family physicians. NPJ Prim Care Respir Med. 2014;24:
14022.
18. Centers for Medicare & Medicaid Services. Proposed Decision Memo
for Screening for Lung Cancer with Low Dose Computed Tomography
(LDCT) (CAG-00439N). https://www.cms.gov/medicare-coverage-database/
details/nca-proposed-decision-memo.aspx?NCAId=274&NcaName=
Screening+for+Lung+Cancer+with+Low+Dose+Computed+Tomography+
(LDCT)&MEDCACId=68&IsPopup=y&bc=AAAAAAAAAgAAAA%3d%3d&.
19. Centers for Medicare & Medicaid Services. Decision Memo for
Screening for Lung Cancer with Low Dose Computed Tomography
(LDCT) (CAG-00439N). https://www.cms.gov/medicare-coveragedatabase/details/nca-decision-memo.aspx?NCAId=274. Accessed February 13, 2016.
e475
ADDRESSING TOBACCO WITH CANCER SCREENING AND TREATMENT
Lung Cancer Screening, Cancer Treatment, and Addressing the

Continuum of Health Risks Caused by Tobacco
Graham W. Warren, MD, PhD, Jamie S. Ostroff, PhD, and John R. Goffin, MD
OVERVIEW
Tobacco use is the largest preventable risk factor for the development of several cancers, and continued tobacco use by
patients with cancer and survivors of cancer causes adverse outcomes. Worldwide tobacco control efforts have reduced
tobacco use and improved health outcomes in many countries, but several countries continue to suffer from increased
tobacco use and associated adverse health effects. Continued tobacco use by patients undergoing cancer screening or
treatment results in continued risk for cancer-related and noncancer-related health conditions. Although integrating tobacco assessment and cessation support into lung cancer screening and cancer care is well justified and feasible, most
patients with cancer unfortunately do not receive evidence-based tobacco cessation support. Combining evidence-based
methods of treating tobacco addiction, such as behavioral counseling and pharmacotherapy, with practical clinical considerations in the setting of lung cancer screening and cancer treatment should result in substantial improvements in access
to evidence-based care and resultant improvements in health risks and cancer treatment outcomes.
obacco use, particularly combustible tobacco, remains

the leading preventable cause of cancer worldwide.
Since the original U.S. Surgeon Generals report (SGR) on
tobacco in 1964 concluding that smoking was causally linked
to lung and laryngeal cancer, there have been numerous
SGRs confirming the adverse effects of tobacco on a spectrum of diseases, including increased risk for developing a
wide range of cancers.1 The 2014 SGR extended findings of
prior reports by concluding that continued smoking by
patients with cancer and survivors of cancer causes increased overall mortality, cancer-specific mortality, risk for
second primary cancer, and associations with increased
cancer treatment toxicity.2 As updated in the 2014 SGR,
smoking and tobacco addiction usually starts in youth,
continues into adulthood, and results in chronic exposure to
chemicals that lead to the development of cancer, heart
disease, pulmonary disease, and several other adverse
health conditions. Many of the effects of tobacco are partly
or entirely reversible, with reduced risks of adverse health
conditions associated with longer periods of smoking abstinence. As it pertains to patients with cancer and the risk of
developing cancer, smoking cessation can reduce the risks of
developing cancer, can improve cancer treatment outcomes, and can reduce or prevent the development of other
tobacco-related diseases.2
Although smoking cessation can reverse many of the

adverse effects of smoking among the general population
and among patients with cancer, most oncologists unfortunately do not address tobacco use by providing
evidence-based tobacco cessation support.3,4 There are
several potential reasons why oncologists and other providers may not provide cessation support,5 but the fundamental consequence is that many people continue to be
exposed to cancer risk and adverse cancer treatment outcomes because they do not have access to evidence-based
tobacco cessation support. The overall purpose of this educational article is to convey the continuum of risk associated with tobacco use cancer risk and cancer treatment
outcomes, discuss recent data as related to lung cancer
screening and cancer treatment in the context of tobacco
use, and provide practical considerations that can be helpful
to increase access to tobacco cessation support. There are
limited studies evaluating methods to implement smoking
cessation into lung cancer screening and cancer care.
However, this article will further elaborate on practical
clinical considerations that could be helpful in the design and
implementation of tobacco cessation approaches across the
continuum of cancer risk and treatment.
There are thousands of references describing the adverse
health risks associated with tobacco use and hundreds of
From the Departments of Radiation Oncology and Cell and Molecular Pharmacology, Medical University of South Carolina, Charleston, SC; Department of Psychiatry and Behavioral
Sciences, Memorial Sloan Kettering Cancer Center, New York, NY; Department of Oncology, McMaster University, Hamilton, Ontario, Canada.
Corresponding author: Graham W. Warren, MD, PhD, Department of Radiation Oncology, Medical University of South Carolina, 169 Ashley Ave., MSC 318, Charleston, SC 29425;
email: warrengw@musc.edu.
223
WARREN, OSTROFF, AND GOFFIN
large reviews describing specific areas of risk and risk mitigation as related to tobacco use. For the purposes of this
educational article, we will highlight substantial review work
presented over the past 50 years by the U.S. Public Health
Service (PHS) through updated SGRs that have focused on
tobacco-associated health risks, addiction, and mitigation,1
with particular emphasis on the 2014 SGR2 that summarizes
many current and former conclusions associated with tobacco use. Whenever possible, references will be provided
to online resources that keep information up to date, which
is important as related to ongoing product, use, and health
information related to tobacco.
WORLDWIDE TOBACCO USE AND THE

CONTINUUM OF RISK
Cigarette smoke is the predominant form of worldwide
tobacco consumption, although in different countries different patterns exist that are highly influenced by culture,
gender, governmental regulation, advertising, and
manufacturing. Readers are encouraged to review the fifth
edition of the Tobacco Atlas, updated in 2015,6 available
online and in print. Produced by the American Cancer Society and the World Lung Foundation, the Tobacco Atlas
highlights patterns of worldwide tobacco use. Highlights
from the Tobacco Atlas and 2014 SGR include the following:
Approximately 100 million people died of tobaccorelated causes in the 20th century and an estimated
1 billion people will die in the 21st century from tobaccorelated causes if current tobacco use patterns persist. An
estimated 50% of people who use tobacco will die of a
tobacco-related disease.
An estimated 5.8 trillion cigarettes were smoked worldwide in 2014. In many high-income countries, tobacco
use has plateaued and in many cases decreased over the
past 30 to 40 years with concomitant decreases in
KEY POINTS
224
Tobacco use confers a continuum of risk across the

general population as well as among patients diagnosed
with cancer, justifying the need to address tobacco use
before, during, and after a cancer diagnosis.
Worldwide tobacco consumption patterns continue to
place a substantial burden on the economy and health of
people in every nation.
Addressing tobacco use is a critical component of lung
cancer screening and cancer treatment.
Although strong evidence exists for methods to help
patients stop tobacco use, clinical implementation is
limited with little evidence-based support.
Considering the clinical setting, available resources, and
method of tobacco assessment, cessation support and
developing a sustainable clinical approach are practical
considerations when implementing change to improve
access to tobacco cessation support.
mortality associated with tobacco-related diseases.

However, in most low- and middle-income countries,
tobacco use (particularly combustible tobacco use
[cigarette smoking]) is increasing. Growing tobacco
consumption in China has offset decreased tobacco
consumption in other developed countries, and onethird of all male individuals who smoke are Chinese. In
many countries, male tobacco consumption is decreasing, whereas female tobacco use has remained
unchanged or is increasing.
Lung cancer, chronic obstructive pulmonary disease,
and heart disease are leading causes of death from
tobacco. More than 80% of lung cancer and lung cancer
deaths are attributed to smoking, and approximately
one-third of all cancer deaths are attributed to tobacco
use. However, tobacco use causes disease in almost
every organ in the body. Thus, continued tobacco use
after a cancer diagnosis results in continued risk to other
organs and decreases the effectiveness of cancer
treatment.
Exposure to secondhand smoke can confer similar disease risks in a dose- and time-dependent manner.
Higher risk is associated with higher and more prolonged
exposure.
Addiction has both behavioral and biologic underpinnings. Most people start using tobacco during childhood
or adolescence, primarily influenced by social pressure,
advertising, and curiosity. Chronic tobacco use and
addiction is reflective of nicotine dependence, which
can affect constitutively expressed nicotinic acetylcholine receptors in the brain and throughout the body.
Tobacco use has strong social associations, in that many
people who smoke often have friends and relatives who
smoke.
Tobacco use is higher among people with lower education, among individuals with lower income, and among
people who are diagnosed with a psychiatric disease.
However, many financially disadvantaged people will
continue to expend resources to continue tobacco use.
Tobacco addiction usually starts early in life. Most
people have tried to stop smoking several times and
most continue to want to stop smoking. Tobacco cessation is often very difficult and requires multiple attempts to achieve success. However, people who have
quit smoking in the past remain at risk for relapse,
particularly if they are exposed to secondhand smoke.
It is estimated that worldwide tobacco use causes $1
trillion annually in economic damage, with $289 billion
spent in treating the health consequences of smoking in
the United States alone. Only a very small portion of
tobacco product revenues (typically much less than 1%)
are used for tobacco control initiatives.
Effective strategies to reduce tobacco use include increased product taxation, smoke-free policies and laws,
legislative regulation, education such as graphic product

labels, limiting or eliminating marketing, increased
pricing, and providing smoking cessation resources.
International treaties and agreements, such as the
World Health Organization Framework Convention on
Tobacco Control, that coordinate these activities can be
highly effective in influencing national and international
change. However, litigation is a widely used method by
the tobacco industry to limit tobacco control initiatives.
Although it was discussed in both the 2014 SGR and the
Tobacco Atlas, an important topic to discuss individually is
continuously evolving tobacco product design. Changes
in cigarette design over the past 50 years have resulted in
cigarettes that are easier to inhale, leading to changes in
disease risk. For instance, lung cancer was once a rare
disease that significantly increased with the advent of mass
cigarette production and changes in cigarette design that
allowed deeper inhalation resulted in the shift of lung cancer
patterns from centrally to peripherally located cancers.7
Unfortunately, changes in tobacco design did not result in
improved health: recent data demonstrate that people who
smoke are now more heavily addicted, with continued increased health risks.8 However, the benefits of cessation
remain intact with reduced health risks associated with
longer periods of cessation.
As new tobacco products are developed, it will take years
or decades to determine the risk. Electronic cigarettes (ecigarettes) or electronic nicotine delivery systems (ENDS)
are relatively new products with largely unknown health
effects and patterns of use. Although many online resources
are promoting the use of ENDS to facilitate tobacco cessation, the effects of ENDS to assist in tobacco cessation are
not yet known and ENDS cannot yet be promoted as a tobacco cessation device.9 It is important to note that the
health effects of ENDS may be dependent on the population
in question: ENDS use by youth who have never used tobacco are likely to result in different health effects compared
with adults who smoke and transition to ENDS use.9 Currently, physicians and clinical providers are limited in providing evidence-based guidance regarding ENDS use
because most of the evidence is simply lacking at this time.
TOBACCO USE IN THE CONTEXT OF LUNG

CANCER SCREENING AND CANCER TREATMENT
When considering integration of tobacco cessation into early
detection of lung cancer, it is important to note that although diagnosis with resectable nonsmall cell lung cancer
can result in 5-year survival exceeding 70%, most patients
with lung cancer are diagnosed with symptoms, are not
operable, and have a dismal long-term prognosis.10 Capturing patients at an early curative stage is the optimal
strategy for managing lung cancer. Early efforts to screen for
lung cancer with chest x-rays or sputum cytology may have
resulted in a modest shift in staging, but with no effect on
survival.11 Subsequent screening of large cohorts with CT
scans suggested improved stage shifting and improved
survival. The large, randomized National Lung Screening Trial

(NLST) provided the first strong evidence of benefit from
screening with low-dose computed tomography (LDCT)
compared with chest x-ray.12 Investigators screened an eligible population (age 5574) who had at least a 30 pack-year
smoking history and reported smoking within the past 15
years, in which participants had a baseline LDCT scan that
was followed by two annual screening LDCT scans. With a
follow-up of 6.5 years, LDCT screening decreased lung
cancerspecific mortality by 20%, overall survival was improved by 6.7%, and CT LDCT screening diagnosed more
stage I to stage II disease (57% vs. 39%). Based on NLST data,
several agencies have recommended LDCT screening for an
eligible population provided that screening is undertaken in
the setting of appropriate expertise.
With the advent of LDCT screening as evidence-based
medicine, it has been increasingly important to consider
the utility of smoking cessation at the time of CT screening.
Several advocates have promoted screening for tobacco use
and providing tobacco cessation support to people who use
tobacco. In general, as supported by current National
Comprehensive Cancer Network (NCCN) guidelines, smoking
cessation is recommended for any individuals who are
undergoing LDCT screening and reporting current tobacco
use.13 Newly released NCCN Smoking Cessation Guidelines
provide additional guidance and details for providing tobacco cessation support to patients with cancer as well as
patients undergoing LDCT screening.14
Perhaps the best evidence for promoting smoking cessation in the context of LDCT screening is the recent report
by Tanner et al,15 who evaluated the effect of smoking and
smoking cessation on survival in the NLST. Current smoking
increased overall and cancer-specific mortality among individuals who participated in LDCT screening, but 7 years of
smoking cessation reduced mortality at a level comparable
to the benefit of LDCT, and 15 years of cessation reduced
mortality by 38%. Given the large population of current
smokers who are eligible for lung cancer screening, costeffectiveness is an important consideration for policy
makers considering implementation. Studies based on NLST
data suggest that screening for lung cancer may have an
incremental cost-effectiveness ratio (ICER) per quality of
life-year saved in the $50,000 to $80,000 range.16 It is already established that smoking cessation programs are
highly cost-effective with an ICER less than $9,000 per
life-year saved.17 Simulation work suggests that the addition
of a smoking cessation program to an LDCT screening
program can substantially improve cost-effectiveness.18 As
such, combining LDCT and smoking cessation programs may
add incremental treatment cost but also enhance cost efficiency and provide the proven health benefits associated
with each.
When considering tobacco use by patients with cancer, the
2014 SGR concluded that smoking by patients with cancer
and survivors of cancer caused adverse outcomes and that
smoking cessation improves cancer treatment outcomes.2
Smoking and tobacco products alter cancer biology, promoting
225
increased proliferation, angiogenesis, migration, invasion,

metastasis, and resistance to cytotoxic cancer therapy.19
Examination of the exhaustive evidence in the 2014 SGR
demonstrates that one or more adverse effects of smoking
extend to virtually all cancer disease sites and cancer
treatments. As a result, the adverse effects of smoking are
generally ubiquitous across cancer as a whole and addressing tobacco use is pertinent to the treatment of all
patients with cancer. The authors are unaware of detailed
cost-benefit analyses for implementing smoking cessation
into the care of patients with cancer; however, given the
extremely high costs of treating cancer and the relatively low
cost of providing smoking cessation support, it is very likely
that implementing smoking cessation practices into cancer
care may be one of the most cost-effective methods to
improve cancer treatment outcomes.
LUNG CANCER SCREENING, CANCER

TREATMENT, AND TOBACCO CESSATION
The question arises as to whether a LDCT screening population is well suited to promoting smoking cessation.
Presumably, individuals participating in LDCT screening may
represent a population motivated to improve their health.
Compared with the general population, participants in the
NLST were more commonly married, better educated, more
frequently ex-smokers, and younger. Persistent smokers in
the NLST were associated with having lower education,
being unmarried, being younger, and having a heavier
smoking history.20 Similarly, in the NELSON screening study,
participants in the control arm were younger, were better
educated, were former smokers, and expressed a greater
willingness to quit smoking.21 Established evidence demonstrates that smoking is associated with having a lower
socioeconomic status, having lower education, being unmarried, and having poor health in part attributed to poorer
health-promoting activities.2,6 Participating in a LDCT program could reflect increased interest in quitting or
following a healthier lifestyle. Data suggest that nonparticipants express feelings of protection based on personal
or family health status, indicate that they are too old for
screening, and express sentiments of fatalism, all characteristics unlikely to compel action.22 Consideration must be
given as to how to recruit hard-to-reach populations.
When considering the effect of LDCT on smoking cessation, data suggest that participation in screening is associated with increased motivation to quit smoking.23 Although
NLST participants perceived a high risk of cancer as a result of
smoking, smokers doubted the benefits of quitting and had
low confidence in their ability to quit.24 A review of nine
cross-sectional, longitudinal and randomized controlled
studies examined the effect of CT screening for lung cancer
on smoking behaviors and found quit rates ranging from
6.6% to 42%.25 Investigators for the Danish Lung Cancer
Screening Trial found no difference between study arms in
smoking rates at 5 years, although the number of ex-smokers
increased significantly in both arms and a stated motivation
226
to quit correlated with actual cessation.26 Consistently

negative results from the Early Lung Cancer Action Program
demonstrated that baseline negative LDCT scan results were
associated with lower point abstinence rates at follow-up
compared with positive LDCT scan results, but they had no
significant effect on prolonged abstinence.27 However, NLST
participants with physicians who assisted with cessation and
arranged follow-up had a 40% to 46% higher rate of abstinence at 1 year, although physician rates of assisting and
arranging were low.28
There have been several thoughtful recent reviews on
addressing tobacco use in the context of cancer care.29-32
The principles of addressing tobacco use mimic evidence-based
PHS guidelines in the general population33 recommending the
use of behavioral counseling and pharmacotherapy to help
patients quit smoking. The 5As model (Ask, Assess, Advise,
Assist, Arrange) is a widely accepted method of addressing
tobacco use. The 2As and R (or 2As and C) model refers to Ask,
Advise, Refer or to Ask, Advise, Connect, respectively, in which
patients are referred or connected to evidence-based smoking
cessation support through a dedicated tobacco cessation
program. As reviewed in these references, patients with
cancer are often receptive to smoking cessation support
but suffer from similar patterns of tobacco use and risk of
relapse compared with the general population. What is
particularly unique to patients with cancer is that many
newly diagnosed patients will receive cancer treatment
requiring multiple visits to an oncology provider over a
relatively short period of time. As such, patients with
cancer have much more frequent contact with medical
providers than the general population, and this provides an
ideal platform to address tobacco use at multiple visits and
support patients with cancer in an effective quit attempt.29-31
The concept of a teachable moment is ideal, in which
patients with newly diagnosed cancer are faced with a lifechanging situation and providers can take advantage of this
opportunity to assist patients in a beneficial change in health
habits. Reviews above and the American Society for Clinical
Oncology tobacco resources34 discuss these issues in greater
detail.
Despite the clinical need and rationale to address tobacco
use, recent surveys demonstrate that although approximately 90% of oncologists ask about tobacco use and approximately 80% advise patients to quit smoking, only 30%
to 40% provide assistance with quitting.3,4 Given the evidence, one might question why many clinicians and oncologists do not readily incorporate tobacco cessation into
effective clinical practice. However, it is important to note
that approximately 90% reported believing that smoking
caused adverse outcomes for patients with cancer and that
smoking cessation should be a standard part of cancer care,
which suggests that efforts to convert physicians knowledge
base to a better understanding of the adverse effects of
smoking may not prove fruitful. A recent survey of LDCT
screening programs shows very similar findings, with 99%
reporting asking about tobacco use and 91% advising to quit,
but only 50% to 60% providing assistance with cessation and
38% discussing cessation medications.35 Analysis of predictive

barriers to providing cessation support suggests that a lack of
time, lack of cessation resources, and lack of education on
cessation methods are dominant barriers to providing cessation support.5 However, the concern that patients do not
want help with tobacco cessation has been refuted in studies
demonstrating that approximately 80% to 90% of patients with
cancer automatically referred to receive smoking cessation
support (an opt-out approach to tobacco cessation) are
receptive to intervention36,37 and patients who quit smoking
had significant improvements in survival.38 Collectively, these
data support a substantial need to provide evidence-based
cessation support to participants in LDCT screening programs
as well as to patients with cancer to manifest in the numerous
health benefits associated with smoking cessation. However,
despite the potential public health benefits for further reduction in tobacco-related morbidity and mortality, it is unknown whether screening and treatment centers will commit
the time and resources needed to ensure delivery of evidencebased tobacco cessation treatment.35
PRACTICAL CONSIDERATIONS TO IMPLEMENT

CHANGE
Simply stated, asking patients about tobacco use and advising them to quit is performed by most clinicians and is
necessary, but patients have higher smoking cessation rates
when providers help them quit smoking. Resources noted
above expound on common evidence-based methods to
address tobacco use and assist patients in quitting
smoking.2,14,29-34 Readers are encouraged to review these
resources as needed. However, implementation of
evidence-based care requires careful consideration of clinical resources, environment, and expertise. Several practical
considerations are important to consider when developing
approached to addressing tobacco use; unfortunately, the
evidence base is lacking as to the best approach. When
addressing tobacco use with LDCT or cancer care, it is important to consider the following:
What is the clinical setting? Consider opportunities to
integrate tobacco treatment with medical management
and engage patients in tobacco treatment efforts. Patients presenting for LDCT screening typically return
annually for repeat assessments unless an abnormal
finding requires further work-up. In this setting, LDCT
participants may be reticent to return for in-person
clinical support for tobacco cessation alone. In contrast, patients diagnosed with cancer often return frequently and multiple times for work-up, management,
and follow-up. Considering delivery of tobacco cessation
in coordination with repeated cancer care visits may be
helpful to both patients and providers.
What resources are available to assess and address
tobacco use? One size does not fit all. Larger institutions
may have more resources to support development of
dedicated tobacco cessation programs, whereas small
clinics and individual providers may need to provide
cessation support themselves or refer to existing community resources. Considering the location where cessation support is provided is important to consider.
Providing cessation support in a busy clinical setting
may be met with resistance, such as where patient wait
times for oncology care are often delayed because of
limited clinical space. This is particularly salient in clinics
where multiple clinicians use a single clinic room.
How will tobacco be assessed? Reviews above discuss
methods to assess tobacco. However, it is important to
consider how tobacco use assessments will be collected.
For instance, tobacco assessments collected by nurses
or other staff can permit identification of tobacco use
that can be presented to a physician prior to entering a
patient room. It also permits consideration of scheduling a
patient for cessation support while a physician is
interviewing a patient for LDCT screening or cancer care.
The use of electronic medical records or paper charts is
also important and clinic specific. Clinical infrastructure
and development of efficient assessment and treatment
systems can be critical to gaining the support of clinical
and administrative staff in a department. However, once a
patients tobacco use has been identified, all clinicians
should advise patients to quit smoking and either provide
cessation assistance or refer/connect patients to dedicated cessation support programs.
How will tobacco cessation support be delivered? The
basic elements of providing tobacco cessation support
include assessing a patients willingness to quit and developing an individualized tobacco cessation approach
that incorporates counseling, pharmacotherapy, and
follow-up. Specialized training is available. However,
cessation support can be delivered by a broad variety of
providers, including physicians, physician assistants, nurse
practitioners, nurses, pharmacists, psychologists, and
other providers who have received tobacco treatment
education. Advice to quit smoking should be repeatedly
delivered by all clinical staff. Support in the United States
and many other countries can also be delivered using
quitlines (1-800-QUIT-NOW in the United States) and
community cessation programs. In the experience of the
authors, many institutions may already have programs or
clinicians available that are often underutilized.
How will a tobacco cessation program be maintained? In
the experience of the authors, successful implementation of a tobacco cessation effort by clinicians or dedicated cessation programs includes buy-in and
participation from patients, clinical staff, providers, and
administration. Educating about the importance of tobacco across the clinic, cancer center, or institution can
increase awareness and help sustain a tobacco cessation
initiative. Considering issues related to billing and reimbursement can facilitate implementation in many
cases, but given the high cost associated with cancer
227
care, the cost savings associated with cessation by

helping prevent costs associated with increased cancer
treatment toxicity or progression to next-line cancer
therapy are likely substantial. Thus, perhaps billing and
reimbursement on the front end of clinical care should
not be the foremost concerns when implementing a
cessation approach. As noted above, most clinicians
fully support addressing tobacco but lack the education,
time, and resources to implement cessation support.5
CONCLUSION
Overall, the principles of tobacco use are relatively simple.
Tobacco causes adverse health effects and continued
tobacco use will continue to promote adverse health risks.

This simple relationship is easy to apply to a chronic disease
state such as cancer. Addressing tobacco by assessing tobacco
use and providing evidence-based tobacco cessation support
is extremely important across the continuum of cancer
screening, diagnosis, and treatment. Although many evidencebased approaches exist, most providers do not assist patients
with quitting smoking. However, smoking cessation is the best
method to reduce or eliminate the adverse health effects of
tobacco. Combining practical clinical considerations with
evidence-based tobacco cessation methods should result in
substantially increased access to cessation support and ultimately result in improved patient outcomes.
References
1. U.S. Department of Health and Human Services. Tobacco. www.
surgeongeneral.gov/priorities/tobacco/. Accessed February 1, 2016.
2. U.S. Department of Health and Human Services. The Health Consequences
of Smoking50 Years of Progress: A Report of the Surgeon General.
Atlanta, GA: U.S. Department of Health and Human Services, Centers for
Disease Control and Prevention, National Center for Chronic Disease
Prevention and Health Promotion, Office on Smoking and Health; 2014.
3. Warren GW, Marshall JR, Cummings KM, et al. Addressing tobacco use
in patients with cancer: a survey of American Society of Clinical Oncology members. J Oncol Pract. 2013;9:258-262.
4. Warren GW, Marshall JR, Cummings KM, et al; IASLC Tobacco Control
and Smoking Cessation Committee. Practice patterns and perceptions
of thoracic oncology providers on tobacco use and cessation in cancer
patients. J Thorac Oncol. 2013;8:543-548.
5. Warren GW, Dibaj S, Hutson A, et al. Identifying targeted strategies to
improve smoking cessation support for cancer patients. J Thorac Oncol.
2015;10:1532-1537.
6. American Cancer Society. The Tobacco Atlas. www.cancer.org/aboutus/
globalhealth/cancerandtobaccocontrolresources/the-tobacco-atlas.
7. Warren GW, Cummings KM. Tobacco and lung cancer: risks, trends, and
outcomes in patients with cancer. Am Soc Clin Oncol Educ Book. 2013;
33:359-364.
8. Thun MJ, Carter BD, Feskanich D, et al. 50-year trends in smokingrelated mortality in the United States. N Engl J Med. 2013;368:351-364.
9. Brandon TH, Goniewicz ML, Hanna NH, et al. Electronic nicotine delivery
systems: a policy statement from the American Association for Cancer
Research and the American Society of Clinical Oncology. J Clin Oncol.
2015;33:952-963.
10. Goldstraw P, Crowley J, Chansky K, et al; International Association for the
Study of Lung Cancer International Staging Committee; Participating Institutions. The IASLC Lung Cancer Staging Project: proposals for the
revision of the TNM stage groupings in the forthcoming (seventh)
edition of the TNM Classification of malignant tumours. J Thorac
Oncol. 2007;2:706-714.
11. Humphrey LL, Teutsch S, Johnson M; U.S. Preventive Services Task
Force. Lung cancer screening with sputum cytologic examination, chest
radiography, and computed tomography: an update for the U.S. Preventive Services Task Force. Ann Intern Med. 2004;140:740-753.
12. Aberle DR, Adams AM, Berg CD, et al; National Lung Screening Trial
Research Team. Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med. 2011;365:395-409.
228
13. National Comprehensive Cancer Center. NCCN Clinical Practice

Guidelines in Oncology (NCCN Guidelines): Lung Cancer Screening.
www.nccn.org/professionals/physician_gls/pdf/lung_screening.pdf.
14. National Comprehensive Cancer Center. NCCN Clinical Practice Guidelines
in Oncology (NCCN Guidelines): Smoking Cessation. www.nccn.org/
professionals/physician_gls/pdf/smoking.pdf. Accessed February 1, 2016.
15. Tanner NT, Kanodra NM, Gebregziabher M, et al. The association between smoking abstinence and mortality in the National Lung Screening
Trial. Am J Respir Crit Care Med. Epub 2015 Oct 26.
16. Black WC, Gareen IF, Soneji SS, et al. Cost-effectiveness of CT screening in
the National Lung Screening Trial. N Engl J Med. 2014;371:1793-1802.
17. Cornuz J, Gilbert A, Pinget C, et al. Cost-effectiveness of pharmacotherapies for nicotine dependence in primary care settings: a multinational comparison. Tob Control. 2006;15:152-159.
18. Goffin JR, Flanagan WM, Miller AB, et al. Cost-effectiveness of lung
cancer screening in Canada. JAMA Oncol. 2015;1:807-813.
19. Sobus SL, Warren GW. The biologic effects of cigarette smoke on cancer
cells. Cancer. 2014;120:3617-3626.
20. Tammemagi MC, Berg CD, Riley TL, et al. Impact of lung cancer screening
results on smoking cessation. J Natl Cancer Inst. 2014;106:dju084.
21. Yousaf-Khan U, Horeweg N, van der Aalst C, et al. Baseline characteristics and mortality outcomes of control group participants and
eligible non-responders in the NELSON lung cancer screening study.
J Thorac Oncol. 2015;10:747-753.
22. Patel D, Akporobaro A, Chinyanganya N, et al; Lung-SEARCH Investigators. Attitudes to participation in a lung cancer screening trial:
a qualitative study. Thorax. 2012;67:418-425.
23. Ostroff JS, Buckshee N, Mancuso CA, et al. Smoking cessation following CT
screening for early detection of lung cancer. Prev Med. 2001;33:613-621.
24. Park ER, Streck JM, Gareen IF, et al. A qualitative study of lung cancer
risk perceptions and smoking beliefs among national lung screening trial
participants. Nicotine Tob Res. 2014;16:166-173.
25. Poghosyan H, Kennedy Sheldon L, Cooley ME. The impact of computed
tomography screening for lung cancer on smoking behaviors:
a teachable moment? Cancer Nurs. 2012;35:466-475.
26. Ashraf H, Saghir Z, Dirksen A, et al. Smoking habits in the randomised
Danish Lung Cancer Screening Trial with low-dose CT: final results
after a 5-year screening programme. Thorax. 2014;69:574-579.
27. Anderson CM, Yip R, Henschke CI, et al. Smoking cessation and relapse
during a lung cancer screening program. Cancer Epidemiol Biomarkers
Prev. 2009;18:3476-3483.
28. Park ER, Gareen IF, Japuntich S, et al. Primary care provider-delivered smoking
cessation interventions and smoking cessation among participants in the
National Lung Screening Trial. JAMA Intern Med. 2015;175:1509-1516.
29. Warren GW, Sobus S, Gritz ER. The biological and clinical effects of
smoking by patients with cancer and strategies to implement evidencebased tobacco cessation support. Lancet Oncol. 2014;15:e568-e580.
30. Gritz ER, Toll BA, Warren GW. Tobacco use in the oncology setting: advancing
clinical practice and research. Cancer Epidemiol Biomarkers Prev. 2014;23:3-9.
31. Warren GW, Ward KD. Integration of tobacco cessation services into
multidisciplinary lung cancer care: rationale, state of the art, and future
directions. Transl Lung Cancer Res. 2015;4:339-352.
32. Karam-Hage M, Cinciripini PM, Gritz ER. Tobacco use and cessation for cancer
survivors: an overview for clinicians. CA Cancer J Clin. 2014;64:272-290.
33. Fiore MC, Jaen CR, Baker TB, et al. Treating Tobacco Use and Dependence: 2008 Update. Clinical Practice Guideline. Rockville, MD: U.S.
Public Health Service; 2008.
34. American Society of Clinical Oncology. Tobacco cessation and control

resources. www.asco.org/practice-research/tobacco-cessation-andcontrol-resources. Accessed February 1, 2016.
35. Ostroff JS, Copeland A, Borderud SP, et al. Readiness of lung cancer
screening sites to deliver smoking cessation treatment: current practices, organizational priority, and perceived barriers. Nicotine Tob Res.
Epub 2015 Sep 7.
36. Warren GW, Marshall JR, Cummings KM, et al. Automated tobacco
assessment and cessation support for cancer patients. Cancer. 2014;
120:562-569.
37. Tang MW, Oakley R, Dale C, et al. A surgeon led smoking cessation
intervention in a head and neck cancer centre. BMC Health Serv Res.
2014;14:636.
38. Dobson Amato KA, Hyland A, Reed R, et al. Tobacco cessation may
improve lung cancer patient survival. J Thorac Oncol. 2015;10:
1014-1019.
229
LUNG CANCER
Lung Cancer, Small Cell Lung

Cancer: On the Move (Again?)
CHAIR
Walter J. Curran Jr., MD
Winship Cancer Institute
Atlanta, GA
SPEAKERS
Solange Peters, MD, PhD
Lausanne University
Lausanne, Switzerland
Maria Catherine Pietanza, MD
New York, NY
NEW APPROACHES TO PATIENTS WITH SMALL CELL LUNG CANCER
Seeking New Approaches to Patients With Small Cell Lung

Cancer
Marie Catherine Pietanza, MD, Stefan Zimmerman, MD, Solange Peters, MD, PhD, and Walter J. Curran Jr., MD
OVERVIEW
The fundamental approach to the treatment of small cell lung cancer (SCLC) has not changed in the last several decades, with
most advances being restricted to improved radiation approaches. The standard first-line chemotherapy regimen in the
United States and Europe remains cisplatin or carboplatin plus etoposide in the treatment of limited stage (LS-SCLC) and
extensive stage (ES-SCLC) disease. Radiation therapy is administered to those patients with LS-SCLC, whose cancer is
confined to the chest in a single tolerable radiation field. This article will summarize a number of exciting observations
regarding the biology of SCLC and how a deeper understanding of newly integrated targets and target pathways may lead to
new and better therapeutic approaches in the near future.
he fundamental approach to the treatment of SCLC has

not changed in the last several decades, with most advances being restricted to improved radiation approaches.
The standard first-line chemotherapy regimen in the United
States and Europe remains cisplatin or carboplatin plus
etoposide in the treatment of LS-SCLC and ES-SCLC disease.
Radiation therapy is administered to those patients with
LS-SCLC, whose cancer is confined to the chest in a single
tolerable radiation field: data support initiation of radiation
during cycle one or two and use of hyperfractionated radiation therapy,1-8 though there is an ongoing large national
cooperative group study addressing the question of standard hyperfractionation versus a higher total dose radiation
using modern radiation techniques (NCT00632853). More
recently, thoracic radiation therapy is being extended to
patients with ES-SCLC after response to chemotherapy, as a
phase III trial showed considerable reduction in intrathoracic recurrence and in 2-year overall survival.9 Prophylactic cranial irradiation is recommended to patients with
LS-SCLC and ES-SCLC demonstrating a response to frontline platinum-based therapy, which has been shown to
decrease the risk of intracranial recurrence and improve
overall survival.10,11
Although the response rates to first-line treatment are as
high as 60%80% in SCLC, cure occurs only in approximately
20% of patients, restricted to those with LS-SCLC.12 Essentially all patients with ES-SCLC, and the majority of patients
with LS-SCLC, suffer relapse within months of completing
initial therapy. Depending on the length of time a response
is maintained, which is the strongest predictor of outcome,
patients are defined as having sensitive or refractory disease.

Those patients who maintain a response to initial treatment
of 3 months or longer are defined as having sensitive disease;
additional chemotherapy yields response rates of 25% and
median survival from the time of relapse of approximately
6 months. Patients with refractory disease are those who
either have no response to initial therapy or progress within
3 months; these patients rarely benefit from additional
treatment, with response rates less than 10% and median
survival of 4 months.
Only one agent has been approved by the U.S. Food and
Drug Administration for recurrent or progressive SCLC:
topotecan.13-15 Unfortunately, for patients with SCLC whose
disease has progressed after first- and second-line treatments, there are no accepted regimens. Although investigators have studied a wide range of rationally based
therapies in the hopes of improving outcomes, results have
not been favorable (reviewed in Pietanza et al16).
However, more recent insights generated from genomic
and proteomic studies, as well as pathway-specific investigations, as indicated above, may lead to the development of
effective therapies, which are critically needed in SCLC.
POTENTIALLY TARGETABLE SCLC MOLECULAR

ALTERATIONS
The MYC gene family members of transcription factors,
which are contributors to oncogenesis, have been described
in approximately 20% of SCLCs.17 As stated above, previous
efforts to inhibit MYC activity proved to be difficult; however,
From the Merck Research Laboratories, Rahway, NJ; Department of Oncology, HFR Fribourg, Fribourg, Switzerland; Department of Oncology, Centre Hospitalier Universitaire Vaudois,
Lausanne, Switzerland; Winship Cancer Institute of Emory University, Atlanta, GA.
Corresponding author: Walter J. Curran Jr., MD, Winship Cancer Institute of Emory University, 1365 Clifton Rd. NE, Room A1356, Atlanta, GA 30322; email: curran04@gmail.com.
e477
PIETANZA ET AL
further understanding of the biology of the disease has

revealed specific mechanisms by which MYC can be potentially targeted. MYC is a transcriptional regulator of
aurora kinases A and B, which, in the absence of p53,
provide a growth advantage,18-21 and as such, favorable
antitumor activity may be demonstrated by inhibiting aurora
kinases.22 Multiple preclinical studies have suggested that
SCLCs with MYC alterations are sensitive to aurora kinase
inhibitors.18,22,23 Alisertib, a selective aurora kinase A
inhibitor, has been evaluated in a phase II clinical trial of
patients with recurrent or progressive SCLC and demonstrated a response rate of 21% among patients with both
sensitive and refractory disease.24 To augment these results,
there is an ongoing clinical trial evaluating paclitaxel with or
without alisertib for the second-line treatment of patients
with SCLC (NCT02038647), as aurora kinase A has a key role in
mitotic spindle assembly.
Inhibiting key regulators of MYC transcription may be
another potential mechanism in SCLC. MYC transcription in
SCLC and other tumors is regulated by the bromodomain and
extraterminal (BET) family of bromodomain-containing
proteins, which bind acetylated lysines on histone tails
and recruit key transcriptional proteins.25,26 Inhibiting BET
has been shown to prevent interaction of BET proteins with
acetylated histones and halt assembly of an active gene
transcriptional complex, leading to focal chromatin remodeling and decreased inhibition of c-Myc. GSK525762 is a
small-molecule inhibitor of BET being evaluated in a phase I
clinical trial that includes patients with SCLC (NCT01587703).
The genomic studies completed in SCLC thus far have
found amplification of fibroblast growth factor receptor
(FGFR) 1 in approximately 6% of cases,27-29 and sensitivity to
FGFR inhibitors has been described in some, but not all, SCLC
tumors.22 Although the magnitude to which this subset of
SCLC is dependent on the FGFR pathway remains unknown,
there are clinical studies evaluating drugs targeting the
FGFR family members for patients with SCLC: including JNJ
KEY POINTS
e478
SCLC is an aggressive malignancy, uniformly linked to

tobacco use, and has a unique biology.
The fundamental approach to the treatment of SCLC has
not changed in nearly 4 decades and incorporates use of
a platinum doublet and, when possible, radiation
therapy.
As the biology of this disease is elucidated, multiple
targets are being identified and new trials have been
developed.
The MYC gene family of transcription factors may be
a targetable entity to test new therapies against, in
particular, aurora kinase inhibitors.
Epigentic processes, such as gene promoter methylation
and histone acetylation, may be candidate targets to
test in SCLC.
42756493 (a pan-FGFR inhibitor; NCT01703481),30 BIBF1120

(a multitargeted drug that inhibits FGFR, vascular endothelial growth factor receptor, and platelet-derived growth
factor receptor; NCT02152059); and lucitinib (a potent
inhibitor of FGFR1/2, vascular endothelial growth factor
receptor 13, and platelet-derived growth factor receptor
a/b; NCT02109016).
PROBING THE EPIGENOME

Epigenetic processes, such as gene promoter methylation
and histone acetylation, are known to be dysregulated in
SCLC and lead to alterations in chromatin and other associated factors that modify the ability of genes to be transcribed.31 Acetyltransferases and histone deacetylases
regulate histone acetylation, which leads to increased accessibility of promoter regions and increased transcription
of genes.32,33 Vorinostat and belinostat, which are histone
deacetylase inhibitors, have synergistic activity when added
to topotecan and cisplatin/etoposide, respectively.34,35
Clinical trials investigating the combination of vorinostat
(NCT00702962) and belinostat (NCT00926640) with platinum and etoposide in the first-line treatment of patients
with ES-SCLC are in progress.
DNA REPAIR
SCLC has been characterized by aberrant expression of
DNA repair proteins, including O6 alkyl-guanine, DNA
alkyltransferase (MGMT), poly (ADP-ribose) polymerase-1
(PARP-1), checkpoint kinase-1 (Chk-1), BRCA-1, and RAD51.36,37
Therefore, multiple DNA repair pathways represent attractive
targets in SCLC.
Temozolomide is an oral alkylating agent with improved
outcomes in the presence of MGMT promoter hypermethylation.38,39 In a phase II clinical trial of patients with
relapsed sensitive or refractory SCLC, temozolomide demonstrated an overall response rate of 20%, and responses
were noted in patients in need of third-line treatment and in
those with brain metastases.39 These findings led to the
addition of temozolomide to compendia of agents recommended for use in the treatment of SCLC.40
PARP activity is essential for the repair of single-stranded
DNA breaks through the base excision repair pathway.41,42 In
preclinical SCLC models, PARP inhibitors have demonstrated
activity whether evaluated alone or in combination with
DNA-damaging agents, including temozolomide.37,43-46
Talazoparib (BMN673), a PARP inhibitor that both mediates
catalytic PARP1 inhibition and traps PARP1 and PARP2 enzymes at damaged DNA sites,47 has shown single-agent activity
in patients with sensitive relapsed SCLC, with a 10% overall
response rate and 25% clinical benefit response rate.48 The
PARP inhibitors, talazoparib, veliparib, and olaparib, are being
evaluated in combination with etoposide/platinum or temozolomide for patients with newly diagnosed or relapsed SCLC,
respectively (NCT01286987, NCT01642251, NCT02289690,
NCT01638546, NCT02446704, and NCT02049593). Phase I
data reveal that veliparib can safely be administered with

etoposide and platinum in patients with untreated SCLC.49
Among the multiple functions of Chk-1 are activation of
S and G2/M phase checkpoints, preventing entry into mitosis, and coordination of homologous repair.50,51 Additional
proteins necessary at the G2/M checkpoint include ataxia
telangiectasia and Rad3-related (ATR) and Wee-1. Inhibition
of Chk-1, ATR, or Wee-1 leads to G2 checkpoint abrogation
and, in p53-negative tumors, such as SCLC, sensitizes the
cancer cells to DNA-damaging agents, leading to mitotic catastrophe and cell death.50 The Chk-1 inhibitors LY2606368
and AZD7762 have demonstrated activity when administered
alone or with olaparib and cisplatin in preclinical SCLC
models.50,52,53 Numerous inhibitors of Chk-1 (LY2606368 and
AZD7762), ATR (AZD6738, VX970, and Wee-1; AZD1775) are in
clinical development, and two phase I trials are actively
recruiting patients with SCLC: topotecan with VX970 in
ES-SCLC (NCT02487095) and AZD1775 with olaparib in refractory solid tumors (NCT02511795).
THE NOTCH DEVELOPMENTAL PATHWAY

The Notch pathway is essential in early lung development
and regulates stem cell self-renewal; thus, when abnormally
activated, it can cause neoplastic proliferation, representing
an early event in tumorigenesis.54-56 Recent comprehensive
genomic analyses have found that the NOTCH family genes
are mutated in 25% of human SCLC. Thus, this pathway is a
potential target in SCLC. Importantly, however, Notch signaling can have oncogenic or tumor-suppressive effects
depending on the cellular content and can influence multiple
other oncogenic pathways.57
Overexpression of Notch2 and Notch3 receptors and
target genes has been noted in SCLC. In preclinical SCLC
models, tarextumab (OMP-59R5), which is a fully human
monoclonal antibody that selectively inhibits Notch2 and
Notch3 receptor function, has demonstrated delayed tumor recurrence following the discontinuation of chemotherapy, as well as a reduction in cancer stem cell frequency
and tumorigenicity.58 A phase I study of tarextumab with
etoposide/platinum in patients with ES-SCLC has been
completed,59 and a randomized phase II study is ongoing
(NCT01859741).
ANTIBODY-DRUG CONJUGATES
Antibody-drug conjugates, composed of an antibody directed to a well-characterized antigen on cancer cells, a
linker, and cytotoxic agent (payload), represent an effective mechanism of targeted drug delivery, resulting in
decreased toxicity and improved therapeutic index.60
Rovalpituzumab tesirine exploits the Notch pathway as the
humanized monoclonal antibody portion targets the cell
surface available Notch ligand delta-like protein 3 (DLL3),
which is overexpressed in SCLC tumor-initiating cells but
not in normal tissue.61 The phase I trial of rovalpituzumab
tesirine enrolled 73 recurrent patients with SCLC in need
of second- or third-line treatment and demonstrated a
response rate of 23% in those evaluable.62 Notably, among

the 27 patients with confirmed DLL3-high tumor expression,
the response rates were 44% and 45% in the second- and
third-line settings, respectively, which were durable.62
A phase II trial evaluating rovalpituzumab tesirine for thirdline treatment of patients with SCLC with DLL3-expressing
tumors (TRINITY; NCT02674568) has begun accrual and, if
positive, will fulfill an unmet need, as there are no approved agents in this setting for this malignancy.
A phase II study evaluating sacituzumab govitecan
(IMMU-132), an antibody-drug conjugate comprising SN-38,
the active metabolite of the topoisomerase I inhibitor,
irinotecan, conjugated to an antiTrop-2 humanized antibody has been enrolling patients with previously treated
SCLC and non-SCLC (NCT01631552). Interim results have
shown a response rate of 30% in 20 evaluable patients with
SCLC who had received a median of 2.5 prior lines of therapy
and median progression-free survival of 2.4 months.63 As
Trop-2 is a human trophoblast cell-surface glycoprotein
overexpressed in multiple epithelial tumors, including SCLC,
this antibody drug conjugate is being studied in other malignancies as well.
IMMUNOTHERAPY
Modulating the immune response may represent another
treatment modality for patients with SCLC, as supported by
clinical and preclinical data indicated above (and reviewed
in Pietanza et al64): the disease is associated with immunogenic effects; PD-L1 expression on SCLC tumors correlates with the presence of tumor-infiltrating lymphocytes;
and the malignancy is nearly entirely related to smoking
and characterized by an elevated mutation burden, which
are known to be markers of response to immune checkpoint inhibitors.65-71
Patients with untreated stage IIIB/IV NSCLC or ES-SCLC
were administered paclitaxel and carboplatin alone or with
the addition of ipilimumab, a humanized IgG1 monoclonal
antibody against CTLA-4, given on two dosing schedules, in a
randomized, double-blind, three-arm phase II trial.72 Based
on the results from the patient with SCLC cohort that received the phased dosing schedule of ipilimumab, administered with cycle three of paclitaxel and carboplatin, who
appeared to have improved immune-related progressionfree survival compared with those treated in the other two
arms,73 a randomized, multicenter, double-blind phase III
trial comparing the efficacy of platinum/etoposide with
or without ipilimumab in patients with newly diagnosed
ES-SCLC, with OS as the primary endpoint, has completed
accrual, and results are anticipated (NCT01450761).
Nivolumab, with and without ipilimumab, has been
evaluated in heavily pretreated patients with relapsed
SCLC as part of a phase I trial evaluating the agents in
various tumor types (NCT01928394). Early results indicate response rates of 13% and 31%, when nivolumab
is administered alone or with ipilimumab, respectively,
regardless of PD-L1 tumor expression, previous lines of
e479
PIETANZA ET AL
therapy, and platinum sensitivity.74 Notably, the responses are durable, with 1-year overall survival of 27%
and 48% for patients treated with nivolumab and nivolumab with ipilumumab, respectively.74 These results
have led to the development of two phase III studies
among patients with SCLC evaluating nivolumab, nivolumab with ipilimumab, or placebo (NCT02538666) in the
maintenance setting, in which there is no defined treatment modality, and nivolumab compared with chemotherapy for those in need of second-line treatment
(NCT02481830).
As part of a phase IB multicohort study, pembrolizumab
has been evaluated among patients with relapsed SCLC with
PD-L1positive tumors (NCT02054806). The response rate
was 29%, with a median duration of response of 29 weeks.75
There are several ongoing trials of pembrolizumab in patients
with SCLC, including a phase II study of the agent as maintenance therapy after the completion of standard, first-line
chemotherapy in extensive-stage disease (NCT02359019),
a phase I trial evaluating pembrolizumab with concurrent
chemoradiation therapy (NCT02402920), and others investigating the drug alone (NCT02628067), with various
chemotherapy agents (NCT02551432, NCT02580994, and
NCT02331251), or with novel biologics (NCT02661100).
Finally, durvalumab (MEDI4736, the humanized IgG1k

monoclonal antibody directed against PD-L1) is being evaluated alone (NCT01693562) and with tremelimumab, a humanized IgG2 monoclonal CTLA-4 antibody (NCT02658214)
among patients with relapsed SCLC.
As seen, the immune system is a reasonable target in SCLC,
with promising results from early-phase trials. Currently,
multiple approaches are being developed to use immune
checkpoint inhibitors in an attempt to best exploit this
treatment modality in SCLC.
CONCLUSION
SCLC has a unique biology, driven by multiple aberrant
pathways and mutations, and distinct clinical features. The
mechanisms that lead to the shift from initial therapeutic
sensitivity to ultimate therapeutic resistance are not well
understood. Recent genomic, proteomic, and preclinical
studies have identified novel therapeutic strategies currently being evaluated in clinical trials. Ongoing efforts to
collect and analyze samples in the setting of these studies,
with use of improved research tools, will allow us to continue
to develop rational clinical trials with the potential to advance the field.
References
1. Amini A, Byers LA, Welsh JW, et al. Progress in the management of
limited-stage small cell lung cancer. Cancer. 2014;120:790-798.
2. Spiro SG, James LE, Rudd RM, et al; London Lung Cancer Group. Early
compared with late radiotherapy in combined modality treatment for
limited disease small-cell lung cancer: a London Lung Cancer Group
multicenter randomized clinical trial and meta-analysis. J Clin Oncol.
2006;24:3823-3830.
3. Fried DB, Morris DE, Poole C, et al. Systematic review evaluating the
timing of thoracic radiation therapy in combined modality therapy for
limited-stage small-cell lung cancer. J Clin Oncol. 2004;22:4837-4845.
4. De Ruysscher D, Pijls-Johannesma M, Bentzen SM, et al. Time between
the first day of chemotherapy and the last day of chest radiation is the
most important predictor of survival in limited-disease small-cell lung
5. De Ruysscher D, Pijls-Johannesma M, Vansteenkiste J, et al. Systematic
review and meta-analysis of randomised, controlled trials of the timing
of chest radiotherapy in patients with limited-stage, small-cell lung
cancer. Ann Oncol. 2006;17:543-552.
6. Pijls-Johannesma MC, De Ruysscher D, Lambin P, et al. Early versus late
chest radiotherapy for limited stage small cell lung cancer. Cochrane
Database Syst Rev. 2005;1:CD004700.
7. Huncharek M, McGarry R. A meta-analysis of the timing of chest irradiation in the combined modality treatment of limited-stage small cell
lung cancer. Oncologist. 2004;9:665-672.
8. Turrisi AT III, Kim K, Blum R, et al. Twice-daily compared with oncedaily thoracic radiotherapy in limited small-cell lung cancer treated
concurrently with cisplatin and etoposide. N Engl J Med. 1999;340:
265-271.
9. Slotman BJ, van Tinteren H, Praag JO, et al. Use of thoracic radiotherapy
for extensive stage small-cell lung cancer: a phase 3 randomised
controlled trial. Lancet. 2015;385:36-42.
e480
10. Auperin A, Arriagada R, Pignon JP, et al; Prophylactic Cranial Irradiation

Overview Collaborative Group. Prophylactic cranial irradiation for patients with small-cell lung cancer in complete remission. N Engl J Med.
1999;341:476-484.
11. Slotman B, Faivre-Finn C, Kramer G, et al; EORTC Radiation Oncology
Group and Lung Cancer Group. Prophylactic cranial irradiation in extensive small-cell lung cancer. N Engl J Med. 2007;357:664-672.
12. Simon GR, Turrisi A. American College of Chest P. Management of small
cell lung cancer: ACCP evidence-based clinical practice guidelines (2nd
edition). Chest. 2007;132:324S-339S (suppl 3).
13. von Pawel J, Schiller JH, Shepherd FA, et al. Topotecan versus cyclophosphamide, doxorubicin, and vincristine for the treatment of recurrent
small-cell lung cancer. J Clin Oncol. 1999;17:658-667.
14. OBrien ME, Ciuleanu TE, Tsekov H, et al. Phase III trial comparing
supportive care alone with supportive care with oral topotecan in
patients with relapsed small-cell lung cancer. J Clin Oncol. 2006;24:
5441-5447.
15. Eckardt JR, von Pawel J, Pujol JL, et al. Phase III study of oral compared
with intravenous topotecan as second-line therapy in small-cell lung
16. Pietanza MC, Krug LM, Wu A, et al. Small cell and neuroendocrine
tumors of the lung. In DeVita VT, Lawrence TS, Rosenberg SA (eds).
DeVita, Hellman, and Rosenbergs Cancer: Principles and Practice of
Oncology. 10th ed. Philadelphia: Lippincott, Williams, and Wilkins;
2015;536-559.
17. Hwang DH, Sun H, Rodig SJ, et al. Myc protein expression correlates with
MYC amplification in small-cell lung carcinoma. Histopathology. 2015;
67:81-89.
18. Brockmann M, Poon E, Berry T, et al. Small molecule inhibitors of
aurora-a induce proteasomal degradation of N-myc in childhood
neuroblastoma. Cancer Cell. 2013;24:75-89.
19. den Hollander J, Rimpi S, Doherty JR, et al. Aurora kinases A and B are
up-regulated by Myc and are essential for maintenance of the malignant state. Blood. 2010;116:1498-1505.
20. Lu LY, Wood JL, Ye L, et al. Aurora A is essential for early embryonic development and tumor suppression. J Biol Chem. 2008;283:
31785-31790.
21. Vader G, Lens SM. The Aurora kinase family in cell division and cancer.
Biochim Biophys Acta. 2008;1786:60-72.
22. Sos ML, Dietlein F, Peifer M, et al. A framework for identification of
actionable cancer genome dependencies in small cell lung cancer. Proc
Natl Acad Sci USA. 2012;109:17034-17039.
23. Hook KE, Garza SJ, Lira ME, et al. An integrated genomic approach to
identify predictive biomarkers of response to the aurora kinase inhibitor
PF-03814735. Mol Cancer Ther. 2012;11:710-719.
24. Melichar B, Adenis A, Lockhart AC, et al. Safety and activity of alisertib,
an investigational aurora kinase A inhibitor, in patients with breast
cancer, small-cell lung cancer, non-small-cell lung cancer, head and neck
squamous-cell carcinoma, and gastro-oesophageal adenocarcinoma:
a five-arm phase 2 study. Lancet Oncol. 2015;16:395-405.
25. Helin K, Dhanak D. Chromatin proteins and modifications as drug
targets. Nature. 2013;502:480-488.
26. Delmore JE, Issa GC, Lemieux ME, et al. BET bromodomain inhibition as a
therapeutic strategy to target c-Myc. Cell. 2011;146:904-917.
27. Rudin CM, Durinck S, Stawiski EW, et al. Comprehensive genomic
analysis identifies SOX2 as a frequently amplified gene in small-cell lung
cancer. Nat Genet. 2012;44:1111-1116.
28. Peifer M, Fernandez-Cuesta L, Sos ML, et al. Integrative genome analyses identify key somatic driver mutations of small-cell lung cancer.
Nat Genet. 2012;44:1104-1110.
29. George J, Lim JS, Jang SJ, et al. Comprehensive genomic profiles of small
cell lung cancer. Nature. 2015;524:47-53.
30. Tabernero J, Bahleda R, Dienstmann R, et al. Phase I dose-escalation
study of JNJ-42756493, an oral pan-fibroblast growth factor receptor
inhibitor, in patients with advanced solid tumors. J Clin Oncol. 2015;33:
3401-3408.
31. Jones PA, Baylin SB. The epigenomics of cancer. Cell. 2007;128:683-692.
32. Gregory PD, Wagner K, Horz W. Histone acetylation and chromatin
remodeling. Exp Cell Res. 2001;265:195-202.
33. Lane AA, Chabner BA. Histone deacetylase inhibitors in cancer therapy.
J Clin Oncol. 2009;27:5459-5468.
34. Bruzzese F, Rocco M, Castelli S, et al. Synergistic antitumor effect
between vorinostat and topotecan in small cell lung cancer cells is
mediated by generation of reactive oxygen species and DNA damageinduced apoptosis. Mol Cancer Ther. 2009;8:3075-3087.
35. Luchenko VL, Salcido CD, Zhang Y, et al. Schedule-dependent synergy of
histone deacetylase inhibitors with DNA damaging agents in small cell
lung cancer. Cell Cycle. 2011;10:3119-3128.
36. Toyooka S, Toyooka KO, Maruyama R, et al. DNA methylation profiles of
lung tumors. Mol Cancer Ther. 2001;1:61-67.
37. Byers LA, Wang J, Nilsson MB, et al. Proteomic profiling identifies
dysregulated pathways in small cell lung cancer and novel therapeutic
targets including PARP1. Cancer Discov. 2012;2:798-811.
38. Hegi ME, Diserens AC, Gorlia T, et al. MGMT gene silencing and
benefit from temozolomide in glioblastoma. N Engl J Med. 2005;352:
997-1003.
39. Pietanza MC, Kadota K, Huberman K, et al. Phase II trial of temozolomide
in patients with relapsed sensitive or refractory small cell lung cancer,
with assessment of methylguanine-DNA methyltransferase as a potential biomarker. Clin Cancer Res. 2012;18:1138-1145.
40. Kalemkerian GP, Loo BW, Akerley W, et al. NCCN Guidelines Version 1.2016
Small Cell Lung Cancer. http://www.nccn.org/professionals/physician_gls/
recently_updated.asp. Accessed April 7, 2016.
41. Schreiber V, Ame JC, Dolle P, et al. Poly(ADP-ribose) polymerase-2

(PARP-2) is required for efficient base excision DNA repair in association
with PARP-1 and XRCC1. J Biol Chem. 2002;277:23028-23036.
42. Memisoglu A, Samson L. Base excision repair in yeast and mammals.
Mutat Res. 2000;451:39-51.
43. Tamborini E, Bonadiman L, Negri T, et al. Detection of overexpressed
and phosphorylated wild-type kit receptor in surgical specimens of
small cell lung cancer. Clin Cancer Res. 2004;10:8214-8219.
44. Palma JP, Wang YC, Rodriguez LE, et al. ABT-888 confers broad in vivo
activity in combination with temozolomide in diverse tumors. Clin
Cancer Res. 2009;15:7277-7290.
45. Cardnell RJ, Feng Y, Diao L, et al. Proteomic markers of DNA repair and
PI3K pathway activation predict response to the PARP inhibitor BMN
673 in small cell lung cancer. Clin Cancer Res. 2013;19:6322-6328.
46. Feng Y, Post LE, Cardnell R, et al. BMN 673 as single agent and in
combination with temozolomide or PI3K pathway inhibitors in small cell
lung cancer and gastric cancer models. Eur J Cancer. 2014;50:81.
47. Murai J, Huang SY, Das BB, et al. Trapping of PARP1 and PARP2 by
Clinical PARP Inhibitors. Cancer Res. 2012;72:5588-5599.
48. Wainberg Z, Rafii S, Ramanathan R. Safety and antitumor activity of the
PARP inhibitor BMN673 in a phase 1 trial recruiting metastatic small-cell
lung cancer (SCLC) and germline BRCA-mutation carrier cancer patients.
49. Owonikoko TK, Dahlberg SE, Khan SA, et al. A phase 1 safety study of
veliparib combined with cisplatin and etoposide in extensive stage small
cell lung cancer: A trial of the ECOG-ACRIN Cancer Research Group
(E2511). Lung Cancer. 2015;89:66-70.
50. Syljuasen RG, Hasvold G, Hauge S, et al. Targeting lung cancer through
inhibition of checkpoint kinases. Front Genet. 2015;6:70.
51. Maugeri-Saccà M, Bartucci M, De Maria R. Checkpoint kinase 1 inhibitors for potentiating systemic anticancer therapy. Cancer Treat Rev.
2013;39:525-533.
52. Valliani AA, Sen T, Masrorpour F, et al. Check point kinase 1 targeting
as a novel therapeutic strategy in small cell lung cancer. Cancer Res.
2015;75 (suppl 15; abstr 5316).
53. Hsu W-H, Hsu S-T, Rao G, et al. Chk1 Inhibition Enhances Cisplatin
Cytotoxicity Regardless of p53 Status in Human Small Cell Lung Cancer
Cells. J Thorac Oncol. 2015;10:S399 (suppl 2).
54. Peacock CD, Watkins DN. Cancer stem cells and the ontogeny of lung
55. Pardal R, Clarke MF, Morrison SJ. Applying the principles of stem-cell
biology to cancer. Nat Rev Cancer. 2003;3:895-902.
56. Rubin LL, de Sauvage FJ. Targeting the Hedgehog pathway in cancer. Nat
Rev Drug Discov. 2006;5:1026-1033.
57. Takebe N, Nguyen D, Yang SX. Targeting notch signaling pathway in
cancer: clinical development advances and challenges. Pharmacol Ther.
2014;141:140-149.
58. Yen W-C, Fischer MM, Axelrod F, et al. Targeting Notch signaling with a
Notch2/Notch3 antagonist (tarextumab) inhibits tumor growth and
decreases tumor-initiating cell frequency. Clin Cancer Res. 2015;21:
2084-2095.
59. Pietanza MC, Spira AI, Jotte RM, et al. Final results of phase Ib of
tarextumab (TRXT, OMP-59R5, anti-Notch2/3) in combination with
etoposide and platinum in patients with untreated extensive-stage
small-cell lung cancer. J Clin Oncol. 2015;33 (suppl; abstr 7508).
60. Zolot RS, Basu S, Million RP. Antibody-drug conjugates. Nat Rev Drug
Discov. 2013;12:259-260.
61. Saunders LR, Bankovich AJ, Anderson WC, et al. A DLL3-targeted
antibody-drug conjugate eradicates high-grade pulmonary neuroendocrine tumor-initiating cells in vivo. Sci Transl Med. 2015;7:302ra136.
62. Pietanza MC, Spigel D, Bauer TM, et al. Safety, activity, and response
durability assessment of single agent rovalpituzumab tesirine, a
e481
PIETANZA ET AL
63.
64.
65.
66.
67.
68.
69.
delta-like protein 3-targeted antibody drug conjugate, in small cell lung

cancer. Eur J Cancer. 2015;51:S712 (suppl 3).
Camidge DR, Starodub AN, Ocean A, et al. Therapy of advanced metastatic lung cancers with an anti-Trop-2-SN-38 antibody-drug conjugate, IMMU-132: interim phase II clinical results. J Thorac Oncol. 2015;
10:S173 (suppl 2).
Pietanza MC, Byers LA, Minna JD, et al. Small cell lung cancer: will recent
progress lead to improved outcomes? Clin Cancer Res. 2015;21:
2244-2255.
Graus F, Dalmou J, Ren~e R, et al. Anti-Hu antibodies in patients with
small-cell lung cancer: association with complete response to therapy
and improved survival. J Clin Oncol. 1997;15:2866-2872.
Schultheis AM, Scheel AH, Ozretic L, et al. PD-L1 expression in small cell
neuroendocrine carcinomas. Eur J Cancer. 2015;51:421-426.
Soria J-C, Marabelle A, Brahmer JR, et al. Immune checkpoint modulation
for non-small cell lung cancer. Clin Cancer Res. 2015;21:2256-2262.
Hellmann MD, Creelan BC, Woo K, et al. Smoking history and response to
nivolumab in patients with advanced NSCLCs. Ann Oncol. 2014;25:iv429.
Garon EB, Rizvi NA, Hui R, et al; KEYNOTE-001 Investigators.
Pembrolizumab for the treatment of non-small-cell lung cancer. N Engl J
Med. 2015;372:2018-2028.
e482
70. Herbst RS, Soria JC, Kowanetz M, et al. Predictive correlates of response
to the anti-PD-L1 antibody MPDL3280A in cancer patients. Nature.
2014;515:563-567.
71. Rizvi NA, Hellmann MD, Snyder A, et al. Cancer immunology. Mutational
cancer. Science. 2015;348:124-128.
72. Lynch TJ, Bondarenko I, Luft A, et al. Ipilimumab in combination with
paclitaxel and carboplatin as first-line treatment in stage IIIB/IV nonsmall-cell lung cancer: results from a randomized, double-blind, multicenter phase II study. J Clin Oncol. 2012;30:2046-2054.
73. Reck M, Bondarenko I, Luft A, et al. Ipilimumab in combination with
paclitaxel and carboplatin as first-line therapy in extensive-diseasesmall-cell lung cancer: results from a randomized, double-blind, multicenter phase 2 trial. Ann Oncol. 2013;24:75-83.
74. Calvo E, Lopez-Martin JA, Bendell J, et al. Nivolumab monotherapy or in
combination with ipilimumab for treatment of recurrent small cell lung
cancer. Eur J Oncol. 2015;51:S633 (suppl 3).
75. Ott PA, Elez E, Hiret S, et al. Pembrolizumab for extensive stage disease
SCLC: efficacy and relationship with PD-L1 expression abstr 3285.
Presented at: 16th World Conference on Lung Cancer; September 2015;
Denver, CO.
LUNG CANCER
Treatment of Lung Cancer in

Medically Compromised Patients
CHAIR
Jeffrey Crawford, MD
Durham, NC
SPEAKERS
Josephine Louella Feliciano, MD
University of Maryland Greenebaum Cancer Center
Baltimore, MD
Paul Wheatley-Price, MBChB, FRCP, MD
Ottawa Hospital Research Institute
Ottawa, ON, Canada
CRAWFORD, WHEATLEY-PRICE, AND FELICIANO

Jeffrey Crawford, MD, Paul Wheatley-Price, MBChB, FRCP, MD, and Josephine Louella Feliciano, MD
OVERVIEW
Outcomes for patients with lung cancer have been improved substantially through the integration of surgery, radiation, and
systemic therapy for patients with early-stage disease. Meanwhile, advances in our understanding of molecular mechanisms
have substantially advanced our treatment of patients with advanced lung cancer through the introduction of targeted
therapies, immune approaches, improvements in chemotherapy, and better supportive care. However, the majority of these
advances have occurred among patients with good functional status, normal organ function, and with the social and
economic support systems to be able to benefit most from these treatments. The aim of this article is to bring greater
attention to management of lung cancer in patients who are medically compromised, which remains a major barrier to care
delivery. Impaired performance status is associated with poor outcomes and correlates with the high prevalence of cachexia
among patients with advanced lung cancer. CT imaging is emerging as a research tool to quantify muscle loss in patients with
cancer, and new therapeutics are on the horizon that may provide important adjunctive therapy in the future. The benefits of
cancer therapy for patients with organ failure are poorly understood because of their exclusion from clinical trials. The
availability of targeted therapy and immunotherapy may provide alternatives that may be easier to deliver in this population, but clinical trials of these new agents in this population are vital. Patients with lower socioeconomic status are
disproportionately affected by lung cancer because of higher rates of tobacco addiction and the impact of socioeconomic
status on delay in diagnosis, treatment, and outcomes. For all patients who are medically compromised with lung cancer,
multidisciplinary approaches are particularly needed to evaluate these patients and to incorporate rapidly changing
therapeutics to improve outcomes.
ince the pioneering work of Karnofsky, the relationship

between impaired functional status and poor outcome
in patients with lung cancer has been well established.1 For
patients with advanced cancer, a grading system including
both body mass index and percent weight loss is highly
predictive of survival outcomes across multiple tumor types
including lung cancer. 2 Weight loss per se is the result of
the process of cancer cachexia, which has been defined
as a multifactorial syndrome associated with ongoing loss
of skeletal muscle mass (with or without fat mass), which
cannot be fully reversed through conventional nutritional
support and leads to progressive functional impairment.3
This definition has been very useful as a framework for
better understanding of cachexia, both at the molecular and
clinical levels, with a focus on muscle and muscle wasting.
CANCER CACHEXIA
Cancer-induced muscle wasting results in accelerated
muscle loss and a decline in physical function. Approximately
half of the patients with advanced lung cancer have severe
muscle wasting at diagnosis, and muscle wasting will
develop in approximately two-thirds of patients during

treatment. It is also clear that muscle wasting, or sarcopenia,
is common in patients with lung cancer, regardless of body
weight, and may exist even among patients with obesity.
This population of patients with sarcopenic obesity is at
particular high risk of complications, including impaired
functional status.4
To better understand the impact of muscle wasting on
patients with cancer techniques have been developed using
standardized CT, which allows quantitative assessment of
skeletal muscle and other body tissues.5 By using this
technique for patients with advanced lung and gastrointestinal cancers, virtually 100% of patients with a body mass
index below 20 will have muscle wasting, 30%50% will have
muscle wasting with a normal body mass index, and as many
as 20% with obesity will have overt muscle wasting. Looking
across all patients with sarcopenia, the incidence of doselimiting chemotherapy toxicity is increased, and time to
progression and overall survival is reduced.
For the patient with cancer the disease process leads to a
decline in muscle mass and associated anorexia, fatigue, and
Department of Medicine and Solid Tumor Therapeutics Program, Duke Cancer Institute, Duke University Medical Center, Durham, NC; Division of Medical Oncology, University of Ottawa,
The Ottawa Hospital Cancer Centre, Ottawa, Ontario, Canada; University of Maryland Greenebaum Cancer Center, Baltimore, MD.
Corresponding author: Jeffrey Crawford, MD, Duke Cancer Institute, DUMC 3476, Durham, NC 27710; email: crawf006@mc.duke.edu.
e484
CACHEXIA, FATIGUE, AND PATIENTS WITH LUNG CANCER
reduced quality of life, with subsequent weakness, declining

muscle strength, and reduced mobility, disability, and impact on physical performance. At a molecular level, this loss
of muscle mass is associated with changes in protein stores,
including enzymes and other regulatory proteins, altered
metabolism, such as insulin resistance, and impaired
immunity.
The molecular mechanisms behind the muscle loss that
occurs in cancer cachexia and other forms of cachexia is an
area of extreme active investigation. Other factors that
may lead to direct muscle atrophy through cytokines, such
as tumor necrosis factoralpha and interleukin-6, as well
as myostatin and activin, are also under active investigation.
A number of agents that affect these cytokine-mediated
pathways are in development.6 A second approach has
been to directly increase muscle growth signals through
androgenic anabolic steroids or selective androgen receptor modulators (SARM), as well as ghrelin mimetics. A firstin-class SARM, enobosarm, showed promising results on
improvement in muscle and physical function among patients with cancer cachexia.7 Subsequent phase III trials
among patients with advanced lung cancer receiving chemotherapy showed an increase in muscle mass in the
enobosarm treatment group compared with overall decline
in muscle mass in the placebo population, but effects on
physical function using stair-climb power were inconsistent
between the trials.8 Phase II trials of anamorelin, a ghrelin
receptor agonist for patients with cancer cachexia, again
showed improvement in skeletal muscle mass.9 Phase III
trials have also been performed with this agent for patients
with advanced lung cancer and cachexia and have shown
improvement in skeletal muscle mass, compared with placebo controls. In addition, this agent has positive effects on
improving appetite and overall weight gain. Improvement
in function as measured by hand-grip strength, however,
was not observed.10 Multiple factors may explain the lack
of correlation between increase in muscle and lack of
KEY POINTS
Patients with lung cancer who are compromised

physically, medically, or socioeconomically are often
excluded from clinical trials and have poorer treatment
outcomes.
Cancer cachexia is clinically recognized by weight loss,
but the primary process is muscle loss, which occurs in
the majority of patients with advanced lung cancer,
regardless of body mass index.
Care for patients with cancer and organ failure requires
close integration with disease subspecialists.
Socioeconomic disparities affect lung cancer risk,
diagnosis, therapy, and outcomes.
New paradigms are needed that include
multidisciplinary approaches to improving outcomes of
medically compromised patients.
functional improvement, including the duration of therapy,

the choice of the functional tests, and other factors. However, these agents and others in development represent
exciting possibilities for future therapeutics in the field of
cancer cachexia.11
At present, most importantly for us as clinicians is recognition of the presence of cachexia in our patients. Weight
loss remains the primary clinical criteria. By definition, any
patient with greater than 5% of body weight loss has cachexia regardless of their overall body weight. Although this
only represents a portion of patients with muscle loss as
defined by CT imaging, weight loss at least allows us to
recognize this subset of patients who require additional
supportive care. Although current pharmacologic interventions
have not been established to be effective, exercise12 and
nutritional support13 continue to be important for the
multimodality treatment of patients with lung cancer at
all stages. It is hoped that in the years ahead, CT imaging
of muscle will move from an investigational tool to a
clinical measurement to enable us to assess the impact of
our interventions for our patients, which should be and
include pharmacologic approaches, nutritional support, and
exercise.
TREATMENT OF PATIENTS WITH ORGAN

FAILURE
For patients with lung cancer, choices of systemic therapy
are routinely informed by, and based on, protocols emerging
from clinical research. These guide the patient and clinician
about the gold standard options when making decisions
regarding optimal therapy. However, in reality, many patients seen daily in the clinic may not be eligible for a clinical
trial because of a variety of factors ranging from poor
functional status, comorbidities, other malignancies, nonmeasurable disease, or blood work falling outside of specific
parameters, to name a few. It is reasonable, therefore, to
question the transferability of clinical trial results to a
general population, and it potentially means that the
standards of care options actually have a less solid evidence
base for many patients. In a recent analysis of 528 patients
who were newly diagnosed with stage IV nonsmall cell lung
cancer (NSCLC) and seen in consultation by medical oncologists, only 55% received systemic treatment.14 Further,
when simple and limited generic clinical trial inclusion criteria were applied to these patients, only 27% would have
been trial-eligible.15 In additional analysis of this dataset,
the average survival of patients who received systemic
palliative chemotherapy was not significantly different
among patients who were considered trial eligible or trial
ineligible. An interpretation may be that clinical trial eligibility is too strict and clinical judgment is more important.
Despite this, there is another small but important cohort of
patients with major medical comorbidities, or a formal diagnosis of organ failure, who appropriately are excluded
from clinical trials and for whom systemic therapy poses
specific and challenging problems. In a review of selected
e485
TABLE 1. Selected Recent Practice-Changing Trials in Lung Cancer and Organ Function Exclusions
Trial
Author
Year
Renal Failure
Liver Failure
Heart Failure
First-Line Chemotherapy
Schiller et al16
2002
Excluded
Excluded
Unstated
17
Second-Line Chemotherapy
Hanna et al
2004
Excluded
Excluded
Unstated
Chemotherapy + Bevacizumab
Sandler et al18
2006
Excluded
Excluded
Excluded
Maintenance
Ciuleanu et al19
2009
Excluded
Excluded
Excluded
20
EGFR + Gefitinib
Mok et al
2009
Unstated
Excluded
Unstated
ALK + Crizotinib
Shaw et al21
2013
Excluded
Excluded
Excluded
Nivolumab
Brahmer et al22
2015
Excluded
Excluded
Unstated
recent practice-changing chemotherapy, targeted therapy,

and immunotherapy trials, patients with renal impairment,
hepatic impairment, or cardiac impairment would have been
excluded (Table 1).16-23 Therefore, how should clinicians
and patients approach decisions about systemic therapy in
the presence of organ failure, given the lack of available
evidence?
This article provides guidance on a reasonable and pragmatic
approach to making systemic therapy decisions for patients
with lung cancer and organ failure. In recognizing and accepting
that there is an absence of published data in the medical literature to provide clarity, these recommendations are the
considered views of the authors, but other reasonable and
well-planned approaches may have equal validity. It is also
important to recognize that there is a spectrum ranging from
normal organ function, through stages of organ impairment
or dysfunction to organ failure. This article is restricted to
discussing implications of systemic therapy for patients with
lung cancer and established organ failure.
The first recommendation is that these issues should
be discussed in a multidisciplinary format, and specific interaction with specialists related to the particular organ
failure is advised (e.g., nephrology, hepatology, cardiology,
etc.). Further, additional consultation with a specialist oncology pharmacist is advised, particularly if the decision is
made to proceed with therapy. Patients should be fully
informed regarding relative benefits and harms from therapy, the consequences of declining therapy, and that proceeding with treatment will almost certainly not be based
on level-one evidence. For patients with advanced disease,
consideration should be given to early palliative care specialist input and advanced care planning.
Understanding the cause and prognosis of organ failure is
self-evidently important. Some patients are diagnosed
with an acute organ failure directly related to the lung
cancer, and, if considered reversible, this should be addressed. For example in a patient presenting with acute
renal failure caused by hydronephrosis from retroperitoneal
lymphadenopathy, by inserting a nephrostomy tube, the
renal failure can be reversed and the barrier to systemic
therapy is removed. However, the remainder of this article
will address patients with pre-existing organ failure, rather
than organ failure secondary to the malignancy. In a recent
review of clinical indicators of 6-month mortality in advanced
e486
noncancer illnesses, Salpeter et al evaluated heart failure,

dementia, geriatric failure-to-thrive syndrome, hepatic cirrhosis, chronic obstructive pulmonary disease, and end-stage
renal disease. This list represented approximately 70% of the
noncancer diagnoses on admission to hospice.24 Clearly, not
all patients with these conditions die within 6 months, but the
authors identified common and disease-specific prognostic
indicators, including poor performance status (PS), malnutrition, comorbid illness, and organ dysfunction.
In the cancer clinic, the clinician must understand the
natural course of the organ failure pathology. For patients
with liver, kidney, or heart failure who may be waiting for
organ transplantation, the diagnosis of lung cancer unfortunately makes them ineligible for the transplant program. Regarding prognosis of advanced organ failure, the
United States Renal Data System Annual Data Report, describes 3-year survival as 52% for patients receiving hemodialysis for end-stage renal disease, and 61% for patients
receiving peritoneal dialysis. The risk of death is particularly
high in the first year of hemodialysis, with rates reported up
to 25%. The Canadian Organ Replacement Register Annual
Report describes a 5-year survival for patients on dialysis of
approximately 43%. For patients with end-stage heart failure, the 1-year survival is approximately 50%,25 which is not
dramatically different from patients with stage IV NSCLC
receiving first-line chemotherapy. The prognosis of patients
with liver cirrhosis is variable, depending on severity, etiology, and the presence or absence of complications. The
Model for End-Stage Liver Disease (MELD) score is used to
assess the severity of chronic liver disease26 as an alternative
to the Child-Pugh scoring system. Salpeter et al24 reported
that patients with decompensated liver failure (the presence
of complications of cirrhosis) may have a median survival of
less than 6 months if associated with high MELD scores.
An understanding of competing morbidities therefore
clearly plays an important role in understanding the role
systemic therapy plays in lung cancer. In assessing the need
for adjuvant chemotherapy for patients with early-stage
disease, it is highly likely that any benefit from chemotherapy (approximately 5%) will be outweighed by the competing
risks of the comorbid condition for patients with organ failure.
As an illustration, the adjuvant online program is used to
model potential benefits from adjuvant chemotherapy in
resected NSCLC. Figure 1, adapted from the adjuvant online
FIGURE 1. The Benefit of Adjuvant Chemotherapy for

a Patient With Stage IIA NonSmall Cell Lung Cancer,
With or Without Organ Failure
website, demonstrated that the absolute benefit of chemotherapy (notwithstanding the challenges of actually administering the treatment) is only modestly reduced but is now
seen in the context of lung cancer no longer being the primary
health threat in any event. In contrast to the curative situation
in which the prize of successful therapy is great, for patients
with metastatic disease the role of competing morbidities
and organ failure is even more striking. When the absolute
benefits, measured in terms of prolonged overall survival
from systemic therapy, may only be measured in a few
months in a healthy population, the benefits from palliative
systemic therapy in patients with significant comorbidities or
organ failure are likely to be further diminished. These issues
must be openly disclosed and discussed with patients prior to
any decision to attempt systemic therapy.
For many (or indeed the majority of) patients with advanced
lung cancer and coexisting organ failure, a palliative approach
may be the most appropriate. Based on the seminal publication
by Temel et al, early palliative care intervention is associated
with improved quality of life and longer survival.27 Therefore
a recommendation would be to initiate a palliative approach
to management of these patients, including discussions regarding advanced care planning and place of care. While
a cliche, for many patients facing an incurable lung cancer,
and barriers to therapy because of organ failure, we maybe
should consider care as being more important than treatment.
Specifically, patients with rapid decline, hospitalization, frailty,
and poor functional status should have serious consideration to
a purely palliative treatment plan.
After assessing patients with lung cancer, in the multidisciplinary context and taking into account the issues discussed, the decision may still be to proceed with therapy.
This should be on the understanding of the relative lack
of data, and then a choice of regimen based on an understanding of the drug metabolism, with appropriate dose
adjustments after dialogue with an oncology pharmacist.
Table 2 outlines common lung cancer drugs and their route
of elimination and recommendations for their use in renal or
hepatic impairment and for patients receiving dialysis.

Tabular information is taken from product monographs and
selected references.28,29 Data on efficacy for these drugs in
these scenarios are largely limited to case reports.
In conclusion, patients with lung cancer and organ failure
represent a population excluded from clinical trials and
with a limited evidence base. The competing morbidity and
mortality mitigate against potential benefits from anticancer systemic therapy. The newer generations of targeted
therapies and immunotherapies may be easier to deliver,
but limited data exist. Clinicians should discuss these cases
in a multidisciplinary environment, and early intervention
from palliative care specialists may be particularly appropriate. Finally, as a community, we should seek to perform
clinical trials in this population of patients with lung cancer.
TREATMENT OF LUNG CANCER IN MEDICALLY

COMPROMISED PATIENTS: COMPROMISED
SOCIAL AND ECONOMIC STATUS IN LUNG
CANCERIS IT REAL AND DOES IT MATTER?
Many factors can make a patient vulnerable to poor outcomes with lung cancer. Biologic and patient-related factors
such as poor nutrition, functional status, or organ function
undoubtedly contribute to vulnerability to poor outcomes.
However, it is critical to recognize that poor socioeconomic
status (SES) can render a patient equally vulnerable to poor
outcomes from lung cancer. Socioeconomic status is often
defined as ones class or standing measured by education,
income, and occupation.30 Poor SES can affect the disease at
various stages including development, diagnosis, treatment,
and, ultimately, outcomes from lung cancer.7 There are
many confounders such as race, education level, and insurance status when evaluating socioeconomic disparities,
but these variables are intimately connected. Interventions
that target these vulnerable populations may minimize the
socioeconomic disparities that exist in lung cancer.
Socioeconomic disparities have also been shown to affect lung cancer risk, diagnosis, therapy, and outcomes internationally. It is often difficult to separate socioeconomic
disparities from other sources of disparities in lung cancer
such as race and insurance. Here, we will discuss the role of
socioeconomic factors, with a focus on income, in the development, treatment, and outcomes of lung cancer in the
United States.
Socioeconomic Factors Contribute to Disparities

in Lung Cancer Risk
Tobacco is the leading risk factor for lung cancer development, and approximately 80%90% of lung cancers are
related to tobacco.31 Earlier age of smoking initiation has
been associated with a higher prevalence of smoking
through adulthood, lower cessation rates, and increased risk
of lung cancer. It is indisputable that these factors disproportionately affect lower-income populations.32-35
Childhood poverty, for example, has been linked to as
much as a 33% higher likelihood of initiation and earlier
e487
TABLE 2. Common Lung Cancer Drugs and Recommendations for Use in Renal or Hepatic Impairment and for
Patients Receiving Dialysis
Drug
Elimination
Liver
Renal
Dialysis
Cisplatin
Renal
Not applicable
Decrease depending on CrCl
50% post-HD or non-HD days
Carboplatin
Renal
Not applicable
Calvert formula
Calverts formula give on non-HD day
Docetaxel
Liver
Adjust
Not applicable
Before or after HD not removed by HD
Pemetrexed
Renal
Caution in severe dysfunction
Avoid if CrCl # 45
Avoid
Paclitaxel
Liver
Adjust
Not applicable
Not removed by HD
Gemcitabine
Urine (inactive
metabolites)
Consider adjustment if bilirubin

. 27
Caution (D/C if HUS)
Give 6-12 hours before HD
Vinorelbine
Liver
Adjust depending on bilirubin
Not applicable
Limited data, consider 20 mg/m2 on

non-HD day
Etoposide
Liver/Renal
Adjust depending on bilirubin
Adjust
50% before or after HD
Gefitinib
Liver
Caution
Caution if CrCl , 20
No information
Erlotinib
Liver
Caution
Not applicable
No information
Afatinib
Liver
Caution
No information
Crizotinib
Liver
Adjust
No information
Ceritnib
Liver
Adjust
No information
Bevacizumab
Reticulo-endothelial
system
Not involved
Not involved
No information
Nivolumab
Biochemical degradation; no liver or kidney

involvement
No effect in mild impairment

(not studied in severe)
No effect if CrCl $ 15 mL/min

(not studied if less)
No information
Abbreviations: CrCl, creatinine clearance; HD, hemodialysis; HUS, hemolytic uremic syndrome.
onset of smoking (odds ratio [OR] 1.33; 95% CI, 1.081.63).34

The prevalence of smoking is significantly higher in people
who live below the poverty versus those who live above the
poverty level. At the census track level in the United States,
smoking rates in populations with incomes below $20,000
per year is estimated at 38.2% compared with 17.3%29.6%
in populations who have incomes above $20,000 per
year.36,37 In addition, it has been recognized that people in
lower-income groups are less likely to use clinically proven
smoking cessation treatments and have less successful quit
attempts.38,39
Many factors contribute to this disproportionate tobacco
epidemic. One factor is the tobacco industrys marketing
campaigns. The tobacco industry has marketed smoking
and tobacco products to target young, lower-income populations for decades, and these campaigns have been remarkably successful.40,41 Adolescents who are receptive to
tobacco advertising, such as those who can recognize tobacco advertising messages or who identify a favorite
tobacco advertisement, are 2.33.9 times as likely to begin
smoking in adolescence compared with those who are not
receptive to tobacco advertising.42 Other factors such as
placement of low-cost tobacco outlets in low-income areas
and environmental or social stressors exacerbate this epidemic as well.43
Despite antitobacco campaigns such as clean environment
bills and advertisements, the decline in smoking rates has
been larger among people above the poverty level. Programs like those that have raised the price of cigarettes,
reduced advertising to lower-income populations, or have
e488
provided extra support to vulnerable populations have had

alleviated, but not eliminated, smoking-related disparities in
low-income populations.44,45 It is crucial to address smoking
uptake and prevalence in lower-income populations to further reduce disparities in the development of lung cancer.
Socioeconomic Status Affects Stage at Diagnosis and

Receipt of Stage-Appropriate Therapy for Lung Cancer
Lung cancer prognosis depends on numerous factors including
stage at diagnosis and receipt of stage-appropriate therapy.
Unfortunately, socioeconomic factors, including lower income,
are associated with diagnosis of lung cancer at later, less
curable stages.46,47 Numerous studies have demonstrated
that patients with lower income and Medicaid insurance or
no insurance are more likely to have advanced disease.7,48
For example, in areas where there is less than 10% poverty,
52%54.6% of patients have distant metastases compared with
areas with more than 20% poverty, where 56.8%59% of
patients have metastatic lung cancer.2 Similarly, in an analysis
of the Florida Cancer Registry, 25% of individuals from areas
with less than 5% poverty have localized disease compared
with only 22% of those from areas with more than 15% poverty
(p , .001).49 Presentation at later stages may be related to
numerous factors, such as a lower likelihood of being under the
care of a health provider who could potentially recognize the
signs of lung cancer or economic stressors that prevent individuals from seeking care.50
One potential factor that may widen the disparity in stage at
diagnosis in lower-income populations may be with the use and
uptake of lung cancer screening. The National Lung Screening

Trial (NLST) has demonstrated that lung cancer screening with
low-dose CT scan can reduce lung cancer mortality by 20% and
that more lung cancer cases were diagnosed at earlier stages of
disease for patients who underwent a low-dose CT.51 However,
this research population of over 53,000 individuals represented a
population of largely white, highly-educated former smokers.51
It is largely unknown how these results translate or can be
applied to lower-income populations who are more likely to be
current smokers and less likely to be under the care of a health
care provider. Furthermore, it is largely unknown what health
care providers beliefs and attitudes are toward lung cancer
screening and how often screening guidelines are adhered to in
the era after the National Lung Screening Trial results were
published in various practice settings.52,53 It is likely that as lung
cancer screening becomes more refined and used, similar socioeconomic disparities in screening will become apparent, and
efforts to empirically reduce this are important.
Lower socioeconomic groups are more likely to experience
delays in therapy from the time of diagnosis and less likely to
undergo stage-appropriate therapy for lung cancer.54,55 For
example, in a large population-based Surveillance, Epidemiology, and End Results (SEER)-Medicare analysis, those with
incomes of less than $25,000 were more likely to have delays in
guideline-concordant lung cancer care (hazard ratio [HR] 0.89;
95% CI, 0.80.98; p # .05).54
As has been observed for other malignancies such as breast
or colon cancer, in which disparities have become apparent,
lack of awareness, lack of access, or mistrust of the health care
system may contribute to socioeconomic and racial disparities
in screening rates, stage at diagnosis, therapy, and survival.
Provider factors, such as whether a health care provider has
knowledge or understanding about the diagnosis or management of lung cancer, can also affect this receipt of therapy.
For example, even after the NLST results were published, many
physicians are not aware of who would be eligible for lung
cancer screening, even at an NLST site.52 In addition, lower
income has been associated with a lower likelihood of referral
to an oncologist, as well as lower likelihood of receipt of
guideline-appropriate therapy for locally advanced and
advanced NSCLC.56 Furthermore, health system barriers
such as lack of access to specialists can also influence the
diagnosis and receipt of therapy for lung cancer.
significantly associated with lower rates of hospice utilization

in elderly patients with advanced lung cancer.60
We Can Learn From Other Cancers Where

Socioeconomic Disparities Have Been Reduced
As health care providers, we have an opportunity to learn
from successful programs that have reduced cancer disparities by targeting vulnerable populations. One highly
successful program implemented by the Delaware Cancer
Consortium, for example, mandated a statewide program
to increase colorectal cancer (CRC) screening in high-risk
populations (African Americans and under- or uninsured
patients) and demonstrated remarkable success.61 This
program used nurse navigators and care coordinators to
recruit vulnerable populations, funded screening for
treatment for CRC, and accomplished its goal of reducing
CRC screening, treatment, and survival disparities. In 9 years,
the disparities in screening rates, stage at diagnosis, and
mortality were almost eliminated. Similarly, because the CRC
program has been so successful, Delaware has also recently
launched a comprehensive program for lung cancer screening.62
In this program, education about screening is targeted
toward health care providers and patients. Patients who
are potentially eligible for screening are given access to
patient navigators and care coordinators who assist them
with receiving counseling and assistance with smoking
cessation. Eligible participants are also able to establish
care with a primary care provider who will order screening if
appropriate and recommend follow-up or referrals to
providers for further workup. Furthermore, participants
who are diagnosed with lung cancer will also be able to
TABLE 3. Contributing Factors to Economic Disparities

and Possible Interventions
Contributing Factor
Intervention
Age at Smoking Initiation
Interventions to target peer

groups 43,63,64
Smoking Prevalence
Clean air laws62

Smoke-free multi-unit housing62
Prohibit tobacco sales near schools62
Prohibit outdoor advertisements62
Prohibit distribution of free or lowcost tobacco 43,62
Lower Income Associated With Poorer Outcomes

From Lung Cancer
Although there are many confounders when evaluating socioeconomic disparities in lung cancer outcomes, such as
smoking status, race, or medical comorbidities, individuals
with lower incomes have inferior outcomes from lung cancer.
In population-based studies, even when controlling for other
prognostic factors, lower income (, $46,000 compared
with . $46,000) was associated with inferior 30-day and
long-term survival after surgical resection.55,57 Low income
has also been independently associated with inferior survival
in advanced lung cancer.58,59 Furthermore, lower income was
Smoking Cessation
Reduce low-cost tobacco sale

outlets43
Financial incentives65
Increased access to smoking cessation
support65,66
Stage at Diagnosis and Receipt Education to providers61

of Therapy
Awareness (Patient and
Provider)
Education on screening61
Access to Care
Payment for therapy61
Fear of Diagnosis
Patient and nurse navigators61,62

Establish care with a provider61,62
e489
receive therapy for their diagnosis with minimal to no

cost.
Despite improvements in early detection of lung cancer
and in therapeutics, the toll from lung cancer remains dismal
and disparities at all stages of lung cancer diagnosis and
management persist and, in many situations, continue to
widen. It is crucial that as health care providers, we acknowledge that socioeconomic disparities contribute to inferior outcomes for lung cancer and that we can learn from
the past and from other malignancies in which disparities
have been reduced to guide current and future identification and management of lung cancer (Table 3).
References
1. Karnofsky DA, Burchenal JH. The clinical evaluation of chemotherapeutic
agents in cancer. In MacLeod CM (ed). Evaluation of Chemotherapeutic
Agents. New York: Columbia University Press, 1949;191-205.
2. Martin L, Senesse P, Gioulbasanis I, et al. Diagnostic criteria for the
classification of cancer-associated weight loss. J Clin Oncol. 2015;33:
90-99.
3. Fearon K, Strasser F, Anker SD, et al. Definition and classification of
cancer cachexia: an international consensus. Lancet Oncol. 2011;12:
489-495.
4. Prado CM, Lieffers JR, McCargar LJ, et al. Prevalence and clinical implications of sarcopenic obesity in patients with solid tumours of the
respiratory and gastrointestinal tracts: a population-based study.
Lancet Oncol. 2008;9:629-635.
5. Prado CM, Antoun S, Sawyer MB, et al. Two faces of drug therapy in
cancer: drug-related lean tissue loss and its adverse consequences to
survival and toxicity. Curr Opin Clin Nutr Metab Care. 2011;14:250-254.
6. Cohen S, Nathan JA, Goldberg AL. Muscle wasting in disease: molecular
mechanisms and promising therapies. Nat Rev Drug Discov. 2015;14:
58-74.
7. Dobs AS, Boccia RV, Croot CC, et al. Effects of enobosarm on muscle
wasting and physical function in patients with cancer: a double-blind,
randomised controlled phase 2 trial. Lancet Oncol. 2013;14:335-345.
8. Crawford J, Dalton JT, Hancock ML, et al. LATE BREAKING ABSTRACT:
Results from two Phase 3 randomized trials of enobosarm, selective
androgen receptor modulator (SARM), for the prevention and treatment of muscle wasting in NSCLC. Presented at: European Cancer
Congress; 2013 Amersterdam; Abstract 21.
9. Garcia JM, Boccia RV, Graham CD, et al. Anamorelin for patients with
cancer cachexia: an integrated analysis of two phase 2, randomised,
placebo-controlled, double-blind trials. Lancet Oncol. 2015;16:108-116.
10. Temel JS, Currow DC, Fearon K, et al. Phase III trials of anamorelin in
patients with advanced non-small cell lung cancer (NSCLC) and cachexia
(ROMANA 1 and 2). J Clin Oncol. 2015;33(suppl; abstr 9500).
11. Fearon K, Argiles JM, Baracos VE, et al. Request for regulatory guidance
for cancer cachexia intervention trials. J Cachexia Sarcopenia Muscle.
2015;6:272-274.
12. Alves CR, da Cunha TF, da Paix~ao NA, et al. Aerobic exercise training as
therapy for cardiac and cancer cachexia. Life Sci. 2015;125:9-14.
13. Bourdel-Marchasson I, Blanc-Bisson C, Doussau A, et al. Nutritional
advice in older patients at risk of malnutrition during treatment for
chemotherapy: a two-year randomized controlled trial. PLoS One. 2014;
9:e108687.
14. Brule S, Al-Baimani K, Jonker H, et al. Palliative chemotherapy (CT) for
advanced non-small cell lung cancer (NSCLC): investigating disparities
between patients who are treated versus those who are not. J Clin
Oncol. 2015;33(suppl; abstr e17681).
15. Al-Baimani K, Jonker H, Zhang T, et al. Are clinical trial eligibility criteria
an accurate reflection of a real world population of advanced lung
cancer patients. Presented at: World Conference on Lung Cancer; 2015
Denver, CO; Abstract 1398.
e490
16. Schiller JH, Harrington D, Belani CP, et al. Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer. N Engl J
Med. 2002;346:92-98.
17. Hanna N, Shepherd FA, Fossella FV, et al. Randomized phase III trial of
pemetrexed versus docetaxel in patients with non-small-cell lung cancer
previously treated with chemotherapy. J Clin Oncol. 2004;22:1589-1597.
18. Sandler A, Gray R, Perry MC, et al. Paclitaxel-carboplatin alone or with
bevacizumab for non-small-cell lung cancer. N Engl J Med. 2006;355:
2542-2550.
19. Ciuleanu T, Brodowicz T, Zielinski C, et al. Maintenance pemetrexed
plus best supportive care versus placebo plus best supportive care for
non-small-cell lung cancer: a randomised, double-blind, phase 3 study.
Lancet Oncol. 2009;374:1432-1440.
20. Mok TS, Wu YL, Thongprasert S, et al. Gefitinib or carboplatin-paclitaxel
in pulmonary adenocarcinoma. N Engl J Med. 2009;361:947-957.
21. Shaw AT, Kim DW, Nakagawa K, et al. Crizotinib versus chemotherapy in advanced ALK-positive lung cancer. N Engl J Med. 2013;
368:2385-2394.
22. Brahmer J, Reckamp KL, Baas P, et al. Nivolumab versus docetaxel in
advanced squamous-cell non-small-cell lung cancer. N Engl J Med. 2015;
373:123-135.
23. Gettinger SN, Horn L, Gandhi L, et al. Overall survival and long-term
safety of nivolumab (anti-programmed death 1 antibody, BMS-936558,
ONO-4538) in patients with previously treated advanced non-small-cell
lung cancer. J Clin Oncol. 2015;33:2004-2012.
24. Salpeter SR, Luo EJ, Malter DS, Stuart B. Systematic review of noncancer
presentations with a median survival of 6 months or less. Am J Med.
125:512 e1-6.
25. Friedrich EB, Bohm M. Management of end stage heart failure. Heart.
2007;93:626-631.
26. Kamath PS, Kim WR; Advanced Liver Disease Study Group. The model
for end-stage liver disease (MELD). Hepatology. 2007;45:797-805.
27. Temel JS, Greer JA, Muzikansky A, et al. Early palliative care for patients with
metastatic nonsmall-cell lung cancer. N Engl J Med. 2010;363:733-742.
28. Janus N, Thariat J, Boulanger H, et al. Proposal for dosage adjustment
and timing of chemotherapy in hemodialyzed patients. Ann Surg Oncol.
2010;21:1395-1403.
29. Brandes JC, Grossman SA, Ahmad H. Alteration of pemetrexed excretion
in the presence of acute renal failure and effusions: presentation of a
case and review of the literature. Cancer Invest. 2006;24:283-287.
30. Baker EH. Socioeconomic status, definition. The Wiley Blackwell
Encyclopedia of Health, Illness, Behavior, and Society. Hoboken, NJ:
Wiley-Blackwell; 2014.
31. U.S. Department of Health and Human Services. The Health Consequences of Smoking: A Report of the Surgeon General. Atlanta, GA: U.S.
Department of Health and Human Services, Centers for Disease Control
and Prevention, National Center for Chronic Disease Prevention and
Health Promotion, Office on Smoking and Health; 2004.
32. Chen J, Millar WJ. Age of smoking initiation: Implications for quitting.
Health Rep. 1998;9:39-46.
33. Wiencke JK, Thurston SW, Kelsey KT, et al. Early age at smoking initiation
and tobacco carcinogen DNA damage in the lung. J Natl Cancer Inst.
1999;91:614-619.
34. Gilman SE, Abrams DB, Buka SL. Socioeconomic status over the life
course and stages of cigarette use: initiation, regular use, and cessation.
J Epidemiol Community Health. 2003;57:802-808.
35. Hegmann KT, Fraser AM, Keaney RP, et al. The effect of age at smoking
initiation on lung cancer risk. Epidemiology. 1993;4:444-448.
36. King BA, Dube SR, Tynan MA. Current tobacco use among adults in the
United States: findings from the National Adult Tobacco Survey. Am J
Public Health. 2012;102:e93-e100.
37. Agaku IT, King BA, Dube SR; Centers for Disease Control and Prevention.
Current cigarette smoking among adults - United States, 2005-2012.
MMWR Morb Mortal Wkly Rep. 2014;63:29-34.
38. Burns ME, Fiore MC. Under-use of tobacco dependence treatment
among Wisconsins fee-for-service Medicaid recipients. WMJ. 2001;
100:54-58.
39. Fu SS, Kodl MM, Joseph AM, et al. Racial/Ethnic disparities in the use of
nicotine replacement therapy and quit ratios in lifetime smokers ages
25 to 44 years. Cancer Epidemiol Biomarkers Prev. 2008;17:1640-1647.
40. Hackbarth DP, Silvestri B, Cosper W. Tobacco and alcohol billboards in
50 Chicago neighborhoods: market segmentation to sell dangerous
products to the poor. J Public Health Policy. 1995;16:213-230.
41. Yerger VB, Przewoznik J, Malone RE. Racialized geography, corporate
activity, and health disparities: tobacco industry targeting of inner
cities. J Health Care Poor Underserved. 2007;18:10-38.
42. Evans N, Farkas A, Gilpin E, et al. Influence of tobacco marketing and
exposure to smokers on adolescent susceptibility to smoking. J Natl
Cancer Inst. 1995;87:1538-1545.
43. Cantrell J, Anesetti-Rothermel A, Pearson JL, et al. The impact of the
tobacco retail outlet environment on adult cessation and differences by
neighborhood poverty. Addiction. 2015;110:152-161.
44. Vijayaraghavan M, Messer K, White MM, et al. The effectiveness of
cigarette price and smoke-free homes on low-income smokers in the
United States. Am J Public Health. 2013;103:2276-2283.
45. Haas JS, Linder JA, Park ER, et al. Proactive tobacco cessation outreach
to smokers of low socioeconomic status: a randomized clinical trial.
JAMA Intern Med. 2015;175:218-226.
46. Halpern MT, Ward EM, Pavluck AL, et al. Association of insurance status
and ethnicity with cancer stage at diagnosis for 12 cancer sites: a retrospective analysis. Lancet Oncol. 2008;9:222-231.
47. Schwartz KL, Crossley-May H, Vigneau FD, et al. Race, socioeconomic
status and stage at diagnosis for five common malignancies. Cancer
Causes Control. 2003;14:761-766.
48. Slatore CG, Au DH, Gould MK; American Thoracic Society Disparities in
Healthcare Group. An official American Thoracic Society systematic
review: insurance status and disparities in lung cancer practices and
outcomes. Am J Respir Crit Care Med. 2010;182:1195-1205.
49. Yang R, Cheung MC, Byrne MM, et al. Do racial or socioeconomic
disparities exist in lung cancer treatment? Cancer. 2010;116:2437-2447.
50. Pateman K, Ford P, Fitzgerald L, et al. Stuck in the catch 22: attitudes
towards smoking cessation among populations vulnerable to social
disadvantage. Addiction. Epub 2015 Nov 28.
51. Aberle DR, Adams AM, Berg CD, et al; National Lung Screening Trial
Research Team. Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med. 2011;365:395-409.
52. Lewis JA, Petty WJ, Tooze JA, et al. Low-dose CT lung cancer
screening practices and attitudes among primary care providers at
an academic medical center. Cancer Epidemil Biomarkers Prev.
2015;24:664-670.
53. Lam VK, Miller M, Dowling L, et al. Community low-dose CT lung cancer
screening: a prospective cohort study. Lung. 2015;193:135-139.
54. Nadpara P, Madhavan SS, Tworek C. Guideline-concordant timely
lung cancer care and prognosis among elderly patients in the United
States: a population-based study. Cancer Epidemiol. 2015;39:
1136-1144.
55. Melvan JN, Sancheti MS, Gillespie T, et al. Nonclinical factors associated
with 30-day mortality after lung cancer resection: an analysis of 215,000
patients using the national cancer data base. J Am Coll Surg. 2015;221:
550-563.
56. Goulart BH, Reyes CM, Fedorenko CR, et al. Referral and treatment
patterns among patients with stages III and IV nonsmall-cell lung
cancer. J Oncol Pract. 2013;9:42-50.
57. Khullar OV, Gillespie T, Nickleach DC, et al. Socioeconomic risk factors
for long-term mortality after pulmonary resection for lung cancer: an
analysis of more than 90,000 patients from the national cancer data
base. J Am Coll Surg. 2015;220:156-168.
58. Tannenbaum SL, Koru-Sengul T, Zhao W, et al. Survival disparities in
non-small cell lung cancer by race, ethnicity, and socioeconomic status.
Cancer J. 2014;20:237-245.
59. Erhunmwunsee L, Joshi MB, Conlon DH, et al. Neighborhood-level
socioeconomic determinants impact outcomes in nonsmall cell lung
cancer patients in the Southeastern United States. Cancer. 2012;118:
5117-5123.
60. Mack JW, Chen K, Boscoe FP, et al. Underuse of hospice care by
Medicaid-insured patients with stage IV lung cancer in New York and
California. J Clin Oncol. 2013;31:2569-2579.
61. Grubbs SS, Polite BN, Carney J Jr, et al. Eliminating racial disparities in
colorectal cancer in the real world: it took a village. J Clin Oncol. 2013;31:
1928-1930.
62. Healthy Delaware. https://www.healthydelaware.org/lung. Accessed
January 13, 2016.
63. Lipperman-Kreda S, Friend KB, Grube JW. Rating the effectiveness of
local tobacco policies for reducing youth smoking. J Prim Prev. 2014;35:
85-91.
64. Lee YO, Jordan JW, Djakaria M, et al. Using peer crowds to segment
Black youth for smoking intervention. Health Promot Pract. 2014;15:
530-537.
65. Parks MJ, Slater JS, Rothman AJ, et al. Interpersonal communication and
smoking cessation in the context of an incentive-based program: survey
evidence from a telehealth intervention in a low-income population.
J Health Commun. 2016;21:125-133.
66. Christiansen BA, Reeder KM, TerBeek EG, et al. Motivating low socioeconomic status smokers to accept evidence-based smoking cessation
treatment: a brief intervention for the community agency setting.
Nicotine Tob Res. 2015;17:1002-1011.
e491
Clinical Conundrums in
Melanoma Therapy
CHAIR
Janice M. Mehnert, MD
Rutgers Cancer Institute of New Jersey
New Brunswick, NJ
SPEAKERS
Robyn Saw, MBBS, FRACS, MS
Melanoma Institute Australia and The University of Sydney
North Sydney, Australia
Sapna Pradyuman Patel, MD
Houston, TX
BIOMARKERS FOR IMMUNOTHERAPY
Biomarkers for Immunotherapy: Current Developments and

Challenges
Kristen R. Spencer, DO, MPH, Jianfeng Wang, MD, Ann W. Silk, MD, Shridar Ganesan, MD, PhD,
Howard L. Kaufman, MD, and Janice M. Mehnert, MD
OVERVIEW
Immunotherapy has revolutionized cancer therapy and has been named the cancer advance of the year for 2016. Checkpoint
inhibitors have demonstrated unprecedented rates of durable responses in some of the most difficult-to-treat cancers;
however, many treated patients do not respond, and the potential for serious side effects exists. There is a growing need to
identify biomarkers that will improve the selection of patients who will best respond to therapy, further elucidate drug
mechanisms of action, and help tailor therapy regimens. Biomarkers are being explored at the soluble, cellular, and genomic
levels, and examples in immunotherapy include serum proteins, tumor-specific receptor expression patterns, factors in the
tumor microenvironment, circulating immune and tumor cells, and host genomic factors. The search for reliable biomarkers
is limited by our incomplete understanding of how immunotherapies modify the already complex immune response to
cancer, as well as the contribution of immuno-editing to a dynamic and inducible tumor microenvironment and immune
milieu. Furthermore, there has been little extension of any candidate assay into large, prospective studies, and the lack of
standardization in measurement and interpretation restricts their validity. Both tumor-infiltrating lymphocytes and PD-L1
expression within the tumor microenvironment have been recognized as having both prognostic and predictive value for
patients treated with immunotherapy. Alternately, a larger panel of gene signatures, chemokines, and other factors that
correlate with response has been proposed. In this article, we will explore the status of current biomarker candidates.
ince ipilimumab entered the treatment landscape in

2011, immunotherapy has continued to revolutionize
cancer therapy. In fact, immunotherapy was recently named
the American Society of Clinical Oncologys top cancer advance of the year for 2016.1 A number of U.S. Food and Drug
Administration (FDA)approved agents are available for an
increasing number of difficult-to-treat cancers, such as
melanoma, renal cell carcinoma (RCC), and lung cancer,
among others. In contrast with most chemotherapy and
targeted therapies, immunotherapy offers the possibility
of durable response, sometimes even without continued
treatment.2-4 However, objective responses among patients
treated with single-agent regimens are seen in less than onehalf of patients treated. Combination immune checkpoint
inhibitor therapy raises response rates but also increases
toxicity and cost.5 Thus, to optimize selection of appropriate
patients for immunotherapy and avoid unnecessary toxicity and health care costs, there is a clear need to identify
truly predictive, and not simply prognostic, biomarkers of
response.
Understanding which factors predict clinical benefit with
immunotherapy can improve the selection of tumor types
and patient subsets who will respond, illuminate the

mechanism of action of novel immunotherapeutic approaches, and potentially inform which patients require
single-agent versus various combination strategies, which
are increasing in availability. This review focuses on the development of biomarkers for immunotherapy at the soluble,
cellular, and genomic levels. Examples of biomarkers in the
immunotherapy landscape include (1) soluble factors such
as serum proteins, (2) tumor-specific factors such as receptor
expression patterns and components of the microenvironment, and (3) host genomic factors.6-8 Despite the interest in
biomarker development for immunotherapy, validated biomarkers have remained an elusive goal.
WHY ARE BIOMARKERS CHALLENGING IN

IMMUNOTHERAPY?
Our incomplete understanding of the mechanisms of action
of specific immunotherapies makes it difficult to identify a
surrogate marker that adequately captures the process
across different classes of drugs.9 Many published analyses
of potential predictive biomarkers for immunotherapy are
From the Rutgers Cancer Institute of New Jersey, New Brunswick, NJ.
Corresponding author: Janice M. Mehnert, MD, Rutgers Cancer Institute of New Jersey, 195 Little Albany St., New Brunswick, NJ 08901; email: mehnerja@cinj.rutgers.edu.
e493
SPENCER ET AL
retrospective, with limited extension into large prospective

trials. In addition, there has been substantial variability in
standardization, measurement, and interpretation of early
biomarker assays.10 Furthermore, biomarker development
in immunotherapy is challenged by the fact that immunotherapy targets are often inducible and dynamic over time
and location. This is a function of the complex tumor microenvironment and the contribution of immuno-editing
to the immune milieu. The tumor microenvironment involves complicated interactions between several types of
infiltrating immune cells such as monocytes, neutrophils,
dendritic cells, T and B cells, eosinophils, basophils, mast
cells, and natural killer (NK) cells as well as the heterogeneous tumor cells themselves and their companion stromal
cells, including tumor-associated macrophages, fibroblasts,
adipocytes, and endothelial and other cells.11 The local
environment is further complicated by microniches created by alterations in perfusion, oxygenation, electrolyte
levels, and the subsequent development of resistant tumor
cells surviving in nutrient- and oxygen-deprived conditions.11,12 Thus, these microniches likely represent distinct
microenvironments with different cell types and factors,
all within one tumor deposit. Finally, incomplete immunoediting may result in selective pressure on tumor cells,
resulting in resistant tumor cell clones and immune
escape. 11
Despite these challenges, clinical research of immunotherapy over the last several years has confirmed the importance of tumor-infiltrating lymphocytes within the tumor
microenvironment as having both prognostic and predictive
value for patients with cancer and for treatment with immunotherapy, respectively. There have also been several
trials of T-cell checkpoint inhibitors in which PD-L1 expression in the tumor microenvironment has been
KEY POINTS
e494
Checkpoint inhibitors have demonstrated

unprecedented rates of durable responses in difficultto-treat tumors; however, many treated patients still do
not respond.
Biomarkers remain a critical missing link in attempting
to identify appropriate candidates for immunotherapy
and tailoring immunotherapy treatment regimens.
Examples of biomarkers being explored in
immunotherapy include serum proteins, tumor-specific
receptor expression patterns, factors in the tumor
microenvironment, circulating immune and tumor cells,
and host genomic factors.
The identification of appropriate biomarkers has been
limited by little extension into large prospective trials
that can validate their use and variability in assay
development and interpretation.
Tumor-infiltrating lymphocytes, tumor PD-L1
expression, tumor mutational burden, and several other
factors are being considered as candidate biomarkers.
associated with more favorable outcomes, although this has

not been uniformly demonstrated. Other groups have used
a larger panel of gene signatures, including T helper (Th)
1 cytokines, chemokines, and other factors, that correlate with
therapeutic responses. These studies collectively suggest
that there may be host, tumor, and immune factors that can
be used for biomarker development. The importance of the
tumor microenvironment has been appropriately stressed,
but, practically, the ability to use serum or peripheral blood
biomarkers is of major clinical utility. We will explore the
status of current biomarker candidates from both compartments (Table 1).
PERIPHERAL BLOOD BIOMARKERS

Candidate biomarkers from serum, plasma, or peripheral
blood must be accurately and reproducibly measurable,
clinically feasible, cost-effective, and prospectively validated
in randomized clinical trials. Putative biomarkers in blood
might be soluble factors such as serum proteins, circulating
tumor DNA, or other cellular factors such as tumor cells,
T-cell subsets, or other immune cell populations. The serum
factors may be single or could include a panel of factors
preferably measured by a single, validated assay. To date,
most published analyses of peripheral blood biomarkers in
immunotherapy have been retrospective and hypothesis
generating, although important information has been gained
that illuminates the mechanisms of clinical benefit with some
approaches and has helped inform subsequent clinical trial
design.
Serum Soluble Biomarkers

The potential role of soluble serum proteins was first suggested by studies of high-dose interleukin-2 (HD IL-2) among
patients with advanced melanoma and RCC in the 1990s.
High pretreatment serum levels of IL-6 and C-reactive
protein (CRP) were identified as possible prognostic markers
for treatment failure and shorter overall survival (OS) in
metastatic RCC after IL-2 therapy.13 Later, a French multiinstitutional study revealed that high pretreatment serum
values of CRP were independent predictors of resistance to
IL-2 therapy among patients with metastatic melanoma.14
More recently, in a retrospective analysis of patients with
advanced melanoma, pretreatment serum VEGF and fibronectin, as measured by a customized proteomics array,
were found to inversely correlate with clinical response to
IL-2 treatment.15 In this trial, high levels of these proteins
were associated with lack of clinical response and decreased
OS. A prospective study of patients with melanoma being
treated with IL-2 (Melanoma SELECT) has been completed
and will help to confirm the validity of these markers as
predictive biomarkers.
Elevated levels of VEGF and CRP have similarly been associated with clinical outcomes for patients treated with
ipilimumab, an antibody that targets the T-cell checkpoint
CTLA-4 and was approved by the FDA in 2011 for the
treatment of patients with advanced melanoma. In an
TABLE 1. Potential Predictive Biomarkers for Immunotherapy

Type
Source
Biomarker
Clinical Significance
Soluble
Serum
IL-6
High levels are prognostic for HD IL-2 treatment failure and shorter OS
in metastatic RCC13
CRP
High levels predict resistance to HD IL-214; decreasing levels during ipilimumab

therapy are associated with disease control and survival18
VEGF
High levels are an independent predictor for lack of response to HD IL-215 and are
associated with decreased OS16
LDH
Low pretreatment levels predict benefit from ipilimumab17; decreasing levels

during ipilimumab therapy are associated with disease control and survival18
sCD25
High levels predict resistance to ipilimumab19
NY-ESO-1 antibody
Seropositivity has greater likelihood to respond to CTLA-4 blockade29,30
Neutrophils/leukocytes
High counts are prognostic for HD IL-2 treatment failure and shorter OS21
Lymphocytes
Immediate lymphocytosis is associated with response to HD IL-2 therapy26
Cellular
Peripheral blood
CD8 T cells
Presence is associated with clinical benefit to CTLA-4 blockade30
ALCs
Increasing counts during ipilimumab therapy are associated with an improved

OS18,23; however, this may occur in all patients regardless of benefit27
Eosinophils
+
CD4 ICOS T cells
Increase in frequency after ipilimumab46
MDSCs
Low frequency predicts benefit from ipilimumab therapy25
Tumor
PD-L1
Refer to Table 2
TILs
CD4+ICOShigh T cells
Increased frequency correlates with clinical benefit in ipilimumab41,44-46
Genomic
Increasing counts during ipilimumab therapy are associated with an improved OS23
Tumor
CD8 T cells
PD-1/PD-L1 expression on these cells predicts response to PD-1 blockade47,55,57
Tumor mutation loads
Predict clinical benefit to ipilimumab8,75 and PD-1 blockade7
MMR
Predicts clinical benefit to PD-1 blockade76,77
Abbreviations: IL, interleukin; HD, high-dose; OS, overall survival; RCC, renal cell carcinoma; CRP, C-reactive protein; LDH, lactate dehydrogenase; NY-ESO-1, NY-esophageal
cancer 1; ALC, absolute lymphocyte count; ICOS, inducible T-cell costimulator; MDSC, myeloid-derived suppressor cell; TIL, of tumor-infiltrating lymphocyte; MMR, mismatch
repair.
analysis of sera collected from 176 patients with melanoma

treated with ipilimumab, pretreatment VEGF of 43 pg/mL
or greater was associated with decreased OS.16 In addition,
elevated serum lactate dehydrogenase (LDH) levels have
demonstrated a negative predictive value, with limited longterm benefit from ipilimumab treatment found among
patients with baseline serum LDH greater than twice the
upper limit of normal in a cohort of 166 patients from the
Netherlands.17 These findings were also reported in an independent cohort of 64 patients from the United Kingdom.
Simeone et al18 also found that decreasing levels of serum
LDH and CRP between baseline and the end of ipilimumab
treatment at week 12 were significantly associated with
both disease control and survival for 95 patients treated with
ipilimumab (p , .0001). Although they are often commercially available, these laboratory assessments are not factored into current clinical algorithms for making treatment
decisions. Another soluble factor that has been proposed is
baseline serum concentrations of soluble CD25, the alpha
chain of the IL-2 receptor, which independently predicted
OS in a large retrospective trial of 262 patients. Low levels
of CD25 were associated with favorable outcomes and elevated levels predicted resistance to ipilimumab therapy, a
finding rooted in preclinical studies demonstrating that
blockade of the IL-2 receptor subunits abrogated the antitumor efficacy of CLTA-4 blockade.19 It is not currently clear
whether these biomarkers are predictive or merely prognostic factors. The prospective SELECT trial may help clarify
this important distinction.
Peripheral Blood Cellular Biomarkers

Similar to serum soluble factors, cells in the peripheral blood,
such as T cells, NK cells, dendritic cells, macrophages, and
tumor cells, have been studied as both prognostic and
predictive factors in multiple clinical trials. Elevated peripheral blood neutrophil and monocyte counts were associated with poor survival among patients with metastatic
melanoma20 and were an independent prognostic factor for
short OS among patients with stage IV melanoma undergoing IL-2based immunotherapy in the retrospective
European Organisation for the Research and Treatment of
Cancer 18951 study.21 Interestingly, neutrophil and monocyte counts did not predict clinical benefit in a recent retrospective trial of 36 patients treated with ipilimumab,
although an early increase in eosinophil count was associated with improved clinical benefit,22 a finding reported
previously23 that can perhaps be explained as a sign of a
developing inflammatory response. In this analysis, myeloidderived suppressor cells (MDSCs), a heterogeneic immunoregulatory population of immune cells, were also negatively
correlated with clinical benefit. This finding was consistent with
reports of poorer prognosis for patients with increased MDSCs
e495
SPENCER ET AL
during the first 24 weeks of treatment and improved outcome

with low baseline MDSC counts.24,25
Lymphocytes are perhaps the most frequently studied
peripheral blood component as predictors of response to
immunotherapy. It is well established that initial lymphopenia and rebound lymphocytosis are hematologic side
effects of IL-2. A strong positive association between clinical
response and the degree of lymphocytosis immediately after
therapy has been reported.26 In a study of ipilimumab
treatment, decreasing regulatory CD4+FoxP3+ T cells and
increasing absolute lymphocyte counts (ALCs) during
treatment were significantly associated with both disease
control and survival.18 This same trend of increasing ALCs
correlating with improved clinical benefit has been reported
elsewhere,23 but a pooled analysis of data from several
studies noted this increase among all patients treated regardless of benefit.27 Martens et al28 recently reported a
composite analysis suggesting that a signature of low LDH,
absolute monocyte counts, high relative lymphocyte counts,
eosinophil counts, and regulatory T cells are associated with
favorable outcomes among 209 patients, suggesting that
evaluating multiple dynamic cell populations may be necessary to achieve predictive power.
Finally, the presence of induced autoimmunity has long
been hypothesized to predict clinical benefit from immunotherapy. Metastatic melanoma has been associated with
spontaneous antibody formation to a variety of common
tumor antigens, including differentiation antigens gp100,
tyrosinase, and Melan-A/melanoma antigen recognized by
T-cells 1 (MART-1), as well as cancer/testis antigens melanoma associated antigen 3 (MAGE-3) or NY-esophageal
cancer 1 (NY-ESO-1). As many as 50% of patients with advanced melanoma with NY-ESO-1expressing tumors spontaneously develop antibody.29 The patients with NY-ESO-1
seropositivity at baseline showing responses or with seronegativity showing a seroconversion had a greater likelihood
of experiencing clinical benefit 24 weeks after treatment
with ipilimumab (an antiCTLA-4 antibody) than their seronegative counterparts.30
TUMOR CELL PHENOTYPE AND TUMOR

MICROENVIRONMENT BIOMARKERS
Although blood is more easily accessible, there is clear
evidence that activity within the tumor microenvironment
provides important clues as to how an individual tumor and
patient may respond to immunotherapy.
Tumor-Infiltrating Immune Cells as Biomarkers

Tumors with an active immune microenvironment, as exemplified by infiltration of activated, effector T cells, may be
better primed to respond to immunotherapy. Indeed, analyses of melanoma samples revealed that two distinct
phenotypic classes of tumor microenvironments exist: those
with a high prevalence of T cells (T-cell predominant or T-cell
inflamed) and those without T cells (T-cell poor or nonT-cell
inflamed; Fig. 1).31,32 The T cellinflamed tumors harbor
e496
large numbers of T cells at the tumor periphery, with increased expression of T-cell activation markers, type 1 interferon signatures, and high levels of Th1 cytokines and
chemokines that can effectuate T-cell recruitment and effector functions. In contrast, the nonT cellinflamed tumor
microenvironment has few, if any, effector T cells, although
these tumors may contain evidence of chronic inflammation
with tumor-associated macrophages, MDSCs, Th2 cytokines,
and chemokines, resulting in an immunosuppressed microenvironment that allows tumor progression. The precise
underlying mechanisms that mediate the type of immune
microenvironment in cancer are not completely understood.
Nonetheless, there have been several tumor-derived soluble
and cell membrane factors that may be responsible for the
observed phenotypic differences. For example, some tumor
cells in an inflamed environment will express high levels of
T-cell checkpoints, such as PD-L1, B7H4, Tim-3, Lag-3, and
others that can disable tumor-infiltrating effector cells.
Furthermore, various metabolic changes within inflamed
tumors, such as beta-catenin expression and intracellular
hypoxia and metabolic demand, or release of soluble
factors, such as indoleamine-2,3-dioxygenase, IL-10, VEGF,
and transforming growth factor-beta may result in inhibition of effector T cells. Some tumors may also exhibit
high levels of CD4+FoxP3+ regulatory T cells. These events
have a cumulative effect of ultimately preventing effector
T-cell function despite their abundance (Fig. 1).31 The
importance of T cells within the tumors has been underscored by the prognostic influence of such cells in a variety of
solid tumors, including colorectal, hepatocellular, gallbladder, pancreatic, esophageal, ovarian, endometrial, cervical,
bladder, nonsmall cell lung, prostate, head and neck, and
breast cancer.33-36
Currently, there is some controversy about which immune
cells are important in terms of promoting antitumor immunity versus tumor progression. The most important cells
for promoting antitumor immunity are likely CD8+ cytotoxic
T cells, CD4+ Th1 cells, NK cells, and mature dendritic cells.
The tumors with this lymphocyte-predominant infiltrate
pattern do seem to benefit more from tumor immunotherapy
than cancers without these tumor-infiltrating lymphocytes
(Fig. 1).31 The elegant work of Galon et al34 suggests that when
colorectal cancer lesions were analyzed for simply CD3+ T cells
or CD8+ T cells, a significant improvement in overall survival
was seen. In a different series, T cellinfiltrated melanomas,
especially those with high CD8+ T-cell content, were more
likely to be associated with high PD-L1 expression, improved
prognosis, and longer time to development of brain metastases.37 The presence of intratumoral CD3+ T cells is also
correlated with improved progression-free survival (PFS)
and OS among patients with advanced ovarian carcinoma.38
High peritumoral densities of infiltration of lymphocytes
expressing the T-cell activation markers CD25 or OX40 were
associated with longer survival of patients with cutaneous
malignant melanoma.39 Concurrent high CD8+ and CD4+ T-cell
infiltration in nonsmall cell lung carcinoma was found to be
an independent favorable prognostic factor.40
FIGURE 1. Tumors Exhibiting a T CellInflamed or Non-T-Cell-Inflamed Phenotype
(Top) Some tumors exhibit a T cellinflamed phenotype. In these tumors, a large number of tumor-infiltrating lymphocytes (TILs) and chemokines that recruit T cells are found.
Negative immune regulators including Fox P3+ regulatory T cells, PD-L1, and indoleamine-2,3-dioxygenase (IDO) are present as well; a type 1 interferon signature may also be present.
(Bottom) In contrast, some tumors exhibit a non-T-cellinflamed phenotype with an inverse pattern in which TILs are absent but chronic inflammation, as evidenced by common
suppressive cytokines, tumor-associated macrophages, and myeloid-derived suppressor cells (MDSCs), exists.31,73
Additional studies of markers of immune activation, such

as inducible T-cell costimulator (ICOS), a T cellspecific
molecular that belongs to the CD28/CTLA-4 family and is
expressed only after T-cell activation, have been performed.
ICOS is thought to play an especially important role in T-cell
survival, proliferation, and generation of memory T cells. In a
neoadjuvant trial of 12 patients with bladder cancer treated
with ipilimumab, investigators demonstrated an increase in
peripheral and intratumoral ICOShighCD4+ T cells, which may
have an association with good clinical responses to ipilimumab therapy.41 Similar findings have been reported in
patients with prostate cancer, breast cancer, or mesothelioma treated with ipilimumab or tremelimumab.42-45 A
statistically significant increase in the frequency of ICOS+
CD4+ T cells measured by flow cytometry in the peripheral
blood from patients with melanoma was found after ipilimumab treatment.46 These data suggest that the frequency
of ICOS+ CD4+ T cells may be useful in monitoring the biologic
activity of antiCTLA-4 therapy.
In contrast with the lymphocytes and activation markers
mentioned thus far, the presence of CD4+FoxP3+ regulatory
T cells, MDSCs, tumor-associated macrophages, and Th2type cytokine profiles has been associated with poor
prognosis or lack of response to immunotherapy. Th17 cells

are also controversial with mixed correlations in clinical
reports.
PD-1/PD-L1 Biomarkers in Development

Among the most important potential biomarkers of the last
few years are PD-1 and PD-L1 (Table 2). PD-1 is expressed in
the majority of tumor-infiltrating T cells, including antigenspecific CD8+ T cells. PD-1 expression occurs after T-cell
activation and has been used as a marker of T-cell exhaustion. The PD-1 receptor binds to two ligands, which have
been designated PD-L1 and PD-L2. When PD-1 is engaged, it
results in T-cell elimination and PD-1 is thus considered a
T-cell checkpoint whose physiologic function is likely designed
to eliminate activated T cells once they have completed their
effector functions. When the interaction between PD-1 and
PD-L1 is inhibited (for example, with antiPD-1 or antiPD-L1
antibodies), T cells remain activated, mediating antitumor
activity.47 Interestingly, PD-L1 expression has been detected
in several different types of tumor cells, including melanoma, nonsmall cell lung cancer, RCC, bladder cancer,
gastric cancer, ovarian cancer, B-cell lymphoma cells, Merkel
cell carcinoma, and Hodgkin lymphoma (Table 2).
e497
SPENCER ET AL
TABLE 2. Summary of PD-L1 Expression and Response to Therapy in Various Clinical Trials
First Author
Tumor Type
Therapy
Cutoff (%)
Biomarker Results
Topalian
Advanced melanoma, NSCLC,

CRPC, RCC, and CRC
Pembrolizumab
0 of 17 patients with PD-L1negative tumors had

objective response
Borghaei51
Advanced nonsquamous NSCLC
Nivolumab vs.
docetaxel
1, 5, and 10
Nivolumab had superior efficacy to docetaxel, greater

with higher tumor membrane PD-L1 expression
Muro80
Gastric
Pembrolizumab
Tumor PD-L1 expression was associated with ORR
Taube53
Melanoma, NSCLC, RCC,

CRC, CRPC
Nivolumab
Tumor cell PD-L1 expression correlated with objective

response
Disis68
Recurrent/refractory ovarian
cancer
Avelumab
Trend toward better response rates in PD-L1positive

tumors
Garon49
Advanced NSCLC
Pembrolizumab
50
PD-L1 expression in at least 50% of tumor cells correlated

with improved efficacy
Powles69
Bladder
Atezolizumab
(antiPD-L1)
1, 5, and 10
PD-L1positive tumors at . 5% had particularly high

response rates
Weber66
Advanced melanoma progressed

on antiCTLA-4 therapy
Nivolumab vs.
investigators
choice
Higher response rates with nivolumab correlated with

positive tumor PD-L1 expression, but patients with
PD-L1negative tumors still had benefit
Weber65
Advanced melanoma progressed

on prior therapy/CTLA-4 therapy
Nivolumab
1 and 5
PD-L1 positivity correlated significantly with better

response but negativity did not rule out response
Kefford67
Melanoma
Pembrolizumab
PD-L1 positivity associated with improved ORR and

PFS, but activity observed in patients with low PD-L1
expression
Robert70
Metastatic melanoma
Nivolumab vs.
dacarbazine
Nivolumab-treated patients had improved objective

response rate and overall survival, regardless of
PD-L1 status
Motzer71
Metastatic RCC
Nivolumab
1 and 5
Response rates were higher with greater PD-L1

expression ($ 5%), but those with lower expression
(, 5%) also had meaningful responses
Brahmer50
Advanced progressed squamous

NSCLC
Nivolumab vs.
docetaxel
1, 5, and 10
Expression of PD-L1 was neither prognostic nor

predictive of benefit
Herbst57
Advanced melanoma, NSCLC,

RCC, and other
Atezolizumab
Response correlated with PD-L1 expression by

tumor-infiltrating immune cells, but correlation
between response and PD-L1 expression by tumor cells
was not significant
48
Abbreviations: NSCLC, nonsmall cell lung cancer; CRPC, castration-resistant prostate cancer; RCC, renal cell carcinoma; CRC, colorectal cancer; ORR, overall response rate; PFS,
progression-free survival.
In early phase I clinical studies of PD-1 blockade, clinical

responses were noted among patients with melanoma, RCC,
and nonsmall cell lung cancer.48 In many studies, a correlation between clinical response and the presence of
PD-L1 expression on tumor cells was reported, although
this was not seen in every trial (Table 2). Based on this
unexpected finding, some clinical trials used baseline
PD-L1 expression as a criterion for study participation.
A clinical trial of pembrolizumab, a monoclonal antibody
that targets PD-1, was conducted in previously treated
patients with PD-L1 expressing nonsmall cell lung cancer
and demonstrated an objective response rate of 19%,
leading to FDA approval of the agent but only for patients
whose tumors stain positive for PD-L1. 49 Other clinical
studies using nivolumab, a comparable antiPD-1 agent,
also showed significant clinical responses in patients
with nonsmall cell lung cancer resulting in FDA approval
without requiring pretreatment PD-L1 expression testing because this was not used as a study eligibility
criterion. 50,51
e498
There are multiple challenges with using PD-1 and PD-L1

expression as biomarkers. First, PD-L1 expression is heterogenous and dynamic within an individual. PD-L1 expression
can be induced by activated tumor-specific T cells, further
demonstrating that PD-L1 expression is a dynamic, not a
static, process.52,53 PD-L1 expression is often heterogeneous
within a sample, thus using focal expression may reflect the
biology of the tumor or may simply be sampling error.53
Furthermore, PD-L1 expression is often discordant between
the primary lesion and its metastases, and a positive marker
in one may not be reflective of the other.48,52,54 Additionally,
multiple cell types within a tumor besides the tumor cells
themselves may express PD-L1, such as lymphocytes and
macrophages, potentially altering readings,55 especially if
these inflammatory cells represent the host immune response to tumor.56 A higher number of PD-1 and PD-L1
expressing tumor-infiltrating CD8+ T cells was associated
with clinical responses among patients with metastatic
melanoma treated with pembrolizumab.55 In a study evaluating MPDL3280A, an antiPD-L1 antibody, in multiple
cancer types, responses were observed in patients with

tumor-infiltrating immune cells expressing high levels of
PD-L1.57 These results suggest that PD-1/PD-L1expressing
tumor-infiltrating immune cells may be predictive of tumors
likely to be responsive to T-cell checkpoint blockade and may
serve as a predictive biomarker. In most of these studies,
however, there was a low rate of clinical response observed
in patients with tumors having low levels of PD-1/PD-L1
expression. Thus, the sensitivity and specificity must be
better defined to avoid overtreatment of patients positive
for PD-1/PD-L1 who will not respond and undertreatment of
patients negative for PD-1/PD-L1 who might respond.
SPECIFIC PD-1/PD-L1 BIOMARKER ASSAYS

Although the immunohistochemistry (IHC) assay for PD-L1 is
affordable and relatively easy to perform on tissue, it is
limited by factors such as adequacy and abundance of the
sample. The PD-L1 ligand has only two hydrophilic regions
where antibodies bind, and IHC antibodies seemingly bind
at a different site than therapeutic antibodies, possibly
compromising the efficacy of the technique.52,58,59 Different
IHC antibodies and staining techniques introduce further
variability into the test and may influence positive or negative test rates.52 Without standardization across assays, it
becomes difficult to compare clinical trial results, especially
if the assays details are not easily available. Indeed, in a
recent study by McLaughlin et al60 testing two antiPD-L1
rabbit monoclonal antibodies, nearly 25% of lung cancer
samples that were positive for PD-L1 with one antibody tested
negative with the second. Especially pertinent to recent investigations, different assessment methods such as reading
tumor cells or tumor-infiltrating lymphocytes or both, different scoring methods such as percent staining positive or
H score, and different cutoff values set for positive or negative
results (1%, 5%, 10%, or even 50%) prevent standardization
across tumor types and same therapies.52
In October 2015, the FDA gave the first regulatory approval
to an immunotherapy biomarker to Mercks PD-L1 companion diagnostic assay. This assay, PD-L1 IHC 22C3
pharmDx, uses the monoclonal antibody clone 22C3 for
membrane staining of tumor and/or infiltrating immune
cells in formalin-fixed, paraffin-embedded samples, with
a positivity cutoff of 1% or greater. The biomarker was
evaluated in the KEYNOTE-001 study of pembrolizumab in
patients with advanced nonsmall cell lung cancer. To
clinically validate the assay, 495 patients were assigned
either to a training group (one-third) or a validation group
(two-thirds). Data obtained from the training group were
used to establish the cutoff for PD-L1 positivity, which was
ultimately defined as expression in 50% of tumor cells. The
outcomes data from the validation group were analyzed
using the 50% cutoff defined by the training group. Patients
with a score of 50% or greater had a higher response rate,
longer PFS, and longer OS. Given the link between PD-L1
expression defined by the PD-L1 IHC 22C3 pharmDx assay
and efficacy with pembrolizumab, the FDA approved the
assay as a companion diagnostic for the indication of

identifying patients with nonsmall cell lung cancer for
treatment with pembrolizumab.49,61,62
The antiPD-1 antibody nivolumab also has a companion
diagnostic PD-L1 assay. PD-L1 IHC 28-8 pharmDx uses the
monoclonal antibody 28-8 to score tumor cell membranes in
formalin-fixed, paraffin-embedded tissue. Although the
original CheckMate 057 study of nivolumab versus docetaxel
was conducted among an unselected population of patients
with advanced nonsquamous nonsmall cell lung cancer, a
retrospective analysis of the data showed that patients
whose tumors expressed PD-L1 at a low threshold value of
1% or greater had a more significant improvement over
docetaxel than PD-L1negative patients across all endpoints. However, the retrospective PD-L1 testing was only
performed on 78% of the patient specimens, which may
introduce selection bias.51,61 This may be reflected in the
premarket approval from the FDA, which states that PD-L1
expression as detected by the 28-8 pharmDx assay in nonsquamous nonsmall cell lung cancer may be associated
with enhanced survival from nivolumab.61,63 This assay was
approved by the FDA in late 2015 in association with the
expanded approval of nivolumab in previously treated
nonsmall cell lung cancer.
The POLAR study examined the efficacy of the PD-L1 inhibitor atezolizumab versus docetaxel among patients with
advanced nonsmall cell lung cancer of squamous and nonsquamous histologies. This study used an SP142 antibody to
the membranes of tumor cells and tumor-infiltrating immune
cells. The data confirmed the efficacy of atezolizumab independently of PD-L1 expression, but the superiority of
atezolizumab over docetaxel was more pronounced with
higher expressions of PD-L1 as measured by the SP142 assay.61,64
As noted elsewhere, clinically meaningful responses have
been seen among patients whose tumors have been defined
as negative for PD-L149,65-71 (Table 2). In addition, among
patients receiving combination checkpoint inhibitor therapy, responses among patients with PD-L1positive and
PD-L1negative results were similar.72 Given that a negative
result may still have considerable variability and may
exclude a patient from a therapy that could confer significant
clinical benefit, perhaps the more appropriate immediate
use of a PD-L1 biomarker is not to exclude patients with
negative results from treatment, but to better refine predictors of response by evaluating PD-L1 expression in
combination with other putative predictive markers to
better improve the sensitivity and specificity.73
TUMOR GENOMIC BIOMARKERS

The ability of T-cell checkpoint inhibitors to induce an effective immune response in some cancers raises the critical
question of what antigen(s) are being targeted by the activated T cells. It was originally hypothesized that the
mechanism of antitumor activity might be through stoking
pre-existing, ineffective tumor-reactive T cells. In several
recent studies, however, it has become clear that clinical
e499
SPENCER ET AL
responses to T-cell checkpoint inhibitors are correlated with

overall mutation burden in the tumor cell.7,8,74,75 This
finding suggested that the T-cell response may be targeted
not to established antigens, but to so-called neoantigens,
that evolve as the mutation rate increases in the tumor cell.
Most of these mutations are likely passenger mutations
that may be influencing the immuno-editing process and
placing selective pressure on the immune system to promote tumor elimination through new, previously unedited
antigens. Although this is an intriguing hypothesis, further
work is needed to firmly establish whether mutation burden
or neoantigen emergence can be predictive biomarkers for
tumor response to immunotherapy agents.
A recent study using whole-exome sequencing (WES) to
analyze the potential effect of cancer genomes on the response to ipilimumab among 64 patients with melanoma
confirmed that a high mutational load and the number of
nonsynonymous mutations per exome were significantly
correlated with improved OS.75 Using a bioinformatics
pipeline, the authors identified 101 tetrapeptide (four amino
acid) motifs within the nonamer (nine amino acid) peptides
sitting in the peptide-binding grooves formed by major
histocompatibility complex class I molecules. These tetrapeptides were shared exclusively among patients in the
discovery cohort who had long-term clinical benefit. A recent correction to the original report clarified that this
analysis did not have a true independent validation set
because all samples were used in the process of identifying
the final set of neoantigens.74
Indeed, mutation burden may be as predictive as neoantigen load in the landscape of response to immunotherapy. Rizvi et al7 observed that a high mutational burden
may be predictive of response to immunotherapy. Using
WES of nonsmall cell lung cancers treated with pembrolizumab, this study revealed higher nonsynomymous
mutation burden in tumors, and clinical responses were
correlated with molecular signature characteristics of tobacco carcinogen-related mutagenesis, higher neoantigen
burden, and DNA repair pathway mutations. In one responder, pembrolizumab enhanced neoantigen-specific
CD8+ T-cell reactivity that was associated with tumor regression. In a series of 110 patients with melanoma, overall
mutation load, neoantigen load, and expression of immune
microenvironment cytolytic markers were associated with
clinical benefit, but no recurrent neoantigen peptide sequences predicted responder patient populations.8 Therefore, although it is intriguing, the neoantigen hypothesis
needs further validation in other cohorts.
Although heavy mutation burdens may be seen in lung and
melanoma tumors, which are often generated by exogenous
environmental insults such as tobacco and ultraviolet light,
gene mutations that beget mutations, such as DNA mismatch repair (MMR) gene deficiency, can also result in a
heavy mutational burden and may be useful to predict response to immunotherapy. A phase II study evaluated the
response to pembrolizumab therapy among patients with
MMR-deficient cancers. Although 0% of patients (zero of 18)
e500
with MMR-proficient colorectal cancer responded to therapy, 40% of patients (four of 10) with MMR-deficient colorectal cancer and 71% of patients (five of seven) with
MMR-deficient noncolorectal cancer had immune-related
objective responses.76,77 Further WES revealed much higher
somatic mutation loads with a mean of 1,782 mutations per
tumor in MMR-deficient tumors, compared with a mean of
only 73 mutations per tumor in MMR-proficient tumors.
Mutations in other enzymes involved in DNA replication and
repair, such as the DNA polymerase epsilon gene (POLE) and
DNA polymerase delta 1 (POLD1), have been implicated in
the generation of a high mutation burden and associated
with response to immunotherapy. The POLE mutation is
associated with disruption of the exonuclease activity required for proofreading function and results in a high mutational burden or ultramutator phenotype.78 Both POLE
and POLD1 mutations have been noted among patients with
nonsmall cell lung cancer who are responsive to pembrolizumab, and POLE and POLD1 are associated with high
mutational burdens.7 In addition, a case of an exceptional
response to pembrolizumab for a patient with endometrial
cancer whose tumor had a somatic POLE mutation was
recently reported.79 These data suggest that the presence of
MMR deficiency or POLE mutation may identify a subset of
seemingly diverse cancers that are especially vulnerable to
immune checkpoint blockade, although this approach must
be validated in further prospective studies.
FUTURE DIRECTIONS
The progress in tumor immunotherapy over the last 5 years
has been truly remarkable, with several monotherapy and
combination therapy regimens demonstrating impressive
clinical benefit across numerous solid and hematologic
malignancies. The ability to have a predictive biomarker to
better select appropriate patients for specific treatment
regimens and help define prognosis and other important
patient outcomes is a high priority in the field. Although
several soluble factors and cell-surface receptors in the
peripheral blood and tumor microenvironments have been
proposed, none have yet been fully validated in prospective,
randomized clinical trials. Putative biomarkers of interest
include the number and location of tumor-infiltrating immune cells, PD-1 and PD-L1 expression, and host genomic
factors, such as mutation burden and neoantigen emergence. Biomarker development also depends on having standardized, reproducible, and feasible assays that can be rapidly
completed with clearly defined cutoff values and with high
sensitivity and specificity. Although it may be premature to
recommend any of the current biomarkers for routine clinical
practice, there should be a high priority to include biomarkers
in clinical trial design for ongoing investigations of tumor immunotherapy regimens. Combination strategies and/or larger
panels may be of special interest for improving the predictive
power of contemporary biomarkers. As the use of immunotherapy is further optimized and novel combination
approaches are developed, biomarker investigation will
assume a critical role to inform the rational selection of

patients for this treatment approach.
ACKNOWLEDGMENT
K.R.S. and J.W. contributed equally to this work.
References
1. Dizon DS, Krilov L, Cohen E, et al. Clinical cancer advances 2016: annual
report on progress against cancer from the American Society of Clinical
Oncology. J Clin Oncol. 2016;34:987-1011.
2. Ansell SM, Lesokhin AM, Borrello I, et al. PD-1 blockade with nivolumab
in relapsed or refractory Hodgkins lymphoma. N Engl J Med. 2015;372:
311-319.
3. Gettinger SN, Horn L, Gandhi L, et al. Overall survival and long-term
safety of nivolumab (anti-programmed death 1 antibody, BMS-936558,
ONO-4538) in patients with previously treated advanced non-small-cell
lung cancer. J Clin Oncol. 2015;33:2004-2012.
4. Larkin J, Lao CD, Urba WJ, et al. Efficacy and safety of nivolumab in
patients with BRAF V600 mutant and BRAF wild-type advanced melanoma: a pooled analysis of 4 clinical trials. JAMA Oncol. 2015;1:
433-440.
5. Postow MA, Chesney J, Pavlick AC, et al. Nivolumab and ipilimumab
versus ipilimumab in untreated melanoma. N Engl J Med. 2015;372:
2006-2017.
6. Schumacher TN, Kesmir C, van Buuren MM. Biomarkers in cancer
immunotherapy. Cancer Cell. 2015;27:12-14.
cancer. Science. 2015;348:124-128.
8. Van Allen EM, Miao D, Schilling B, et al. Genomic correlates of response
to CTLA-4 blockade in metastatic melanoma. Science. 2015;350:
207-211.
9. Strimbu K, Tavel JA. What are biomarkers? Curr Opin HIV AIDS. 2010;5:
463-466.
10. Fox BA, Schendel DJ, Butterfield LH, et al. Defining the critical hurdles in
cancer immunotherapy. J Transl Med. 2011;9:214.
11. Nelson D, Fisher S, Robinson B. The Trojan Horse approach to tumor
immunotherapy: targeting the tumor microenvironment. J Immunol
Res. 2014;2014:789069.
12. Das B, Tsuchida R, Malkin D, et al. Hypoxia enhances tumor stemness by
increasing the invasive and tumorigenic side population fraction. Stem
Cells. 2008;26:1818-1830.
13. Blay JY, Negrier S, Combaret V, et al. Serum level of interleukin 6 as a
prognosis factor in metastatic renal cell carcinoma. Cancer Res. 1992;
52:3317-3322.
14. Tartour E, Blay JY, Dorval T, et al. Predictors of clinical response to
interleukin-2based immunotherapy in melanoma patients: a French
multiinstitutional study. J Clin Oncol. 1996;14:1697-1703.
15. Sabatino M, Kim-Schulze S, Panelli MC, et al. Serum vascular endothelial
growth factor and fibronectin predict clinical response to high-dose
interleukin-2 therapy. J Clin Oncol. 2009;27:2645-2652.
16. Yuan J, Zhou J, Dong Z, et al. Pretreatment serum VEGF is associated
with clinical response and overall survival in advanced melanoma
patients treated with ipilimumab. Cancer Immunol Res. 2014;2:
127-132.
17. Kelderman S, Heemskerk B, van Tinteren H, et al. Lactate dehydrogenase as a selection criterion for ipilimumab treatment in
metastatic melanoma. Cancer Immunol Immunother. 2014;63:449-458.
18. Simeone E, Gentilcore G, Giannarelli D, et al. Immunological and biological changes during ipilimumab treatment and their potential
correlation with clinical response and survival in patients with advanced
melanoma. Cancer Immunol Immunother. 2014;63:675-683.
19. Hannani D, Vetizou M, Enot D, et al. Anticancer immunotherapy by

CTLA-4 blockade: obligatory contribution of IL-2 receptors and negative
prognostic impact of soluble CD25. Cell Res. 2015;25:208-224.
20. Schmidt H, Bastholt L, Geertsen P, et al. Elevated neutrophil and
monocyte counts in peripheral blood are associated with poor survival
in patients with metastatic melanoma: a prognostic model. Br J Cancer.
2005;93:273-278.
21. Schmidt H, Suciu S, Punt CJ, et al; American Joint Committee on Cancer
Stage IV Melanoma; EORTC 18951. Pretreatment levels of peripheral
neutrophils and leukocytes as independent predictors of overall survival
in patients with American Joint Committee on Cancer Stage IV Melanoma: results of the EORTC 18951 Biochemotherapy Trial. J Clin Oncol.
2007;25:1562-1569.
22. Gebhardt C, Sevko A, Jiang H, et al. Myeloid cells and related chronic
inflammatory factors as novel predictive markers in melanoma treatment with ipilimumab. Clin Cancer Res. 2015;21:5453-5459.
23. Delyon J, Mateus C, Lefeuvre D, et al. Experience in daily practice with
ipilimumab for the treatment of patients with metastatic melanoma: an
early increase in lymphocyte and eosinophil counts is associated with
improved survival. Ann Oncol. 2013;24:1697-1703.
24. Kitano S, Postow MA, Cortez C, et al. Myeloid-derived suppressor cell
quantity prior to treatment with ipilimumab at 10mg/kg to predict for
overall survival in patients with metastatic melanoma. J Clin Oncol.
2012;30 (suppl; abstr 2518).
25. Meyer C, Cagnon L, Costa-Nunes CM, et al. Frequencies of circulating
MDSC correlate with clinical outcome of melanoma patients treated
with ipilimumab. Cancer Immunol Immunother. 2014;63:247-257.
26. Phan GQ, Attia P, Steinberg SM, et al. Factors associated with response
to high-dose interleukin-2 in patients with metastatic melanoma. J Clin
Oncol. 2001;19:3477-3482.
27. Postow MA, Chasalow SD, Yuan J, et al. Pharmacodynamic effect of
ipilimumab on absolute lymphocyte count (ALC) and association with
overall survival in patients with advanced melanoma. J Clin Oncol. 2013;
31 (suppl; abstr 9052).
28. Martens A, Wistuba-Hamprecht K, Geukes Foppen MH, et al. Baseline
peripheral blood biomarkers associated with clinical outcome of advanced melanoma patients treated with ipilimumab. Clin Cancer Res.
Epub 2016 Jan 19.
29. Gnjatic S, Nishikawa H, Jungbluth AA, et al. NY-ESO-1: review of an
immunogenic tumor antigen. Adv Cancer Res. 2006;95:1-30.
30. Yuan J, Adamow M, Ginsberg BA, et al. Integrated NY-ESO-1 antibody
and CD8+ T-cell responses correlate with clinical benefit in advanced
melanoma patients treated with ipilimumab. Proc Natl Acad Sci USA.
2011;108:16723-16728.
31. Gajewski TF, Schreiber H, Fu YX. Innate and adaptive immune cells in the
tumor microenvironment. Nat Immunol. 2013;14:1014-1022.
32. Spranger S, Spaapen RM, Zha Y, et al. Up-regulation of PD-L1, IDO, and T
(regs) in the melanoma tumor microenvironment is driven by CD8(+)
T cells. Sci Transl Med. 2013;5:200ra116.
33. Mlecnik B, Tosolini M, Kirilovsky A, et al. Histopathologic-based
prognostic factors of colorectal cancers are associated with the state
of the local immune reaction. J Clin Oncol. 2011;29:610-618.
34. Galon J, Costes A, Sanchez-Cabo F, et al. Type, density, and location of
immune cells within human colorectal tumors predict clinical outcome.
Science. 2006;313:1960-1964.
e501
SPENCER ET AL
35. Pagès F, Berger A, Camus M, et al. Effector memory T cells, early

metastasis, and survival in colorectal cancer. N Engl J Med. 2005;353:
2654-2666.
36. Jochems C, Schlom J. Tumor-infiltrating immune cells and prognosis: the
potential link between conventional cancer therapy and immunity. Exp
Biol Med (Maywood). 2011;236:567-579.
37. Kluger HM, Zito CR, Barr ML, et al. Characterization of PD-L1 expression
and associated T-cell infiltrates in metastatic melanoma samples from
variable anatomic sites. Clin Cancer Res. 2015;21:3052-3060.
38. Zhang L, Conejo-Garcia JR, Katsaros D, et al. Intratumoral T cells, recurrence, and survival in epithelial ovarian cancer. N Engl J Med. 2003;
348:203-213.
39. Ladanyi A, Somlai B, Gilde K, et al. T-cell activation marker expression on
tumor-infiltrating lymphocytes as prognostic factor in cutaneous malignant melanoma. Clin Cancer Res. 2004;10:521-530.
40. Hiraoka K, Miyamoto M, Cho Y, et al. Concurrent infiltration by CD8+
T cells and CD4+ T cells is a favourable prognostic factor in non-small-cell
lung carcinoma. Br J Cancer. 2006;94:275-280.
41. Carthon BC, Wolchok JD, Yuan J, et al. Preoperative CTLA-4 blockade:
tolerability and immune monitoring in the setting of a presurgical
clinical trial. Clin Cancer Res. 2010;16:2861-2871.
42. Calabro` L, Maio M. Immune checkpoint blockade in malignant mesothelioma. Semin Oncol. 2015;42:418-422.
43. Vonderheide RH, LoRusso PM, Khalil M, et al. Tremelimumab in
combination with exemestane in patients with advanced breast cancer
and treatment-associated modulation of inducible costimulator expression on patient T cells. Clin Cancer Res. 2010;16:3485-3494.
44. Chen H, Liakou CI, Kamat A, et al. Anti-CTLA-4 therapy results in higher
CD4+ICOShi T cell frequency and IFN-g levels in both nonmalignant and
malignant prostate tissues. Proc Natl Acad Sci USA. 2009;106:
2729-2734.
45. Liakou CI, Kamat A, Tang DN, et al. CTLA-4 blockade increases
IFNgamma-producing CD4+ICOShi cells to shift the ratio of effector to
regulatory T cells in cancer patients. Proc Natl Acad Sci USA. 2008;105:
14987-14992.
46. Ng Tang D, Shen Y, Sun J, et al. Increased frequency of ICOS+ CD4 T cells
as a pharmacodynamic biomarker for anti-CTLA-4 therapy. Cancer
Immunol Res. 2013;1:229-234.
47. Ahmadzadeh M, Johnson LA, Heemskerk B, et al. Tumor antigen-specific
CD8 T cells infiltrating the tumor express high levels of PD-1 and are
functionally impaired. Blood. 2009;114:1537-1544.
2443-2454.
49. Garon EB, Rizvi NA, Hui R, et al; KEYNOTE-001 Investigators. Pembrolizumab for the treatment of non-small-cell lung cancer. N Engl J
Med. 2015;372:2018-2028.
advanced squamous-cell nonsmall-cell lung cancer. N Engl J Med.
2015;373:123-135.
advanced nonsquamous nonsmall-cell lung cancer. N Engl J Med.
2015;373:1627-1639.
52. Meng X, Huang Z, Teng F, et al. Predictive biomarkers in PD-1/PD-L1
checkpoint blockade immunotherapy. Cancer Treat Rev. 2015;41:
868-876.
53. Taube JM, Anders RA, Young GD, et al. Colocalization of inflammatory
response with B7-h1 expression in human melanocytic lesions supports
an adaptive resistance mechanism of immune escape. Sci Transl Med.
2012;4:127ra37.
54. Madore J, Vilain RE, Menzies AM, et al. PD-L1 expression in melanoma
shows marked heterogeneity within and between patients: implications
e502
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
for anti-PD-1/PD-L1 clinical trials. Pigment Cell Melanoma Res. 2015;28:

245-253.
Tumeh PC, Harview CL, Yearley JH, et al. PD-1 blockade induces responses by inhibiting adaptive immune resistance. Nature. 2014;515:
568-571.
Bhaijee F, Anders RA. PD-L1 expression as a predictive biomarker: is
absence of proof the same as proof of absence? JAMA Oncol. 2016;2:
54-55.
Herbst RS, Soria J-C, Kowanetz M, et al. Predictive correlates of response
to the anti-PD-L1 antibody MPDL3280A in cancer patients. Nature.
2014;515:563-567.
Patel SP, Kurzrock R. PD-L1 expression as a predictive biomarker in
cancer immunotherapy. Mol Cancer Ther. 2015;14:847-856.
Sznol M, Chen L. Antagonist antibodies to PD-1 and B7-H1 (PD-L1) in the
treatment of advanced human cancer. Clin Cancer Res. 2013;19:
1021-1034.
McLaughlin J, Han G, Schalper KA, et al. Quantitative assessment of the
heterogeneity of PD-L1 expression in non-small-cell lung cancer. JAMA
Oncol. 2016;2:46-54.
Jrgensen JT. Companion diagnostic assays for PD-1/PD-L1 checkpoint
inhibitors in NSCLC. Expert Rev Mol Diagn. 2016;16:131-133.
U.S. Food and Drug Administration. Premarket approval (PMA) for
PD-L1 IHC 22C3 pharmDx. http://www.accessdata.fda.gov/scripts/
cdrh/cfdocs/cfpma/pma.cfm?id=P150013. Accessed October 2, 2015.
U.S. Food and Drug Administration. Premarket approval (PMA) for
PD-L1 IHC nivolumab pharmDx. http://www.accessdata.fda.gov/scripts/
cdrh/cfdocs/cfpma/pma.cfm?id=P150025. Accessed October 9, 2015.
Spira AI, Park K, Mazieres J, et al. Efficacy, safety and predictive biomarker results from a randomized phase II study comparing
MPDL3280A vs docetaxel in 2L/3L NSCLC (POPLAR). J Clin Oncol. 2015;
Weber JS, Kudchadkar RR, Yu B, et al. Safety, efficacy, and biomarkers of
nivolumab with vaccine in ipilimumab-refractory or -naive melanoma.
J Clin Oncol. 2013;31:4311-4318.
Weber JS, DAngelo SP, Minor D, et al. Nivolumab versus chemotherapy
Kefford R, Ribas A, Hamid O, et al. Clinical efficacy and correlation with
tumor PD-L1 expression in patients (pts) with melanoma (MEL) treated
with the anti-PD-1 monoclonal antibody MK-3475. J Clin Oncol. 2014;
32:5s (suppl; abstr 3005).
Disis ML, Patel MR, Pant S, et al. Avelumab (MSB0010718C), an anti-PDL1 antibody, in patients with previously treated, recurrent or refractory
ovarian cancer: a phase Ib, open-label expansion trial. J Clin Oncol. 2015;
Powles T, Eder JP, Fine GD, et al. MPDL3280A (anti-PD-L1) treatment
leads to clinical activity in metastatic bladder cancer. Nature. 2014;515:
558-562.
Robert C, Long GV, Brady B, et al. Nivolumab in previously untreated
Motzer RJ, Rini BI, McDermott DF, et al. Nivolumab for metastatic renal
cell carcinoma: results of a randomized phase II trial. J Clin Oncol. 2015;
33:1430-1437.
Wolchok JD, Kluger H, Callahan MK, et al. Nivolumab plus ipilimumab in
advanced melanoma. N Engl J Med. 2013;369:122-133.
Mahoney KM, Atkins MB. Prognostic and predictive markers for the new
immunotherapies. Oncology (Williston Park). 2014;28:39-48.
Chan TA, Wolchok JD, Snyder A. Genetic basis for clinical response to
CTLA-4 blockade in melanoma. N Engl J Med. 2015;373:1984.
Snyder A, Makarov V, Merghoub T, et al. Genetic basis for clinical response
to CTLA-4 blockade in melanoma. N Engl J Med. 2014;371:2189-2199.
76. Kelderman S, Schumacher TN, Kvistborg P. Mismatch repair-deficient

cancers are targets for anti-PD-1 therapy. Cancer Cell. 2015;28:11-13.
78. Hussein YR, Weigelt B, Levine DA, et al. Clinicopathological analysis of
endometrial carcinomas harboring somatic POLE exonuclease domain
mutations. Mod Pathol. 2015;28:505-514.
79. Mehnert JM, Panda A, Zhong H, et al. Exceptional response to PD-1

antibody treatment in a POLE-mutant endometrial cancer (Abstract).
Mol Cancer Ther. 2015;14:PR05.
80. Muro K, Bang YJ, Shankaran V, et al. Relationship between PD-L1 expression
and clinical outcomes in patients (Pts) with advanced gastric cancer treated
with the anti-PD-1 monoclonal antibody pembrolizumab (Pembro;
MK-3475) in KEYNOTE-012. J Clin Oncol. 2015;33 (suppl 3; abstr 3).
e503
Curing High-Risk Melanoma: Are

We There Yet?
CHAIR
Sanjiv S. Agarwala, MD
St. Lukes Medical Center
Easton, PA
SPEAKERS
Alexander C.J. van Akkooi, MD, PhD
Netherlands Cancer Institute Antoni van Leeuwenhoek
Amsterdam, Netherlands
Michael B. Atkins, MD
Lombardi Comprehensive Cancer Center
Washington, DC
Paul Lorigan, MD
Institute of Cancer Sciences, University of Manchester
Manchester, United Kingdom
SURGICAL MANAGEMENT OF MELANOMA
Surgical Management and Adjuvant Therapy for High-Risk and

Metastatic Melanoma
Alexander C.J. van Akkooi, MD, PhD, Michael B. Atkins, MD, Sanjiv S. Agarwala, MD, and Paul Lorigan, MD
OVERVIEW
Wide local excision is considered routine therapy after initial diagnosis of primary melanoma to reduce local recurrences, but
it does not impact survival. Sentinel node staging is recommended for melanomas of intermediate thickness, but it has also
not demonstrated any indisputable therapeutic effect on survival. The prognostic value of sentinel node staging has been
long established and is therefore considered routine, especially in light of the eligibility criteria for adjuvant therapy (trials).
Whether completion lymph node dissection after a positive sentinel node biopsy improves survival is the question of current
trials. The MSLT-2 study is best powered to show a potential benefit, but it has not yet reported any data. Another study, the
German DECOG study, presented at the 2015 American Society of Clinical Oncology (ASCO) Annual Meeting did not show any
benefit but is criticized for the underpowered design and insufficient follow-up. There is no consensus on the use of adjuvant
interferon in melanoma. This topic has been the focus of many studies with different regimens (low-, intermediate-, or highdose and/or short- or long-term treatment). Adjuvant interferon has been shown to improve relapse-free survival but failed
to improve overall survival. More recently, adjuvant ipilimumab has also demonstrated an improved relapse-free survival.
Overall survival data have not yet been reported due to insufficient follow-up. Currently, studies are ongoing to analyze the
use of adjuvant antiPD-1 and molecular targeted therapies (vemurafenib, dabrafenib, and trametinib). In the absence of
unambiguously positive approved agents, clinical trial participation remains a priority. This could change in the near future.
urvival rates differ according to the stage of melanoma,

and, even for the low stages of melanoma, the 10-year
survival rates differ largely, from 93% in stage IA to 39%
in stage IIC.1 Since the last update in the American Joint
Committee on Cancer (AJCC) staging system, a number of
treatment breakthroughs have occurred in terms of improvement in systemic therapy (e.g., immune checkpoint
inhibitors, BRAF/MEK inhibitors [BRAFi/MEKi]),2-12 which
may ultimately affect these survival rates. Nevertheless,
these survival rates illustrate the fact that already some
patients with melanoma and only primary tumors without
(nodal) metastases will be likely to progress and can be
considered high risk. Moreover, it demonstrates the fact that
surgery can be curative for the majority of patients with
early-stage melanoma.
SURGICAL TREATMENT OF THE HIGH-RISK

MELANOMA PATIENT: IS THERE A STANDARD?
Primary Tumor
After diagnosis, most physicians recommend performing
a wide local excision with a certain margin. An important
question is whether an increasing margin will improve
overall survival (OS). This was the topic of a few prospective

randomized controlled trials comparing 1-cm with 3-cm,
2-cm with 5-cm, and 2-cm with 4-cm resection margins.13-15
The World Health Organization trial by Cascinelli et al
compared 1-cm margins with 3-cm margins for two Breslow
thickness groups, less than 1-mm Breslow and 1.012.00-mm
Breslow, and did not find any significant difference in local
recurrence rate or in OS after a median follow-up of 12 years
among a total of 612 patients.13,14 The Swedish group
compared 2-cm with 5-cm margins for 0.82.0 mm Breslow
thickness tumors in 989 patients and found a nonsignificant
hazard ratio (HR) of 0.96 (95% CI, 0.751.24) for OS and 1.02
(95% CI, 0.801.30) for recurrence-free survival after a median follow-up of 11 years (range 717 years).15 Similar results
are reported in the other studies by Khayat et al, Karakousis
et al, Thomas et al, and Gillgren et al.16-19
In contrast to these randomized controlled trials, Haydu
et al analyzed 2,131 patients with pT2 tumors from a single
institution.20 They found that there was a significantly worse
disease-free survival for patients with a margin less than
8 mm on pathologic analysis (corresponding to , 1 cm
clinically) compared with a 8- to 16-mm margin (corresponding
From the Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital, Amsterdam, Netherlands; Georgetown Lombardi Comprehensive Cancer Center, Washington, DC; St. Lukes
University Hospital, Temple University, Allentown, PA; University of Manchester, The Christie NHS Foundation Trust, Manchester, United Kingdom.
Corresponding author: Alexander C.J. van Akkooi, MD, PhD, Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX, Amsterdam, Netherlands;
email: a.v.akkooi@nki.nl.
e505
VAN AKKOOI ET AL
to 12 cm clinically; p = .044), however, this did not translate

into any OS difference.20
Interestingly, an update of the prospective randomized
controlled trial by Thomas19 discussed at the 2015 ASCO
Annual Meeting by Hayes et al21,22 did show, for the first
time, a difference in melanoma-specific survival (HR 1.27;
95% CI, 1.021.59; p = .036) in favor of the 3-cm margin
group compared with the 1-cm margin group. All patients in
this U.K. study had melanomas with a Breslow thickness
of greater than 2 mm, and 453 patients underwent a 1-cm
resection margin compared with 457 patients undergoing a 3-cm resection margin. Although there were more
melanoma-related deaths in the 1-cm resection margin
group, the number of nonmelanoma-related deaths increased
in the 3-cm group. Therefore, the OS did not significantly differ
between the two groups.22 At the same time, this study has
also been criticized for the fact that patients were not staged
by sentinel node procedure and, therefore, an undetected
misbalance in this important staging factor might explain
the imbalance in the survival outcomes.23 Thus, the authors
practice-changing conclusion that a lesser margin has a detrimental effect on a patients survival cannot be simply accepted yet.
Sentinel Node
Multiple institutional and multicenter studies have verified
the prognostic information collected from the sentinel node
status of patients, not only for intermediate thickness but
also for thick melanoma.24-30 There is ongoing debate on the
interpretation of the results with respect to the interim and
final data of the Multicenter Selective Lymphadenectomy
Trial-1 (MSLT-1).28,31-34 The authors of the MSLT-1 study
concluded that: Biopsy-based staging of intermediatethickness or thick primary melanomas provides important
prognostic information and identifies patients with nodal
metastases who may benefit from immediate complete
lymphadenectomy. Biopsy-based management prolongs
disease-free survival for all patients and prolongs distant
KEY POINTS
e506
Sentinel node staging provides optimal staging

information and is required for participation in adjuvant
therapy trials.
Surgery alone can be curative for primary melanomas,
locoregional disease, and in selected stage IV cases.
Adjuvant interferon has not demonstrated unequivocal
results but is the only approved adjuvant therapy in
many countries.
Adjuvant therapy with checkpoint inhibitors (antiCLTA4 and antiPD-1) and targeted therapies are ongoing,
and the results are pending.
AntiCTLA-4 has demonstrated a progression-free
survival benefit, but its effect on overall survival is not
clear yet.
disease-free survival and melanoma-specific survival for

patients with nodal metastases from intermediate-thickness
melanomas.28 This is formally a completely correct conclusion because there is a clear benefit of 12.3% (69.8%
[6 4.4%] for sentinel nodepositive versus 57.5% [6 5.4%]
for observation nodepositive; HR 0.56; 95% CI, 0.370.84;
p = .006) in melanoma-specific survival after 10 years.28
However, the study has been criticized for a number of
reasons. First, the benefit was only found in a post hoc
subgroup analysis, but no difference in survival was seen
between the primary randomized study groups (wide local
excision plus observation versus wide local excision plus
sentinel node biopsy), with a 10-year melanoma-specific
survival of 78.3% (6 2.0%) compared with 81.4% (6 1.5%),
respectively (p = .18).28,34 This is reflected in potential
overtreatment (minimal sentinel node tumor burden; prognostic false positivity) and under-treatment (worse outcome
of patients with false-negative disease compared with patients with sentinel nodepositive disease), which do not
influence the subgroup analysis but do influence the primary randomized study groups and primary endpoint of the
study.32-34
Therefore, the true therapeutic benefit of a sentinel node
procedure remains a matter of debate. Nonetheless, today
most surgeons, dermatologists, and other melanoma specialists worldwide consider sentinel node biopsy a standardof-care procedure for optimal staging and, therefore, part of
the workup of intermediate thickness and thick melanomas
($ pT1b = $ 1.00-mm Breslow or in case of , 1.00-mm
Breslow when the tumor is ulcerated or if there is one or
more dermal mitosis present), who do not yet have clinically apparent detectable metastases. With the potential
for more effective adjuvant therapy options as approved
therapies in the stage IV setting complete testing in the
stage III population, this need for optimal staging will only
increase.
Lymph Node Dissections

Lymph node dissections can be divided into three categories:
(1) elective (or prophylactic) lymph node dissection, (2) therapeutic lymph node dissection, and (3) completion lymph
node dissection.
Elective lymph node dissection is currently an abandoned
procedure. Four prospective randomized controlled trials
from the 1970s and 1990s have sought to find a survival
benefit for elective lymph node dissection but have
failed.35-38 The morbidity of elective lymph node dissection
was high, and, in the case of trunk or head and neck melanomas, it could be unclear which nodal basin should be
prophylactically removed. Subgroup analyses from the
Cascinelli et al and Balch et al studies did seem to indicate a
potential benefit for patients with intermediate thickness
melanoma, who did indeed have nodal involvement.37,38
This was the starting point for the MSLT-1 study.
Therapeutic lymph node dissection is performed for clinically apparent detectable nodal metastases at presentation
or during follow-up. If staging imaging examinations (CT,

PET-CT, and/or MRI) do not indicate any distant metastases,
surgeons worldwide will propose this to patients. Five-year
survival rates differ depending on the type of dissection (neck,
axilla, or groin), the number of lymph nodes involved, and the
presence of extracapsular extension. The AJCC database reports survival between 59% for stage IIIB and 40% for stage
IIIC.1 Some institutional databases report even worse outcome of 20%40%.39-45
One prospective randomized controlled trial, MSLT-2
(NCT00297895), is evaluating the therapeutic value of completion lymph node dissection after a positive sentinel node
biopsy, which compares with nodal observation by periodic
ultrasound.46,47 This study has completed accrual but needs
to complete follow-up before initial outcomes can be reported. Therefore, there is no consensus yet on this topic,
which is illustrated by several studies demonstrating a distribution of around 50% for completion lymph node dissection compared with 50% for nodal observation in the United
States/North America, Europe, and Australia.48,49
At the 2015 ASCO Annual Meeting, Leiter et al reported on
the results of the prospective randomized controlled trial
German DECOG group study, which also compared completion lymph node dissection with nodal observation. In a
total of 483 patients (242 of whom underwent completion
lymph node dissection versus 241 of whom underwent nodal
observation), with a median follow-up of 34 months, there
were no differences in recurrence-free survival, distant
metastasesfree survival, or melanoma-specific survival.50
However, this trial was criticized for the lack of mature
follow-up and was likely to be underpowered compared with
the MSLT-2 trial, which accrued nearly 2,000 patients. The
very recent results of the Sunbelt Melanoma Trial confirm
the DECOG results, by showing again that there was no
difference in overall survival after CLND. However, this was
not the primary focus of this study and potentially underpowered to look at that.51 Finally, there is a question if all
patients with tumor-bearing sentinel nodes might benefit
from completion lymph node dissection or if some patients
with minimal sentinel node tumor burden might not benefit
as their prognosis is comparable to patients with sentinel
nodenegative disease.52-54 This topic is being addressed by
the EORTC 1208 Minitub (NCT01942603) prospective
registry.
In-Transit/Distant Metastases
The latest AJCC staging system considers in-transit metastases as stage IIIC, and they have a poor 5-year survival rate
of 46% or 69% (with and without regional lymph node
involvement, respectively).1 Read et al reported rates of
40% or 59% (with and without regional lymph node involvement, respectively),55 and Pawlik et al reported a rate
of 54%.56 None of these studies reported the type of intransit metastases (cutaneous or subcutaneous) or number
of metastases (single or multiple), and they also did not
report on the type of treatment (e.g., surgery, amputation,
radiotherapy, isolated limb perfusion/infusion, systemic

therapy). However, regardless of this heterogeneity, most
surgeons and other melanoma-treating specialists will
concur that, in the absence of distant metastases on dissemination examination (CT, PET-CT, or MRI), a locoregional
treatment with curative intent is the preferred approach.
However, with the advent of more effective systemic
therapies, it may become reasonable to first treat such
patients with systemic treatment (particularly immunotherapy) and save surgical resection as a salvage procedure
for residual disease.
Surgical treatment of distant metastases is a rare phenomenon because usually the amount of lesions is too
many for complete resection of all with any potential clinical benefit. However, there are a few scenarios where one
might consider surgery for a patient with stage IV disease.
Obvious reasons are in the case of symptoms such as bowel
obstruction, bleeding, and pain for palliative reasons. But
there is potential for cure in a highly selected patient cohort.
Typical selection criteria are one, two, or three or more
lesions; M1a, M1b, or M1c disease; and the disease-free
interval before the development of these distant metastases. Howard et al demonstrated in these cases a 4-year
survival rate of 20.8% for patients undergoing surgery as part
of their treatment compared with 7.0% for patients who
were not eligible for surgery and received systemic therapy
only (prior to checkpoint inhibitors/BRAFi/MEKi).57 A selected subgroup of patients who only received surgery could
achieve even a 4-year survival of 45.7%.57 Similar rates have
been shown by Sosman et al (31% at 4 years).58 In general,
surgical resections with curative intent are reserved for this
highly selected (biased) patient cohort.
Future Projections
With the introduction of effective systemic therapy (e.g.,
immune checkpoint inhibitors, BRAFi/MEKi), the role of
surgery will be subject to change. Primarily, staging for
adjuvant therapy through standard of care sentinel-node
biopsy will no longer be a matter of debate. The question of
whether completion lymph node dissection is necessary
after a positive sentinel node will be a matter of debate at
first (upcoming 1020 years) but is likely to evolvesimilarly
to the situation in breast cancer, where it is reserved for
cases who did not respond to the systemic therapy or as a
salvage procedure after progression. At the same time,
surgery for patients with stage IV disease will also evolve.
Patients with solitary residual disease (near complete response to systemic therapy) or with a mixed response
(progression of solitary lesion, when all others show response) will be the target of surgical resection. Finally, locally
advanced (unresectable) disease will be the target of neoadjuvant induction treatment to allow complete resection.
Surgical Conclusions
Wide local excision is recommended to reduce local recurrence rates, but it does not influence survival. Worldwide,
e507
VAN AKKOOI ET AL
there is no consensus on whether performing a sentinel node

procedure and/or completion lymph node dissection has a
proven therapeutic benefit. Sentinel node staging and subsequent completion lymph node dissection in the case of
positive sentinel node biopsy provides optimal staging information and is required to allow patients to be included in
adjuvant therapy trials. Locoregional treatment is considered
with curative intent for in-transit metastases. Highly selected
patients with stage IV disease might benefit from surgical
resection. The role of surgery for melanoma will evolve because of the introduction of effective systemic therapy, with
an increasing need for optimal staging, increasing salvage
surgery after near complete response or in case of mixed
response, and neoadjuvant approaches after induction treatment of locally advanced disease.
ADJUVANT THERAPY FOR PATIENTS WITH

RESECTED HIGH-RISK AND STAGE IV MELANOMA
The past 5 years have seen dramatic improvements in the
outcomes for patients with metastatic melanoma, evidenced by an improvement in the expected median survival
of patients from 9 months to approximately 2530 months
and durable treatment-free responses from 10% to as high
as 50% in this short period of time.59,60 Linked to this, a
number of new immunotherapy and targeted therapy agents
have been licensed both as single agents and in combinations.
There is a high expectation that outcomes for patients will
continue to improve based on the evaluation of new agents
and the rapidly increasing understanding of the molecular
basis of resistance to both targeted therapy and immunotherapy and how this can be circumvented.
Advances in the field of adjuvant therapy of melanoma
have been significantly less dramatic than for metastatic
melanoma. Despite the approval of two new agents (ipilimumab and pegylated interferon [PEG-IFN]), the past
5 years have not seen any improvement in OS in any adjuvant
therapy study. However, it is too early to draw any conclusions about the use of targeted therapy or checkpoint
inhibitors as adjuvant therapy, as the studies are either still
recruiting or have not yet been fully reported. Until then, IFN
remains a treatment option in the United States and in many
European countries, although there is no consensus on the
exact indication group, dose, and/or duration of treatment.
The U.S. Food and Drug Administration (FDA) approved
ipilimumab in October 2015 based on the improvement in
relapse-free survival (RFS), but with absent OS data at this
moment, a definitive recommendation for its use cannot yet
be made.
We discuss the current role of adjuvant therapy in melanoma and highlight potential developments in the next
5 years.
Challenges in Evaluating Adjuvant Trials

A number of challenges arise when examining individual
studies and comparing different adjuvant therapy trials. The
AJCC staging system reflects the understanding of staging at
e508
the time of its publication; however, it is at the mercy of

rapidly advancing technology and the inconsistent application of this both across and within trials. Sentinel node
biopsy has been clearly established as a very strong prognostic factor for primary melanoma.1,28 However, the
emergence of molecular-based diagnostic techniques means
that much a smaller tumor burden can be detected in the
lymph node, and the clinical implications and optimum
treatment of microscopic nodal disease has not yet been
established.40 At a more basic level, not all older trials mandated sentinel node biopsy, so the prognosis for patients with
stage II disease varies greatly, making this the most heterogeneous stage group. The use of sentinel node biopsy and
more sensitive imaging for detection of distant metastatic
disease have resulted in stage migration with improvement in
the median survival for patients with stage II, III, and IV disease,
making it difficult to compare trials where different staging
techniques were performed.
Tumor ulceration is an independent prognostic factor for
survival across all stage groups. Although there is an accepted definition of tumor ulceration, there remains the
potential for differences in interpretation of the morphologic features defining tumor ulceration. Newer clinical trials
require central review of pathology specimens to minimize
uncertainty and standardize staging; this is particularly important for ulceration and sentinel node biopsy involvement.
However, older studies on which many management guidelines are based did not require this.
Identifying the appropriate clinical endpoint for an adjuvant therapy study in melanoma has become increasingly
difficult. Until recently, there was no concern that a potential
benefit in an adjuvant therapy study could be confounded by
subsequent therapy treatment in the advanced disease
setting; this is clearly not the case at present. There remains
major discussion about the validity of recurrence-free survival as a clinically useful endpoint in its own right.61 It is clear
that patients do value time without disease, even if this is
at the expense of treatment-related toxicity.62 Increasingly, there is a phenomenon of patients wishing to remain
disease-free as long as possible in an attempt to bridge to
emerging new and potentially more effective agents and/or
clinical trials for the treatment of metastatic disease. Other
questions that must be considered include the potential
toxicity of adjuvant therapy, the cost implications of this, and
whether these treatments are more effective and/or costeffective in the adjuvant or metastatic disease settings.
Although some of these challenges are easily addressed,
others will require clinical trials that include quality of life
and cost-effectiveness endpoints.
Adjuvant Interferon
The role of IFN as adjuvant treatment of melanoma remains
unclear, leading to a lack of consensus across Europe and the
United States. A number of randomized trials have investigated the use of adjuvant IFN-a for the treatment
of patients with high-risk melanoma following surgical
resection. The results of these trials have, however, been

inconsistent, with some suggesting a survival benefit for
IFN-a and others showing no difference. High-dose IFN-a
has been approved since 1996 based on the results of the
ECOG 1684 trial of 287 patients with stage IIB/III melanoma,
which showed a benefit for high-dose IFN-a on both RFS and
OS.63 Updated results with a median follow-up of 12.6 years
showed that the RFS benefit was maintained (HR 0.72;
p = .02), but the HR for OS had decreased from 0.67 to 0.82 and
was no longer significant (p = .18), though clearly the effect
of competing causes of death on survival may have contributed to this.64 The ECOG 1690 trial compared high- and
low-dose IFN-a with observation in 624 patients with stage
IIB/III disease and showed only a marginal benefit of RFS for
both regimens and no OS benefit for either treatment.65 As
this trial included patients with T4 primary tumors (stage II)
and did not require regional nodal staging, many patients
relapsed in regional nodes. Crossover of patients in the
observation arm to the then FDA-approved high-dose IFN
regimen after lymph node recurrence was cited as a possible
contributing cause for the lack of difference. Intergroup
E1694 compared high-dose IFN-a-2b with the GM2-KLH
vaccine in patients with resected stage IIB and III
melanomathe same population studied in ECOG 1684 and
1690.66 The trial was closed prematurely when interim
analysis disclosed a significantly worse RFS and OS for patients treated with the GM2-KLH vaccine compared with the
high-dose IFN-a (HR 1.47 and 1.52, respectively). Although
there was some initial concern that patients who were
administered the GM2-KLH vaccine fared worse than they
would have on observation, a subsequent trial comparing
the vaccine with observation in patients with stage II melanoma showed no significant difference in OS, perhaps
mitigating this concern.
Very recently, the results of the Sunbelt Melanoma Trial
were published. This trial examined high-dose adjuvant IFN
after sentinel nodepositive disease in one node versus
observation versus high-dose IFN in the case of more than
one positive lymph node. Also, patients with sentinel node
immunohistochemistrynegative but reverse transcriptionpolymerase chain reactionpositive disease were randomly
assigned into three groups: observation, complete lymph
node dissection, or complete lymph node dissection plus
TABLE 1. Outcome According to Dose of Interferon

Overall Risk (95% CI)
Dose
EFS
OS
High (1,196 patients)
0.83 (0.720.96)
0.93 (0.801.08)
PEG-IFN (1,256 patients)
0.83 (0.761.00)
0.96 (0.821.11)
Intermediate (2,243 patients)
0.84 (0.740.95)
0.91 (0.791.04)
Low (2,732 patients)
0.85 (0.770.94)
0.86 (0.770.96)
Very low (484 patients)
0.99 (0.801.23)
0.96 (0.761.21)
Overall (95% CI)
0.86 (0.810.91)
0.90 (0.850.97)
Abbreviations: EFS, event-free survival; OS, overall survival; PEG-IFN, pegylatedinterferon.
high-dose IFN. The concept of the study, when designed in

1997, was to find out if patients with minimal disease
(sentinel node negative, but RT-PCR positive) would be the
group with the most benefit. The trial did not demonstrate
any OS benefit.51 The subgroup of patients with ulceration
did show a benefit for disease-free survival but not for OS.
The amount of ulcerated patients in this study was minimal,
which might be the explanation for the lack of OS benefit in
this group.
In Europe, low-dose IFN-a was approved based on a
French trial of 499 patients with stage II disease, which
showed an RFS benefit (p = .035) and a trend toward an
improvement in OS (p = .059).67 More recently, in 2011, the
FDA approved PEG-IFN for patients with stage III melanoma
based on the EORTC 18991 trial of 1,256 patients.68 The trial
first reported in 2007 with a median follow-up of 3.8 years.
There was a significant improvement in RFS in favor of PEGIFN-a-2b (HR 0.82; p = .011), however, there was no significant effect on either distant metastasis-free survival or
OS. Subgroup analysis showed that the benefit was only seen
in patients with microscopic nodal disease, and there was
also an improvement in distant metastasis-free survival.
Updated results with a median follow-up of 7.6 years
showed the effect on RFS had deteriorated from an HR of
0.82 to 0.87.69 In addition, the greatest benefit was seen for
the patients with microscopic nodal disease (N1) who had
ulcerated primary tumors (RFS HR 0.72; OS HR 0.59), with no
benefit seen in nonulcerated tumors.
Other studies have looked at dose and duration of
treatment, and many did not use a no-treatment control
arm. Shortening the regimen to a month of high-dose therapy
administered either once or multiple times has not been
successful.70-73 Five meta-analyses have been reported.74-78
The two meta-analyses by Mocellin included studies without
an active treatment control arm.74,75 A recent meta-analysis
reported on 17 trials published between 1995 and 2011.76
Adjuvant IFN was associated with a significantly improved
disease-free survival (HR 0.83; 95% CI, 0.7887; p = .00001)
and OS (HR 0.91; 95% CI, 0.8597; p = .003). This translated
into an improvement in OS from 46.1% to 49.1% at 5 years and
from 37.1% to 39.9% at 10 years. There was no clear evidence
that a particular dose or duration of treatment or patient
subgroup benefitted differently. However, ulceration status
affected IFN benefit on event-free survival and OS. For
ulcerated melanoma, IFN increased event-free survival from
26.9% to 32.9% at 5 years and from 20.4% to 27.3% at 10 years
(Table 1).
The interaction between ulceration and response to IFN is
not universally accepted. No interaction has been seen in the
studies from ECOG (Kirkwood, personal communication) or
from the Nordic group.79 This is being examined prospectively in an EORTC trial 18081 (NCT01502696).
To look further at whether progression-free survival
predicts for OS with IFN treatment, Suciu and colleagues
examined 12 studies involving 6,708 patients for which individual patient data were available to produce a training
set. This was then validated on published data from 5,774
e509
VAN AKKOOI ET AL
patients from a further 13 studies, including studies with an

active treatment control arm. The results showed that both
individual and trial level correlations were close, suggesting
that for adjuvant IFN trials, there is a strong surrogacy of RFS
for OS.80
Non-IFNBased Adjuvant Therapy Approaches

Other approaches to adjuvant treatment have been pursued
over the past two decades with most showing mixed results. These have included biochemotherapy, granulocytemacrophage colony-stimulating factor with or without
peptide vaccinations, and bevacizumab. More encouraging
data are emerging with the checkpoint inhibitor ipilimumab,
and there is great optimism surrounding the potential value
of the PD-1 pathway inhibitors. These approaches are described below.
Biochemotherapy. Cisplatin and interleukin-2 (IL-2)based
biochemotherapy produced encouraging results in a series
of phase II trials in patients with stage IV melanoma. These
results prompted the Southwest Oncology Group (SWOG) to
perform a phase III cooperative group trial (S0008) in which
biochemotherapy was compared with high-dose IFN.81 The
trial randomly assigned 432 patients with high-risk melanoma to either three cycles of cisplatin, vinblastine,
dacarbazine, IL-2, and IFN-a given over a 9-week period or
to high-dose IFN-a for 1 year. At a median follow-up of
7.2 years, the biochemotherapy regimen significantly prolonged RFS (median 4.0 vs. 1.9 years, and 5-year RFS rate 48%
vs. 39%; HR 0.75; 95% CI, 0.580.97).81 However, there was no
significant difference in the OS (5-year rate 56% for both
treatment arms; HR 0.98; 95% CI, 0.741.31).81 The biochemotherapy regimen was substantially more toxic, with
grade 3 or 4 side effects (consisting primarily of hematologic
and gastrointestinal toxicity) observed in 76% of participants.
Neurologic, psychiatric, and hepatic toxicities were the most
frequent with high-dose IFN-a. Biochemotherapy toxicities
were limited to the 9-week treatment period, while IFN-a
toxicities were distributed across the year of treatment. Even
though the biochemotherapy regimen was the first to
produce a significant improvement in RFS compared with an
active control arm, the lack of OS benefit coupled with the
failure of this regimen to show an OS benefit relative to
chemotherapy alone in patients with stage IV disease has
limited its acceptance in the adjuvant setting.
Granulocyte-macrophage colony-stimulating factor.
Granulocyte-macrophage colony-stimulating factor (GM-CSF)
has been reported to prolong survival in patients with
resected stage IV disease in a single-arm trial,82 prompting its
evaluation in a randomized controlled trial (E4697).83 In this
multicenter trial, 815 patients with high-risk stage III or
completely resected stage IV melanoma were randomly
assigned to receive GM-CSF or placebo. Patients who were
HLA-A2positive were randomly assigned to peptide vaccination or placebo, in combination with GM-CSF or placebo;
e510
patients who were HLA-A2negative received only GM-CSF or

placebo. There was no benefit associated with the use of the
peptide vaccine.83 There was no significant difference in either OS (5-year survival rate 52% vs. 49%; HR 0.94; 95% CI,
0.771.15) or RFS (median RFS 11.4 vs. 8.8 months, and 5-year
RFS rate 31% vs. 27%; HR 0.88; 95% CI, 0.741.04; p = .131),
when GM-CSF was compared with placebo. Exploratory analyses showed a trend toward improved OS in patients with
resected visceral metastases who received GM-CSF, but this
observation required independent validation especially given
that this was not a protocol stratification factor.83 Given these
mixed results, GM-CSF is generally not recommended for use
in the adjuvant setting.
Bevacizumab. Bevacizumab was reported to have some
activity in patients with advanced melanoma, prompting its
study in the adjuvant setting. In a phase III multicenter trial
conducted in the United Kingdom, 1,343 patients with
resected stage IIB, IIC, or III disease were randomly assigned
to receive 1 year of bevacizumab treatment (7.5 mg/kg every
3 weeks) or to observation.84 At a median follow-up of
25 months, there was no significant difference in OS, which
was the primary endpoint of the trial (HR 0.97; 95% CI,
0.781.22; p = .76).84 However, there was a significant increase in the disease-free interval (1-year and 2-year
disease-free rates 77% vs. 70% and 59% vs. 57%, respectively; HR 0.83; 95% CI, 0.700.98), and an insignificant
trend toward improved distant metastasis-free survival (HR
0.88; 95% CI, 0.731.06; p = .18).84 Interpretation of the trial
and the potential role of bevacizumab will require further
follow-up to assess the 5-year OS rate. Until then, the use of
bevacizumab in the adjuvant setting for patients with advanced melanoma is not recommended.
Ipilimumab. Ipilimumab was approved for use in patients
with stage IV disease in 2011 based on its being able to
prolong median OS.2 The approved regimen in the stage IV
setting (3 mg/kg every 3 weeks for four doses) was also
shown to produce a durable effect on OS extending out to
10 years.85 The encouraging initial studies with ipilimumab
in the stage IV setting prompted its study in the adjuvant
setting. Two randomized phase III trials compared with
placebo (EORTC 18071) or high-dose IFN (ECOG 1609) have
completed accrual. In the EORTC 18071 trial, patients with
resected stage III melanoma were randomly assigned to
receive ipilimumab 10 mg/kg or placebo every 3 weeks for
four doses and then every 3 months for up to 3 years.86
Results showed improved median recurrence-free survival
of 26.1 months with ipilimumab compared with 17.1 months
with placebo (HR 0.75; p = .0013).86 In subgroup analyses,
patients with microscopic lymph node disease or ulcerated
primary lesions demonstrated the most benefit. Importantly, there was a high rate of grade 3-5 immune-related
adverse events in patients receiving ipilimumab (43%
compared with 2% with placebo).86 These included five
treatment-related deaths, despite treatment with immunomodulatory therapy. Although these data are provocative
and have led to the recent FDA approval of ipilimumab for

patients with resected stage III melanoma, it is as yet unclear
whether this reduction in recurrence rate with ipilimumab
represents an improvement over adjuvant IFN therapy and
whether this benefit will translate into an improvement in
OS. Furthermore, the approved adjuvant dose of 10 mg/kg is
over three times higher than the approved dose in metastatic melanoma, with the potential for higher toxicity. These
questions are being addressed by the U.S. Intergroup E1609
study, which randomly assigned patients with resected stage
III-IV melanoma to ipilimumab 10 mg/kg or 3 mg/kg or highdose IFN.87 The study completed accrual of more than 1,500
patients in the summer of 2014 and is pending analysis for
coprimary endpoints of RFS and OS. Long-term survival data
from both of these adjuvant studies will ultimately be necessary to determine the true effectiveness of adjuvant
ipilimumab therapy relative to either high-dose IFN or
observation.
MAGE-A3 vaccine. The DERMA (ADjuvant ImmunothERapy
with MAGE3 in MelanomA) was a double-blind, randomized,
placebo-controlled phase III study of recombinant MAGE-A3
with AS15 antigen-specific cancer immunotherapeutic (ASCI)
in patients with MAGE-A3positive resected stage IIIB/C
disease. MAGE-A3 is expressed on approximately 60% of
melanoma specimens. The study was based on an EORTC
phase II study of patients with stage IV M1A disease that
identified a superior survival benefit for patients receiving
the MAGE-A3 vaccine treated with the AS15 rather than
the ASO2B adjuvant (HR 0.55; 95% CI, 0.281.06).88
Furthermore, a genetic predictor identified a group of patients receiving MAGE-A3 with AS15 ASCI with a better OS
(HR 0.27; 95% CI, 0.080.89).88 The study screened 3,914
patients and randomly assigned 1,344 patients 2:1 to 13
intramuscular injections of vaccine or placebo. The study
failed to meet its coprimary endpoint of disease-free survival
in either the overall population of patients studied or in
those with the potential predictive gene signature.88
Ongoing Adjuvant Studies

PD-1 pathway blockers. In the meantime, the field has
moved on to examine potentially more exciting checkpoint
inhibitors: those blocking the PD-1 pathway. PD-1 inhibitors
have proven to be less toxic and more active than ipilimumab in patients with established, unresectable metastatic melanoma.9,89 Given their favorable therapeutic index
and ability to produce durable off-treatment benefit in
patients with advanced melanoma, there is much interest in
developing this class of therapies as adjuvant treatment for
patients with high-risk resected melanoma. Results from a
phase I trial of nivolumab plus a multipeptide vaccine in
33 patients with resected stage IIIC or IV melanoma showed a
relatively low relapse rate (30%) during a median follow-up
period of 32.1 months from trial enrolment. Median RFS
was estimated to be 47.1 months.90 Phase III trials with
nivolumab and pembrolizumab in patients with resected
stage III and IV melanoma are currently underway. These

include Checkmate-238 (NCT02388906) comparing ipilimumab 10 mg/kg (according to the EORTC 18071 schedule) to
nivolumab 3 mg/kg administered intravenously every
2 weeks; EORTC 1325 (KEYNOTE-054; NCT02362594) protocol
comparing pembrolizumab (200-mg flat dose) with placebo;
and the SWOG S1404 protocol comparing pembrolizumab
(200-mg flat dose) with high-dose IFN (NCT02506153).
Checkmate-238 completed accrual in September 2015.
EORTC 1325 and SWOG S1404 are actively accruing patients.
Molecularly targeted therapies. The selective inhibitors of
the mutant BRAF kinase (vemurafenib and dabrafenib) have
been shown to be superior to dacarbazine in patients with
stage IV BRAFV600-mutant melanoma, leading to their FDA
approval.3,8 More recently, the combination of dabrafenib
and the MEK inhibitor trametinib have been proven to
produce superior response rate, median progression-free
survival, and median OS compared with single-agent dabrafenib or vemurafenib in this patient population.7,8 Based
on these encouraging results, clinical trials have begun testing
these agents in patients with high-risk resected BRAF-mutant
melanoma. In the COMBI-AD trial, approximately 850 patients
are to be randomly assigned to receive either dabrafenib
with trametinib or placebo and treated for up to 1 year
(NCT01682083). This trial completed accrual in December
2014 and is pending analysis. In the BRIM 8 study, up to
775 patients with stage IIC to stage IIIC resected melanoma
will be randomly assigned to receive vemurafenib or placebo
for up to 1 year. The study is currently recruiting at 187 sites in
29 countries (NCT01667419).
The results of these trials are eagerly awaited, especially
the assessment of whether these active agents can eliminate
the last tumor cell in the low-volume adjuvant setting and
thereby actually prevent, not just delay, disease recurrence.
Adjuvant Therapy Conclusions

The role of adjuvant therapy in patients with resected highrisk melanoma remains unsettled. Although several approaches have shown improvements in RFS, only the high-dose
IFN regimen has shown an improvement in OS. The extent of
this survival benefit remains controversial. Although multiple
meta-analyses have suggested that the benefit of IFN on OS is
small (under 10%), these usually (1) include studies using
different dose levels of IFN rather than simply the high-dose
regimen, (2) include the E1690 study data despite its survival
endpoint being potentially confounded by crossover to then
FDA-approved IFN use, and (3) exclude the overwhelmingly
positive E1694 trial because it involved a vaccine control arm.
The unresolved question is: do you base a decision on the
results of the individual trials or the meta-analyses? The choice
for some is further complicated by the toxicity of the highdose IFN regimen, which is clearly more than desirable for the
adjuvant setting.
The benefit of ipilimumab on RFS appears to be equivalent
at least to that of high-dose IFN and was sufficiently robust as
e511
VAN AKKOOI ET AL
to garner FDA approval. Given that the major effect of

ipilimumab in patients with stage IV disease is on OS, it is
likely that the RFS benefit of ipilimumab will ultimately
translate into a survival benefit. Nonetheless, many will be
unwilling to prescribe ipilimumab given the high incidence of
high-grade toxicity, including five deathsan outcome that
is difficult to swallow in a population in whom as much
as half might remain disease-free in the absence of any
treatment.
Absent an unambiguously positive approved agent for
adjuvant therapy in melanoma, clinical trial participation
remains the priority. The most promising approaches currently under study include BRAF inhibitors and PD-1 pathway
blockers. The BRAF inhibitors have powerful antitumor effects in patients with BRAF-mutant melanoma. Thus, they
are highly likely to prolong RFS compared with placebo in the

ongoing clinical trials. However, in contrast to immunotherapy, they typically do not produce durable off-treatment
responses in the metastatic setting and, therefore, may only
delay, rather than prevent, relapse in the adjuvant setting.
The current trials will eventually answer these questions,
but, until then, the use of these agents should be reserved
for patients with measurable disease where the effect of
tumor volume and its duration can be monitored. Although
questions remain about the ability of PD-1 blockers to effectively activate T cells in the absence of measurable
metastatic disease, it is hoped that, as in the stage IV setting,
these agents will show more improvement over existing
treatments and finally yield universally accepted adjuvant
treatment approaches for patients with high-risk melanoma.
References
1. Balch CM, Gershenwald JE, Soong SJ, et al. Final version of 2009 AJCC
melanoma staging and classification. J Clin Oncol. 2009;27:6199-6206.
711-723.
3. Chapman PB, Hauschild A, Robert C, et al. Improved survival with
vemurafenib in melanoma with BRAF V600E mutation. N Engl J Med.
2011;364:2507-2516.
4. Flaherty KT, Infante JR, Daud A, et al. Combined BRAF and MEK inhibition in melanoma with BRAF V600 mutations. N Engl J Med. 2012;
367:1694-1703.
5. Flaherty KT, Robert C, Hersey P, et al. Improved survival with MEK
inhibition in BRAF-mutated melanoma. N Engl J Med. 2012;367:
107-114.
6. Hauschild A, Grob JJ, Demidov LV, et al. Dabrafenib in BRAF-mutated
metastatic melanoma: a multicentre, open-label, phase 3 randomised
7. Long GV, Stroyakovskiy D, Gogas H, et al. Combined BRAF and MEK
inhibition versus BRAF inhibition alone in melanoma. N Engl J Med.
2014;371:1877-1888.
8. Robert C, Karaszewska B, Schachter J, et al. Improved overall survival in
melanoma with combined dabrafenib and trametinib. N Engl J Med.
2015;372:30-39.
9. Larkin J, Chiarion-Sileni V, Gonzalez R, et al. Combined nivolumab and
ipilimumab or monotherapy in untreated melanoma. N Engl J Med.
2015;373:23-34.
2006-2017.
11. Robert C, Long GV, Brady B, et al. Nivolumab in previously untreated
12. Wolchok JD, Kluger H, Callahan MK, et al. Nivolumab plus ipilimumab in
13. Bono A, Bartoli C, Clemente C, et al. Ambulatory narrow excision for thin
melanoma (, or = 2 mm): results of a prospective study. Eur J Cancer.
1997;33:1330-1332.
14. Cascinelli N. Margin of resection in the management of primary melanoma. Semin Surg Oncol. 1998;14:272-275.
15. Cohn-Cedermark G, Rutqvist LE, Andersson R, et al. Long term results
of a randomized study by the Swedish Melanoma Study Group on 2-cm
e512
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
versus 5-cm resection margins for patients with cutaneous melanoma

with a tumor thickness of 0.8-2.0 mm. Cancer. 2000;89:1495-1501.
Gillgren P, Drzewiecki KT, Niin M, et al. 2-cm versus 4-cm surgical
excision margins for primary cutaneous melanoma thicker than 2 mm:
a randomised, multicentre trial. Lancet. 2011;378:1635-1642.
Karakousis CP, Balch CM, Urist MM, et al. Local recurrence in malignant
melanoma: long-term results of the multiinstitutional randomized
surgical trial. Ann Surg Oncol. 1996;3:446-452.
Khayat D, Rixe O, Martin G, et al. Surgical margins in cutaneous melanoma (2 cm versus 5 cm for lesions measuring less than 2.1-mm thick).
Cancer. 2003;97:1941-1946.
Thomas JM, Newton-Bishop J, AHern R, et al. Excision margins in highrisk malignant melanoma. N Engl J Med. 2004;350:757-766.
Haydu LE, Stollman JT, Scolyer RA, et al. Minimum safe pathologic
excision margins for primary cutaneous melanomas (1-2 mm in
thickness): analysis of 2131 patients treated at a single center. Ann Surg
Oncol. Epub 2015 May 9.
Hayes AJ, Maynard L, AHern R, et al. Long term follow up of survival in a
randomised trial of wide or narrow excision margins in high risk primary
melanoma. J Clin Oncol. 2015;33 (suppl; abstr 9001).
Hayes AJ, Maynard L, Coombes G, et al. Wide versus narrow excision
margins for high-risk, primary cutaneous melanomas: long-term
follow-up of survival in a randomised trial. Lancet Oncol. 2016;17:
184-192.
Moncrieff M. Excision margins for melanomas: how wide is enough?
Lancet Oncol. 2016;17:127-128.
Cascinelli N, Bombardieri E, Bufalino R, et al. Sentinel and nonsentinel
node status in stage IB and II melanoma patients: two-step prognostic
indicators of survival. J Clin Oncol. 2006;24:4464-4471.
Doubrovsky A, De Wilt JH, Scolyer RA, et al. Sentinel node biopsy
provides more accurate staging than elective lymph node dissection in
patients with cutaneous melanoma. Ann Surg Oncol. 2004;11:
829-836.
Estourgie SH, Nieweg OE, Valdes Olmos RA, et al. Review and evaluation
of sentinel node procedures in 250 melanoma patients with a median
follow-up of 6 years. Ann Surg Oncol. 2003;10:681-688.
Gershenwald JE, Thompson W, Mansfield PF, et al. Multi-institutional
melanoma lymphatic mapping experience: the prognostic value of
sentinel lymph node status in 612 stage I or II melanoma patients. J Clin
Oncol. 1999;17:976-983.
28. Morton DL, Thompson JF, Cochran AJ, et al. Final trial report of sentinelnode biopsy versus nodal observation in melanoma. N Engl J Med. 2014;
370:599-609.
29. Roulin D, Matter M, Bady P, et al. Prognostic value of sentinel node
biopsy in 327 prospective melanoma patients from a single institution.
Eur J Surg Oncol. 2008;34:673-679.
30. van der Ploeg AP, van Akkooi AC, Schmitz PI, et al. EORTC Melanoma
Group sentinel node protocol identifies high rate of submicrometastases according to Rotterdam Criteria. Eur J Cancer. 2010;
46:2414-2421.
31. Morton DL, Thompson JF, Cochran AJ, et al. Sentinel-node biopsy or
nodal observation in melanoma. N Engl J Med. 2006;355:1307-1317.
32. Thomas JM. Prognostic false-positivity of the sentinel node in melanoma. Nat Clin Pract Oncol. 2008;5:18-23.
33. Torjesen I. Sentinel node biopsy for melanoma: unnecessary treatment? BMJ. 2013;346:e8645.
34. van Akkooi AC. Sentinel node followed by completion lymph node
dissection versus nodal observation: staging or therapeutic? Controversy continues despite final results of MSLT-1. Melanoma Res. 2014;
24:291-294.
35. Veronesi U, Adamus J, Bandiera DC, et al. Inefficacy of immediate node
dissection in stage 1 melanoma of the limbs. N Engl J Med. 1977;297:
627-630.
36. Sim FH, Taylor WF, Ivins JC, et al. A prospective randomized study of the
efficacy of routine elective lymphadenectomy in management of
malignant melanoma. Preliminary results. Cancer. 1978;41:948-956.
37. Cascinelli N, Morabito A, Santinami M, et al. Immediate or delayed
dissection of regional nodes in patients with melanoma of the trunk:
a randomised trial. WHO Melanoma Programme. Lancet. 1998;351:
793-796.
38. Balch CM, Soong S, Ross MI, et al. Long-term results of a multiinstitutional randomized trial comparing prognostic factors and surgical results for intermediate thickness melanomas (1.0 to 4.0 mm).
Intergroup Melanoma Surgical Trial. Ann Surg Oncol. 2000;7:87-97.
39. Badgwell B, Xing Y, Gershenwald JE, et al. Pelvic lymph node dissection is
beneficial in subsets of patients with node-positive melanoma. Ann
Surg Oncol. 2007;14:2867-2875.
40. Balch CM, Gershenwald JE, Soong SJ, et al. Multivariate analysis of
prognostic factors among 2,313 patients with stage III melanoma:
comparison of nodal micrometastases versus macrometastases. J Clin
Oncol. 2010;28:2452-2459.
41. Jonk A, Kroon BB, Rumke P, et al. Results of radical dissection of the
groin in patients with stage II melanoma and histologically proved
metastases of the iliac or obturator lymph nodes, or both. Surg Gynecol
Obstet. 1988;167:28-32.
42. Karakousis CP, Driscoll DL. Positive deep nodes in the groin and survival
in malignant melanoma. Am J Surg. 1996;171:421-422.
43. Karakousis CP, Driscoll DL, Rose B, et al. Groin dissection in malignant
melanoma. Ann Surg Oncol. 1994;1:271-277.
44. Spillane AJ, Pasquali S, Haydu LE, et al. Patterns of recurrence and
survival after lymphadenectomy in melanoma patients: clarifying the
effects of timing of surgery and lymph node tumor burden. Ann Surg
Oncol. 2014;21:292-299.
45. van der Ploeg AP, van Akkooi AC, Schmitz PI, et al. Therapeutic surgical
management of palpable melanoma groin metastases: superficial or
combined superficial and deep groin lymph node dissection. Ann Surg
Oncol. 2011;18:3300-3308.
46. Morton DL. Overview and update of the phase III Multicenter Selective
Lymphadenectomy Trials (MSLT-I and MSLT-II) in melanoma. Clin Exp
Metastasis. 2012;29:699-706.
47. van Akkooi AC. Comments and controversies: sentinel node and lymphadenectomy in melanoma: staging or therapeutical?: what to expect
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
from the multicenter selective lymphadenectomy trial-1 results in

melanoma? Melanoma Res. 2013;23:181-184.
Bilimoria KY, Balch CM, Bentrem DJ, et al. Complete lymph node dissection for sentinel node-positive melanoma: assessment of practice
patterns in the United States. Ann Surg Oncol. 2008;15:1566-1576.
Pasquali S, Spillane AJ, de Wilt JH, et al. Surgeons opinions on lymphadenectomy in melanoma patients with positive sentinel nodes:
a worldwide web-based survey. Ann Surg Oncol. 2012;19:4322-4329.
Leiter U, Stadler R, Mauch C, et al. Survival of SLNB-positive melanoma
patients with and without complete lymph node dissection: a multicenter, randomized DECOG trial. J Clin Oncol. 2015;33 (suppl; abstr
LBA9002).
McMasters KM, Egger ME, Edwards MJ, et al. Fianl results of the Sunbelt
Melanoma Trial: a multi-institutional prospective randomized phase III
study evaluating the role of adjuvant high-dose interferon alfa-2b and
completion lymph node dissection for patients staged by sentinel lymph
node biopsy. J Clin Oncol. Epub 2016 Feb 8.
van Akkooi AC, de Wilt JH, Verhoef C, et al. Clinical relevance of
melanoma micrometastases (,0.1 mm) in sentinel nodes: are these
nodes to be considered negative? Ann Oncol. 2006;17:1578-1585.
van der Ploeg AP, van Akkooi AC, Haydu LE, et al. The prognostic
significance of sentinel node tumour burden in melanoma patients: an
international, multicenter study of 1539 sentinel node-positive melanoma patients. Eur J Cancer. 2014;50:111-120.
van der Ploeg AP, van Akkooi AC, Rutkowski P, et al. Prognosis in patients with sentinel node-positive melanoma is accurately defined by
the combined Rotterdam tumor load and Dewar topography criteria.
J Clin Oncol. 2011;29:2206-2214.
Read RL, Haydu L, Saw RP, et al. In-transit melanoma metastases: incidence, prognosis, and the role of lymphadenectomy. Ann Surg Oncol.
2015;22:475-481.
Pawlik TM, Ross MI, Johnson MM, et al. Predictors and natural history of
in-transit melanoma after sentinel lymphadenectomy. Ann Surg Oncol.
2005;12:587-596.
Howard JH, Thompson JF, Mozzillo N, et al. Metastasectomy for distant
metastatic melanoma: analysis of data from the first Multicenter Selective Lymphadenectomy Trial (MSLT-I). Ann Surg Oncol. 2012;19:
2547-2555.
Sosman JA, Moon J, Tuthill RJ, et al. A phase 2 trial of complete resection
for stage IV melanoma: results of Southwest Oncology Group Clinical
Trial S9430. Cancer. 2011;117:4740-06.
Daud A, Ribas A, Robert C, et al. Long-term efficacy of pembrolizumab
(pembro; MK-3475) in a pooled analysis of 655 patients (pts) with
advanced melanoma (MEL) enrolled in KEYNOTE-001. J Clin Oncol. 2015;
Grob JJ, Amonkar MM, Karaszewska B, et al. Comparison of dabrafenib
and trametinib combination therapy with vemurafenib monotherapy
on health-related quality of life in patients with unresectable or
metastatic cutaneous BRAF Val600-mutation-positive melanoma
(COMBI-v): results of a phase 3, open-label, randomised trial. Lancet
Oncol. 2015;16:1389-1398.
Flaherty KT, Hennig M, Lee SJ, et al. Surrogate endpoints for overall
survival in metastatic melanoma: a meta-analysis of randomised
controlled trials. Lancet Oncol. 2014;15:297-304.
Kilbridge KL, Cole BF, Kirkwood JM, et al. Quality-of-life-adjusted survival analysis of high-dose adjuvant interferon alpha-2b for high-risk
melanoma patients using intergroup clinical trial data. J Clin Oncol.
2002;20:1311-1318.
Kirkwood JM, Strawderman MH, Ernstoff MS, et al. Interferon alfa-2b
adjuvant therapy of high-risk resected cutaneous melanoma: the
Eastern Cooperative Oncology Group Trial EST 1684. J Clin Oncol. 1996;
14:7-17.
e513
VAN AKKOOI ET AL
64. Kirkwood JM, Manola J, Ibrahim J, et al. A pooled analysis of eastern

cooperative oncology group and intergroup trials of adjuvant high-dose
interferon for melanoma. Clin Cancer Res. 2004;10:1670-1677.
65. Kirkwood JM, Ibrahim JG, Sondak VK, et al. High- and low-dose interferon alfa-2b in high-risk melanoma: first analysis of intergroup trial
E1690/S9111/C9190. J Clin Oncol. 2000;18:2444-2458.
66. Kirkwood JM, Ibrahim JG, Sosman JA, et al. High-dose interferon alfa-2b
significantly prolongs relapse-free and overall survival compared with
the GM2-KLH/QS-21 vaccine in patients with resected stage IIB-III
melanoma: results of intergroup trial E1694/S9512/C509801. J Clin
Oncol. 2001;19:2370-2380.
67. Grob JJ, Dreno B, de la Salmonière P, et al. Randomised trial of interferon alpha-2a as adjuvant therapy in resected primary melanoma
thicker than 1.5 mm without clinically detectable node metastases.
Lancet. 1998;351:1905-1910.
68. Eggermont AM, Suciu S, Santinami M, et al. Adjuvant therapy with
pegylated interferon alfa-2b versus observation alone in resected stage
III melanoma: final results of EORTC 18991, a randomised phase III trial.
Lancet. 2008;372:117-126.
69. Eggermont AM, Suciu S, Testori A, et al. Long-term results of the
randomized phase III trial EORTC 18991 of adjuvant therapy with
pegylated interferon alfa-2b versus observation in resected stage III
melanoma. J Clin Oncol. 2012;30:3810-3818.
70. Agarwala SS, Lee SJ, Flaherty K, et al. Randomized phase III trial of highdose interferon alfa-2b (HDI) for 4 weeks induction only in patients with
intermediate- and high-risk melanoma (Intergroup trial E 1697). J Clin
71. Payne MJ, Argyropoulou K, Lorigan P, et al. Phase II pilot study of intravenous high-dose interferon with or without maintenance treatment
in melanoma at high risk of recurrence. J Clin Oncol. 2014;32:185-190.
72. Pectasides D, Dafni U, Bafaloukos D, et al. Randomized phase III study of
1 month versus 1 year of adjuvant high-dose interferon alfa-2b in
patients with resected high-risk melanoma. J Clin Oncol. 2009;27:
939-944.
73. Hauschild A, Weichenthal M, Rass K, et al. Efficacy of low-dose interferon alpha2a 18 versus 60 months of treatment in patients with
primary melanoma of .= 1.5 mm tumor thickness: results of a randomized phase III DeCOG trial. J Clin Oncol. 2010;28:841-846.
74. Mocellin S, Lens MB, Pasquali S, et al. Interferon alpha for the adjuvant
treatment of cutaneous melanoma. Cochrane Database Syst Rev. 2013;
6:CD008955.
75. Mocellin S, Pasquali S, Rossi CR, et al. Interferon alpha adjuvant therapy
in patients with high-risk melanoma: a systematic review and metaanalysis. J Natl Cancer Inst. 2010;102:493-501.
76. Suciu S, Ives N, Eggermont A, et al. Predictive importance of ulceration
on the efficacy of adjuvant interferon-a (IFN): An individual patient data
(IPD) meta-analysis of 15 randomized trials in more than 7,500 melanoma patients (pts). J Clin Oncol. 2014;32:5s (suppl; abstr 9067).
77. Wheatley K, Ives N, Eggermont A, et al. Interferon-alpha as adjuvant
therapy for melanoma: an individual patient data meta-analysis of
randomised trials. J Clin Oncol. 2007;25 (suppl; abstr 8526).
78. Wheatley K, Ives N, Hancock B, et al. Does adjuvant interferon-alpha for
high-risk melanoma provide a worthwhile benefit? A meta-analysis of
the randomised trials. Cancer Treat Rev. 2003;29:241-252.
e514
79. Hansson J, Aamdal S, Bastholt L, et al. Two different durations of adjuvant therapy with intermediate-dose interferon alfa-2b in patients
with high-risk melanoma (Nordic IFN trial): a randomised phase 3 trial.
Lancet Oncol. 2011;12:144-152.
80. Suciu S, Eggermont A, Lorigan P, et al. Relapse-free survival (RFS) as a
surrogate endpoint for overall survival in adjuvant Interferon trials in
patients with resectable cutaneous melanoma: an individual patient
data meta-analysis. In: ESMO 2014 Congress. Madrid, Spain; 2014.
Abstract 1089PD.
81. Flaherty LE, Othus M, Atkins MB, et al. Southwest Oncology Group
S0008: a phase III trial of high-dose interferon Alfa-2b versus cisplatin,
vinblastine, and dacarbazine, plus interleukin-2 and interferon in patients with high-risk melanomaan intergroup study of cancer and
leukemia Group B, Childrens Oncology Group, Eastern Cooperative
Oncology Group, and Southwest Oncology Group. J Clin Oncol. 2014;32:
3771-3778.
82. Spitler LE, Weber RW, Allen RE, et al. Recombinant human granulocytemacrophage colony-stimulating factor (GM-CSF, sargramostim) administered for 3 years as adjuvant therapy of stages II(T4), III, and IV
melanoma. J Immunother. 2009;32:632-637.
83. Lawson DH, Lee S, Zhao F, et al. Randomized, placebo-controlled, phase
III trial of yeast-derived granulocyte-macrophage colony-stimulating
factor (GM-CSF) versus peptide vaccination versus GM-CSF plus peptide
vaccination versus placebo in patients with no evidence of disease after
complete surgical resection of locally advanced and/or stage IV melanoma: a trial of the Eastern Cooperative Oncology Group-American
College of Radiology Imaging Network Cancer Research Group (E4697).
J Clin Oncol. 2015;33:4066-4076.
84. Corrie PG, Marshall A, Dunn JA, et al. Adjuvant bevacizumab in patients
with melanoma at high risk of recurrence (AVAST-M): preplanned interim results from a multicentre, open-label, randomised controlled
phase 3 study. Lancet Oncol. 2014;15:620-630.
85. Schadendorf D, Hodi FS, Robert C, et al. Pooled analysis of long-term
survival data from phase II and phase III trials of ipilimumab in unresectable or metastatic melanoma. J Clin Oncol. 2015;33:1889-1894.
86. Eggermont AM, Chiarion-Sileni V, Grob JJ, et al. Adjuvant ipilimumab
versus placebo after complete resection of high-risk stage III melanoma
(EORTC 18071): a randomised, double-blind, phase 3 trial. Lancet Oncol.
2015;16:522-530.
87. Tarhini AA. Melanoma and Skin Cancer: Adjuvant Therapy for High-Risk
Melanoma. Physicians Education Resource, LLC. http://www.gotoper.
com/publications/ajho/2014/2014nov/adjuvant-therapy-for-high-riskmelanoma. Accessed March 16, 2016.
88. Dreno B, Thompson JA, Smithers BM. DERMA, a double-blind, placebo
controlled Phase III study to assess the efficacy of MAGE-A3 cancer
immunotherapeutic adjuvant therapy in patients with resected MAGEA3 positive Stage III melanoma. Proc SMR. 2015.
89. Robert C, Schachter J, Long GV, et al; KEYNOTE-006 investigators.
Pembrolizumab versus ipilimumab in advanced melanoma. N Engl J
Med. 2015;372:2521-2532.
90. Gibney GT, Kudchadkar RR, DeConti RC, et al. Safety, correlative
markers, and clinical results of adjuvant nivolumab in combination with
vaccine in resected high-risk metastatic melanoma. Clin Cancer Res.
2015;21:712-720.
Cancer Survivorship: Is Every

Patients Journey the Same?
CHAIR
Michael T. Halpern, MD, PhD
University of Arizona College of Public Health
Tucson, AZ
SPEAKERS
Mary S. McCabe, RN, MA
New York, NY
Mary Ann Burg, PhD, MSW, LCSW
University of Florida
Gainesville, FL
THE CANCER SURVIVORSHIP JOURNEY
The Cancer Survivorship Journey: Models of Care, Disparities,

Barriers, and Future Directions
Michael T. Halpern, MD, PhD, Mary S. McCabe, RN, MA, and Mary Ann Burg, PhD, MSW, LCSW
OVERVIEW
Although the number of long-term cancer survivors has increased substantially over past years, the journey of survivorship
does not always include high-quality, patient-centered care. A variety of survivorship care models have evolved based on
who provides this care, the survivor population, the site of care, and/or the capacity for delivering specific services. Other
areas of survivorship care being explored include how long follow-up care is needed, application of a risk-based approach to
survivorship care, and the role of the survivor in his or her own recovery. However, there is little evidence indicating whether
any models improve clinical or patient-reported outcomes. A newer focus in survivorship care has included assessment of
potential disparities; the sociodemographic characteristics of population subgroups associated with barriers to receiving
high-quality cancer treatment may also affect the survivorship period. Developing policies and programs to address disparities in survivorship care is not simple, and examining how financial hardship affects cancer outcomes, reducing economic
barriers to care, and increasing incorporation of patient-centered strategies may be important components. Here too, there
is little evidence regarding the best strategies to address these disparities. Barriers to providing high-quality, patientcentered survivorship care include lack of evidence, lack of a trained survivorship workforce, lack of reimbursement
structures/insurance coverage, and lack of a health care system that reduces fragmented care. Future research needs to
focus on developing a survivorship care evidence base, exploring strategies to facilitate provision of survivorship care, and
disseminating best survivorship care practices to diverse and international audiences.
s of January 2014, there were approximately 14.5 million

cancer survivors in the United States; this number is
expected to grow to 19 million by 2024.1 Substantial increases in the number of cancer survivors have also been
reported in other countries2-4; for example Maddams et al
projected that the number of cancer survivors in the United
Kingdom will increase by approximately one million per
decade.5 The rapidly increasing numbers of cancer survivors
is the intended consequence of improvements in early
cancer diagnosis and cancer treatment. As these two aspects
of cancer care will likely continue to improve, the number of
cancer survivors will correspondingly continue to increase.
The title of the corresponding American Society of Clinical
Oncology (ASCO) 2016 Annual Meeting Education Session,
Cancer Survivorship: Is Every Patients Journey the Same?
is largely rhetorical. Clearly, the journey for every cancer
survivor is not the same. However, this raises the question:
why doesnt every cancer survivor receive high-quality and
comprehensive care tailored to his or her needs, situation,
and risk factors? The goal of this article is to explore some of
the factors that lead to differences in the survivorship journey,
to identify challenges or barriers to providing high-quality

survivorship care, and to suggest potential directions for
future research to enhance the survivorship experience.
DEFINITION OF SURVIVORSHIP
In exploring causes of variation in the survivorship journey, a
key issue is how the term cancer survivor is defined. As
with other aspects of survivorship and survivorship care,
there is not consensus on this definition. Some organizations, such as the National Coalition for Cancer Survivorship
and National Cancer Institutes Office of Cancer Survivorship, define survivors as individuals with cancer from the
time of diagnosis and for the balance of life, and include in
their definition family, friends, and caregivers of individuals
with cancer. In contrast, others define survivors as individuals who are diagnosed with cancer and have completed potentially curative treatment, until either death or
recurrence. The Institute of Medicines (IOMs) 2005 report
From Cancer Care to Cancer Survivor: Lost in Transition
acknowledges the definition of survivors used by the National
From the Mel and Enid Zuckerman College of Public Health, University of Arizona, Tucson, AZ; Memorial Sloan Kettering Cancer Center, New York, NY; University of Central Florida School
of Social Work, Orlando, FL.
Corresponding author: Michael T. Halpern, MD, PhD, University of Arizona, Mel and Enid Zuckerman College of Public Health, 1295 North Martin Ave., Tucson, AZ 85724; email:
mthalpern@email.arizona.edu.
231
HALPERN, MCCABE, AND BURG
Cancer Institutes Office of Cancer Survivorship but then

focuses on the period following first diagnosis and treatment
and prior to the development of a recurrence of the initial
cancer or death.6 This report highlights the potential
distinction between a survivor and the period of care referred to as survivorship. Jefford et al reports that the
definition of survivorship used in the United Kingdom is the
period after completion of initial treatment, regardless of
whether the person is free from cancer at that time.7
More broadly, there is disagreement regarding whether
the term cancer survivor should be used at all. As discussed
by Bell and Ristovski-Slijepcevic, some disagree with the
term survivor being used for individuals who have experienced cancer.8 This term is not used in other medical
conditions; people who have had heart attacks are not called
heart attack survivors.8 As an alternative, the U.K. National
Health Services has used the term living beyond cancer in
some of their materials.
MODELS OF SURVIVORSHIP CARE

As survivorship takes its place as a formal part of the cancer
care continuum, there are increasing international efforts directed toward the development and, most recently, the
evaluation of models of care delivery. This activity has resulted
in a growing body of literature describing and comparing
various configurations of health care services and models of
KEY POINTS
232
There are increasing international efforts directed

toward models of care delivery for survivorship. Models
may be based on the care provider, survivor population,
care site, service delivery capacity, follow-up care
duration, and the survivors role in recovery.
There are four categories of care: nurse-led survivorship
care models, risk-based care plans, rehabilitationfocused care models, and self-management techniques
for post-treatment recovery.
The burden of cancer is disproportionate across
subpopulations, with gender, race, ethnicity, age,
income, health insurance, place of residence, and access
to health care contributing to cancer disparities. These
factors may lead to disparities in survivorship care and
influence survivorship experiences.
Developing policies and programs to address disparities
in survivorship care may include examining how
financial hardship affects cancer outcomes, reducing
economic barriers to care, and increasing incorporation
of patient-centered, integrated strategies into oncology
care.
There are four main barriers to the provision of
survivorship care: lack of evidence, lack of a trained
survivorship care workforce, lack of reimbursement
structures/insurance coverage for survivorship care
services, and lack of a cohesive health care system that
would reduce fragmented care.
follow-up care for the period following treatment.9,10 As one

might expect, a variety of models have evolved based on the
provider, survivor population, site of care, and capacity for the
delivery of specific services. In addition, the type of survivorship care model is influenced by the design of the health
care delivery system(s) and payment mechanisms for health
care. Despite these differences, the major goals of care for
cancer survivors transcend national health care systems and
focus on improved survival, long-term health, and the highest
quality of life possible, including both physical and psychosocial
well-being.11 However, many recommendations for survivorship care are largely based on expert consensus. At present,
there is only limited evidence regarding services that improve
outcomes for cancer survivors, and there is essentially no information regarding the costs of providing these services
relative to their benefits.
In the United States and elsewhere, the IOM report, From
Cancer Patient to Cancer Survivor: Lost in Transition, has
served as the foundational document for establishing the
essential elements of care to be included in any survivorship
care model: surveillance for recurrence, screening for second cancers, assessment and intervention of long-term/late
effects, counseling for healthy living, and communication/
coordination with primary care.6 However, since this document was published in 2006, a number of countries, such as
the United Kingdom and Canada, have developed a national
strategy for cancer survivorship. In the United Kingdom, the
National Cancer Survivorship Initiative was launched in 2008
as a partnership between the Department of Health, which
has responsibility for the National Health Service, and
Macmillan Cancer Support, a large British charity.12 To date,
the National Cancer Survivorship Initiative has promoted
novel, proactive approaches to identifying groups at increased risk and customizing interventions for them.12 In
Canada, although the provision of health care is a provincial responsibility, the Canadian Partnership Against Cancer
implemented a national strategy in 2007.13 Similar to the
United Kingdom, Canadian Partnership Against Cancer has
focused on the concept of risk stratification to inform care
pathways. Other nations, such as Australia, have developed
regional plans with the state government organizing the
initiative. This allows for pilot projects that are specific to the
patient population of the region.7 In both examples of national and regional planning, there has been a population
focus on survivorship care with an expectation and often a
requirement for the implementation of pilot models that will
be assessed and revised as necessary.
In addition to the growing international acknowledgment
that individuals treated for cancer continue to have health
care needs, there has been an evolution in thinking about
who provides this care, how long follow-up is needed, and
the role of the survivor in his or her own recovery. Initially,
the focus of post-treatment services was generic, with a
spotlight on surveillance for recurrence and the oncology
specialist providing the follow up. But more recently, there
have been proposals and studies conducted across the globe
where models apply a risk-based approach to care so that
the unique needs of survivors are addressed while efficiently applying health care resources. Although there are an
expanding number of survivorship models to consider, some
important types of models are worth highlighting because
they are applicable across countries and have the potential
to transform the delivery of survivorship care. This article
will briefly highlight four important categories of care: (1)
nurse-led survivorship care models as an important way to
redesign the survivorship work force; (2) risk-based care
plans where survivorship services are stratified based on
need and the primary care provider (PCP) is a partner in care;
(3) care models focused on a rehabilitation; and (4) efforts to
introduce self-management techniques for post-treatment
recovery.
First, there has been an increasing interest internationally
in the development of nurse-led survivorship clinics where
nurses are responsible for a variety of services. Although the
nurse is the health care provider common to this model, the
services range from active responsibility for the medical
follow up to providing counseling and education about late
effects. In the United States, these clinics have been focused
primarily on the nurse practitioner (NP) as either an independent provider of the post-treatment care or as an
individual who provides a onetime consultation visit. In the
independent model, the survivor transitions from the oncologist to the NP who is then responsible for all the elements included in the IOM report and provides the patient
and PCP with the treatment summary and survivorship care
plan. In contrast, the consultation visit allows the oncologist
to continue providing follow-up care, but the NP provides a
one-time visit focused on the psychosocial issues, makes
appropriate referrals, and provides a survivorship care plan
to the survivor and PCP.
Watts and colleagues have published an important description of the NP role in implementing Wagners chronic
care model that includes shared visits and group visits,
both of which are relevant to oncology follow-up care.14 In
Australia, research is ongoing to evaluate nurse-led survivorship care for specific patient populations. For example,
researchers at the Peter MacCallum Cancer Centre conducted a
randomized study comparing the effect of their SurvivorCare
(SC) intervention with usual care for colorectal cancer survivors. This new intervention includes the provision of educational materials, a survivorship care plan, and a tailored, nurseled post-treatment consultation followed by three telephone
calls. The addition of SurvivorCare to usual care did not have a
beneficial effect on distress, supportive care needs, or quality
of life, but patients who received SurvivorCare were more
satisfied.15 Researchers at Peter MacCallum Cancer Centre also
implemented and evaluated a nurse-led consultation that
includes a tailored education package, a psychosocial assessment, and the provision of a tailored survivorship care plan for
long-term survivors of Hodgkin lymphoma.16 In the United
Kingdom, a nurse-led service model has been successfully
implemented for men treated for prostate cancer. During the
evaluation period, the researchers found that 90% of the invited men participated in the survivorship services, which
included face-to-face supportive care visits with the nurse

and a 6-week education program.17
Another important approach to survivorship care is one
that focuses not on the oncology providers but rather on the
PCP. Using a risk-based approach to care where survivors are
stratified as being at low, medium, or high risk for recurrence
and late effects, one can integrate the PCP in a shared-care
model or transition all the post-treatment care to the PCP at
varying points based on the potential for cancer-related
risks. Having the PCP as a focus of the post-treatment care
has been integral to survivorship planning countries
throughout Europe and in Canada where the pan-Canadian
Guideline for Survivorship services recommends that the
PCP be integrated early to provide survivorship services.18
Increasingly, this model is gaining traction in the United
States for reasons of manpower concerns, a better understanding of the shared care model successfully used by
other specialties, and the growing number of older cancer
survivors for whom the PCP is essential.
As with other survivorship models, the availability and
uptake of a rehabilitation care model will depend on the
health care system of the country and how well established
this service is as usual care. Because rehabilitation services
after illness are well developed throughout much of Europe,
survivorship services can and are applied within this construct and include psychosocial care, exercise, and physical
therapy. For example, in Germany, rehabilitation has long
been established as standard care to promote recovery after
acute illness, but it has focused on returning to work and is
done as an inpatient service. However, studies have shown
that patients prefer outpatient services, which may make
them more accessible in the future.19 In Belgium, a study was
conducted to offer adult cancer survivors access to a rehabilitation program that included physical training and
psychosocial support during a 12-week program. Repeated
measures showed improvements in physical ability and
quality of life and a decrease in fatigue.20 In contrast, in the
United States, rehabilitation services are separate and referrals are made primarily to meet physical and vocational
needs. The current state of rehabilitation across the globe
requires greater evaluation where these services are well
developed but not yet routinely offered to individuals after
cancer treatment.
Finally, self-management is increasingly discussed in relationship to cancer survivorship care. It is a component of
chronic care management that offers methods for patient
empowerment and prepares individuals by providing them
with the skills and knowledge to play a role in optimizing
their health. In both the United Kingdom and Canada, national guidance has promoted such an approach to care
using a variety of interventions such as counseling, psychoeducation, and other forms of professional ad peer support.18,21 Many of the self-management programs in the
United States build on the successful chronic disease selfmanagement model developed by Lorig and Barlow, with a
focus on promoting lifestyle change and psychosocial
health.22 Although it is a concept well suited to survivorship
233
care, self-management, to date, has not become a wellevaluated component of survivorship care in the United
States.
CONTRIBUTING FACTORS TO DISPARITIES IN

CANCER SURVIVORSHIP
We know that the burden of cancer is disproportionate in
the population. Evidence for population health disparities
in cancer incidence, morbidity, and mortality are indisputable, and some of these disparities are projected to increase
in the future.23-26 Gender, race, ethnicity, age, income, health
insurance, place of residence, and access to health care are all
factors that contribute to cancer disparities. Being black in
America is particularly associated with higher risk for poor
outcomes in cancer. For example, it is well documented that
black women present with later-stage breast cancers and
more aggressive tumors, which contributes to higher mortality and morbidity compared with white women, and black
men have higher prostate cancer incidence, more aggressive
disease, and the highest mortality of any population group
from prostate cancer.24 Overall, compared with the white
populations, the black population is less likely to survive most
forms of cancer regardless of stage of diagnosis, tumor
characteristics, and comorbidity.24
Ethnic identity is associated with disparities in cancer
morbidity and mortality, but the trends are not straightforward. Cervical cancer is an example of the complex
contribution of ethnic status to cancer rates and outcomes.
Cervical cancer incidence is higher in Hispanic women64%
higher than for non-Hispanic white women. Cervical cancer
incidence is also 30% higher in Native Hawaiian/Pacific Islander women compared with non-Hispanic white women.
Although American Indian/Alaskan Native women have
lower incidence of cervical cancer than white women, they
experience increased mortality from cervical cancer.27
Cancer survival for almost all cancers is less for Alaska Natives than for the white population for almost all cancers.28
The reasons for race and ethnic disparities in cancer rates
and outcomes are complex, with the main causes including
both biologic and social factors. For example, the dramatic
problem of cancer disparities in the Alaska Native population
is believed to be associated with this ethnic groups unique
exposure to multiple infectious and carcinogenic agents,
their unique geographic problems with access to care, and
their not so unique problems with obesity, smoking, physical
inactivity, and alcohol consumption.28
DISPARITIES IN THE SURVIVORSHIP EXPERIENCE

Are there reasons to believe that there is diversity in how
cancer is experienced across population subgroups, similar
to the diversity in cancer incidence and outcomes? We still
know very little about how racial and ethnic minorities
experience life after cancer compared with the white majority, but there is evidence that side effects from cancer
treatment are different to some degree. For example, side
effect profiles for men with prostate cancer may differ by
234
type of treatment received. Black men are known to have

lower odds of having radical prostatectomy than white
men, and they also report lower quality of life related to
physical (urinary, bowel, and sexual dysfunction) and psychological functioning post-treatment compared with white
men.26,29 The type of treatment a patient with cancer receives is an important factor in what survivors experience
after treatment, and the type of treatment received is a
function of clinical considerations and current treatment
options. However, treatment choice also varies by more
modifiable factors, including what the provider communicates to the patient about treatment options, patient
preferences, patient access to treatment centers, and
patient ability to cover the costs of treatment. In the case of
race disparities in prostate cancer post-treatment, symptoms apparently cannot be explained by clinical factors as
much as lower socioeconomic status, being uninsured, and
having treatment in a county hospital compared with a
private hospital.29
Race and ethnic differences in the survivorship experience
may be more common than we currently know, and much
more research is needed to fully understand variability in
cancer outcomes. For example, there is now evidence that
black pediatric cancer survivors experience increased rates
of anthracycline-related cardiotoxicities compared with
white pediatric cancer survivors30; vitamin D deficiency rates
in black and Hispanic breast cancer survivors are more
prevalent than among white breast cancer survivors; and
minority women breast cancer survivors suffer from disturbed sleep at higher rates than white breast cancer
survivors.31
Obesity may be an important crosscutting factor in differential cancer outcomes and side effects. Overall, cancer
survival rates are lowest among the black population and
Native American/Pacific Islanderstwo population subgroups
with a high prevalence of obesity.28,32 Obesity may also
contribute to racial and ethnic differences in certain persistent
adverse effects of cancer treatment, including lymphedema,
cancer-related fatigue, and chemotherapy-induced peripheral
neuropathy.32
REDUCING DISPARITIES OVER THE

SURVIVORSHIP CONTINUUM
Although it is clear that there are subgroups of patients with
cancer who are experiencing poorer outcomes than others,
solutions to closing this gap are not simple. One important
avenue of research is examining how financial hardship
affects risk for poor cancer outcomes. Minority patients are
more likely to experience greater financial hardship in the
wake of cancer than other groups.33 Cancer treatment is
expensive, and out-of-pocket costs are particularly problematic for those with public health insurance or no health
insurance, those with incomes under 100% of the poverty
level, and those age 50 to 64.34 Financial circumstances may
constitute a barrier to good follow-up care among survivors.
For example, decreased and discontinued hormonal therapy
adherence in women with breast cancer, delayed medical

care and prescription renewals, and higher levels of unmet
needs in cancer survivors are associated with low income
status.35-37
National oncology organizations wide endorsement of
cancer treatment summaries and cancer survivor care plans
was based on the hope that these tools would improve the
post-treatment quality of care and increase patient self-care
following cancer treatment.6 The study conducted by Kent
et al in 2013 asked survivors if they had received a treatment
summary; only 17.6% had received one whereas 62.6%
wanted but did not receive one. Those who did not receive a
treatment summary had a higher number of health information needs.37 Would the receipt of cancer care summaries or cancer treatment plans help reduce race and
ethnic disparities in cancer outcomes through improved
follow-up care? In qualitative studies, minority cancer survivors believed that survivorship care plans would be useful
in navigating post-treatment health care.38,39 However, to
date, research has not conclusively demonstrated effect of
survivorship care plans on survivors outcomes or on disparities in survivorship care. One systematic review of
outcomes of survivorship care plan found they had no effect
on survivor distress, satisfaction with care, cancer care coordination, or oncologic outcomes.40 And ultimately, there is
limited use of cancer care plans in oncology settings for a
variety of practical reasons.41
ASCOs Recommendations for Elimination of Cancer Care
Disparities in the United States called for reducing economic
barriers to care along with increasing research on disparities,
increasing minority enrollment into clinical trials, and increasing diversity of the oncology workforce.42 How have we
done so far? The expansion of insured minority Americans
under the Affordable Health Care for America Act has not
solved the problem of out-of-pocket and rising costs of cancer
care as a barrier to receiving high-quality survivorship care.
Women and minorities remain severely under-represented in
cancer clinical trials.43 There have been gains in the numbers
of minority medical school graduates, but there is still a severe
under-representation of minority physicians in oncology: in
2013, only 2.3% of oncologists were black.44
The 2008 IOM report, Cancer Care for the Whole Patient:
Meeting Psychosocial Health Needs, promoted a standard
of care that incorporates patient-centered and integrated
strategies into routine oncology care, including identifying
psychosocial health needs, connecting patients and families
to needed services, supporting patients in managing illness,
and coordinating the psychosocial and biomedical health
care services.45 New transdisciplinary research on cancer
disparities supports the notion that environmental factors
are important in cancer outcomes, specifically the role of
accumulated social stress (such as chronic exposure to
poverty and violence) in the alteration of an individuals
neuroendocrine stress response and how it influences tumor
biology and cancer progression.46,47 We cannot reverse our
patients life experiences, but we can incorporate knowledge
of the role of the social environment and social stress in
new models of care. We need models of care that support

patient self-management within continued, patient-centered
follow-up care services. To begin to reverse disparities in
cancer outcomes, we must provide cancer survivors with
rational and practical supportspecifically the services of
integrated social work professionalsto help them secure
social services and basic need referrals and to provide
psychosocial support to increase their opportunity to have
similar experiences in cancer as the most privileged patients
with cancer.
BARRIERS TO PROVIDING HIGH-QUALITY

SURVIVORSHIP CARE
There are multiple reasons that every survivors journey
is not the same. Earle and Ganz discuss an illuminating
framework for understanding the barriers to providing highquality survivorship care.48 In this framework, physicians
incorporate something new into their practice only if at least
one of three conditions is met: (1) it has clearly been shown
to be better for their patients; (2) they are specifically remunerated for it; or (3) it is more efficient for them in their
practice. Earle and Ganz state that at the time of their
manuscript, survivorship care as envisioned by the IOM
report does not readily meet any of these criteria.
In the context of this framework, we discuss below four
barriers to the provision of survivorship care: lack of evidence, lack of a trained survivorship care workforce, lack of
reimbursement structures/insurance coverage for survivorship care services, and lack of a cohesive health care
system that would reduce fragmented care.
Lack of Evidence for Survivorship Care

The basis for many of the barriers to providing high-quality
care to all survivors is the lack of evidence regarding best
practices for survivorship care. As discussed by McCabe et al,
much of the literature on survivorship care is based on
analyses of existing datasets or surveys.49 Few prospective
comparative studies have examined the benefits (in health
or economic terms) of alternative programs or models of
care for cancer survivors. Furthermore, even the available
databases are limited. In describing programs for childhood
cancer survivors, many of which have been in existence
substantially longer than have programs for adult cancer
survivors, Eshelman-Kent et al reported that only 41% of
participating Childrens Oncology Group centers indicated
having a database that tracked survivors health outcomes.50
The evidence base demonstrated that improved outcomes
associated with survivorship care are extremely limited.
More importantly, there is almost no information on
whether one type of survivorship care model results in
better outcomes compared with another type of model, or
even what a survivorship care model is.51 When assessment of survivorship programs have been performed, these
assessments have generally focused on patient-reported
outcomes such as quality of life. Although these are
clearly important, there is a critical lack of information on
235
long-term outcomes, including survival and costs, associated

with survivorship care.51 In addition, although efforts to
incorporate survivorship care plans into survivorship care
are becoming more widespread, there is little evidence that
these plans have substantial or sustained benefits among
survivors.40
Lack of a Trained Survivorship Care Workforce

Reports have discussed current and projected future shortages among the oncology workforce.52 This shortage is made
more acute by the increase over time in resource utilization
intensity for cancer care. As discussed by Bunnell and Shulman, among individuals diagnosed with cancer, there has
been an increasing number of physician visits and infusion
visits per patient.53 Related to the lack of reimbursement
for survivorship services (discussed later), this workforce
shortage highlights the opportunity costs associated with
survivorship care; that is, oncologists have insufficient time
available to engage in all patient activities and may be able
to earn greater reimbursement by providing cancer treatment rather than survivorship care. Furthermore, developing,
implementing, and operating survivorship programs is time
consuming. Among institutions providing care for survivors
of childhood cancer, the most commonly reported barrier
was lack of dedicated time for developing late effects care
programs.50
Other models of survivorship care, including shared care
with primary care physicians and involvement of advanced
practice providers (physician assistants and NPs), have been
discussed as strategies to improve survivorship care in the
context of oncologist shortages. However, shortages among
primary care physicians and increased numbers of patients
with health insurance (following implementation of the
Affordable Care Act) who need primary care services may
limit the ability to implement shared care survivorship
models.53 There are also substantial differences between
PCPs and oncologists regarding the perceived role of PCPS in
survivorship care. In a U.S. physician survey, most PCPs
supported a shared care model, whereas most oncologists
favored an oncologist-based model of survivorship care.54 A
majority of oncologists (87%) did not feel that PCPs should
take on the primary role of cancer follow up. Similarly,
Potosky et al reported that oncologists were less likely than
PCPs to believe that PCPs could conduct appropriate testing
for breast cancer recurrence or could provide care for late
effects of breast cancer.55 These findings suggest that despite workforce shortages, U.S. oncologists may not be
accepting of opportunities to share care or transition care for
survivors to PCPs. However, involvement of other health
care professionals in survivorship care may be more common in other countries. For example, in a survey of Italian
oncologists, Numico et al reported that only 55% and 30% of
oncologists in Italy indefinitely follow patients with breast
cancer and patients with colorectal cancer, respectively,
while other oncologists discharge survivors to PCPs usually
after 5 years.56
236
A potential strategy to address oncology workforce shortages related to survivorship care is to enhance training in
this area. Rowland et al highlighted the need to train the
next generation of health care providers about survivorship
issues and to attract clinicians and researchers to this area.57
McCabe et al also discussed expanding and coordinating educational opportunities for medical professionals
as a strategy toward improving overall survivorship care.49
Essential topics specified for survivorship education curricula
included survivors quality of life, health promotion, palliative
care, and prevention, diagnosis, and treatment of recurrences,
secondary cancers, and late-occurring complications.
Lack of Reimbursement for Survivorship Care Services

Lack of reimbursement for survivorship care is largely a
problem in the United States, which lacks a national health
care system. However, many of the issues related to this
barrier apply globally to physicians, when there is a need to
prioritize the cancer care services provided. Survivorship
care can be time consuming. Eshelman-Kent et al reported
that an initial visit with survivors of childhood cancer required 122 minutes, whereas subsequent annual visits were
91 minutes.50 In a survey of Australian oncologists and
nurses who provided breast cancer survivorship care, onethird of these specialists reported spending more than 25%
of their clinical time providing follow-up care.58
Most U.S. insurers do not provide additional reimbursement
for survivorship care, which may include psychosocial support
and care planning services. As presented in the 2009 IOM
report Ensuring Quality Cancer Care Through the Oncology
Workforce: Sustaining Research and Care in the 21st Century:
Reimbursement for survivorship care does not cover the
costs of providing appropriate monitoring and support for
the physical, social, and emotional effects in the short and
long term course for the disease and treatment of cancer. In
order to offer these services, institutions that do provide this
care absorb the cost or must seek funding from other
sources. Many community-level institutions and smallerscale providers simply cannot afford to do so.59, p. 32
McCabe et al also discuss these barriers, commenting

on the limited reimbursement for components of survivorship care.49 These authors highlight that this may result
in a misalignment of resource utilization, with revenuegenerating services (such as surveillance testing) being
overused and non-revenue-generating survivorship care
services (such as care coordination) being underused. These
nonreimbursed services may be time intensive. However,
although care coordination between oncologists and PCPs
is almost always highlighted as an important component of
survivorship care, definitions for care coordination vary
greatly, and little evidence is available showing the health
or economic benefits of care coordination for cancer
survivors.
Furthermore, it may be more efficient (and equally effective) to provide certain survivorship care services by email or
telephone rather than in person, but many insurers reimburse
only for face-to-face delivery of care, resulting in a financial

disincentive for provision of remote services.
Lack of a Cohesive Health Care System That Would

Reduce Fragmented Care
Approximately 75% of survivors experience long-term or late
effects of cancer and cancer treatment.4 These long-term
and late effects can involve all body systems and can affect
all aspects of health. As such, we need a diverse range of
health care providers to support cancer survivors. As discussed by Landier, the complexity of survivorship care corresponds to the diversity of patient characteristics, diagnoses,
treatments, and potential sequelae:
Consequences of cancer treatment span the medical, psychosocial, socioeconomic and spiritual domains, and survivors and their healthcare providers often lack awareness of
treatment-related risks and the potential interactions of
these risks with common comorbidities, such as obesity,
diabetes, cardiovascular disease, anxiety, and depression. 60
This diversity in symptoms experienced and health care

services needed among cancer survivors almost guarantees
that their health care will be fragmented. Even dedicated
and well-staffed cancer survivorship programs and health
care systems may not be able to provide care for all the
possible needs of survivors in a timely manner.
FUTURE AREAS OF FOCUS FOR SURVIVORSHIP

CARE RESEARCH
Based on the information summarized above, we recommend consideration of several areas for future research on
survivorship care. We must focus on the continued development of an evidence base for survivorship care. This
should include rigorous assessments of the following:
Effect of different types of models of care, including
personnel providing survivorship care, use of health
information technologies and care not administered
face to face, and comorbidities and risk adjustment
Needed resources (including funding, time, and personnel) to implement different types of survivorship care
programs among an oncology workforce that is already

experiencing substantial shortages
Effectiveness of care coordination (and development
of a common specification or specifications of care
coordination) for survivors
Costs and cost-effectiveness for survivorship care
Key structures and settings, processes and services, and
outcomes for survivorship programs, including quality of
care metrics that can potentially be associated with
reimbursement.
Other areas of focus for survivorship care research include
the following:
Development of strategies to improve provision of
patient-centered survivorship care to individuals from
underserved populations and address survivorship care
disparities
Empowerment of cancer survivors to be active participants in their care
Approaches to integrate survivorship care in physician
practices, health care institutions, health care systems,
and broader communities, and to include diverse groups
of health care providers
Improvement of tools and support to facilitate efficient
provisioning of survivorship care by oncologists, PCPs,
and other health care professionals
Consideration of how evidence-based interventions can
be disseminated to and implemented in other settings,
in the United States and internationally
Dissemination of information to government and private sector health care decision makers regarding the
need for high-quality survivorship care and for development of policies that facilitate implementation and
evaluation of evidence-based programs
As cancer care continues to improve and the survivor
population continues to grow, conducting research to address knowledge gaps and barriers in survivorship care will
become even more urgent. This will be critical to enhance
health outcomes and quality of life among survivors and to
help ensure that every survivor experiences an optimal and
patient-centered journey.
References
1. DeSantis CE, Lin CC, Mariotto AB, et al. Cancer treatment and survivorship statistics, 2014. CA Cancer J Clin. 2014;64:252-271.
2. De Angelis R, Grande E, Inghelmann R, et al. Cancer prevalence estimates in Italy from 1970 to 2010. Tumori. 2007;93:392-397.
3. Herrmann C, Cerny T, Savidan A, et al. Cancer survivors in Switzerland:
a rapidly growing population to care for. BMC Cancer. 2013;13:
287.
4. Mattioli V, Montanaro R, Romito F. The Italian response to cancer
survivorship research and practice: developing an evidence base for
reform. J Cancer Surviv. 2010;4:284-289.
5. Maddams J, Utley M, Mller H. Projections of cancer prevalence in the
United Kingdom, 2010-2040. Br J Cancer. 2012;107:1195-1202.
6. Hewitt M, Greenfield S, Stovall E (eds). From Cancer Patient to Cancer

Survivor: Lost in Transition. Washington, DC: National Academies Press; 2005.
7. Jefford M, Rowland J, Grunfeld E, et al. Implementing improved posttreatment care for cancer survivors in England, with reflections from
Australia, Canada and the USA. Br J Cancer. 2013;108:14-20.
8. Bell K, Ristovski-Slijepcevic S. Cancer survivorship: why labels matter.
J Clin Oncol. 2013;31:409-411.
9. McCabe MS, Jacobs LA. Clinical update: survivorship caremodels and
programs. Semin Oncol Nurs. 2012;28:e1-e8.
10. McCabe MS, Partridge AH, Grunfeld E, et al. Risk-based health care, the
cancer survivor, the oncologist, and the primary care physician. Semin
Oncol. 2013;40:804-812.
237
11. McCabe MS, Faithfull S, Makin W, et al. Survivorship programs and care
planning. Cancer. 2013;119:2179-2186.
12. Richards M, Corner J, Maher J. The National Cancer Survivorship Initiative: new and emerging evidence on the ongoing needs of cancer
survivors. Br J Cancer. 2011;105:S1-S4.
13. Canadian Partnership Again Cancer. www.partnershipagainstcancer.ca.
14. Watts SA, Gee J, ODay ME, et al. Nurse practitioner-led multidisciplinary teams to improve chronic illness care: the unique strengths of
nurse practitioners applied to shared medical appointments/group
visits. J Am Acad Nurse Pract. 2009;21:167-172.
15. Jefford M, Gough K, Drosdowsky A, et al. A randomized controlled trial
of a supportive care package (SurvivorCare) for survivors of colorectal
cancer. J Clin Oncol. 2016;34:3s (suppl; abstr 79).
16. Gates P, Seymour JF, Krishnasamy M. Insights into the development of a
nurse-led survivorship care intervention for long-term survivors of
Hodgkins lymphoma. Aust J Cancer Nurs. 2012;13:4-10.
17. Ferguson J, Aning J. Prostate cancer survivorship: a nurse-led service
model. Br J Nurs. 2015;24:S14-S21.
18. Howell D, Hack TF, Oliver TK, et al. Survivorship services for adult cancer
populations: a pan-Canadian guideline. Curr Oncol. 2011;18:e265-e281.
19. Bergelt C, Lehmann C, Beierlein V, et al. Patients assessment of inpatient
and outpatient rehabilitation. Psychooncology. 2009;18:S48.
20. Dergent T, Troch E, Vandebroek A, et al. Finding a new balance: a cancer
rehabilitation program. Psychooncology. 2009;18:S120.
21. Department of Health. The Expert Patient: A New Approach to Chronic
Disease Management for the 21st Century. London: Her Majestys
Stationery Office; 2001.
22. Barlow J, Wright C, Sheasby J, et al. Self-management approaches for
people with chronic conditions: a review. Patient Educ Couns. 2002;48:
177-187.
23. Aizer AA, Wilhite TJ, Chen MH, et al. Lack of reduction in racial disparities in cancer-specific mortality over a 20-year period. Cancer. 2014;
120:1532-1539.
24. American Cancer Society. Cancer Facts and Figures for African
Americans 2013-2014. http://www.cancer.org/acs/groups/content/
@epidemiologysurveilance/documents/document/acspc-036921.
pdf. Accessed February 7, 2016.
25. Blinder VS, Griggs JJ. Health disparities and the cancer survivor. Semin
Oncol. 2013;40:796-803.
26. Borysova ME, Sultan DH, Chornokur G, et al. Prostate cancer disparities
through the cancer control continuum. Soc Sci. 2013;2:247-269.
27. National Cancer Institute, Division of Cancer Control and Population
Sciences, Surveillance Research Program, Surveillance Systems Branch.
Surveillance, Epidemiology, and End Results (SEER) Program. http://
seer.cancer.gov/statfacts/html/cervix.html. Accessed March 13, 2016.
28. White MC, Espey DK, Swan J, et al. Disparities in cancer mortality and
incidence among American Indians and Alaska Natives in the United
States. Am J Public Health. 2014;104:S377-S387.
29. Jenkins R, Schover LR, Fouladi RT, et al. Sexuality and health-related
quality of life after prostate cancer in African-American and white men
treated for localized disease. J Sex Marital Ther. 2004;30:79-93.
30. Aminkeng F, Ross CJ, Rassekh SR, et al. Higher frequency of genetic
variants conferring increased risk for ADRs for commonly used drugs
treating cancer, AIDS and tuberculosis in persons of African descent.
Pharmacogenomics J. 2014;14:160-170.
31. Yao S, Ambrosone CB. Associations between vitamin D deficiency and
risk of aggressive breast cancer in African-American women. J Steroid
Biochem Mol Biol. 2013;136:337-341.
32. Schmitz KH, Agurs-Collins T, Neuhouse ML, et al. Impact of obesity, race
and ethnicity on cancer survivorship. In Bowen DJ, Denis GV, Berger NA
(eds). Impact of Energy Balance on Cancer Disparities, Energy Balance
238
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
and Cancer 9. Switzerland: Springer International Publishing, 2014;

63-90.
Yabroff KR, Dowling EC, Guy GP Jr, et al. Financial hardship associated
with cancer in the United States: findings from a population-based
sample of adult cancer survivors. J Clin Oncol. 2016;34:259-267.
Guy GP Jr, Yabroff KR, Ekwueme DU, et al. Healthcare expenditure
burden among non-elderly cancer survivors, 2008-2012. Am J Prev Med.
2015;49:S489-S497.
Burg MA, Adorno G, Lopez ED, et al. Current unmet needs of cancer
survivors: analysis of open-ended responses to the American Cancer
Society Study of Cancer Survivors II. Cancer. 2015;121:623-630.
Hershman DL, Tsui J, Wright JD, et al. Household net worth, racial
disparities, and hormonal therapy adherence among women with earlystage breast cancer. J Clin Oncol. 2015;33:1053-1059.
Kent EE, Forsythe LP, Yabroff KR, et al. Are survivors who report cancerrelated financial problems more likely to forgo or delay medical care?
Cancer. 2013;119:3710-3717.
Burg MA, Lopez ED, Dailey A, et al. The potential of survivorship care
plans in primary care follow-up of minority breast cancer patients. J Gen
Intern Med. 2009;24:S467-S471.
Keesing S, McNamara B, Rosenwax L. Cancer survivors experiences of
using survivorship care plans: a systematic review of qualitative studies.
J Cancer Surviv. 2015;9:260-268.
Brennan ME, Gormally JF, Butow P, et al. Survivorship care plans in
cancer: a systematic review of care plan outcomes. Br J Cancer. 2014;
111:1899-1908.
Salz T, McCabe MS, Onstad EE, et al. Survivorship care plans: is there
buy-in from community oncology providers? Cancer. 2014;120:
722-730.
Moy B, Polite BN, Halpern MT, et al. American Society of Clinical
Oncology policy statement: opportunities in the patient protection and
affordable care act to reduce cancer care disparities. J Clin Oncol. 2011;
29:3816-3824.
Kwiatkowski K, Coe K, Bailar JC, et al. Inclusion of minorities and women
in cancer clinical trials, a decade later: have we improved? Cancer. 2013;
119:2956-2963.
Hamel LM, Chapman R, Malloy M, et al. Critical shortage of African
American medical oncologists in the United States. J Clin Oncol. 2015;
33:3697-3700.
Institute of Medicine. Adler NE, Page AEK (eds). Cancer Care for the
Whole Patient: Meeting Psychosocial Health Needs. Washington, DC:
The National Academies Press, 2008.
Gehlert S, Sohmer D, Sacks T, et al. Targeting health disparities: a model
linking upstream determinants to downstream interventions. Health
Aff (Millwood). 2008;27:339-349.
Volden PA, Conzen SD. The influence of glucocorticoid signaling on
tumor progression. Brain Behav Immun. 2013;30:S26-S31.
Earle CC, Ganz PA. Cancer survivorship care: dont let the perfect be the
enemy of the good. J Clin Oncol. 2012;30:3764-3768.
McCabe MS, Bhatia S, Oeffinger KC, et al. American Society of Clinical
Oncology statement: achieving high-quality cancer survivorship care.
J Clin Oncol. 2013;31:631-640.
Eshelman-Kent D, Kinahan KE, Hobbie W, et al. Cancer survivorship
practices, services, and delivery: a report from the Childrens Oncology
Group (COG) nursing discipline, adolescent/young adult, and late effects committees. J Cancer Surviv. 2011;5:345-357.
Halpern MT, Viswanathan M, Evans TS, et al. Models of cancer survivorship care: overview and summary of current evidence. J Oncol Pract.
Epub 2014 Sep 9.
Erikson C, Salsberg E, Forte G, et al. Future supply and demand for
oncologists : challenges to assuring access to oncology services. J Oncol
Pract. 2007;3:79-86.
53. Bunnell CA, Shulman LN. Will we be able to care for cancer patients in
the future? Oncology (Williston Park). 2010;24:1343-1348.
54. Cheung WY, Aziz N, Noone AM, et al. Physician preferences and attitudes regarding different models of cancer survivorship care: a comparison of primary care providers and oncologists. J Cancer Surviv. 2013;
7:343-354.
55. Potosky AL, Han PK, Rowland J, et al. Differences between primary care
physicians and oncologists knowledge, attitudes and practices regarding the care of cancer survivors. J Gen Intern Med. 2011;26:
1403-1410.
56. Numico G, Pinto C, Gori S, et al. Clinical and organizational issues in the
management of surviving breast and colorectal cancer patients: attitudes and feelings of medical oncologists. PLoS One. 2014;9:e101170.
57. Rowland JH, Hewitt M, Ganz PA. Cancer survivorship: a new challenge in
delivering quality cancer care. J Clin Oncol. 2006;24:5101-5104.
58. Brennan ME, Butow P, Spillane AJ, et al. Survivorship care after breast
cancer: follow-up practices of Australian health professionals and attitudes to a survivorship care plan. Asia Pac J Clin Oncol. 2010;6:116-125.
59. Institute of Medicine. Ensuring Quality Cancer Care Through the Oncology Workforce: Sustaining Research and Care in the 21st Century:
Workshop Summary. Washington, DC; The National Academies Press,
2009.
60. Landier W. Survivorship care: essential components and models of delivery.
Oncology (Williston Park). 2009;23:46-53. http://www.cancernetwork.
com/oncology-nursing/survivorship-care-essential-components-and-modelsdelivery. Accessed March 14, 2016.
239
Cancer Treatment as an
Accelerated Aging Process
CHAIR
Arti Hurria, MD
City of Hope
Duarte, CA
SPEAKERS
Hyman B. Muss, MD
Chapel Hill, NC
Lee Jones, PhD
New York, NY
HURRIA, JONES, AND MUSS
Cancer Treatment as an Accelerated Aging Process:

Assessment, Biomarkers, and Interventions
Arti Hurria, MD, Lee Jones, PhD, and Hyman B. Muss, MD
OVERVIEW
An accumulating body of evidence supports the hypothesis that cancer and/or cancer treatment is associated with
accelerated aging. The majority of these data come from the pediatric literature; however, a smaller yet growing body of
literature points toward similar findings in the geriatric population. This is a key survivorship issue the growing number of
older adults with cancer face, along with the short- and long-term impact of cancer therapy on the aging process. This article
will review clinical and biologic markers of aging in older adults with cancer, use cardiovascular disease as a model of
accelerated aging, and discuss potential interventions to decrease the risk.
he U.S. population is aging with the number of individuals age 65 or older anticipated to double between
2010 and 2030.1 This growing population is at risk for cancer
because the majority of cancer incidence and mortality occurs
in individuals age 65 and older. Together, the aging of the U.S.
population and the association of cancer and aging is culminating in a 67% increase in cancer incidence in individuals
age 65 or older in the United States from 2010 to 2030.2
However, many of these individuals will survive cancer, and
the majority of cancer survivors are older adults. Presently,
there are 8 million cancer survivors age 65 or older in the
United States, and this number is anticipated to continue to
grow to 11 million by 2020.3
A key survivorship issue facing these older adults is the shortand long-term impact of cancer therapy on the aging process.
It has been suggested that cancer and/or its treatment may
contribute to an accelerated aging phenotype.4 The majority of
these data come from the pediatric literature, in which cancer survivors are more predisposed to the development of
frailty as well as chronic conditions such as myocardial infarction, congestive heart failure, and second cancers.5-7 There
is a smaller yet growing body of literature pointing toward
similar findings in the geriatric population. This article will
review clinical and biologic markers of aging in older adults with
cancer, use cardiovascular disease as a model of accelerated
aging, and discuss potential interventions to decrease the risk.
THE CLINICAL ASSESSMENT OF AGING

The aging process is unique to the individual, and chronological age is a poor descriptor of an older adult. For example,
two individuals who are chronologically age 75 can have very

different functional ages. At the extremes, one individual
could be wheelchair-bound in a nursing home, and another
may be a marathon runner; distinguishing the difference in
functional age between these two individuals can be performed with the eyeball test from the door of the examination room. However, for most clinical situations, an
individuals outward appearance can be deceiving, and an
individuals functional age can be quite difficult to determine
without a more detailed evaluation. There are two main
ways in which an oncologist can get a better sense of the
functional age of an older adult. The first is through performing
a geriatric assessment, and the second is by assessing frailty.
Both of these methods are discussed below.
A geriatric assessment identifies factors other than
chronological age that can predict the risk of morbidity and
mortality in older adults. These include functional status,
cognition, comorbidity, psychological state, social support,
and nutritional status. There is a growing body of literature
regarding the benefits of performing a geriatric assessment
in older adults with cancer.8 In patients with cancer, this
assessment can identify areas of vulnerability not otherwise
detected in a routine history and physical examination,
predict cancer treatment toxicity and survival, and serve as a
platform for interventions to decrease toxicity risk.8 Short
geriatric assessment tools for oncologists have been developed that are feasible to implement in both daily clinical
practice and among patients enrolled in clinical trials. Geriatric assessment tools that can be mailed to the patient and
completed prior to the office visit, as well as computerized
From the City of Hope Comprehensive Cancer Center and Beckman Research Institute, Duarte, CA; Memorial Sloan Kettering Cancer Center, New York, NY; The University of North
Carolina at Chapel Hill, Chapel Hill, NC.
Corresponding author: Arti Hurria, MD, City of Hope, 1500 East Duarte Rd., Duarte, CA 91010; email: Ahurria@coh.org.
e516
CANCER TREATMENT AND ACCELERATED AGING
geriatric assessment tools, have also been evaluated.9-11

Abbreviated geriatric assessment screening tools are available; however, a consensus has not been reached regarding
which tool could identify those patients who would benefit
from a more detailed geriatric assessment.12
There are compelling geriatric assessment findings in
cancer survivors that support the hypothesis that cancer and
its treatment may impact the aging process. Cancer survivors
are more likely to report poorer physical and mental healthrelated quality of life compared with adults with cancer.13,14
Older cancer survivors are more likely to have limitations in
performing activities of daily living as well as mobility limitations compared with older adults without a history of
cancer.15 There is an increase in the number of comorbidities
in individuals who are cancer survivors compared with
those without a history of cancer.16 Furthermore, specific
comorbid conditions may be linked to the treatment the
patient has receivedfor example, congestive heart failure
(among patients receiving anthracycline-based therapies),17
peripheral neuropathy (among patients receiving taxanes),18
and declines in bone health (among patients receiving aromatase inhibitors).19,20 A study of the neuropsychological
function of older patients (age 6070) with breast cancer
demonstrated that those who were exposed to chemotherapy were at higher risk for posttreatment cognitive decline and factors associated with cognitive agingsuch as
lower cognitive capacity (low cognitive reserve) and apolipoprotein (ApoE4+) statusinteract with chemotherapy
treatment to increase the risk of cognitive decline.8
Another means of assessing the aging process, which
comes primarily from the geriatric literature, is measuring
frailty. Frailty can be defined as a decrease in physiologic
reserve that places an individual at increased risk for adverse
events such as hospitalization, falls, and poorer overall
survival. There are two main methods of assessing frailty in
the geriatric population. The first was proposed by Fried
KEY POINTS
Cancer and/or its treatment may contribute to an

accelerated aging phenotype.
A key survivorship issue facing older adults is the shortand long-term impact of cancer therapy on the aging
process.
Clinical (geriatric assessment) and biologic markers of
aging hold great promise as independent predictors of
patient outcomes including toxicity, functional reserve,
and survival.
Standard cancer treatment causes significant and
marked impairments in global cardiovascular function,
which may persist years after the cessation of primary
therapy.
Structured exercise training may be an effective strategy
to mitigate acute and long-term impairments in
cardiovascular function in patients both during and
following primary adjuvant therapy.
TABLE 1. The Frailty Phenotype21

Categorizations
Criteria
Frailty Phenotype: Presents

With 3 Criteria
Unintentional weight loss

($ 10 pounds in past year)
Self-reported exhaustion
Intermediate or Prefrail
Weakness (lowest 20th percentile in
Phenotype: Presents
grip strength)
With 1 or 2 of the Criteria
Slow walking speed (lowest
20th percentile on a timed walk of
15 feet)
Low physical activity (lowest quintile of
kilocalories per week)
et al21 utilizing data from the Cardiovascular Health Study.

They identified a phenotype for frailty among over 5,000
community-dwelling men and women age 65 or older
(Table 1), which consists of weight loss, exhaustion, slow
walking speed, weakness, and low physical activity. Patients
who were defined as frail or prefrail, compared with nonfrail,
were at increased risk for hospitalization, falls, decreased
mobility, decline in the ability to complete activities of daily
living, and mortality over the subsequent 3 and 7 years.
Rockwood and Mitnitski22 defined another method of
assessing frailty that is based on the accumulation of deficits.
Deficits captured in a geriatric assessment are tallied and
provide a composite index of the degree of frailty. These
methods of describing frailty have mainly been used as
research tools rather than tools used in daily clinical practice.
BIOMARKERS OF AGING AND CANCER THERAPY

Geriatric assessment is the cornerstone for assessing
function in patients with cancer prior to treatment. It can be
helpful in predicting survival, treatment-related toxicity, and
other outcomes. However, geriatric assessment can be time
consuming, and many clinicians do not have the resources to
perform a geriatric assessment in daily practice. Biomarkers
of aging may help fill this gap.23 The term biomarker has
many definitions; the World Health Organization has
defined a biomarker as almost any measurement reflecting
an interaction between a biological system and a potential
hazard, which may be chemical, physical, or biological. The
measured response may be functional and physiological,
biochemical at the cellular level, or a molecular interaction.
A list of potential biomarkers is provided in Table 2. For this
review, we will focus on several categories of potential
biomarkers, including chronic inflammatory markers,
markers of cellular senescence, and sarcopenia to explore
how they might be further evaluated with the goal of defining markers that can independently predict outcomes in
older patients with cancer. Several excellent reviews of this
topic have been recently published.23,24
Inflammatory markers have been extensively studied, and
increased levels have been shown to correlate with frailty,
functional decline, and survival.24 These markers now are
receiving wide attention, as there is good evidence that
chronically elevated levels may accelerate or exacerbate the
e517
TABLE 2. Potential Biomarkers of Treatment-Related Toxicity and Aging

Association With
Frailty/Function
Association With
Mortality
ELISA
Yes (CRP, IL-6, TNF-a,

D-dimer, IL1RA)
Yes (CRP, IL-6,

D-dimer, s-VCAM)
q-PCR or southern blot
Yes
Yes
Marker
Source
Test
Chronic Inflammatory
Markers
Serum or plasma
Telomere Length
Leukocyte DNA
FISH
STELA
p16INK4a
T lymphocyte RNA
RT-qPCR
Maybe
Unknown
Sarcopenia
DEXA scan
Commercially available software

for body composition analysis
Yes
Yes
Incremental exercise testing
Yes
Yes
CT scan
Bioelectrical
Impedance
Maximal Oxygen
Consumption
(VO2max)
O2 and CO2 of inhaled and exhaled air
Abbreviations: ELISA, enzyme-linked immunosorbent assay; CRP, C-reactive protein; TNF-a, tumor necrosis factor a; IL1RA, IL-1 receptor agonist; s-VCAM, soluble vascular cell
adhesion molecule; q-PCR, quantitative polymerase chain reaction; FISH, fluorescent in situ hybridization; STELA, single telomere length analysis; RT-qPCR, quantitative reverse
transcription polymerase chain reaction; DEXA, dual-energy x-ray absorptiometry.
Modified from Hubbard et al.23
aging process. These markers, which include interleukins,

tumor necrosis factors, and others, have been studied
extensively in frail patients in whom they independently
correlate with other measures of physical function.
Interleukin-6 (IL-6) has probably been the most extensively
studied cytokine and has been shown to predict functional
decline, including a diminution in the ability to perform
activities of daily living, poor ambulation, and decreased
mobility.25 There also appears to be a major relationship
between inflammatory markers and cell senescence. Senescent cells are viable and capable of secreting proinflammatory markers that have led to the definition of a
senescence-associated secretory phenotype.26 To date,
however, none of these markers has assumed a major role in
clinical care or further studies designed to see if any single
marker or combination might have an independent role in
the management of the older patient with cancer. These studies
would test whether such markers could be independent predictors of treatment tolerance, including acute and chronic
toxicities, functional loss, and cognitive decline.
Telomere length as well as the proteins that play a role in
telomere length are also of great interest as markers that
may predict survival, functional status, and toxicity,27,28 and
studies are underway to further define the potential role of
telomere length as a predictor of cancer-related outcomes.
Another emerging marker of great potential benefit as a
predictor of toxicity and outcomes is p16ink4a. This gene, in
which expression increases 10-fold with aging, codes for a
protein that blocks cyclin-dependent kinase, which leads to
cell senescence.29 In animal models, there is a clear direct
relationship between p16ink4a expression and organ age.30
In addition, human studies have shown a strong association
of p16ink4a expression and T-cell aging in patients infected
with HIV,31 senescence of human mesenchymal stem cells,32
and hospital stay after coronary artery bypass.33
e518
Other markers of potential interest include the measurement of sarcopenia; recent data show the utility of
using CT scans of the abdomen as a valid measure of
muscle mass and a tool to predict cancer clinical outcomes.34 This is of great interest, as CT scans are widely
used in management of patients with cancer. The role of
maximum oxygen consumption (VO2max) as a biomarker of
aging and treatment effect is discussed elsewhere in this
review.
THE EFFECT OF CANCER TREATMENT ON

BIOMARKERS OF AGING
There is little doubt that the treatment of cancer, especially
radiation therapy and chemotherapy, greatly accelerates
aging.35,36 A recent overview of survivors of childhood
cancer showed that these individuals were at greatly increased risk for substantial comorbidity and premature
death.35 Data from one of the large cohorts described in this
review demonstrated the cumulative prevalence for a serious or life-threatening chronic condition of 81% by age 45;
in addition, there was an extremely high incidence of second
neoplasms that was directly related to the radiation dose. In
another study of survivors of childhood cancer, the prevalence of prefrailty and frailty were 31.5 and 13.1% among
women and 12.9 and 2.7% among men, respectively. This
prevalence of frailty among young adult survivors of cancer
with a mean age of 34 years was similar to that of adults age
65 or older.37
Operative procedures result in a cascade of cytokine and
acute-phase responses including IL-6, white cell count, and
C-reactive protein. In a systematic review that included 164
studies involving 14,362 patients, IL-6 and C-reactive protein
responses were clearly associated with the magnitude and
invasiveness of the operative procedure. Colorectal cancer
surgery resection was associated with the highest acute

increase in cytokines.38 Although of concern in the short run,
it is not likely that surgical procedures directly accelerate
aging. Radiation therapy is clearly related to the formation
of proinflammatory cytokines and may result in progressive and long-term tissue damage.39 As shown in survivors
of childhood cancer, radiation is associated with the accelerated development of second malignancies as well as
other comorbidities. Studies testing how biomarkers might
be used to predict radiation toxicity in individual patients
and possible interventions to ameliorate radiation effects
are needed. As the childhood cancer survivorship studies
show, chemotherapy also is a major cause of accelerated
aging. This has been well demonstrated in vitro40 and in
animal models.30 Chemotherapy has a major effect on
telomere length41 and has been associated with telomere
shortening in hematopoietic stem cells42 as well as in peripheral blood mononuclear cells in patients given repetitive
standard-dose chemotherapy for solid tumors.43 Furthermore, telomere shortening was shown to be greater in older
patients given combined chemotherapy and radiation for
head and neck cancer when compared with younger
patients. 44
p16ink4a has major promise as a biomarker of chemotherapy toxicity. p16ink4a expression increases approximately 10-fold between ages 20 and 80, and this dynamic
range provides for a more robust marker as a predictor of
molecular aging. In one study of women receiving adjuvant
chemotherapy for early-stage breast cancer, p16ink4a
expression measured in peripheral blood T cells increased
by almost one log2 order of magnitude immediately after
treatment and remained elevated 12 months after treatment. 45 This change corresponds to almost a 15-year
increase in chronologic age. In this study, the cytokines
VEGFA and monocyte chemotactic protein-1 also significantly increased and remained elevated at 12 months,
but telomere length was not affected. In a cross-sectional
cohort of patients in the same study, prior chemotherapy exposure was independently associated with increased p16ink4a expression comparable to 10 years of
chronologic aging. Current studies are underway exploring the potential role of p16ink4a as a predictor of toxicity and subsequent comorbidity in patients receiving
chemotherapy.
There is no doubt that chemotherapy and radiation
therapy accelerate aging. This is an especially concerning
observation in younger patients, for whom the dramatically
improved survival rates for many childhood cancers are
now associated with the early development of comorbidities and an increased risk of second cancers, and in
older patients, in whom such changes may result in the
development of new comorbidities or accelerate previous
noncancer-related illness. Several biomarkers of aging are
now available that might prove to identify those patients
most vulnerable to cancer treatment and allow for the earlier
testing of interventions that might minimize treatment related
toxicity.
ACCELERATED AGING IN PATIENTS WITH

CANCER: CARDIOVASCULAR AGING AS A MODEL
As described in the preceding sections, older patients with
cancer are subject to the deleterious effects associated with
normal aging but experience the confounding effects of
cancer treatment that can lead to an accelerated aging
phenotype, characterized by a notable impairment in
physical functioning and other parameters. In current
practice, the extent of functional decline or physical functioning is typically assessed using performance status
scoring systems, evaluated either by the Karnofsky Performance Scale (KPS) or the Eastern Cooperative Oncology
Group (ECOG) scale. However, performance status scoring
systems are subjective and lack sensitivity to discriminate
between individuals, particularly those defined as having a
good (i.e., KPS . 70; ECOG 0 to 1) performance status.46,47
As a result, performance status measurements are often
supplemented with the use of other more objective measures of overall physical functioning such as the comprehensive geriatric assessment or measurement of cardiac
and lung function via resting assessment of left ventricular
ejection fraction and pulmonary function testing, respectively. Although these tools provide valuable information
regarding physical functioning, treatment eligibility, and risk
stratification, they do not provide evaluation of global cardiovascular function and reserve. Indeed, cardiac function is
only one organ component that contributes to the integrative
capacity of the cardiovascular and musculoskeletal system to
transport and use oxygen (O2) for adenosine triphosphate
resynthesis.48 The efficiency of O2 transport and utilization
determines an individuals cardiovascular performance (or
exercise capacity). An incremental cardiopulmonary exercise
test with gas-exchange measurement, to assess peak oxygen
consumption (VO2peak), provides the gold-standard assessment of exercise capacity. VO2peak is inversely correlated with
cardiovascular and all-cause mortality in a broad range of
adult populations.49-53 On this basis, several groups have
started to examine whether the cardiopulmonary exercise
test provides a marker of physiologic aging in the oncology
setting both during and following primary adjuvant therapy.
Cancer Therapy-Induced Changes in Exercise Capacity

No prospective, observational studies have examined longitudinal changes in exercise capacity during and following
adjuvant therapy in patients with solid tumors; nevertheless,
it is possible to glean initial data in the context of randomized
controlled trials of exercise training among patients assigned
to the usual care condition. For example, Courneya et al54
performed an exercise randomized controlled trial among
242 operable patients with breast cancer receiving standard
adjuvant chemotherapy. Among the 82 women assigned to
usual care (i.e., chemotherapy only), exercise capacity as
measured by VO2peak declined approximately 5% from
baseline (following the first cycle of chemotherapy) to following the completion of chemotherapy (median 17 weeks).
Similarly, van Waart et al55 reported that exercise capacity,
e519
as measured by a maximal cycle ergometer test, decreased

approximately 18% from prechemotherapy to immediately
postchemotherapy. Finally, Hornsby et al56 reported that
12 weeks of standard neoadjuvant chemotherapy was associated with a 9.5% decline in VO2peak from prechemotherapy to postchemotherapy. The impairment in
exercise capacity also appears to extend to other tumor sites
receiving other anticancer regimens. For example, Segal
et al57 reported that 6 months of androgen deprivation
therapy with or without radiation therapy was associated
with an approximately 10% decline in VO2peak among men
with advanced prostate cancer. Similarly, West et al58 found
that standard neoadjuvant chemoradiation in patients with
rectal cancer caused a 16% decline in VO2peak. The long-term
clinical importance of this decline is not known; however,
VO2peak typically declines 10% every decade in healthy
women, indicating that short-term chemotherapy may
cause the equivalent of a decade of physiological aging.59
Of importance, the decline in exercise capacity may not
recover, even years following the cessation of primary
therapy. For example, Jones et al59 found that despite
normal resting cardiac function (i.e., left ventricular ejection
fraction $ 50%), VO2peak was, on average, 22% below that of
age-matched sedentary women in 140 patients with earlystage breast cancer a mean of 27 months following the
completion of primary adjuvant therapy. In corroboration,
Khouri et al60 found that VO2peak was, on average, 20% below
that of age-matched sedentary women in 57 patients with
early-stage breast cancer a mean of 26 months following the
completion of primary therapy. The persistent impairment
in exercise capacity also appears to extend beyond operable
breast cancer to other cancer sites. For example, Adams
et al61 performed a study of survivors with Hodgkin disease
(48 patients, mean of 14 years after diagnosis) and found
that VO2peak was significantly reduced in 30% of survivors.
Again, the clinical and prognostic importance of these decrements is currently not known, but because exercise capacity
is a strong independent predictor of both cardiovascular as
well as all-cause mortality in noncancer populations, the
observed impairments are alarming and create a strong rationale for the development and testing of interventions to
prevent and/or treat the observed impairments.
Efficacy of Exercise Training Countermeasures

Aerobic (exercise) training is the most effective therapy to
improve VO2peak in healthy individuals given that it improves
the reserve capacity of all O2 transport organs, which together lead to favorable improvements in exercise capacity,48 although fewer trials have examined the efficacy of
exercise on exercise capacity, as measured by VO2peak, in
patients with cancer, with the vast majority of work to date
in women with early-stage breast cancer. In a recent metaanalysis of six exercise training randomized controlled trials
(involving 571 patients) that assessed the effects of exercise
training in adults with cancer, exercise training led to a
significant improvement in VO2peak (mean weighted
e520
difference = +2.90 mL $ kg21min21; 95% CI, 1.164.64),

compared with nonexercising sedentary control participants.62 However, the data from individual studies are more
heterogeneous. For instance, Courneya et al54 reported that
supervised aerobic training was superior to usual care
(chemotherapy only) for improving VO2peak in 242 patients
with operable breast cancer receiving standard adjuvant
chemotherapy. Interestingly, aerobic training was associated with a nonsignificant improvement in VO2peak (+0.5 mL
$ kg21min21) but completely abrogated the VO2peak decline
observed in the usual-care group. Further work from this
group found that supervised exercise training following
standard (i.e., three times per week, at 25 to 30 minutes/
session), higher volume (i.e., three times per week, at 50 to
60 minutes/session), or combined (i.e., three times per week
of combined aerobic and resistance training) prescriptions
did not mitigate the considerable declines in VO2peak among
301 patients with breast cancer receiving conventional
adjuvant therapy. In contrast, Jones et al63 tested the efficacy of supervised aerobic training consisting of three
sessions per week at 55%100% of VO2peak for 2060 min per
session following a nonlinear prescription in patients with
breast cancer receiving neoadjuvant chemotherapy; specifically, in nonlinear prescriptions, aerobic training sessions
are sequenced in such a fashion that training-induced
physiologic stress is continually altered in terms of intensity and duration in conjunction with appropriate rest
and recovery sessions to optimize cardiovascular adaptation. Attendance and adherence rates to aerobic training
were 82 and 66%, respectively. Intention-to-treat analysis
indicated that VO2peak increased by 2.6 6 3.5 ml/kg/min
(+13.3%) in the chemotherapy plus aerobic training group,
whereas it decreased by 1.5 6 2.2 ml/kg/min (28.6%) in the
chemotherapy only group (between-group difference, p =
.001). In the oncology setting, approximately five additional
studies, both during and following adjuvant therapy, have
examined the safety, tolerability, and preliminary efficacy of
nonlinear aerobic training, compared with a usual care (no
exercise training) control group. Overall, exercise prescriptions adhering to a nonlinear approach appear to be
safe (low adverse event rate), tolerable (mean adherence $
75% of prescribed sessions both during and after primary
adjuvant therapy), and efficacious, conferring favorable
improvements in VO2peak, quality of life, and other physiologic outcomes.64
In summary, the extant evidence indicates that patients
with cancer experience considerable and marked impairments in exercise capacity during cancer therapy that appear
to persist even years following the completion of primary
treatmentsuch decrements are consistent with an
accelerated cardiovascular aging phenotype and may, in
part, contribute to the increased risk of cardiovascular
disease, frailty, and functional dependence in certain cancer
populations. Based on current data, supervised aerobic
exercise training appears to be a safe, tolerable, and efficacious intervention strategy to potentially offset as well as
recover impaired cardiopulmonary function in a broad range
of patients with cancer. The mechanisms, optimal timing,

type, and schedule of exercise training, as well as the longterm clinical implications of declines and/or improvements
in exercise capacity, are a high research priority in geriatric
oncology.
CONCLUSION
An accumulating body of evidence is supporting the hypothesis that cancer and/or cancer treatment is associated
with accelerated aging; however, several gaps in knowledge
remain, and future research is needed to understand the
implications of these findings, as well as ways to decrease
the risk. This unmet need formed the basis for a research
conference of the Cancer and Aging Research Group, National Institute on Aging, and National Cancer Institute, titled Design and Implementation of Intervention Studies to
Improve or Maintain Quality of Survivorship in Older and/or

Frail Adults with Cancer, in which gaps in knowledge and
research priorities to fill these gaps were recommended.
Among the key recommendations was the need to expand
studies focusing on the survivorship issues facing older
adults with cancer, the impact of cancer on the aging process, as well as interventions to decrease the risk. Inclusion
of a geriatric assessment and biomarkers of aging in research
studies will be needed to accomplish these goals. Interventions are needed to halt or modify the accelerated
aging phenotype seen in survivors of cancer. The compelling
data with regard to exercise can serve as a model for future
studies in the years to come.
ACKNOWLEDGMENT
All authors contributed equally.
References
1. Howlader N, Noone AM, Krapcho M, et al. SEER Cancer Statistics Review, 1975-2012, National Cancer Institute. Bethesda, MD, http://seer.
cancer.gov/csr/1975_2012/, based on November 2014 SEER data
submission, posted to the SEER web site, April 2015.
2. Smith BD, Smith GL, Hurria A, et al. Future of cancer incidence in the
United States: burdens upon an aging, changing nation. J Clin Oncol.
2009;27:2758-2765.
3. Parry C, Kent EE, Mariotto AB, et al. Cancer survivors: a booming
population. Cancer Epidemiol Biomarkers Prev. 2011;20:1996-2005.
4. Henderson TO, Ness KK, Cohen HJ. Accelerated aging among cancer
survivors: from pediatrics to geriatrics. Am Soc Clin Oncol Educ Book.
2014;34:e423-e430.
5. Mulrooney DA, Yeazel MW, Kawashima T, et al. Cardiac outcomes in a
cohort of adult survivors of childhood and adolescent cancer: retrospective analysis of the Childhood Cancer Survivor Study cohort. BMJ.
2009;339:b4606.
6. Oeffinger KC, Mertens AC, Sklar CA, et al; Childhood Cancer Survivor
Study. Chronic health conditions in adult survivors of childhood cancer.
N Engl J Med. 2006;355:1572-1582.
7. Mohile SG, Xian Y, Dale W, et al. Association of a cancer diagnosis with
vulnerability and frailty in older Medicare beneficiaries. J Natl Cancer
Inst. 2009;101:1206-1215.
9. Ingram SS, Seo PH, Martell RE, et al. Comprehensive assessment of the
elderly cancer patient: the feasibility of self-report methodology. J Clin
Oncol. 2002;20:770-775.
10. Hurria A, Lichtman SM, Gardes J, et al. Identifying vulnerable older
adults with cancer: integrating geriatric assessment into oncology
practice. J Am Geriatr Soc. 2007;55:1604-1608.
11. McCleary NJ, Wigler D, Berry D, et al. Feasibility of computer-based selfadministered cancer-specific geriatric assessment in older patients with
gastrointestinal malignancy. Oncologist. 2013;18:64-72.
12. Decoster L, Van Puyvelde K, Mohile S, et al. Screening tools for multidimensional health problems warranting a geriatric assessment in
older cancer patients: an update on SIOG recommendations. Ann Oncol.
2015;26:288-300.
13. Reeve BB, Potosky AL, Smith AW, et al. Impact of cancer on healthrelated quality of life of older Americans. J Natl Cancer Inst. 2009;101:
860-868.
14. Weaver KE, Forsythe LP, Reeve BB, et al. Mental and physical healthrelated quality of life among U.S. cancer survivors: population estimates
from the 2010 National Health Interview Survey. Cancer Epidemiol
15. Keating NL, Nrredam M, Landrum MB, et al. Physical and mental health
status of older long-term cancer survivors. J Am Geriatr Soc. 2005;53:
2145-2152.
16. Baker F, Haffer SC, Denniston M. Health-related quality of life of cancer
and noncancer patients in Medicare managed care. Cancer. 2003;97:
674-681.
17. Pinder MC, Duan Z, Goodwin JS, et al. Congestive heart failure in older
women treated with adjuvant anthracycline chemotherapy for breast
18. Lichtman SM, Hurria A, Cirrincione CT, et al; Cancer and Leukemia Group
B. Paclitaxel efficacy and toxicity in older women with metastatic breast
cancer: combined analysis of CALGB 9342 and 9840. Ann Oncol. 2012;
23:632-638.
19. Eastell R, Adams JE, Coleman RE, et al. Effect of anastrozole on bone
mineral density: 5-year results from the anastrozole, tamoxifen, alone
or in combination trial 18233230. J Clin Oncol. 2008;26:1051-1057.
20. Choksi P, Williams M, Clark PM, et al. Skeletal manifestations of
treatment of breast cancer. Curr Osteoporos Rep. 2013;11:319-328.
22. Rockwood K, Mitnitski A. Frailty in relation to the accumulation of
deficits. J Gerontol A Biol Sci Med Sci. 2007;62:722-727.
23. Hubbard JM, Cohen HJ, Muss HB. Incorporating biomarkers into cancer
and aging research. J Clin Oncol. 2014;32:2611-2616.
24. Franceschi C, Campisi J. Chronic inflammation (inflammaging) and its
potential contribution to age-associated diseases. J Gerontol A Biol Sci
Med Sci. 2014;69(Suppl 1):S4-S9.
25. Ferrucci L, Penninx BW, Volpato S, et al. Change in muscle strength
explains accelerated decline of physical function in older women with
high interleukin-6 serum levels. J Am Geriatr Soc. 2002;50:1947-1954.
e521
26. Campisi J, Robert L. Cell senescence: role in aging and age-related

diseases. Interdiscip Top Gerontol. 2014;39:45-61.
27. Chiodi I, Mondello C. Telomere and telomerase stability in human
diseases and cancer. Front Biosci (Landmark Ed). 2016;21:203-224.
28. Jia H, Wang Z. Telomere Length as a Prognostic Factor for Overall Survival
in Colorectal Cancer Patients. Cell Physiol Biochem. 2016;38:122-128.
29. Liu Y, Sanoff HK, Cho H, et al. Expression of p16(INK4a) in peripheral
blood T-cells is a biomarker of human aging. Aging Cell. 2009;8:439-448.
30. Krishnamurthy J, Torrice C, Ramsey MR, et al. Ink4a/Arf expression is a
biomarker of aging. J Clin Invest. 2004;114:1299-1307.
31. Nelson JA, Krishnamurthy J, Menezes P, et al. Expression of p16(INK4a)
as a biomarker of T-cell aging in HIV-infected patients prior to and
during antiretroviral therapy. Aging Cell. 2012;11:916-918.
32. Shang J, Yao Y, Fan X, et al. miR-29c-3p promotes senescence of human
mesenchymal stem cells by targeting CNOT6 through p53-p21 and p16pRB pathways. Biochim Biophys Acta. 2016;1863:520-532.
33. Pustavoitau A, Barodka V, Sharpless NE, et al. Role of senescence marker
p16(INK4a) measured in peripheral blood T-lymphocytes in predicting
length of hospital stay after coronary artery bypass surgery in older
adults. Exp Gerontol. 2016;74:29-36.
34. Kazemi-Bajestani SM, Mazurak VC, Baracos V. Computed tomographydefined muscle and fat wasting are associated with cancer clinical
outcomes. Semin Cell Dev Biol. Epub 2015 Sep 3.
35. Bhatia S, Armenian SH, Armstrong GT, et al. Collaborative Research in
Childhood Cancer Survivorship: The Current Landscape. J Clin Oncol.
2015;33:3055-3064.
36. Armstrong GT, Kawashima T, Leisenring W, et al. Aging and risk of
severe, disabling, life-threatening, and fatal events in the childhood
cancer survivor study. J Clin Oncol. 2014;32:1218-1227.
37. Ness KK, Krull KR, Jones KE, et al. Physiologic frailty as a sign of
accelerated aging among adult survivors of childhood cancer: a report
from the St Jude Lifetime cohort study. J Clin Oncol. 2013;31:4496-4503.
38. Watt DG, Horgan PG, McMillan DC. Routine clinical markers of the
magnitude of the systemic inflammatory response after elective operation: a systematic review. Surgery. 2015;157:362-380.
39. Kim JH, Jenrow KA, Brown SL. Mechanisms of radiation-induced normal
tissue toxicity and implications for future clinical trials. Radiat Oncol J.
2014;32:103-115.
40. Buttiglieri S, Ruella M, Risso A, et al. The aging effect of chemotherapy
on cultured human mesenchymal stem cells. Exp Hematol. 2011;39:
1171-1181.
41. Beeharry N, Broccoli D. Telomere dynamics in response to chemotherapy. Curr Mol Med. 2005;5:187-196.
42. Diker-Cohen T, Uziel O, Szyper-Kravitz M, et al. The effect of chemotherapy on telomere dynamics: clinical results and possible mechanisms. Leuk Lymphoma. 2013;54:2023-2029.
43. Yoon SY, Sung HJ, Park KH, et al. Telomere length shortening of peripheral blood mononuclear cells in solid-cancer patients undergoing
standard-dose chemotherapy might be correlated with good treatment
response and neutropenia severity. Acta Haematol. 2007;118:30-37.
44. Unryn BM, Hao D, Gluck S, et al. Acceleration of telomere loss by
chemotherapy is greater in older patients with locally advanced head
and neck cancer. Clin Cancer Res. 2006;12:6345-6350.
45. Sanoff HK, Deal AM, Krishnamurthy J, et al. Effect of cytotoxic chemotherapy on markers of molecular age in patients with breast cancer.
46. Jones LW, Cohen RR, Mabe SK, et al. Assessment of physical functioning
in recurrent glioma: preliminary comparison of performance status to
functional capacity testing. J Neurooncol. 2009;94:79-85.
47. Abernethy AP, Shelby-James T, Fazekas BS, et al. The Australia-modified
Karnofsky Performance Status (AKPS) scale: a revised scale for
e522
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
contemporary palliative care clinical practice [ISRCTN81117481]. BMC

Palliat Care. 2005;4:7.
Jones LW, Eves ND, Haykowsky M, et al. Exercise intolerance in cancer
and the role of exercise therapy to reverse dysfunction. Lancet Oncol.
2009;10:598-605.
Kavanagh T, Mertens DJ, Hamm LF, et al. Prediction of long-term
prognosis in 12 169 men referred for cardiac rehabilitation. Circulation. 2002;106:666-671.
Kavanagh T, Mertens DJ, Hamm LF, et al. Peak oxygen intake and cardiac
mortality in women referred for cardiac rehabilitation. J Am Coll Cardiol.
2003;42:2139-2143.
Kubozono T, Itoh H, Oikawa K, et al. Peak VO(2) is more potent than Btype natriuretic peptide as a prognostic parameter in cardiac patients.
Circ J. 2008;72:575-581.
Aaronson KD, Mancini DM. Is percentage of predicted maximal exercise
oxygen consumption a better predictor of survival than peak exercise
oxygen consumption for patients with severe heart failure? J Heart Lung
Transplant. 1995;14:981-989.
Jones LW, Watson D, Herndon JE II, et al. Peak oxygen consumption and
long-term all-cause mortality in nonsmall cell lung cancer. Cancer. 2010;
116:4825-4832.
Courneya KS, Segal RJ, Mackey JR, et al. Effects of aerobic and resistance
exercise in breast cancer patients receiving adjuvant chemotherapy:
a multicenter randomized controlled trial. J Clin Oncol. 2007;25:
4396-4404.
van Waart H, Stuiver MM, van Harten WH, et al. Effect of low-intensity
physical activity and moderate- to high-intensity physical exercise
during adjuvant chemotherapy on physical fitness, fatigue, and chemotherapy completion rates: results of the PACES randomized clinical
Hornsby WE, Douglas PS, West MJ, et al. Safety and efficacy of aerobic
training in operable breast cancer patients receiving neoadjuvant
chemotherapy: a phase II randomized trial. Acta Oncol. 2014;53:65-74.
Segal RJ, Reid RD, Courneya KS, et al. Randomized controlled trial of
resistance or aerobic exercise in men receiving radiation therapy for
West MA, Loughney L, Lythgoe D, et al. Effect of prehabilitation on
objectively measured physical fitness after neoadjuvant treatment in
preoperative rectal cancer patients: a blinded interventional pilot study.
Br J Anaesth. 2015;114:244-251.
Jones LW, Courneya KS, Mackey JR, et al. Cardiopulmonary function and
age-related decline across the breast cancer survivorship continuum.
J Clin Oncol. 2012;30:2530-2537.
Khouri MG, Hornsby WE, Risum N, et al. Utility of 3-dimensional
echocardiography, global longitudinal strain, and exercise stress
echocardiography to detect cardiac dysfunction in breast cancer patients treated with doxorubicin-containing adjuvant therapy. Breast
Adams MJ, Lipsitz SR, Colan SD, et al. Cardiovascular status in long-term
survivors of Hodgkins disease treated with chest radiotherapy. J Clin
Oncol. 2004;22:3139-3148.
Jones LW, Liang Y, Pituskin EN, et al. Effect of exercise training on peak
oxygen consumption in patients with cancer: a meta-analysis. Oncologist. 2011;16:112-120.
Jones LW, Fels DR, West M, et al. Modulation of circulating angiogenic
factors and tumor biology by aerobic training in breast cancer patients
receiving neoadjuvant chemotherapy. Cancer Prev Res (Phila). 2013;6:
925-937.
Sasso JP, Eves ND, Christensen JF, et al. A framework for prescription in
exercise-oncology research. J Cachexia Sarcopenia Muscle. 2015;6:
115-124.
New and Improved? Use of White

Blood Cell Growth Factors in
Oncology Practice
CHAIR
Thomas J. Smith, MD, FACP, FASCO, FAAHPM
Baltimore, MD
SPEAKERS
Gary H. Lyman, MD, MPH, FASCO, FACP, FRCP
Seattle, WA
James O. Armitage, MD
University of Nebraska Medical Center
Omaha, NE
SMITH AND HILLNER
Real-World Conundrums and Biases in the Use of White Cell

Growth Factors
Thomas J. Smith, MD, FACP, FASCO, FAAHPM, and Bruce E. Hillner, MD, FASCO
OVERVIEW
We present the 2015 American Society of Clinical Oncology (ASCO) white cell growth factors, or colony-stimulating factor (CSF),
guidelines, updated from 2006. One new indication has been addeddose-intense chemotherapy for bladder cancerto
accompany the existing use for dose-dense breast cancer chemotherapy. Colony-stimulating factors remain appropriate for
any regimen where the risk of febrile neutropenia is about 20% per cycle and dose reduction is not appropriate. Based on new
evidence from multiple trials, CSF use is no longer indicated in treatment of lymphoma unless there are special risk factors. The
United States accounts for 78% of the sales of CSF. The panel approved the use of all biosimilars, but the cost savings will
be small as the price is about 80% of the branded CSFs. More biosimilars at lower cost are awaited. Methods to reduce use
without harm to patients, by requiring justification according to accepted guidelines, are ongoing.
n 2015, the ASCO guideline on the use of CSFs was updated.1

The panel reviewed 66 publications that met eligibility criteria,
including 41 randomized trials.
Since the preceding update in 2006, what has changed? Also,
what is likely to change in the near future? The short answer is
that, unless you treat bladder cancer or lymphoma, not much
has changed, and there are no major practice-changing implications for most oncologists. The Sidebar summarizes the
guideline report. The subsequent commentary is our own
perspective and is not necessarily that of ASCO or the panel.
NEW REASONS AND SITUATIONS WHEN

COLONY-STIMULATING FACTORS ARE
INDICATED
There is now one additional regimen to add to the short list
of indicated combinations for dose-dense chemotherapy,
high-dose methotrexate/vinblastine/doxorubicin/cisplatin
(M-VAC) chemotherapy in urogenital cancers. In a phase
III clinical trial with 7.3 years of follow-up time, the median
and 5-year overall survivals were 15.1 months and 21.8% in
the high-dose M-VAC arm, compared with 14.9 months and
13.5% in the M-VAC arm (hazard ratio for mortality, 0.76).2
As the authors noted, the toxicity was not worse, fewer
people died of their cancers, and high-dose M-VAC
produces a borderline statistically significant relative reduction in the risk of progression and death compared to
M-VAC.2 A similar result of equal efficacy and response
rates with less toxicity (major toxicity rate, 32% vs. 55% for
regular M-VAC) was found with the use of neoadjuvant

M-VAC with CSF support.3,4
There are now two clinical settings in which dose-dense
chemotherapy is indicated instead of usual-dose chemotherapy, breast cancer and uroepithelial cancer. As noted in
the guideline, dose-dense regimens that require a CSF
should only be used within an appropriately designed clinical
trial or if supported by convincing efficacy data.1
For those oncologists who use high-dose chemotherapy
and stem-cell transplantation, CSFs may be used alone, after
chemotherapy, or with plerixafor to help mobilize stem cells.
This has been standard practice for many years.5,6
NEW REASONS AND SITUATIONS WHEN

COLONY-STIMULATING FACTORS ARE NO
LONGER INDICATED
The major change in the 2015 guideline was a recommendation to not routinely use CSFs in the treatment of diffuse
B-cell non-Hodgkin lymphoma in nonelderly patients (younger than age 65 years with no major comorbidities) on the
basis of the lack of compelling benefit for 14-day CHOP
(cyclophosphamide, doxorubicin, vincristine, and prednisolone)
versus 21-day CHOP in two large definitive trials that had
median follow-up times of 5 and 4 years.7,8 The same
conclusion held for a trial in older patients, in which there
was no benefit of 14-day versus 21-day R-CHOP (CHOP +
rituximab).9 This guidance is in line with expert opinion expressed elsewhere and with worldwide common practice.10
From the Harry J. Duffey Family Palliative Care Program of the Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, Baltimore, MD; Hillner Consulting, LLC, Richmond, VA;
Massey Cancer Center, Virginia Commonwealth University, Richmond, VA.
Corresponding author: Thomas J. Smith, MD, Sidney Kimmel Comprehensive Cancer Center, 600 North Wolfe St., Blalock 369, Baltimore, MD 21287; email: tsmit136@jhmi.edu.
e524
ISSUES ON THE USE OF WHITE CELL GROWTH FACTORS
FIGURE 1. Amgen Sales of Colony-Stimulating Factors Compared With World Population
NEW PRODUCTS THAT MAY REDUCE THE AMOUNT

SOCIETY PAYS FOR COLONY-STIMULATING
FACTORS
Filgrastim and peg-filgrastim are expensive. Despite introductions to the market more than 15 years ago, their
prices are actually increasing. As reported in the Amgen
annual report, the 2014 Neulasta sales increased 7 percent
year-over-year for the fourth quarter and 5 percent for the
full year driven mainly by price.11 In 2014, the U.S. sales for
Amgen were $4.488 billion from peg-filgrastim and filgrastim; in contrast, the sales to the rest of the world were
just $1.267 billion (Fig. 1).11
These are just revenues to Amgen and do not represent
what patients and payers actually paid. The United States
represents 4.5% of the world population12 and uses 20% of
the worlds supply of energy13; the United States purchases
KEY POINTS
CSFs are indicated to give dose-intense chemotherapy

for adjuvant breast cancer and neoadjuvant and
metastatic uroepithelial cancers using intensified
M-VAC.
CSFs are indicated when the risk of febrile neutropenia is
about 20% and dose reduction is not an appropriate
strategy; most regimens have a risk of febrile
neutropenia that is less than 20%.
CSFs are not indicated in the routine treatment of
lymphoma with regimens such as R-CHOP.
The United States uses nearly 80% of the worlds supply
of CSFs, far more than any nation, and the price
continues to rise, therefore attempts to reduce use will
increase.
The use of biosimilars for CSFs was strongly endorsed by
the committee and the U.S. Food and Drug
Administration.
78% of the worlds CSF sales. How much of the current U.S.
use is related to income generation from CSF sales compared
with patient demand or physician training may shift and is
worthy of study as large institutional providers and practices
begin switching to accountable care models with global
payments.
What is the potential impact on the U.S. oncology marketplace if we accept (as the ASCO panel and the U.S. Food
and Drug Administration [FDA] have) that biosimilars are
indeed biosimilar? As the guideline stated, pegfilgrastim,
filgrastim, tbo-filgrastim, and filgrastim-sndz (and other
biosimilars, as they become available) can be used for the
prevention of treatment-related febrile neutropenia. The
choice of agent depends on convenience, cost, and the
clinical situation. Tbo-filgrastim has captured approximately 20% of the worldwide market for filgrastim.14 The
manufacturer, Teva, has stated that tbo-filgrastim now
claims 38% of the market share for in-hospital use in the
United States.15 Balugrastim (manufactured by Teva), a
recombinant protein that combines albumin and human
granulocyte CSF (G-CSF), was compared with pegfilgrastim
in a trial with 256 patients who had breast cancer, and
balugrastim had noninferior results, maintained a 1-day
duration of severe neutropenia, and had a slightly shorter
recovery of absolute neutrophil counts in cycle 1.16 However, Teva withdrew its U.S. application for FDA approval for
balugrastim. Teva has released another long-acting biosimilar CSF in Europe, lipegfilgrastim, that will be in direct
competition with pegfilgrastim. Preliminary results showed
equal efficacy.17,18 We could not find any estimates of the
cost impact.
The economic impact of the biosimilars for this indication
has been modest; tbo-filgrastim was priced at approximately
80% of the price of its direct competitor.19 More direct ways
to reduce overall CSF use (specifically, peer-to-peer review
anytime a CSF is ordered) are also successful,20 but most
U.S.insurers and Medicare have not taken that approach
to date. ASCO, in its initial wave of Choosing Wisely topic
e525
SMITH AND HILLNER
SIDEBAR. 2015 Recommendations for the Use of White Blood Cell Growth Factors: ASCO Clinical Practice
Guideline Update*
Primary prophylaxis with a CSF starting with the first cycle and continuing through subsequent cycles of chemotherapy is recommended in patients who have an approximately 20% or higher risk for febrile neutropenia based on patient-, disease- and
treatment-related factors. Primary CSF prophylaxis should also be given in patients receiving dose-dense chemotherapy when
considered appropriate. Consideration should be given to alternative, equally effective and safe chemotherapy regimens not requiring CSF support when available. (Type: evidence-based, benefits outweigh harms. Evidence quality: high. Strength of recommendation: strong.)
Secondary prophylaxis with CSFs is recommended for patients who experienced a neutropenic complication from a prior cycle of
chemotherapy (for which primary prophylaxis was not received), in which a reduced dose or treatment delay may compromise
disease-free or overall survival or treatment outcome. In many clinical situations, dose reduction or delay may be a reasonable
alternative. (Type: evidence-based, benefits outweigh harms. Evidence quality: high. Strength of recommendation: strong.)
CSFs should not be routinely used for patients with neutropenia who are afebrile. (Type: evidence-based, benefits outweigh harms.
evidence quality: high. Strength of recommendation: strong.)
CSFs should not be routinely used as adjunctive treatment with antibiotic therapy for patients with fever and neutropenia. However,
CSFs should be considered in patients with fever and neutropenia who are at high-risk for infection-associated complications, or who
have prognostic factors that are predictive of poor clinical outcomes. (Type: evidence-based, benefits outweigh harms. Evidence
quality: high. Strength of recommendation: strong.)
Dose-dense regimens with CSF support should only be used if supported by convincing efficacy data or within an appropriately
designed clinical trial. Efficacy data support the use of dose-dense chemotherapy in the adjuvant treatment of high-risk breast cancer,
and the use of high-dose intensity methotrexate, vinblastine, doxorubicin, and cisplatin in urothelial cancer. There are limited and
conflicting data on the value of dose-dense regimens with CSF support in non-Hodgkin lymphoma, and it cannot routinely be
recommended at this time. (Type: evidence-based, benefits outweigh harms. Evidence quality: high for breast cancer and lymphoma,
intermediate for urothelial cancer. Strength of recommendation: strong for breast cancer and lymphoma, moderate for urothelial
cancer.)
CSFs may be used alone, after chemotherapy, or in combination with plerixafor to mobilize peripheral-blood progenitor cells.
Choice of mobilization strategy depends in part on type of cancer and type of transplantation. (Type: evidence-based, benefits
outweigh harms. Evidence quality: strong. Strength of recommendation: high.)
CSFs should be administered after autologous stem-cell transplantation to reduce the duration of severe neutropenia. (Type:
evidence-based, benefits outweigh harms. Evidence quality: high; Strength of recommendation: strong.)
CSFs may be administered after allogeneic stem-cell transplantation to reduce the duration of severe neutropenia. (Type: evidencebased. Evidence quality: low. Strength of recommendation: weak).
Prophylactic CSF for patients with diffuse aggressive lymphoma age 65 years and older treated with curative chemotherapy (cyclophosphamide, doxorubicin, vincristine, prednisone and rituximab,) should be considered, particularly in the presence of
comorbidities. (Type: evidence-based, benefits outweigh harms. Evidence quality: intermediate. Strength of recommendation:
moderate.)
The use of CSFs in pediatric patients will almost always be guided by clinical protocols. As in adults, the use of CSFs is reasonable for the
primary prophylaxis of pediatric patients with a high likelihood of febrile neutropenia. Similarly, the use of CSFs for secondary
prophylaxis or for therapy should be limited to high-risk patients. (Type: evidence-based, benefits outweigh harms. Evidence quality:
high. Strength of recommendation: strong.)
For pediatric indications in which dose-intense chemotherapy is known to have a survival benefit, such as Ewing sarcoma, CSFs should
be used to enable the administration of these regimens. (Type: evidence-based, benefits outweigh harms. Evidence quality: high.
Strength of recommendation: strong.)
CSFs should not be used in pediatric patients with nonrelapsed acute lymphoblastic leukemia or nonrelapsed acute myeloid leukemia
who do not have an infection. (Type: informal consensus. Evidence quality: intermediate. Strength of recommendation: moderate.)
Pegfilgrastim, filgrastim, tbo-filgrastim, and filgrastim-sndz (and other biosimilars, as they become available) can be used for the
prevention of treatment-related febrile neutropenia. The choice of agent depends on convenience, cost, and the clinical situation.
There have been no further data comparing granulocyte-CSF (G-CSF) and granulocyte macrophageCSF since the 2006 update;
therefore, there is no change in the recommendation regarding their therapeutic equivalency.(Type: evidence-based, benefits
outweigh harms. Evidence quality: high. Strength of recommendation: strong.)
Current recommendations for the management of patients exposed to lethal doses of total-body radiotherapy, but not doses high
enough to lead to certain death due to injury to other organs, includes the prompt administration of CSF or pegylated G-CSF. (Type:
formal consensus (by others), benefits outweigh harms. Evidence quality: intermediate. Strength of recommendation: moderate.)
*Reproduced from the ASCO guidelines; bold and italics emphasis added by the authors.
e526
ISSUES ON THE USE OF WHITE CELL GROWTH FACTORS
recommendations, has promoted a reduction in use of

CSFs in support of noncurative-intent chemotherapy.21
Overall, there has been no concerted effort by ASCO to
reduce use among its members; we could envision a
concerted social marketing effort that involves patients,
providers, and payers, much like survivorship initiatives.
A recent review noted that physician training, experience, and compensation all appeared to play large roles
in growth factor overuseuse in patients at low risk of

febrile neutropenia.22 Less is known about the more
consequential problem of CSF underusethat is, not
using them in patients at high risk of febrile neutropenia.
At a time when all other drug costs are rising exponentially, it may be tempting to leave U.S. consumption of
CSFs as is, but there are multiple opportunities to improve
value without inducing harm.
References
1. Smith TJ, Bohlke K, Lyman GH, et al. Recommendations for the use of
WBC growth factors: American Society of Clinical Oncology Clinical
Practice Guideline Update. J Clin Oncol. 2015;33:3199-3212.
2. Sternberg CN, de Mulder P, Schornagel JH, et al; EORTC Genito-Urinary
Cancer Group. Seven year update of an EORTC phase III trial of highdose intensity M-VAC chemotherapy and G-CSF versus classic M-VAC in
advanced urothelial tract tumours. Eur J Cancer. 2006;42:50-54.
3. van de Putte EE, Mertens LS, Meijer RP, et al. Neoadjuvant induction
dose-dense MVAC for muscle invasive bladder cancer: efficacy and
safety compared with classic MVAC and gemcitabine/cisplatin. World J
Urol. 2016;34:157-162.
4. Choueiri TK, Jacobus S, Bellmunt J, et al. Neoadjuvant dose-dense
methotrexate, vinblastine, doxorubicin, and cisplatin with pegfilgrastim support in muscle-invasive urothelial cancer: pathologic, radiologic, and biomarker correlates. J Clin Oncol. 2014;32:1889-1894.
5. DiPersio JF, Micallef IN, Stiff PJ, et al; 3101 Investigators. Phase III
prospective randomized double-blind placebo-controlled trial of plerixafor plus granulocyte colony-stimulating factor compared with placebo plus granulocyte colony-stimulating factor for autologous stemcell mobilization and transplantation for patients with non-Hodgkins
lymphoma. J Clin Oncol. 2009;27:4767-4773.
6. DiPersio JF, Stadtmauer EA, Nademanee A, et al; 3102 Investigators.
Plerixafor and G-CSF versus placebo and G-CSF to mobilize hematopoietic stem cells for autologous stem cell transplantation in patients
with multiple myeloma. Blood. 2009;113:5720-5726.
7. Watanabe T, Tobinai K, Shibata T, et al. Phase II/III study of R-CHOP-21
versus R-CHOP-14 for untreated indolent B-cell non-Hodgkins lymphoma: JCOG 0203 trial. J Clin Oncol. 2011;29:3990-3998.
8. Cunningham D, Hawkes EA, Jack A, et al. Rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisolone in patients with
newly diagnosed diffuse large B-cell non-Hodgkin lymphoma: a phase 3
comparison of dose intensification with 14-day versus 21-day cycles.
Lancet. 2013;381:1817-1826.
9. Delarue R, Tilly H, Mounier N, et al. Dose-dense rituximab-CHOP
compared with standard rituximab-CHOP in elderly patients with diffuse large B-cell lymphoma (the LNH03-6B study): a randomised phase 3
10. Bennett CL, Djulbegovic B, Norris LB, et al. Colony-stimulating factors for
febrile neutropenia during cancer therapy. N Engl J Med. 2013;368:
1131-1139.
11. Amgen. Amgens 2014 Revenues Increased 7 Percent To $20.1 Billion

And Adjusted Earnings Per Share (EPS) Increased 14 Percent To
$8.70. http://investors.amgen.com/phoenix.zhtml?c=61656&p=irolnewsArticle&ID=2010978. Accessed February 8, 2016.
12. Worldometers. Current World Population. http://www.worldometers.
info/world-population/#top20. Accessed February 11, 2016, and March
15, 2016.
13. World Population Balance. Population Energy and Consumption. www.
worldpopulationbalance.org/population_energy. Accessed February 11,
2016, and March 15, 2016.
14. Renwick MJ, Smolina K, Gladstone EJ, et al. Postmarket policy considerations for biosimilar oncology drugs. Lancet Oncol. 2016;17:
e31-e38.
15. Trusted to take a bite out of G-CSF acquisition costs. U.S. Pharmacist.
January 2016. p HS 7. http://bt.e-ditionsbyfry.com/publication/?i=287196.
Accessed February 11, 2016, and March 15, 2016.
16. Gladkov O, Moiseyenko V, Bondarenko IN, et al. A Phase III study of
balugrastim versus pegfilgrastim in breast cancer patients receiving
chemotherapy with doxorubicin and docetaxel. Oncologist. 2016;21:
7-15.
17. Ratti M, Tomasello G. Lipegfilgrastim for the prophylaxis and treatment of chemotherapy-induced neutropenia. Expert Rev Clin Pharmacol.
2015;8:15-24.
18. Bondarenko I, Gladkov OA, Elsaesser R, et al. Efficacy and safety of
lipegfilgrastim versus pegfilgrastim: a randomized, multicenter, activecontrol phase 3 trial in patients with breast cancer receiving doxorubicin/
docetaxel chemotherapy. BMC Cancer. 2013;13:386.
19. Royzman I. United States: the Value of Being Highly Similar: First U.S. Biosimilar. http://www.mondaq.com/unitedstates/x/391742/food+drugs+law/
The+Value+of+Being+Highly+Similar+First+US+Biosimilar. Accessed February
12, 2016, and March 15, 2016.
20. Fishman ML, Kumar A, Davis S, et al. Guideline-based peer-to-peer
consultation optimizes pegfilgrastim use with no adverse clinical consequences. J Oncol Pract. 2012;8:e14s-e17s (suppl).
22. Barnes G, Pathak A, Schwartzberg L. G-CSF utilization rate and prescribing patterns in United States: associations between physician and
patient factors and GCSF use. Cancer Med. 2014;3:1477-1484.
e527
GARY H. LYMAN
Issues on the Use of White Blood Cell Growth Factors in

Oncology Practice
Gary H. Lyman, MD, MPH
OVERVIEW
Appropriate use of myeloid growth factors may reduce the risk of neutropenic complications including febrile neutropenia (FN)
in patients receiving cancer chemotherapy. The recently updated American Society of Clinical Oncology (ASCO) Guidelines on
the Use of the White Blood Cell Growth Factors recommends routine prophylaxis with these agents starting in the first cycle
when the risk of FN is 20% or greater. However, the risks for neutropenic complications and the risk of serious adverse
consequences from FN vary considerably with different chemotherapy regimens as well as other disease-, treatment-, and
patient-specific risk factors. Considerably more information is now available on the major risk factors for FN. Multivariable risk
models combining factors look promising but require further validation. Most clinical studies of myeloid growth factor
prophylaxis assessed relative risk (RR) of FN but were not powered to evaluate the effect of prophylaxis on disease-free or
overall survival. Accumulating evidence suggests, however, that the appropriate use of these agents in selected patients may
improve both short-term and long-term survival by reducing the immediate risk of mortality accompanying patients with highrisk disease developing FN as well as improving disease-free and overall survival by enabling the delivery of full dose intensity
chemotherapy and reducing the risk of disease recurrence in patients treated with curative intent. Further studies to evaluate
risk factors and models for FN are needed to guide clinical and shared decision making for the optimal personalized use of these
agents and offer patients at increased risk the best chance of long-term disease control.
yelosuppression including neutropenia represents a

major toxicity associated with systemic cancer chemotherapy and is associated with considerable morbidity,
mortality, and costs.1 Such complications may result in dose
reductions or treatment delays, potentially compromising
disease control and survival outcomes.2-6 Prophylactic myeloid growth factors may be used to reduce the risk of
neutropenic complications and enable safe delivery of
planned chemotherapy dose intensity on schedule when
indicated. Guidelines on the use of myeloid growth factors
were recently updated by ASCO.7 The first question posed to
the panel related to what factors clinicians should consider
when selecting patients for primary prophylaxis of FN with a
colony-stimulating factor (CSF). Several studies demonstrated that the risk of an initial episode of FN is greatest
during the first cycle of treatment when patients are generally receiving full dose intensity.8-10
RISK FACTORS AND MODELS FOR NEUTROPENIA

COMPLICATONS
The risk of FN for patients with cancer who are receiving
systemic chemotherapy is generally based on reported
rates in randomized controlled clinical trials (RCTs).11 A

threshold risk of FN of 20% for routine use of prophylactic
CSFs has been recommended by guidelines from ASCO,
the National Comprehensive Cancer Network (NCCN), and
the European Organisation for the Research and Treatment of Cancer (EORTC) based on results from RCTs of
prophylactic use.7,12-14 However, the risk of chemotherapyinduced neutropenia and its complications are likely underreported in RCTs, with reported rates varying considerably
for common chemotherapy regimens.15,16 Rates of FN observed in observational or real-world patient populations
appear to be consistently greater than those reported
among selected patients accrued to RCTs.17 Additional risk
factors for chemotherapy-induced neutropenia and its
complications have been identified based on patient characteristics, cancer type, and variation in chemotherapy
treatment intensity and intent (cure vs. palliation).18-21 The
updated ASCO CSF guidelines distinguish between risk
factors other than the chemotherapy regimen that increase
the likelihood of FN (Sidebar 1) and those that also increase
the risk of serious medical consequences or death for patients who develop FN (Sidebar 2). Although there is some
From the Hutchinson Institute for Cancer Outcomes Research, Fred Hutchinson Cancer Research Center, and the University of Washington, Seattle, WA.
Corresponding author: Gary H. Lyman, MD, MPH, Hutchinson Institute for Cancer Outcomes Research, Fred Hutchinson Cancer Research Center, 1100 Fairview Ave. North, M3-B232,
P.O. Box 19024, Seattle, WA 98109-1024; email: glyman@fhcrc.org.
e528
RISK FACTORS AND RISK MODELS FOR FEBRILE NEUTROPENIA
SIDEBAR 1. Patient Risk Factors for Febrile

Neutropenia7
In addition to chemotherapy regimen and type of malignancy, consider these factors when estimating a patients
overall risk of febrile neutropenia19,26,34:
Age 65 or older
Advanced disease
Previous chemotherapy or radiation therapy
Pre-existing neutropenia or bone marrow involvement
with tumor
Infection
Open wounds or recent surgery
Poor performance status or poor nutritional status
Poor renal function
Liver dysfunction, most notably elevated bilirubin
Cardiovascular disease
HIV infection
Multiple comorbid conditions
overlap between the individual factors, the latter are important to note even among patients who are considered to
be at low or intermediate risk for developing FN. The NCCN
and EORTC guidelines also identify risk factors for developing
FN for clinicians to consider in addition to the chemotherapy
regimen when estimating a patients individual risk of FN,
supporting the concept of personalized supportive care in
oncology.12,14,22
Because of the number of identified risk factors, uncertainty about their relative importance, and the potential
KEY POINTS
Routine prophylaxis with the white blood cell growth

factors is recommended starting in the first cycle of
cancer chemotherapy when the risk of febrile
neutropenia (FN) is 20% or greater.
The risk of neutropenic complications and associated
infectious complications vary with different
chemotherapy regimens and other disease-, treatment-,
and patient-specific risk factors.
Risk models combining multiple individual risk factors
look promising but require further validation and
improvement.
Appropriate use of the white blood cell growth factors
may improve survival by both reducing the risk of death
associated with FN and improving overall survival by
enhancing delivery of full-dose chemotherapy on
schedule reducing the risk of disease recurrence and
disease-specific mortality.
Further investigation is needed on risk factors and
models for FN to guide clinicians and patients on the
appropriate use of these agents and to provide optimal
treatment to affect long-term disease control and
survival.
SIDEBAR 2. Patient Risk Factors for Poor Clinical

Outcomes From Febrile Neutropenia or
Infection7
Sepsis syndrome
Age 65 or older
Severe neutropenia
Neutropenia expected to be longer than 10 days in
duration
Pneumonia
Invasive fungal infection
Other clinically documented infections
Hospitalization at the time of fever
Prior episode of febrile neutropenia
for intercorrelation between risk factors, there has been

an increasing interest in the development and validation
of multivariable risk models for chemotherapy-associated
neutropenia and its complications. There have been a
number of modeling efforts to predict the risk of serious
complications including death for patients with established
FN.23-25 Although model performance has generally been
limited and has challenged broad utilization of these models,
some institutions and clinicians have used these to identify
patients at low risk for FN for possible outpatient management. Attention has recently shifted to creating risk
models for the development of FN in ambulatory patients
receiving cancer chemotherapy.26,27 Clinical practice guidelines discuss the importance of patient-specific risk factors
and the need for prospective data on the risk of FN in the
unselected general cancer population.7,12
Lyman et al26 reported on a prospective cohort study of
patients with cancer who were treated at oncology practices
throughout the United States. The study was undertaken to
explicitly evaluate the incidence of and risk factors for
neutropenic events among patients with cancer who were
receiving systemic chemotherapy. Consecutive eligible patients with solid tumors or malignant lymphoma initiating a
new chemotherapy regimen were enrolled. Patient demographics and clinical variables included age, sex, ethnicity, employment and educational status, performance
status, body surface area, cancer type, disease stage, history
of prior cancer and treatment, concomitant medications,
baseline hematology and chemistry results, and planned
chemotherapy treatment including drugs, dose, and schedule. Neutropenic events reported included neutropenia
(absolute neutrophil count [ANC] nadir , 1.0 3 103/mm3),
severe neutropenia (ANC nadir , 0.5 3 103/mm3), and FN
(fever/infection and ANC nadir , 1.0 3 103/mm3). The
primary outcome of the study was severe or FN in cycle 1
owing to the dominant risk of events in the first cycle
and their major effect on subsequent risk and treatment
decisions.2,3,10 In the multivariable model based on nearly
2,500 patients, important prognostic factors for neutropenic
complications included a history of previous chemotherapy
as well as baseline leukopenia, hepatic or renal dysfunction,
e529
GARY H. LYMAN
FIGURE 1. Febrile Neutropenia in Patients With HighRisk and Low-Risk Disease Receiving Chemotherapy
potentially life-threatening complications of cancer treatment while allowing for the safe and adequate delivery of
effective chemotherapy dose intensity. Alternatively, the
identification of patients at low risk for early neutropenic
complications may provide reassurance and cost savings in
settings in which more aggressive supportive care is not
warranted. Importantly, among patients receiving chemotherapy who were considered to have an intermediate risk
for FN based on clinical guidelines, approximately onehalf were classified as high risk and one-half were classified
as low risk based on the risk model incorporating other
patient-, disease-, and treatment-related factors. In a subsequent prospective cohort study of nearly 1,000 patients
being treated by 124 community oncologists, a priori
physician-assessed risk of FN as well as the decision to use
prophylactic CSF correlated poorly with the FN risk estimated by the risk model.28
Personalization of supportive care strategies such as the
appropriate and targeted use of myeloid growth factors in a
fashion similar to the personalization of targeted cancer
therapies offers considerable potential for more effective,
safe, and cost-effective cancer care along with improved
survival and quality of life for patients with cancer.22
MYELOID GROWTH FACTORS AND SURVIVAL
(A) KaplanMeier plot displaying the cumulative proportion of patients experiencing

one or more episodes of febrile neutropenia over time in days after chemotherapy
initiation for patients at high risk and patients at low risk based on the risk model. (B)
Hazard function plot displaying the distribution of hazard rates for febrile neutropenia
over time in days after chemotherapy initiation for patients at high risk and patients at
low risk based on the risk model.
Abbreviation: HR, hazard ratio.
Reprinted from Lyman et al.26
delivered chemotherapy relative dose intensity, and the use

of a prophylactic myeloid growth factor. Based on the
median predicted risk of neutropenic events, 34% of patients
classified as high risk experienced events in cycle 1 compared
with 4% among patients at low risk. The cumulative risk of FN
with repeated cycles of chemotherapy for patients at high
and low risk is shown in Fig. 1. KaplanMeier estimates of the
cumulative risk of FN over the first three to four cycles of
chemotherapy were approximately 20% for patients at high
risk compared with 5% for patients at low risk. Additional
external validation of the risk model is currently being
pursued in different settings and patient populations. It is
anticipated that targeted application of a prophylactic
myeloid growth factor starting in cycle 1 among patients
identified as being high risk for early events based on this
model will likely reduce the risk of these serious and
e530
Myeloid growth factors have been developed and approved

for reducing the risk of neutropenic complications including
severe neutropenia and FN. Nevertheless, FN can be a lifethreatening complication of cancer chemotherapy by directly leading to infectious complications and death or
indirectly by prompting a reduction in chemotherapy intensity or duration, resulting in greater risk of disease recurrence and cancer-related mortality (Fig. 2). In a recent
analysis of patients with cancer receiving systemic chemotherapy who were included in a large U.S. health claims
database, Lyman et al29 estimated a 35% increase in early
FIGURE 2. Dual Impact of Neutropenia on Risk of

Infection and Chemotherapy Dose Intensity
Schematic diagram of the dual effects of neutropenic complications including

febrile neutropenia on both early and overall survival based on reductions in the
risk of febrile neutropenia and early infectious complications while enabling the
delivery of full-dose chemotherapy on schedule and potentially reducing the risk
of disease recurrence and cancer-associated mortality.
RISK FACTORS AND RISK MODELS FOR FEBRILE NEUTROPENIA
mortality and a 15% increased risk of all-cause mortality

overall among patients experiencing FN. Although a favorable
effect of myeloid growth factors on survival of patients with
cancer is therefore plausible, data directly addressing this
issue are limited. Most pivotal clinical trials of myeloid growth
factors have been designed to evaluate their effect on severe
neutropenia or FN during the period of chemotherapy delivery, with limited follow-up beyond the treatment period.
Therefore, none of these trials were individually powered to
address the issue of disease-free or overall survival, and only
early mortality was reported. In a systematic review and
meta-analysis of the 17 RCTs of primary prophylaxis with
granulocyte-CSF (G-CSF) including about 3,500 patients with
solid tumors or lymphoma, a significant reduction in the risk of
FN as seen in each individual trial was confirmed (RR, 0.54;
95% CI, 0.43-0.67; p , .001).30 Importantly, in this pooled
analysis, reductions in the risk of infection-related mortality
(RR 0.55; 95% CI, 0.330.90; p = .018) as well as early all-cause
mortality (RR, 0.60; 95% CI, 0.430.83; p = .002) were observed. Although a fourfold increase in reported bone pain
was noted, average delivered chemotherapy relative dose
intensity was significantly greater for patients who were
randomly assigned to receive prophylactic G-CSF.
Concern has been raised about the potential for an increased risk for developing acute myeloid leukemia (AML) or
myelodysplastic syndrome (MDS) among patients treated
with myeloid growth factors. In an analysis of data from the
Surveillance, Epidemiology, and End ResultsMedicare linked
files, Hershman et al31 observed a doubling of the risk of
subsequent AML or MDS, from 0.7% to 1.8% with CSF support
among older women diagnosed with early-stage breast
cancer from 1991 to 1999 who received myeloid growth
factors with their adjuvant chemotherapy. A more recent
systematic review of RCTs among patients receiving systemic
chemotherapy and randomly assigned to receive G-CSF
support or not and reporting on the occurrence of AML/
MDS or all second malignancies and a minimum follow-up of
24 months or more identified 25 eligible trials involving more
than 12,000 patients.32 With an average follow-up of
60 months in these RCTs, AML, or MDS was observed in 0.36%
of controls and 0.79% of patients treated with G-CSF, with a
RR of 1.92 (95% CI, 1.193.07; p = .007) and an absolute risk
difference (ARD) of 0.41% (95% CI, 0.10%0.72%; p = .009). At
the same time, the RR for all-cause mortality with and without
G-CSF support was 0.897 (95% CI, 0.8570.938; p , .001) with
an ARD of 3.40% (95% CI, 2.00%4.80%; p , .001),
representing a nearly 10-fold greater reduction in mortality
than the estimated increase in AML/MDS. Finally, among the
16 RCTs reporting to have delivered chemotherapy intensity
by treatment arm, significant associations were observed
between reduced mortality and increases in both relative
(p = .0148) and absolute (p = .0266) delivered dose intensity
in patients treated with G-CSF compared with controls. These
results were subsequently extended in a systematic review of
RCTs of patients receiving cancer chemotherapy and randomly assigned to G-CSF or control and reporting all-cause
mortality with a minimum follow-up of 24 months but removing the requirement for reporting of AML or MDS.33
Among about 61 eligible RCTs involving nearly 10,000 patients, the RR for all-cause mortality observed was 0.93 (95%
CI, 0.900.96; p , .001). Greater reductions in mortality were
observed among trials with longer follow-up, when treatment
was clearly for curative intent, and when survival was the
primary study outcome. In subgroup analyses by study design,
greater reductions in all-cause mortality were observed when
G-CSF support enabled chemotherapy treatment with a dosedense schedule, greater treatment intensity, or the addition
of one or more additional myelosuppressive agents than
when both study arms were intended to receive the same
chemotherapy drugs, dose, and schedule.
Although the potential effect of myeloid growth factors on
overall survival has not been completely evaluated, the
available evidence suggests no deleterious effects of CSF
support of cancer chemotherapy and raises the potential
for a favorable effect on both early- and overall all-cause
mortality, most notably for patients treated with curative
intent and when CSF support enables treatment enhancement for patients with responsive and potentially curable
malignancies. Although an increased risk of AML or MDS
remains, available studies cannot clearly differentiate between the direct effects of the myeloid growth factor and
the enabling effects of delivering more intensive treatment
with known leukemogenic chemotherapeutic agents. In
total, the apparent reduction in all-cause mortality appears
to overshadow the potential small increased risk associated
with AML or MDS in this setting.
References
1. Kuderer NM, Dale DC, CrawfordJ, et al. Mortality, morbidity, and cost associated
with febrile neutropenia in adult cancer patients. Cancer. 2006;106:2258-2266.
2. Lyman GH, Dale DC, Crawford J. Incidence and predictors of low doseintensity in adjuvant breast cancer chemotherapy: a nationwide study
of community practices. J Clin Oncol. 2003;21:4524-4531.
3. Lyman GH, Dale DC, Friedberg J, et al. Incidence and predictors of low
chemotherapy dose-intensity in aggressive non-Hodgkins lymphoma:
a nationwide study. J Clin Oncol. 2004;22:4302-4311.
4. Lyman GH. Impact of chemotherapy dose intensity on cancer patient
outcomes. J Natl Compr Canc Netw. 2009;7:99-108.
5. Shayne M, Culakova E, Wolff D, et al. Dose intensity and hematologic

toxicity in older breast cancer patients receiving systemic chemotherapy. Cancer. 2009;115:5319-5328.
6. Shayne M, Culakova E, Poniewierski MS, et al. Dose intensity and
hematologic toxicity in older cancer patients receiving systemic chemotherapy. Cancer. 2007;110:1611-1620.
7. Smith TJ, Bohlke K, Lyman GH, et al; American Society of Clinical Oncology. Recommendations for the use of WBC growth factors: American
Society of Clinical Oncology clinical practice guideline update. J Clin
Oncol. 2015;33:3199-3212.
e531
GARY H. LYMAN
8. Lyman GH, Morrison VA, Dale DC, et al; OPPS Working Group; ANC
Study Group. Risk of febrile neutropenia among patients with
intermediate-grade non-Hodgkins lymphoma receiving CHOP chemotherapy. Leuk Lymphoma. 2003;44:2069-2076.
9. Lyman GH, Delgado DJ. Risk and timing of hospitalization for febrile
neutropenia in patients receiving CHOP, CHOP-R, or CNOP chemotherapy for intermediate-grade non-Hodgkin lymphoma. Cancer. 2003;
98:2402-2409.
10. Crawford J, Dale DC, Kuderer NM, et al. Risk and timing of neutropenic
events in adult cancer patients receiving chemotherapy: the results of a
prospective nationwide study of oncology practice. J Natl Compr Canc
Netw. 2008;6:109-118.
11. Lyman GH. Risk assessment in oncology clinical practice. From risk factors
to risk models. Oncology (Williston Park). 2003; 17(Suppl 11):8-13.
12. National Comprehensive Cancer Network (ed). NCCN Clinical Practice
Guidelines in Oncology: Myeloid Growth Factors. Fort Washington, PA:
National Comprehensive Cancer Network Inc; 2015.
14. Aapro MS, Bohlius J, Cameron DA, et al; European Organisation for
Research and Treatment of Cancer. 2010 update of EORTC guidelines
for the use of granulocyte-colony stimulating factor to reduce the
incidence of chemotherapy-induced febrile neutropenia in adult patients with lymphoproliferative disorders and solid tumours. Eur J
Cancer. 2011;47:8-32.
15. Dale DC, McCarter GC, Crawford J, et al. Myelotoxicity and dose intensity of chemotherapy: reporting practices from randomized clinical
trials. J Natl Compr Canc Netw. 2003;1:440-454.
16. Kuderer NM, Wolff AC. Enhancing therapeutic decision making when
options abound: toxicities matter. J Clin Oncol. 2014;32:1990-1993.
17. Truong J, Lee EK, Trudeau ME, et al. Interpreting febrile neutropenia
rates from randomized, controlled trials for consideration of primary
prophylaxis in the real world: a systematic review and meta-analysis.
Ann Oncol. Epub 2015 Dec 27.
18. Laskey RA, Poniewierski MS, Lopez MA, et al. Predictors of severe and
febrile neutropenia during primary chemotherapy for ovarian cancer.
Gynecol Oncol. 2012;125:625-630.
19. Lyman GH, Abella E, Pettengell R. Risk factors for febrile neutropenia
among patients with cancer receiving chemotherapy: a systematic
review. Crit Rev Oncol Hematol. 2014;90:190-199.
20. Weycker D, Li X, Barron R, et al. Importance of risk factors for febrile
neutropenia among patients receiving chemotherapy regimens not
classified as high-risk in guidelines for myeloid growth factor use. J Natl
Compr Canc Netw. 2015;13:979-986.
21. Weycker D, Li X, Edelsberg J, et al. Risk of febrile neutropenia in patients
receiving emerging chemotherapy regimens. Support Care Cancer.
2014;22:3275-3285.
e532
22. Kuderer NM, Lyman GH. Personalized medicine and cancer supportive
care: appropriate use of colony-stimulating factor support of chemotherapy. J Natl Cancer Inst. 2011;103:910-913.
23. Klastersky J, Paesmans M. The Multinational Association for Supportive
Care in Cancer (MASCC) risk index score: 10 years of use for identifying
low-risk febrile neutropenic cancer patients. Support Care Cancer. 2013;
21:1487-1495.
24. Talcott JA, Siegel RD, Finberg R, et al. Risk assessment in cancer patients
with fever and neutropenia: a prospective, two-center validation of a
prediction rule. J Clin Oncol. 1992;10:316-322.
25. Carmona-Bayonas A, Jimenez-Fonseca P, Virizuela Echaburu J, et al.
Prediction of serious complications in patients with seemingly stable
febrile neutropenia: validation of the Clinical Index of Stable Febrile
Neutropenia in a prospective cohort of patients from the FINITE study.
J Clin Oncol. 2015;33:465-471.
26. Lyman GH, Kuderer NM, Crawford J, et al. Predicting individual risk of
neutropenic complications in patients receiving cancer chemotherapy.
Cancer. 2011;117:1917-1927.
27. Jenkins P, Scaife J, Freeman S. Validation of a predictive model that
identifies patients at high risk of developing febrile neutropaenia following chemotherapy for breast cancer. Ann Oncol. 2012;23:
1766-1771.
28. Lyman GH, Dale DC, Legg JC, et al. Assessing patients risk of febrile
neutropenia: is there a correlation between physician-assessed risk and
model-predicted risk? Cancer Med. 2015;4:1153-1160.
29. Lyman GH, Michels SL, Reynolds MW, et al. Risk of mortality in patients
with cancer who experience febrile neutropenia. Cancer. 2010;116:
5555-5563.
30. Kuderer NM, Dale DC, Crawford J, et al. Impact of primary prophylaxis
with granulocyte colony-stimulating factor on febrile neutropenia and
mortality in adult cancer patients receiving chemotherapy: a systematic
review. J Clin Oncol. 2007;25:3158-3167.
31. Hershman D, Neugut AI, Jacobson JS, et al. Acute myeloid leukemia or
myelodysplastic syndrome following use of granulocyte colonystimulating factors during breast cancer adjuvant chemotherapy.
32. Lyman GH, Dale DC, Wolff DA, et al. Acute myeloid leukemia or
myelodysplastic syndrome in randomized controlled clinical trials of
cancer chemotherapy with granulocyte colony-stimulating factor:
a systematic review. J Clin Oncol. 2010;28:2914-2924.
33. Lyman GH, Dale DC, Culakova E, et al. The impact of the granulocyte
colony-stimulating factor on chemotherapy dose intensity and cancer
survival: a systematic review and meta-analysis of randomized controlled trials. Ann Oncol. 2013;24:2475-2484.
34. Hosmer W, Malin J, Wong M. Development and validation of a prediction model for the risk of developing febrile neutropenia in the first
cycle of chemotherapy among elderly patients with breast, lung, colorectal, and prostate cancer. Support Care Cancer. 2011;19:333-341.
Optimizing Palliative and

End-of-Life Care: Evidence-Based
Practice Improvement
CHAIR
Thomas W. LeBlanc, MD, MA
Durham, NC
SPEAKERS
Jennifer S. Temel, MD
Boston, MA
Timothy D. Gilligan, MD, MSc
Cleveland Clinic
Cleveland, OH
NICKOLICH ET AL
Discussing the Evidence for Upstream Palliative Care in

Improving Outcomes in Advanced Cancer
Myles S. Nickolich, MD, Areej El-Jawahri, MD, Jennifer S. Temel, MD, and Thomas W. LeBlanc, MD, MA
OVERVIEW
Palliative care has received increasing attention at the American Society of Clinical Oncology (ASCO) Annual Meeting since
the publication of its provisional clinical opinion on the topic in 2012. Despite frequent discussion, palliative care remains a
source of some controversy and confusion in clinical practice, especially concerning who should provide it, what it encompasses, and when and how it can help patients and their families. In this article, we provide a formal definition of
palliative care and review the state of the science of palliative care in oncology. Several randomized controlled trials now
show that palliative care improves important outcomes for patients with cancer. Related outcome improvements include a
reduction in symptoms, improved quality of life, better prognostic understanding, less depressed mood, less aggressive endof-life care, reduced resource utilization, and even prolonged survival. As such, ASCO recommends early integration of
palliative care into comprehensive cancer care for all patients with advanced disease and/or significant symptom burden.
Our aim is that this summary will facilitate greater understanding about palliative care and encourage further integration of
palliative care services into cancer care. More research is needed to illuminate the mechanisms of action of palliative care
and to improve the specificity of palliative care applications to unique scenarios and populations in oncology.
alliative care has received increasing attention at the

ASCO Annual Meeting since the publication of its
Provisional Clinical Opinion: The Integration of Palliative
Care Into Standard Oncology Care in 2012.1 In that paper,
ASCO recommended early concurrent palliative care as part
of standard cancer care for all patients with metastatic
disease and/or significant symptom burden. This recommendation was based on a growing amount of evidence
about the many benefits of adding specialist palliative care
to standard oncology care, including data from several
large randomized controlled trials.
Despite frequent discussion at ASCO Annual Meetings
since then, palliative care remains a source of some controversy and confusion in clinical practice, especially concerning who should provide it, what it encompasses, and
when and how it can help patients and their families.
WHAT IS PALLIATIVE CARE?

The term palliative care is sometimes misunderstood as
being synonymous with end-of-life care. To address this
misperception, the Center to Advance Palliative Care (CAPC)
conducted a public opinion polling study in 2011, to develop a new definition of palliative care. According to this
definition:
Palliative care is specialized medical care for people with

serious illness. It focuses on providing patients with relief
from the symptoms, pain, and stress of a serious illness
whatever the diagnosis. The goal is to improve quality of life
for both the patient and the family. It is appropriate at any
age and at any stage of a serious illness. It can be provided at
the same time as disease treatment to help people live as
well as possible while facing illness.2
Although other organizations, such as the National Comprehensive Cancer Network and the World Health Organization, have also posed definitions of palliative care, we prefer
the clarity and inclusivity of the CAPC definition. Palliative care
seeks to improve quality of life by providing additional support to the patient and caregiver(s), to assist with symptom
burden, psychosocial needs, spiritual well being, communication, and understanding of prognosis, treatment decision
making, transitions of care, existential issues, and end-of-life
scenarios. Although palliative care is appropriate for many
patients with serious illness, this article focuses on patients
with cancer.
There are different levels of palliative care, each of which is
provided by different members of the care team. Oncologists often provide basic palliative care, including pain and
other symptom management. This type of palliative care is
referred to as primary palliative care, generalist palliative
From Duke University Hospital, Durham, NC; Massachusetts General Hospital, Boston, MA; Duke Cancer Institute, Durham, NC.
Corresponding author: Thomas W. LeBlanc, MD, MA, Duke University Medical Center, P.O. Box 2715, Durham, NC 27710; email: thomas.leblanc@duke.edu.
e534
WHATS THE EVIDENCE BASE FOR PALLIATIVE CARE IN ONCOLOGY?
care,3 or sometimes just supportive care.4 Although it is entirely

appropriate that the oncology care team provides primary
palliative care, this is not always sufficient to meet the needs of
patients and families. In cases of complex and/or unmet needs,
patients may benefit from a higher level of palliative care, often
called specialty palliative care or secondary palliative care. This
type of palliative care is provided by specialists with formal
training in palliative care and related specialties. Specialty
palliative care is akin to the expert care a cardiologist provides
to a patient with severe heart failure, even though the patients primary care physician may also be monitoring the
cardiac status of the patient.
This distinction is critical because some oncologists misunderstand palliative care to suggest that they should pass off
their primary palliative care responsibilities to someone else.
Rather, when we talk about specialty palliative care, we are
referring to care that is provided in addition to the primary
palliative care already being provided by the oncologist. This
type of care is additive, not duplicative, and should not be
misconstrued as any sort of transferal of responsibility. Indeed, data from randomized controlled trials suggest that that
the focus of palliative care specialists is quite different from
that of oncologists.5 Together, oncologists and palliative care
specialists working collaboratively can provide more comprehensive care for patients with cancer and their families.
WHICH OUTCOMES IMPROVE WITH UPSTREAM

PALLIATIVE CARE?
Symptom Management
In a large cluster-randomized controlled trial by Zimmermann
et al6 with patients with advanced solid tumors (lung, gastrointestinal, genitourinary, breast, and gynecologic), patients
who were randomly selected to receive early palliative care
had a significant reduction in symptom intensity at 4 months
by the Edmonton Symptom Assessment Scale, compared
with those randomly assigned to receive standard care

(21.34 vs. +3.23; p = .05).6
Quality of Life
Early palliative care also improves quality of life for patients
with advanced cancer. In the ENABLE II study by Bakitas
et al,7 patients with advanced gastrointestinal, breast, lung,
or genitourinary cancer were randomly assigned to receive a
nurse-led palliative care intervention or standard care within
8 to 12 weeks of diagnosis. Patients who were randomly
assigned to receive early palliative care had better quality of
life per the Functional Assessment of Cancer Therapy
Palliative Care scale (mean difference of 4.6; p = .02).7
Similarly, in a large randomized controlled trial by Temel
et al,8 patients with advanced nonsmall cell lung cancer
(NSCLC) who were randomly assigned to receive early
outpatient palliative care within 8 weeks of diagnosis had a
significant improvement in quality of life, per the trial
outcome index subscale of the Functional Assessment of
Cancer TherapyLung scale at 12 weeks, compared with
those assigned to usual care (59.0 vs. 53.0; p = .009).8 This
finding was also confirmed in the trial by Zimmermann
et al6 where patients assigned to receive the palliative care
intervention per the Quality of Life at the End of Life scale
experienced improvements seen at 3 months, compared
with those receiving standard care (2.33 vs. 0.06; p = .05).6
Early palliative care also improves quality of life for patients in the emergency department setting.9 In a study by
Grudzen et al,9 patients with advanced cancer who presented to the emergency department at a quaternary care
center were randomly assigned to receive a palliative care
consult from the inpatient team and subsequent followup visits in the outpatient palliative care clinic compared
with usual care. Patients who were randomly assigned
to palliative care experienced greater improvements in
quality of life scores by the Functional Assessment of
Cancer TherapyGeneral (FACT-G) scale at 12 weeks
(5.91 vs. 1.08; p = .03).9
KEY POINTS
Palliative care improves important outcomes for

patients with cancer, as shown in at least five
randomized controlled trials to date.
ASCO recommends early integration of palliative care
into comprehensive cancer care for all patients with
advanced disease and/or significant symptom burden.
Outcome improvements seen with early palliative care
include a reduction in symptoms, improved quality of
life, better prognostic understanding, less depressed
mood, improved end-of-life outcomes, reduced
resource utilization, and even prolonged survival.
More research is needed to illuminate the mechanisms
of action of palliative care and to improve the specificity
of palliative care applications to unique scenarios and
populations, such as patients with hematologic
malignancies, caregivers, and across various tumor
types and treatments.
Prognostic Understanding
Unfortunately, many patients with advanced, incurable
cancer misunderstand their prognoses and the goals of their
cancer treatments. In a study by Weeks et al,10 patients with
stage IV lung or colorectal cancer were asked to identify the
goal of their prescribed chemotherapies. Sixty-nine percent
of patients with lung cancer and 81% of those with colorectal
cancer mistakenly thought that their palliative chemotherapy
regimens were prescribed with intent to cure.10 Interestingly,
early palliative care mitigates this problem. In the randomized
controlled trial by Temel et al,11 patients with advanced
NSCLC who were randomly assigned to receive early palliative
care were more likely to retain or develop an accurate understanding of their prognoses compared with patients who
received standard care (82.5% vs. 59.6%; p = .02).11 Of note,
patients in the early palliative care arm who reported an
accurate understanding of prognosis were also less likely to
e535
NICKOLICH ET AL
receive chemotherapy near the end of life (9.4% vs. 50%;

p = .02), an indicator of higher-quality end-of-life care
among patients with advanced cancer.
Caregiver Outcomes
In addition to direct patient benefits of early palliative care,
caregivers also appear to benefit from the early introduction
of palliative care. In the ENABLE III study by Bakitas et al,7 the
caregivers of patients assigned to early palliative care received
an additional caregiver-focused intervention. These caregivers reported better depression scores at 3 months, per
the Center for Epidemiologic StudyDepression scale (mean
difference, 23.4; p = .02). Furthermore, caregivers whose
loved ones died during the study also had reduced stress
burden, as measured on the Montgomery-Borgatta Caregiver
Burden stress subscale.12 The inclusion of a caregiverspecific intervention in this study is unique compared with
other randomized studies of early palliative care done to
date, and results suggest that it is an important area for
future research and intervention development.
Mood
Mood disorders are major contributors to morbidity in
patients with advanced cancer. Palliative care clinicians are
more likely than oncologists to focus on coping skills and
psychosocial concerns, which suggests that a role for early
palliative care is to improve mood among patients with
advanced cancer.5 Indeed, decreases in median survival
have been observed in patients with advanced NSCLC and
comorbid mood disturbance, with median survival times of
only 5.4 months for those with major depression compared
with 10 months for those without (p = .001).13 In the randomized controlled trial by Temel et al,8 patients with advanced NSCLC who received early palliative care had lower
rates of depression than patients who received usual care,
as measured by both the Hospital Anxiety and Depression
Scale (16% vs. 38%; p = .01) and the Patient Health
Questionnaire9 (4% vs. 17%; p = .04). Interestingly, 42.9%
of patients who received early palliative care also showed an
improved depression response compared with 0% of patients in the usual-care control arm at 12 weeks. Of note,
rates of new antidepressant prescriptions and mental health
visits did not differ significantly between the two groups,
which suggests that early specialist palliative care indeed
affects mood.8 Similarly, in the ENABLE II study by Bakitas
et al,7 patients who received early palliative care had
less depressed mood by Center for Epidemiologic Study
Depression scale (mean difference, 21.8; p = .02).7
End-of-Life Outcomes
In the randomized controlled trial by Temel et al,8 patients
who received early palliative care also received less aggressive end-of-life care (palliative care vs. control, 33% vs. 54%;
p = .05). Here, aggressive end-of-life care was a composite
outcome defined by the presence of at least one of the
following criteria: receipt of chemotherapy within 14 days
e536
before death, no hospice care, or admission to hospice 3 or

fewer days before death. In addition, more patients in the
palliative care arm had documented resuscitation preferences in the outpatient setting (53% vs. 28%; p = .05).8
The use of intensive care unit (ICU) services for patients
with advanced cancer is another useful marker of aggressive
end-of-life care, and palliative care appears to improve this
outcome as well. For example, in a retrospective study in the
Veterans Affairs system, patients who were observed by an
inpatient palliative care team had a 42% lower ICU admission
rate than control patients.14 However, in the study by
Grudzen et al,9 emergency departmentinitiated palliative care consultation (with outpatient follow-up thereafter) did not yield any reduction in ICU admissions.9
In the ENABLE II study, there were no differences in
number of days hospitalized, days in the ICU, or frequency of
emergency department visits.7 Similarly, ENABLE III found no
differences in relative rates of days hospitalized, days admitted to an ICU, emergency department visits, chemotherapy within the last 14 days of life, or death at home.15
However, both groups in this study received palliative care,
so it may be that late palliative care improves these outcomes similarly to earlier palliative care. Because there was
not a control arm of patients who received no palliative care
in this study, a direct measure of the impact of palliative care
on this outcome is not possible.
Survival
Beyond improvements in symptom burden and quality of
life, early palliative care is also associated with improved
survival. In the randomized controlled trial by Temel et al,8
patients with metastatic NSCLC who received early palliative
care lived nearly 3 months longer than patients who received
standard care (median, 11.6 vs. 8.9 months; p = .02). This
difference was despite less aggressive end-of-life care.8
Similarly, in the ENABLE III study, patients who received early
palliative care had significantly longer 1-year survival rates
than those who received delayed palliative care; the difference was 15% at 1 year (63% vs. 48%; p = .038).15
Resource Utilization
Palliative care also appears to improve resource utilization
among patients with advanced cancer. However, these data
come from studies of hospitalized patients who received
inpatient palliative care consultations. For example, in
retrospective a study by May et al,16 inpatient palliative care
consultation was associated with significant cost savings
among patients with advanced cancer (p # .01), particularly
for those with more comorbidities. Overall, palliative care
consultation yielded a 32% reduction in hospital costs when
initiated within 2 days of admission for patients with cancer
and multiple comorbidities.16 A prospective cohort study, also
by May et al,17 demonstrated that cost savings may vary on
the basis of the timing of consultation. In this study, earlier
palliative care consultation for hospitalized patients with
advanced cancer resulted in more cost reduction than did later
WHATS THE EVIDENCE BASE FOR PALLIATIVE CARE IN ONCOLOGY?
consultation, with mean reductions in costs of 24% (day 2) and

14% (day 6). Inpatient palliative care consultation within at least
6 days of admission had an estimated mean treatment effect
of 2$1,312 (p = .04) compared with usual care, and, within
2 days of admission, palliative care consultation had a more
pronounced cost reduction (2$2,280; p = .002).17 Within the
Veterans Affairs setting, inpatient palliative care consultation
was associated with similar cost reductions in retrospective
analyses. Unadjusted total direct per day costs for patients
who received a palliative care consultation were $891
versus $1,287.60 for usual care (p , .0001).14
HOW DOES PALLIATIVE CARE IMPROVE THESE

OUTCOMES?
There has been much discussion about the mechanism of
action of early palliative care in patients with advanced
cancer. Part of the difficulty in understanding the mechanism stems from the fact that the interventions studied thus
far have varied significantly. Some, such as ENABLE III, were
largely telephone based and provided by nurses,15 whereas
others were based in the outpatient clinic and provided by
specialist palliative care clinicians.6,8 Additional challenges
stem from the heterogeneity of populations included in
studies to date. Although palliative care mechanisms of
action are ultimately unknown, clues from published studies
point toward key ingredients in the interventions. For
example, a qualitative analysis of documentation from one
early palliative care study demonstrated that palliative care
clinicians spend much of their time on three primary areas:
(1) managing symptoms, (2) facilitating coping, and (3)
serving as a bridge between the patient and oncologist.18
These elements map relatively nicely to the associated
outcome improvements in symptoms and quality of life,
mood, and prognostic understanding that are seen with
early palliative care, and these areas are different from the
usual focus of oncologists in the clinic. However, more research is needed to better understand the most important
components of early palliative care interventions, which will
thereby allow for more streamlined implementation and
dissemination across cancer types and practices.
FUTURE DIRECTIONS
Although specialist palliative care is clearly helpful for patients with advanced solid tumors, it is likely that there are
differences in needs across tumor types and stages. Furthermore, differences in patient characteristics may herald
different palliative care needs. For example, recent evidence
suggests that early palliative care in advanced lung cancer
may be more effective among younger than older patients.19
Differential efficacy may be true across other patient types
and diseases as well. However, although it seems obvious
that patients with heart failure likely have different needs
than patients with advanced lung cancer, it is less clear whether
such differences exist among patients with gastrointestinal
cancers versus lung cancers, for example. More research is
needed in this area.
The hematologic malignancies population is a particularly

unique area for additional exploration. A growing amount of
literature demonstrates sizeable gaps in end-of-life quality
measures and unmet palliative care needs among patients
with hematologic malignancies. For example, patients with
hematologic malignancies are more likely than patients with
solid tumors to be admitted to hospice services within the last
3 days of life.20 Similarly, patients with hematologic malignancies are more likely than patients with solid tumors to
receive inpatient or nursing facilitybased hospice care at the
end of life.21 Unique challenges exist to providing palliative
care to patients with hematologic malignancies because of
several complex factors, including differences in prognoses,
disease trajectories, treatment types and intensities, and
treating clinicians.22 Additional study is needed in the area of
hematologic malignancies and bone marrow transplantation,
and novel interventions to address the unique needs and
issues of these populations also are needed.
Although more research is needed to better understand
how palliative care improves outcomes and how to integrate
these services in different cancer populations, there is
simultaneously a workforce crunch in the field. The palliative
care workforce is unfortunately not large enough to meet
the needs of all patients with cancer. To date, there are
greater than 7,000 board-certified palliative care specialists
in the United States, and there are greater than 100 fellowship training programs. However, projections suggest
that this structure is inadequate to meet the needs of the
U.S. population.23,24 Furthermore, additional education is
needed in the provision of primary palliative care by oncologists and others.25-27 Simply put, specialist palliative
care is a scarce resource. As we develop a better understanding of how palliative care improves outcomes, we
can better apply this scarce resource judiciously.
Caregivers represent another key area of unmet need. To
date, there is little caregiver-focused work in the palliative care
literature, yet there are many reasons to think that caregivers
have unique needs that require novel solutions. In other
words, it is likely that specific interventions are needed to
support these important and often overlooked members of a
patients support system. The ENABLE III study is a wonderful
example of the promise for caregiver-specific interventions,12
however, much more research is needed in this area.28
We must also develop innovative delivery models and
reimbursement structures to facilitate early palliative care
for patients with cancer. Constraints within current models
for providing palliative care through hospice services often
creates an either/or mentality, wherein patients are required to forego cancer therapies to receive these services.
This is particularly problematic in hematologic malignancies,
for which beneficial palliative treatments, such as blood
product transfusions and antimicrobials, are often not
feasible within the hospice care setting.29-31 However, amid
the recent growth of targeted biologic therapies for advanced
solid tumors, this problem is increasing among patients with
solid tumors as well. For example, we know that EGFR inhibitors may palliate bothersome symptoms and improve
e537
NICKOLICH ET AL
quality of life for patients with advanced lung cancer, yet

these are not generally reimbursed under hospice care
models. As data increasingly demonstrate the beneficial
impact of early concurrent palliative care, we must develop
better models to facilitate its delivery as part of standard
cancer care, to supersede the barriers inherent in current
reimbursement constraints. New, developing models of socalled open access care promise to do just this and are a
beacon of hope that allow patients to receive both cancer
treatment and hospice care.
CONCLUSION
Palliative care is specialized care for people with serious
illnesses. When added to standard cancer care, specialty
palliative care improves many outcomes for patients and
their caregivers, including symptom burden, quality of life,
mood, prognostic understanding, end-of-life outcomes,
resource utilization, and even survival. However, more research is needed to illuminate mechanisms of action and to
improve the specificity of palliative care applications to
unique scenarios and populations.
References
1. Smith TJ, Temin S, Alesi ER, et al. American Society of Clinical Oncology
provisional clinical opinion: the integration of palliative care into
standard oncology care. J Clin Oncol. 2012;30:880-887.
2. Center to Advance Palliative Care. 2011 Public Opinion Research on
Palliative Care. https://www.capc.org/media/filer_public/18/ab/
18ab708c-f835-4380-921d-fbf729702e36/2011-public-opinionresearch-on-palliative-care.pdf. Accessed March 20, 2016.
3. Quill TE, Abernethy AP. Generalist plus specialist palliative care:
creating a more sustainable model. N Engl J Med. 2013;368:1173-1175.
4. Hui D, Finlay E, Buss MK, et al. Palliative oncologists: specialists in the
science and art of patient care. J Clin Oncol. 2015;33:2314-2317.
5. Yoong J, Park ER, Greer JA, et al. Early palliative care in advanced lung
cancer: a qualitative study. JAMA Intern Med. 2013;173:283-290.
6. Zimmermann C, Swami N, Krzyzanowska M, et al. Early palliative care
for patients with advanced cancer: a cluster-randomised controlled
trial. Lancet. 2014;383:1721-1730.
7. Bakitas M, Lyons KD, Hegel MT, et al. Effects of a palliative care intervention on clinical outcomes in patients with advanced cancer: the
Project ENABLE II randomized controlled trial. JAMA. 2009;302:
741-749.
8. Temel JS, Greer JA, Muzikansky A, et al. Early palliative care for patients
with metastatic non-small-cell lung cancer. N Engl J Med. 2010;363:
733-742.
9. Grudzen CR, Richardson LD, Johnson PN, et al. Emergency
departmentinitiated palliative care in advanced cancer: a randomized
clinical trial. JAMA Oncol. Epub 2016 Jan 14.
10. Weeks JC, Catalano PJ, Cronin A, et al. Patients expectations about
effects of chemotherapy for advanced cancer. N Engl J Med. 2012;367:
1616-1625.
11. Temel JS, Greer JA, Admane S, et al. Longitudinal perceptions of
prognosis and goals of therapy in patients with metastatic non-smallcell lung cancer: results of a randomized study of early palliative care.
J Clin Oncol. 2011;29:2319-2326.
12. Dionne-Odom JN, Azuero A, Lyons KD, et al. Benefits of early versus
delayed palliative care to informal family caregivers of patients with
advanced cancer: outcomes from the ENABLE III randomized controlled
13. Pirl WF, Greer JA, Traeger L, et al. Depression and survival in metastatic
non-small-cell lung cancer: effects of early palliative care. J Clin Oncol.
2012;30:1310-1315.
14. Penrod JD, Deb P, Luhrs C, et al. Cost and utilization outcomes of
patients receiving hospital-based palliative care consultation. J Palliat
Med. 2006;9:855-860.
15. Bakitas MA, Tosteson TD, Li Z, et al. Early versus delayed initiation of
concurrent palliative oncology care: patient outcomes in the ENABLE III
randomized controlled trial. J Clin Oncol. 2015;33:1438-1445.
e538
16. May P, Garrido MM, Cassel JB, et al. Palliative care teams cost-saving
effect is larger for cancer patients with higher numbers of comorbidities. Health Aff (Millwood). 2016;35:44-53.
17. May P, Garrido MM, Cassel JB, et al. Prospective cohort study of hospital
palliative care teams for inpatients with advanced cancer: earlier
consultation is associated with larger cost-saving effect. J Clin Oncol.
2015;33:2745-2752.
18. Back AL, Park ER, Greer JA, et al. Clinician roles in early integrated
palliative care for patients with advanced cancer: a qualitative study.
J Palliat Med. 2014;17:1244-1248.
19. Nipp RD, Greer JA, El-Jawahri A, et al. Age and gender moderate the
impact of early palliative care in metastatic nonsmall-cell lung cancer.
Oncologist. 2016;21:119-126.
20. OConnor NR, Hu R, Harris PS, et al. Hospice admissions for cancer in the
final days of life: independent predictors and implications for quality
measures. J Clin Oncol. 2014;32:3184-3189.
21. LeBlanc TW, Abernethy AP, Casarett DJ. What is different about patients
with hematologic malignancies? A retrospective cohort study of cancer
patients referred to a hospice research network. J Pain Symptom
Manage. 2015;49:505-512.
22. LeBlanc TW, El-Jawahri A. When and why should patients with hematologic malignancies see a palliative care specialist? Hematology (Am
Soc Hematol Educ Program). 2015;2015:471-478.
23. Lupu D, Friedman L, Alderman J, et al; American Academy of Hospice
and Palliative Medicine Workforce Task Force. Estimate of current
hospice and palliative medicine physician workforce shortage. J Pain
Symptom Manage. 2010;40:899-911.
24. Kamal AH, Bull J, Wolf S, et al. Characterizing the hospice and palliative
care workforce in the U.S.: clinician demographics and professional
responsibilities. J Pain Symptom Manage. 2016;51:597-603.
25. Ramchandran K, Tribett E, Dietrich B, et al. Integrating palliative care
into oncology: a way forward. Cancer Contr. 2015;22:386-395.
26. Buss MK, Lessen DS, Sullivan AM, et al. A study of oncology fellows
training in end-of-life care. J Support Oncol. 2007;5:237-242.
27. Buss MK, Lessen DS, Sullivan AM, et al. Hematology/oncology fellows
training in palliative care: results of a national survey. Cancer. 2011;117:
4304-4311.
28. Kamal AH, Dionne-Odom JN. A blue ocean strategy for palliative care:
focus on family caregivers. J Pain Symptom Manage. 2016;51:e1-e3.
29. Odejide OO. A policy prescription for hospice care. JAMA. 2016;315:
257-258.
30. LeBlanc TW. Addressing end-of-life quality gaps in hematologic cancers:
the importance of early concurrent palliative care. JAMA Intern Med.
2016;176:265-266.
31. LeBlanc TW. Palliative care and hematologic malignancies: old dog, new
tricks? J Oncol Pract. 2014;10:e404-e407.
Rehabilitation of Patients and

Survivors: Seizing the Opportunity
CHAIR
Catherine M. Alfano, PhD
Washington, DC
SPEAKERS
Andrea L. Cheville, MD, MSCE
Mayo Clinic
Rochester, MN
Karen Mustian, PhD, MPH
University of Rochester Medical Center
Rochester, NY
DEVELOPING HIGH-QUALITY CANCER REHABILITATION PROGRAMS
Developing High-Quality Cancer Rehabilitation Programs: A

Timely Need
Catherine M. Alfano, PhD, Andrea L. Cheville, MD, MSCE, and Karen Mustian, PhD, MPH
OVERVIEW
The number of survivors of cancer in the United States, already 14.5 million, is growing with improved cancer treatment and
aging of the population. Two-thirds of cancer survivors will be older than age 65 and are likely to enter cancer treatment
already deconditioned and with multiple comorbidities. Survivors of cancer face numerous adverse consequences of cancer
treatment that add to or exacerbate the effects of existing comorbidities and increase risk of functional decline. Many of
these problems are amenable to rehabilitation interventions, but referral to cancer rehabilitation professionals is not a
standard part of care. We present an expanded prospective model of surveillance, cancer rehabilitation assessment, and
referral efforts using a multidisciplinary team approach. In this model, cancer rehabilitation begins at the time of cancer
diagnosis and continues through and beyond cancer treatment. Physical impairments and psychosocial symptoms are
assessed and treated, and lifestyle and exercise interventions are provided to optimize functioning, health, and quality of
life. We present a stepped-care framework to guide decisions on when, how, and where to refer survivors to cancer
rehabilitation specialists depending on safety requirements and needs. This model has the potential to result in early
identification of symptoms and impairments, appropriate referral and timely treatment, and, in turn, will better address and
minimize both acute and long-term cancer morbidity.
n estimated 14.5 million survivors of cancer currently

live in the United States.1 This population will increase
dramatically in the coming decade because of increased
uptake of cancer screening, improved methods of early
detection, better multimodal cancer treatments, and the
aging of the population.2,3 National reports have documented that survivors of cancer have complex needs.4-8
These include management of chronic and late effects of
cancer and comorbid conditions; surveillance and treatment
of recurrence and second cancers; help with psychological,
social, economic, and family concerns; support to improve
lifestyle behaviors; and interventions to increase adherence
to long-term treatment and follow-up care guidelines. Given
that most cancers are diagnosed in older adults, an estimated two-thirds of all survivors of cancer will be age 65 or
older by 2020.2 This growing number of older survivors of
cancer presents a unique challenge to the health care system
because they are more likely to have multiple chronic diseases and experience poorer physical functioning than
younger survivors of cancer.9,10 Cancer and chronic illness
also may have interactive adverse effects on health and
psychosocial outcomes among older adult survivors of
cancer. Chronic conditions may limit the intensity and duration of cancer treatment,11-14 contributing to poorer
survival15 and increased risk for exacerbation of comorbid

conditions or further declines in physical functioning posttreatment.11,16,17
The United States is currently struggling to identify a
coordinated medical and public health approach to meet
the needs of our growing population of cancer survivors.
Predicted shortages in oncologists,18 primary care practitioners,19 and other medical health professionals underscore the need for effective and efficient care for cancer
survivors. We have argued that a solution to this health care
crisis lies in defining a new model of comprehensive cancer
rehabilitation, involving a multidisciplinary team of providers that aims to optimize the patients physical, psychological, vocational, and social functioning given the limits
imposed by the chronic or late-effects of cancer treatment
and other comorbid conditions.20
THE CASE FOR COMPREHENSIVE CANCER

REHABILITATION
The Institute of Medicine has defined the four pillars of
survivorship care: (1) surveillance for recurrence, second
cancers, and late effects; (2) intervention for treatment
consequences; (3) prevention of recurrence, new cancers,
From American Cancer Society, Washington, DC; Department of Physical Medicine and Rehabilitation, Program to Enhance Care Experiences Through Research, Center for the Science of
Health Care Delivery, Mayo Clinic, Rochester, MN; University of Rochester Medical Center, Wilmot Cancer Institute, Rochester, NY.
Corresponding author: Catherine M. Alfano, PhD, American Cancer Society, 555 11th St. NW, Suite 300, Washington, DC 20004; email: catherine.alfano@cancer.org.
241
ALFANO, CHEVILLE, AND MUSTIAN
late effects, and comorbidities where possible; and (4) coordination of care between primary care, oncology, and
other specialists.4 Comprehensive cancer rehabilitation
delivered using a multidisciplinary team is an essential
component of survivorship care that addresses these four
pillars. First, it treats the acute and chronic effects of cancer
and prevents or mitigates the effects of late-occurring
problems. Survivors of cancer can face numerous adverse
consequences of cancer treatment, many of which are
amenable to rehabilitation interventions. These include
fatigue, cognitive dysfunction, pain syndromes, peripheral
neuropathy, sexual dysfunction, balance and gait problems,
upper- or lower-quadrant mobility issues, lymphedema,
bladder and bowel problems, stoma care, problems with
swallowing or dysphagia, communication difficulty, and
psychosocial problems such as depression, anxiety, or fear of
recurrence, among others.21,22 In a comprehensive rehabilitation model, the multidisciplinary team evaluates the
sum total of problems that a survivor faces and coordinates
treatment rather than other fragmented models of care that
treat each symptom or impairment separately. Second, the
comprehensive cancer rehabilitation team can address preexisting or treatment-related comorbid conditions. Bone
loss, diabetes, cardiovascular disease, congestive heart failure,
adverse body composition, and renal disease are common in
cancer survivors.4 These can be managed through rehabilitation interventions that include medication, counseling,
behavior change and promotion of healthy diets, physical
activity, and weight control.21 Third, these health promotion
interventions, along with self-management skills provided in
the context of comprehensive cancer rehabilitation, have the
KEY POINTS
242
Survivors of cancer can face numerous adverse

consequences of cancer treatment, many of which are
amenable to rehabilitation interventions.
Using an expanded prospective model of surveillance for
comprehensive cancer rehabilitation, assessment and
referral efforts begin at the time of cancer diagnosis,
continue through and beyond treatment, and use a
multidisciplinary team approach.
The model facilitates the identification and treatment of
physical impairments and psychosocial problems and
other problematic symptoms, and provides exercise,
nutrition, and weight management interventions to
optimize functioning, health, and quality of life.
Survivors of cancer can be treated in one of four levels of
stepped care or inpatient rehabilitation depending on
safety requirements and needs.
This expanded prospective model of surveillance for
cancer rehabilitation has the potential to result in early
identification of symptoms and impairments and
appropriate referral and timely treatment and, in turn,
will better address and minimize both acute and longterm cancer morbidity.
potential to decrease the risk of further late effects (e.g.,

cardiac, pulmonary, endocrine, or bone complications of
cancer treatment)21 and may even reduce the risk of
second malignancies.23-29 Finally, comprehensive cancer
rehabilitation focuses jointly on optimizing functional
status and quality of life, preserving the ability to remain in
life roles (including the workforce),30 and maximizing
health and longevity.
A PROSPECTIVE MODEL OF SURVEILLANCE FOR

COMPREHENSIVE CANCER REHABILITATION
A prospective model of surveillance has been developed by a
team of stakeholders as a best practice model for cancer
rehabilitation.31 In this model, rehabilitative efforts begin at
the time of cancer diagnosis and continue through treatment and after treatment ends, using a multidisciplinary
team approach. The model specifies that a multidimensional, comprehensive assessment is conducted at the
preoperative evaluation to establish baseline functioning,
identify patients with pre-existing conditions that may place
them at higher risk for the development of treatment
toxicities and impairments during/after treatment, and refer
patients with current problems for interventions to improve
their symptoms and function. Ideally, this is a clinical
evaluation that includes a history focusing on symptoms and
current exercise and a physical examination that objectively
assesses the patients range of motion, strength, sensation,
reflexes, and gait accompanied by performance testing (e.g.,
The Timed Up and Go, Repeated Sit to Stand tests, and Short
Physical Performance Battery). This may be obtained or
supplemented with patient-reported outcome (PRO)based
measurement of survivors symptoms and functioning, although limited information is available on the sensitivity and
specificity of PROs in identifying cancer survivors in need of
cancer rehabilitation services.32,33 After the preoperative
assessment, the prospective surveillance model specifies
ongoing surveillance efforts with repeated assessments
postoperatively, throughout cancer therapies, and after
treatment ends. This allows for monitoring of the development of treatment toxicities or impairments and facilitation of appropriate and timely referrals to cancer
rehabilitation providers. The reassessments also provide
opportunities for assessment and referral for exercise,
nutrition, and weight management interventions. The
ongoing surveillance also allows the treatment team to
ensure the resolution of problems over the long term.
Though not explicitly stated by the original stakeholder
group, a prospective surveillance model for comprehensive
cancer rehabilitation must incorporate assessment and referral for mental health problems and palliative care needs in
addition to impairment-driven rehabilitation and health
behavior interventions to meet survivors ongoing needs.
Although this model has been outlined by a multidisciplinary
group of stakeholder experts, it has not been extensively
tested for effects on clinical outcome or health care costs.34
The model assumes that prospective surveillance will result
in early identification of symptoms and impairments, appropriate referral and timely treatment, and, in turn, will
better address and minimize both acute and long-term
cancer morbidity. The model has the potential to achieve
the triple aim set forth by the Institute for Healthcare Improvement35 of facilitating better care for individuals, better
health for populations, and lower per capita costs. However,
future health care delivery research must test these assumptions. Although early research suggests substantial
cost savings of implementing this model early compared
with treating advanced-stage problems, at least in the case
of breast cancerrelated lymphedema,36 future research
must include assessment of cost outcomes as previously
described.34
LEVELS OF CANCER REHABILITATION STEPPED

CARE
The expanded prospective model of surveillance presents a
framework for evaluating the needs of people with cancer
and triaging them into appropriate pathways of care based
on their need for cancer rehabilitation. But what type of care
they should receive and who should deliver the care are
issues that continue to challenge the rehabilitation community. Below we present a stepped-care framework to
guide decisions on when, how, and where to refer survivors
to cancer rehabilitation specialists depending on safety
requirements and needs. Note that all levels may involve
physical, occupational, or speech therapy; psychosocial and
cognitive interventions; exercise prescription; and lifestyle
recommendations as dictated by the needs of the individual.
The disciplines and services that provide these different
interventions, the environment in which they are provided,
and the degree of integration of the team will differ
depending on numerous factors. These include the strengths
and expertise of individual providers and existing services,
geographic proximity and communication among services,
and facility and system resources.
Level I: General Conditioning Activities, Unspecialized

In the absence of any cancer-specific morbidity, virtually all
survivors of cancer emerge from their cancer treatment with
poorer aerobic fitness and muscle quality than when they
began.37 For this reason aloneapart from secondary
cancer prevention38,39education and/or prescriptive
counseling on progressive aerobic conditioning, resistance
training, and the benefits of exercise in general should be
integral to any cancer survivorship program. This should be
supplemented by psychosocial counseling to enhance participation in activities and address participation barriers
where needed. This represents the most basic level of
stepped cancer rehabilitation. In the future, efforts to create
more precision medicine approaches to exercise prescription will help guide the type and dose of exercise chosen. For
now, for survivors of cancer who have exercised in the
past and lack concerning impairments or comorbidities,
counseling can be quite rudimentary, taking the form of
educational print materials or videos. More formal counseling by an exercise or rehabilitation professional, enrollment in a conditioning program with clinical oversight, and
more intensive behavioral counseling represent the next
two levels of increasingly time- and resource-intensive rehabilitation. Efficacious models of cardiac and pulmonary
rehabilitation offer a potential framework for the latter, with
pilot studies of their utility for survivors of cancer showing
promise.40,41 As yet, center-based conditioning programs for
cancer survivors have not been developed in the United
States, but successes in Canada and Europe attest to their
value.42,43
Several options are available for practitioners and institutions interested in making general conditioning guidance and training available to survivors of cancer. Physical
therapists without specialized cancer rehabilitation training
routinely deliver deconditioning-directed care, and they are
well equipped to individualize programs and transition
cancer survivors to home-based maintenance programs.
However, in the absence of functional impairments or
steady, ongoing improvement, payer coverage for physical
therapy may be limited. For survivors of cancer without
concerning impairments or comorbidities that would require
physical therapy or medical oversight, the LIVESTRONG at the
YMCA 12-week programs offer services and support delivered by athletic trainers and exercise professionals. These
services extend well beyond conditioning, although the recovery of strength and aerobic fitness are principle program
goals. There are currently approximately 460 LIVESTRONG
programs in the continental United States. Referral to
medically oriented gyms offers another option. Medically
oriented gyms are facilities typically staffed by physical
therapists and/or athletic trainers with expertise in chronic
illness and the care of frail patients, but they may lack
cancer specialization. Medically oriented gyms require outof-pocket payments. However, some payers will cover the
initial exercise prescription, professional oversight and
monitoring, and other services provided in these facilities.
These programs can be supplemented with counseling to
support behavioral changes as needed.
Level II: General Conditioning Activities, Specialized

Higher or more specialized levels of cancer rehabilitation
along the stepped-care continuum require care from clinicians with some degree of specialization in cancer rehabilitation. The most prevalent need at this level is for
general conditioning among survivors of cancer who lack
specific cancer- or cancer treatmentrelated impairments
but have sufficient vulnerability to warrant specialist oversight. Survivors of cancer frequently have or are at risk for
conditions potentially exacerbated by exercise. Theoretically,
these individuals should have a rehabilitation team with at
least one cancer rehabilitation specialist, although empirical
validation of this need is lacking. Psychosocial counseling
to address barriers to participation or self-management
strategies may be needed. Research has established the
243
beneficial and safe nature of resistive training when conducted under the close supervision of exercise professionals
trained in lymphedema and other cancer treatmentrelated
risks.44,45 In some clinical settings, there may not be therapists
with specific cancer rehabilitation training. In these cases, a
rehabilitation savvy oncologist or cancer rehabilitation physician may provide generally trained therapists with a detailed
physical or occupational therapy prescription with clear
precautions. They also may serve as a resource for these
noncancer trained treating professionals.
Level III: Impairment-Directed Care, Uncomplicated

This level of cancer rehabilitation need is broad because it
encompasses focal cancer- and cancer treatmentrelated
impairments that limit function but are uncomplicated by
symptoms or other systemic concerns. These impairments
generally can be managed by physical and occupational
therapists with a moderate degree of cancer specialization.
Care focuses on treating the impairment(s) to optimize
function and also on increasing activity levels with an end
goal of enabling the cancer survivors to safely participate in
lifestyle behaviors without oversight if possible. This level of
care also is more likely to require psychosocial counseling to
help survivors cope with problematic symptoms, body image
concerns, or other mental health issues and support behavior change efforts. The impairments that fall into this
category are as varied as they are numerous. They include,
but are not limited to, cranial nerve 11 palsy, radiationinduced fibrosis, axillary web syndrome (variant Mondor
disease), contractures from sclerodermiform graft versus
host disease, persistent chemotherapy-induced peripheral
neuropathy (CIPN), surgical donor site morbidity (e.g., post
transverse rectus abdominis [TRAM] flap reconstruction),
and secondary myofascial dysfunction. A rapidly growing
body of literature describes the diagnosis and evidencebased treatment of many of these impairments.46-48 Efforts
are underway to make treatment guidelines widely available
to practicing therapists. Fortunately, specialty conferences
and other educational opportunities are a rapidly increasing
means by which therapists can refine their expertise and
benefit from collective experience. Ideally, a cancer rehabilitation physician specialist is available to consult on
complex or nonresponding cases, although in less complex
cases, a physician need not evaluate these impairments and
thereby potentially delay their timely treatment.
Level IV: Impairment-Directed Care, Complicated

The most resource-intensive form of outpatient cancer rehabilitation along the stepped-care continuum is specialist
physiciandirected evaluation and treatment. This should be
reserved for patients with problematic symptoms (e.g., intensely painful axillary cording or severe or refractory impairments) because such care may prove to be time-consuming
and costly for cancer survivors with depleted energy and
resources. At times, the sheer number of co-occurring impairments may necessitate physician involvement in cases
244
such as allogeneic bone marrow transplant recipients who

commonly have CIPN, steroid myopathy, joint contractures,
and generalized deconditioning.49 The need for a cancer
rehabilitation physician inevitably will vary across settings
because many quaternary cancer centers have seasoned
teams with therapists accustomed to managing challenging
impairments. Level IV care is likely to involve coordinating
care among more members of the multidisciplinary team as
physical, occupational, speech and language, and psychosocial therapies all may be needed. As in level III care, level IV
care focuses on treating impairments, ongoing symptom
management, and increasing activity levels to enable the
survivor of cancer to safely transition to participate in
lifestyle behaviors without oversight if possible.
Inpatient Cancer Rehabilitation. Inpatient cancer rehabilitation delivered by a multidisciplinary team of providers through consultation and liaison and postacute care
services clearly is a vital component of comprehensive
cancer care. In fact, the overwhelming majority of physical
impairments, at least among patients with stage IV disease,
are detected when patients are hospitalized.50 Goals of care
are similar to higher outpatient levels: treating impairments,
psychosocial and other problematic symptoms, optimizing
functioning, and supporting exercise and dietary changes to
the extent possible. All hospitals and inpatient cancer
centers offer physical and occupational therapy services.
Most have dietitians and some have rehabilitation-oriented
mental health specialists, but the availability of a rehabilitation physician specialized in cancer is less consistent.
Even when lacking cancer rehabilitation expertise, the involvement of a rehabilitation physician in the management
of hospitalized patients offers salient advantages to a cancer
rehabilitation program, providing particularly effective and
expeditious advocacy for appropriate postacute care services.51 Such optimized discharge planning with the involvement of rehabilitation specialists has been shown to
reduce 30-day readmissions and costs per episode of care,
issues that are becoming increasingly important with the
rapid adoption of bundled payment.
IMPLEMENTATION OF QUALITY CANCER

REHABILITATION CARE
The drive to prevent and reverse disablement has spurred
care providers to develop cancer rehabilitation programs
across diverse delivery settings. This drive has been further
intensified by the American College of Surgeons Commission
on Cancer standard that accredited institutions provide
cancer rehabilitation services.52 Despite increased awareness, patient access to appropriate cancer rehabilitation
services remains hampered by an unfortunate legacy of the
decades-long marginalization of cancer rehabilitationthe
fragmentation of rehabilitation disciplines and the persistence of a severely limited workforce that has proven slow to
expand. Because of low referral rates from oncology clinical
practices, rehabilitation medicine has not been incentivized
to develop cancer rehabilitation positions or training programs. The limited availability of cancer rehabilitation
programs, even in large, high-volume cancer centers,20 and
the varied services provided by the few established programs has afforded oncology clinicians minimal opportunity
to learn how to effectively capitalize on the expertise of their
rehabilitation medicine colleagues to improve patient outcomes. This self-perpetuating cycle has been hard to break.
Even today, oncology trainees seldom learn how and when
to refer their patients for rehabilitation services.
Several approaches have been established and/or are
being actively tested to overcome workforce limitations. The
first, structured rehabilitation provider trainings and certifications, have gained broad traction. CARF International
introduced accreditation standards for cancer rehabilitation
specialty programs in 2014 that can be applied to hospitals,
health care systems, outpatient clinics, and communitybased programs. The Survivorship Training and Rehab
(STAR) program, a commercial entity, offers web-based
trainings to physical and occupational therapists linked to
examinations that, when passed, lead to institutional STAR
certification. The impact of CARF standards or STAR certification on care quality and effectiveness have not been
characterized. The American College of Sports Medicine
(ACSM) also has developed educational guidelines and a
certifying examination for cancer exercise trainers who work
with survivors. The certification provides a professional
competency benchmark for oncology professionals who are
looking for exercise-referral options in their local communities.53 Again, the care implications of practitioner certification are not known. However, the fact that a practitioner
or institution has gone to the trouble of seeking, identifying,
and achieving CARF, STAR, or ACSM certification attests to
their interest and potential skill in cancer rehabilitation.
Another approach to workforce expansion is the enhancement of noncancer physical or occupational therapy services
available in most U.S. communities through the direction
and support of physician and therapist cancer rehabilitation
specialists at tertiary cancer centers. Data collection for the
federally funded Collaborative Care to Preserve Performance in Cancer (COPE) trial finished in 2015.54 COPE is
rigorously assessing whether a telemedicine approach that
leverages limited cancer rehabilitation clinical expertise to
deliver evidence-based treatment by local generalist physical therapists is a useful near-term and cost-sensitive means
of addressing current workforce limitations.
EXERCISE AS AN ESSENTIAL COMPONENT OF

CANCER REHABILITATION, ESPECIALLY FOR
OLDER ADULTS
The team that created the prospective model of surveillance
for cancer rehabilitation described exercise as an essential
component of cancer rehabilitation.31 Therefore, exercise is
integrated into each of the stepped-care levels described
above. This importance of exercise is especially true for
deconditioned and older adult survivors of cancer. A growing
body of literature demonstrates that maintaining moderate

to high levels of physical activity through formal exercise
interventions is an effective method for treating both the
physical and psychological declines experienced by older
patients with cancer as well as younger cancer survivors.55-77
Research shows that adults with cancer decrease their
physical activity levels after diagnosis and during treatment
and often do not return to their pretreatment activity levels
without formal exercise interventions.78-80 This tendency
toward declining physical activity is reinforced as patients often are advised to limit activities when they are
older, fatigued, or experiencing other toxicities and side
effects, especially cardiac, orthopedic, or functional
comorbidities.9,75,77,81-83
Exercise improves a wide array of physical and psychological symptoms, including muscle atrophy and weakness,
fatigue, obesity, immune function, insomnia, anxiety, cognitive
decline, and impaired quality of life, among others.55-74,84-88
Epidemiologic data also suggest that increased physical activity
through regular exercise reduces the risk of cancer recurrence
and cancer mortality.39,89-90 The ACSM published public health
recommendations for exercise among survivors of cancer.91
Geriatric survivors of cancer should start low and progress to
150 minutes of moderate-intensity or 75 minutes of vigorousintensity aerobic exercise per week; 20 to 30 minutes of
strength training across all the major muscle groups two to
three times per week; and regular stretching daily each
week.91-94 Despite these published public health guidelines, it is estimated up to 70% of survivors do not meet
these ACSM public health recommendations.95 This lack of
regular exercise is especially problematic for older survivors of cancer when combined with the considerable
physical and psychological comorbidities, symptoms, and
side effects they experience.
Despite the numerous benefits of exercise as part of
cancer rehabilitation, the majority of survivors of cancer
report that they do not discuss exercise with their treating
oncologist or primary care physician during their cancer
treatment and recovery.96-98 However, research shows that
cancer survivors of all ages want their oncologists to initiate
discussions about exercise and make appropriate referrals.98
Survivors prefer to receive information on exercise early in
the treatment trajectory and throughout the post-treatment
period.99 Given the benefits of exercise during and after
cancer treatments, routine discussions about exercise between oncologists and their older patients who are survivors
of cancer, along with appropriate referrals to a qualified
exercise physiologist or cancer rehabilitation therapist as
needed, could significantly improve prognosis, recovery, and
multiple domains of quality of life for older individuals.
Managing Risk and Contraindications for Exercise

As part of the prospective surveillance model for cancer
rehabilitation, it is essential to identify and address potential
contraindications (e.g., orthopedic, cardiopulmonary, oncologic) that might affect exercise safety and tolerance.53
245
Contraindications do not mean that a geriatric survivor of

cancer cannot exercise at all. As described above in the levels
of stepped care, contraindications simply require specific
modifications to the exercise regimen or care environment
(e.g., facility-based vs. home) so that the individual can
exercise safely and still achieve the desired physical and
mental health benefits.53 Although active exercisers or those
at low to moderate risk based on the ACSM guidelines are
not required to obtain medical clearance to continue or
begin a low to moderate exercise program, most geriatric
survivors of cancer will need and benefit from further
medical assessment because their exercise risk will be
moderate to high.53 These geriatric survivors of cancer will
benefit from additional medical assessment by qualified
medical professionals including oncologists, surgeons,
cardiologists, orthopedists, and neurologists, among
others, because these consultations will provide important
information required by exercise physiologists to appropriately modify an exercise prescription and account for
cancer-specific cardiovascular, pulmonary, metabolic, orthopedic, neurologic, or other comorbidities.53 These
additional medical evaluations should be conducted
before initiation of any baseline exercise testing that
precedes developing an exercise prescription and subsequent exercise participation.53
Exercise Prescription for Older Adult Survivors of

Cancer
Testing, prescription, and monitoring of exercise should be
done by qualified exercise professionals, especially for geriatric survivors of cancer at moderate risk beginning or
continuing vigorous exercise and those at high risk beginning
or continuing any level of exercise.53 Exercise prescriptions
for geriatric survivors of cancer should be individualized and
tailored based on the older individuals health status, disease
trajectory, previous and/or current treatment, current fitness
level, past and present exercise participation, and individual
preferences to be safe and effective.91,92 Moderately intense
aerobic exercise prescriptions (55%75% of heart rate maximum100) scheduled over as little as 10 minutes and no more
than 90 minutes a day 3 to 7 days per week are effective at
reducing symptom burden and improving quality of life among
older patients with cancer and survivors of cancer.23,72,82,101

Accumulating 30 minutes of daily activity through short bouts
of aerobic exercise (3 to 10 minutes each), with rest in between,
also can reduce symptoms in these patients.91 A moderate to
vigorously intense anaerobic resistance exercise prescriptions
schedule of three times per week and progressively increasing
up to as few as two sets and no more than four sets of 8 to 15
repetitions are effective at reducing symptoms.102-106 Combination exercise prescriptions that include both aerobic and
anaerobic exercise are safe and very effective for geriatric
survivors of cancer. Low to moderately intense mindfulnessbased exercise prescriptions, including yoga and tai chi chuan,
scheduled one to three times a week for 6090 minutes, are
also highly effective in reducing symptom burden among older
adults.64,69,71,74,107,108 Additionally, those with advanced
disease can safely perform, tolerate, and benefit from lowintensity exercise.109-114
CONCLUSION
We have described an expanded prospective model of
surveillance for cancer rehabilitation in which assessment
and referral efforts begin at the time of cancer diagnosis,
continue through and beyond treatment, and use a multidisciplinary team approach. This model focuses on identification and treatment of physical impairments and
psychosocial problems and the provision of lifestyle and
exercise interventions. The model proposes that patients
can be triaged into four levels of stepped care in addition to
inpatient rehabilitation, depending on the degree of
comorbidities and impairments noted, safety concerns,
and other needs. This model has the potential to result in
earlier identification of symptoms, appropriate referral,
timely symptom management, and ultimately lower
morbidity and mortality. However, several key research
questions remain. Assessment methods must be developed
to create an appropriate comprehensive measure of
symptoms, functioning, impairments, and needs. Modeling
efforts must focus on identifying the algorithm to triage
survivors of cancer into the levels of care. Once these are in
place, future research efforts must also test the impact of
this model on patient outcomes, care coordination, and
health care costs.
References
1.
2.
3.
4.
246
American Cancer Society. Cancer Treatment and Survivorship Facts &

Figures 2014-2015. Atlanta: American Cancer Society; 2014.
Parry C, Kent EE, Mariotto AB, et al. Cancer survivors: a booming population. Cancer Epidemiol Biomarkers Prev. 2011;20:
1996-2005.
de Moor JS, Mariotto AB, Parry C, et al. Cancer survivors in the United
States: prevalence across the survivorship trajectory and implications
for care. Cancer Epidemiol Biomarkers Prev. 2013;22:561-570.
HewittM, GreenfieldS, Stovall E (eds). From Cancer Patient to Cancer
Survivor: Lost in Transition. Washington, DC: National Academies
Press; 2006.
5.
6.
7.
Hewitt M, Weiner SL, Simone JV. Childhood Cancer Survivorship:

Improving Care and Quality of Life. Washington, DC: National
Academies Press; 2003.
Adler NE, Page A, National Institute of Medicine. (U. S.) Committee
on Psychosocial Services to Cancer Patients/Families in a Community Setting. Cancer Care for the Whole Patient: Meeting
Psychosocial Health Needs. Washington, DC: National Academies
Press; 2008.
Centers for Disease Control and Prevention, Lance Armstrong
Foundation, U.S. Department of Health and Human Services. A
National Action Plan for Cancer Survivorship: Advancing Public Health
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
Strategies. http://www.cdc.gov/cancer/survivorship/pdf/plan.pdf.
Presidents Cancer Panel. Living Beyond Cancer: Finding a New
Balance. Bethesda, MD: National Cancer Institute: 2004.
Bellizzi KM, Mustian KM, Palesh OG, et al. Cancer survivorship and
aging: moving the science forward. Cancer. 2008;113(Suppl 12):
3530-3539.
Avis NE, Deimling GT. Cancer survivorship and aging. Cancer. 2008;113
(Suppl 12):3519-3529.
Yancik R, Wesley MN, Ries LA, et al. Effect of age and comorbidity in
postmenopausal breast cancer patients aged 55 years and older.
JAMA. 2001;285:885-892.
Extermann M. Interaction between comorbidity and cancer. Cancer
Contr. 2007;14:13-22.
Greenfield S, Blanco DM, Elashoff RM, et al. Patterns of care related to
age of breast cancer patients. JAMA. 1987;257:2766-2770.
Ballard-Barbash R, Potosky AL, Harlan LC, et al. Factors associated with
surgical and radiation therapy for early stage breast cancer in older
women. J Natl Cancer Inst. 1996;88:716-726.
Piccirillo JF, Tierney RM, Costas I, et al. Prognostic importance of
comorbidity in a hospital-based cancer registry. JAMA. 2004;291:
2441-2447.
Given CW, Given B, Azzouz F, et al. Comparison of changes in physical
functioning of elderly patients with new diagnoses of cancer. Med
Care. 2000;38:482-493.
Hewitt M, Rowland JH, Yancik R. Cancer survivors in the United States:
age, health, and disability. J Gerontol A Biol Sci Med Sci. 2003;58:
M82-M91.
Erikson C, Salsberg E, Forte G, et al. Future supply and demand for
Pract. 2007;3:79-86.
Salsberg E, Grover A. Physician workforce shortages: implications and
issues for academic health centers and policymakers. Acad Med. 2006;
81:782-787.
Alfano CM, Ganz PA, Rowland JH, et al. Cancer survivorship and cancer
rehabilitation: revitalizing the link. J Clin Oncol. 2012;30:904-906.
Stubblefield MD, ODell, MW. Cancer Rehabilitation Principles and
Practice. New York: Demos Medical Publishing; 2009.
Silver JK, Baima J, Mayer RS. Impairment-driven cancer rehabilitation:
an essential component of quality care and survivorship. CA Cancer J
Clin. 2013;63:295-317.
Holmes MD, Chen WY, Feskanich D, et al. Physical activity and survival
after breast cancer diagnosis. JAMA. 2005;293:2479-2486.
Pierce JP, Stefanick ML, Flatt SW, et al. Greater survival after breast
cancer in physically active women with high vegetable-fruit intake
regardless of obesity. J Clin Oncol. 2007;25:2345-2351.
Irwin ML, Smith AW, McTiernan A, et al. Influence of pre- and
postdiagnosis physical activity on mortality in breast cancer survivors:
the health, eating, activity, and lifestyle study. J Clin Oncol. 2008;26:
3958-3964.
Giovannucci E, Liu Y, Platz EA, et al. Risk factors for prostate cancer
incidence and progression in the health professionals follow-up study.
Int J Cancer. 2007;121:1571-1578.
Meyerhardt JA, Giovannucci EL, Holmes MD, et al. Physical activity and
survival after colorectal cancer diagnosis. J Clin Oncol. 2006;24:
3527-3534.
Meyerhardt JA, Heseltine D, Niedzwiecki D, et al. Impact of physical
activity on cancer recurrence and survival in patients with stage III colon
cancer: findings from CALGB 89803. J Clin Oncol. 2006;24:3535-3541.
Irwin ML, McTiernan A, Manson JE, et al. Physical activity and survival
in postmenopausal women with breast cancer: results from the
Womens Health Initiative. Cancer Prev Res (Phila). 2011;4:522-529.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
de Boer AG, Taskila T, Tamminga SJ, et al. Interventions to enhance

return-to-work for cancer patients. Cochrane Database Syst Rev.
2011;2:CD007569.
Stout NL, Binkley JM, Schmitz KH, et al. A prospective surveillance
model for rehabilitation for women with breast cancer. Cancer. 2012;
118(Suppl 8):2191-2200.
Cheville AL, Yost KJ, Larson DR, et al. Performance of an item response
theory-based computer adaptive test in identifying functional decline.
Arch Phys Med Rehabil. 2012;93:1153-1160.
Hahn EA, Devellis RF, Bode RK, et al; PROMIS Cooperative Group.
Measuring social health in the patient-reported outcomes measurement information system (PROMIS): item bank development and
testing. Qual Life Res. 2010;19:1035-1044.
Cheville AL, Nyman JA, Pruthi S, et al. Cost considerations regarding
the prospective surveillance model for breast cancer survivors.
Cancer. 2012;118(Suppl 8):2325-2330.
Institute for Healthcare Improvement. IHI Triple Aim Initiative. http://
www.ihi.org/engage/initiatives/tripleaim. Accessed February 12, 2016
Stout NL, Pfalzer LA, Springer B, et al. Breast cancer-related
lymphedema: comparing direct costs of a prospective surveillance
model and a traditional model of care. Phys Ther. 2012;92:152-163.
Gilliam LA, St Clair DK. Chemotherapy-induced weakness and fatigue
in skeletal muscle: the role of oxidative stress. Antioxid Redox Signal.
2011;15:2543-2563.
Lahart IM, Metsios GS, Nevill AM, et al. Physical activity, risk of death
and recurrence in breast cancer survivors: a systematic review and
meta-analysis of epidemiological studies. Acta Oncol. 2015;54:635-654.
Ballard-Barbash R, Friedenreich CM, Courneya KS, et al. Physical
activity, biomarkers, and disease outcomes in cancer survivors:
a systematic review. J Natl Cancer Inst. 2012;104:815-840.
Munro J, Adams R, Campbell A, et al. CRIBthe use of cardiac rehabilitation services to aid the recovery of patients with bowel cancer:
a pilot randomised controlled trial (RCT) with embedded feasibility
study. BMJ Open. 2014;4:e004684.
Hubbard G, Adams R, Campbell A, et al. Is referral of postsurgical
colorectal cancer survivors to cardiac rehabilitation feasible and acceptable? A pragmatic pilot randomised controlled trial with embedded qualitative study. BMJ Open. 2016;6:e009284.
Chasen MR, Bhargava R. A rehabilitation program for patients with
gastroesophageal cancera pilot study. Support Care Cancer. 2010;18
(Suppl 2):S35-S40.
Kjaer TK, Johansen C, Ibfelt E, et al. Impact of symptom burden on
health related quality of life of cancer survivors in a Danish cancer
rehabilitation program: a longitudinal study. Acta Oncol. 2011;50:
223-232.
Schmitz KH, Ahmed RL, Troxel A, et al. Weight lifting in women with
breast-cancer-related lymphedema. N Engl J Med. 2009;361:664-673.
Schmitz KH, Ahmed RL, Troxel AB, et al. Weight lifting for women at
risk for breast cancer-related lymphedema: a randomized trial. JAMA.
2010;304:2699-2705.
McNeely ML, Parliament MB, Seikaly H, et al. Effect of exercise on
upper extremity pain and dysfunction in head and neck cancer survivors: a randomized controlled trial. Cancer. 2008;113:214-222.
Cho Y, Do J, Jung S, et al. Effects of a physical therapy program
combined with manual lymphatic drainage on shoulder function,
quality of life, lymphedema incidence, and pain in breast cancer
patients with axillary web syndrome following axillary dissection.
Support Care Cancer. Epub 2015 Nov 5.
De Groef A, Van Kampen M, Dieltjens E, et al. Effectiveness of
postoperative physical therapy for upper-limb impairments after
breast cancer treatment: a systematic review. Arch Phys Med Rehabil.
2015;96:1140-1153.
247
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
248
Smith SR, Haig AJ, Couriel DR. Musculoskeletal, neurologic, and

cardiopulmonary aspects of physical rehabilitation in patients with
chronic graft-versus-host disease. Biol Blood Marrow Transplant.
2015;21:799-808.
Cheville AL, Shen T, Chang M, et al. Appropriateness of the treatment
of fatigued patients with stage IV cancer. Support Care Cancer. 2013;
21:229-233.
Yadav R. Rehabilitation of surgical cancer patients at University
of Texas M. D. Anderson Cancer Center. J Surg Oncol. 2007;95:
361-369.
American College of Surgeons. The Commission on Cancers (CoC)
Cancer Program Standards: Ensuring Patient-Centered Care. Chicago:
American College of Surgeons; 2016.
American College of Sports Medicine. ACSMs Guidelines for Exercise
Testing and Prescription, 8th Ed. Philadelphia: Lippincott Williams &
Wilkins; 2016.
Cheville A. Collaborative Targeted Case Management in Improving
Functional Status in Patients with Stage III-IV Cancer. https://
clinicaltrials.gov/ct2/show/NCT01721343?term=cheville&rank=3.
Accessed February 12, 2016
Demark-Wahnefried W, Clipp EC, Morey MC, et al. Lifestyle intervention development study to improve physical function in older
adults with cancer: outcomes from Project LEAD. J Clin Oncol. 2006;24:
3465-3473.
Morey MC, Snyder DC, Sloane R, et al. Effects of home-based diet and
exercise on functional outcomes among older, overweight long-term
cancer survivors: RENEW: a randomized controlled trial. JAMA. 2009;
301:1883-1891.
Campo RA, Agarwal N, LaStayo PC, et al. Levels of fatigue and distress
in senior prostate cancer survivors enrolled in a 12-week randomized
controlled trial of Qigong. J Cancer Surviv. 2014;8:60-69.
Winters-Stone KM, Leo MC, Schwartz A. Exercise effects on hip bone
mineral density in older, post-menopausal breast cancer survivors are
age dependent. Arch Osteoporos. 2012;7:301-306.
Cormie P, Galv~ao DA, Spry N, et al. Can supervised exercise prevent
treatment toxicity in patients with prostate cancer initiating
androgen-deprivation therapy: a randomised controlled trial. BJU Int.
2015;115:256-266.
Winters-Stone KM, Lyons KS, Dobek J, et al. Benefits of partnered
strength training for prostate cancer survivors and spouses: results
from a randomized controlled trial of the Exercising Together project.
J Cancer Surviv. Epub 2015 Dec 29.
Winters-Stone KM, Dieckmann N, Maddalozzo GF, et al. Resistance exercise reduces body fat and insulin during androgendeprivation therapy for prostate cancer. Oncol Nurs Forum. 2015;42:
348-356.
Sprod LK, Fernandez ID, Janelsins MC, et al. Effects of yoga on cancerrelated fatigue and global side-effect burden in older cancer survivors.
J Geriatr Oncol. 2015;6:8-14.
Sprod LK, Mohile SG, Demark-Wahnefried W, et al. Exercise and
cancer treatment symptoms in 408 newly diagnosed older cancer
patients. J Geriatr Oncol. 2012;3:90-97.
Sprod LK, Janelsins MC, Palesh OG, et al. Health-related quality of life
and biomarkers in breast cancer survivors participating in tai chi
chuan. J Cancer Surviv. 2012;6:146-154.
Sprod LK, Palesh OG, Janelsins MC, et al. Exercise, sleep quality, and
mediators of sleep in breast and prostate cancer patients receiving
radiation therapy. Community Oncol. 2010;7:463-471.
Janelsins MC, Peppone LJ, Heckler CE, et al. YOCAS Yoga reduces
self-reported memory difficulty in cancer survivors in a nationwide
randomized clinical trial: investigating relationships between memory
and sleep. Integr Cancer Ther. Epub 2015 Nov 29.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
Janelsins MC, Davis PG, Wideman L, et al. Effects of tai chi chuan on
insulin and cytokine levels in a randomized controlled pilot study on
breast cancer survivors. Clin Breast Cancer. 2011;11:161-170.
Peppone LJ, Janelsins MC, Kamen C, et al. The effect of YOCAS yoga
for musculoskeletal symptoms among breast cancer survivors on
hormonal therapy. Breast Cancer Res Treat. 2015;150:597-604.
Peppone LJ, Mustian KM, Janelsins MC, et al. Effects of a structured
weight-bearing exercise program on bone metabolism among breast
cancer survivors: a feasibility trial. Clin Breast Cancer. 2010;10:
224-229.
Kamen C, Heckler C, Janelsins MC, et al. A dyadic exercise intervention
to reduce psychological distress among lesbian, gay, and heterosexual
cancer survivors. LGBT Health. 2016;3:57-64.
Mustian KM, Katula JA, Gill DL, et al. Tai chi chuan, health-related
quality of life and self-esteem: a randomized trial with breast cancer
survivors. Support Care Cancer. 2004;12:871-876.
Mustian KM, Peppone L, Darling TV, et al. A 4-week home-based
aerobic and resistance exercise program during radiation therapy:
a pilot randomized clinical trial. J Support Oncol. 2009;7:158-167.
Mustian KM, Sprod LK, Janelsins M, et al. Multicenter, randomized
controlled trial of yoga for sleep quality among cancer survivors. J Clin
Oncol. 2013;31:3233-3241.
Mustian KM, Katula JA, Zhao H. A pilot study to assess the influence of
tai chi chuan on functional capacity among breast cancer survivors.
J Support Oncol. 2006;4:139-145.
Mustian KM, Sprod LK, Palesh OG, et al. Exercise for the management
of side effects and quality of life among cancer survivors. Curr Sports
Med Rep. 2009;8:325-330.
Spence RR, Heesch KC, Brown WJ. Exercise and cancer rehabilitation:
a systematic review. Cancer Treat Rev. 2010;36:185-194.
Mustian KM, Sprod LK, Janelsins M, et al. Exercise recommendations
for cancer-related fatigue, cognitive impairment, sleep problems,
depression, pain, anxiety, and physical dysfunction: a review. Oncol
Hematol Rev. 2012;8:81-88.
Irwin ML, Crumley D, McTiernan A, et al. Physical activity levels before
and after a diagnosis of breast carcinoma: the Health, Eating, Activity,
and Lifestyle (HEAL) study. Cancer. 2003;97:1746-1757.
Courneya KS, Friedenreich CM. Relationship between exercise pattern
across the cancer experience and current quality of life in colorectal
cancer survivors. J Altern Complement Med. 1997;3:215-226.
Mustian KM, Griggs JJ, Morrow GR, et al. Exercise and side effects
among 749 patients during and after treatment for cancer: a University of Rochester Cancer Center Community Clinical Oncology
Program Study. Support Care Cancer. 2006;14:732-741.
Mustian KM, Peppone LJ, Palesh OG, et al. Exercise and cancer-related
fatigue. US Oncol. 2009;5:20-23.
Mustian KM, Morrow GR, Carroll JK, et al. Integrative nonpharmacologic behavioral interventions for the management of
cancer-related fatigue. Oncologist. 2007;12(Suppl 1):52-67.
Bellizzi KM, Mustian KM, Bowen DJ, et al. Aging in the context of
cancer prevention and control: perspectives from behavioral medicine. Cancer. 2008;113(Suppl 12)3479-3483.
Brown JC, Huedo-Medina TB, Pescatello LS, et al. Efficacy of exercise
interventions in modulating cancer-related fatigue among adult
cancer survivors: a meta-analysis. Cancer Epidemiol Biomarkers Prev.
2011;20:123-133.
Irwin ML, Alvarez-Reeves M, Cadmus L, et al. Exercise improves body
fat, lean mass, and bone mass in breast cancer survivors. Obesity
(Silver Spring). 2009;17:1534-1541.
Gleeson M, Bishop NC, Stensel DJ, et al. The anti-inflammatory effects
of exercise: mechanisms and implications for the prevention and
treatment of disease. Nat Rev Immunol. 2011;11:607-615.
87.
Tang M-F, Liou T-H, Lin C-C. Improving sleep quality for cancer patients: benefits of a home-based exercise intervention. Support Care
Cancer. 2010;18:1329-1339.
88. Salmon P. Effects of physical exercise on anxiety, depression, and
sensitivity to stress: a unifying theory. Clin Psychol Rev. 2001;21:33-61.
89. Betof AS, Dewhirst MW, Jones LW. Effects and potential mechanisms
of exercise training on cancer progression: a translational perspective.
Brain Behav Immun. 2013;30(Suppl):S75-S87.
90. Bittoni MA, Harris RE, Buckworth J, et al. Abstract 5043: Physical
activity and the risk of lung cancer death: results from the Third
National Health and Nutrition Examination Survey. Cancer Res. 2014;
74:5043.
91. Schmitz KH, Courneya KS, Matthews C, et al; American College of
Sports Medicine. American College of Sports Medicine roundtable on
exercise guidelines for cancer survivors. Med Sci Sports Exerc. 2010;
42:1409-1426.
92. U.S. Department of Health and Human Services. Physical activity
guidelines Advisory Committee Report, 2008. Part A: executive
summary. Nutr Rev. 2009;67:114-120.
93. Haskell WL, Lee IM, Pate RR, et al. Physical activity and public health:
updated recommendation for adults from the American College of
Sports Medicine and the American Heart Association. Med Sci Sports
Exerc. 2007;39:1423-1434.
94. Haskell WL, Lee IM, Pate RR, et al; American College of Sports
Medicine; American Heart Association. Physical activity and public
health: updated recommendation for adults from the American
College of Sports Medicine and the American Heart Association.
Circulation. 2007;116:1081-1093.
95. Irwin ML. Physical activity interventions for cancer survivors. Br J
Sports Med. 2009;43:32-38.
96. Jones LW, Courneya KS. Exercise discussions during cancer treatment
consultations. Cancer Pract. 2002;10:66-74.
97. Mustian KM, Griggs JJ, Morrow GR, et al. Exercise and side effects
among 749 patients during and after treatment for cancer: a University of Rochester cancer center community clinical oncology
program study. Support Care Cancer. 2006;14:732-741.
98. Yates JS, Mustian KM, Morrow GR, et al. Prevalence of complementary
and alternative medicine use in cancer patients during treatment.
Support Care Cancer. 2005;13:806-811.
99. Sprod LK, Peppone LJ, Palesh OG, et al. Effect of timing of information
on exercise behavior during cancer treatment: a URCC CCOP protocol.
J Clin Oncol. 28:15s (suppl; abstr 9138).
100. Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports
Exerc. 1982;14:377-381.
101. Courneya KS, Segal RJ, Mackey JR, et al. Effects of aerobic and
resistance exercise in breast cancer patients receiving adjuvant
102.
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
chemotherapy: a multicenter randomized controlled trial. J Clin

Oncol. 2007;25:4396-4404.
Ahmed RL, Thomas W, Yee D, et al. Randomized controlled trial of
weight training and lymphedema in breast cancer survivors. J Clin
Oncol. 2006;24:2765-2772.
Courneya KS, Segal RJ, Gelmon K, et al. Six-month follow-up of patientrated outcomes in a randomized controlled trial of exercise training
during breast cancer chemotherapy. Cancer Epidemiol Biomarkers
Prev. 2007;16:2572-2578.
Schmitz KH, Ahmed RL, Hannan PJ, et al. Safety and efficacy of weight
training in recent breast cancer survivors to alter body composition,
insulin, and insulin-like growth factor axis proteins. Cancer Epidemiol
Segal RJ, Reid RD, Courneya KS, et al. Resistance exercise in men
receiving androgen deprivation therapy for prostate cancer. J Clin
Oncol. 2003;21:1653-1659.
Winters-Stone KM, Dobek J, Nail L, et al. Strength training stops bone
loss and builds muscle in postmenopausal breast cancer survivors:
a randomized, controlled trial. Breast Cancer Res Treat. 2011;127:
447-456.
Mustian KM, Janelsins M, Sprod L, et al. YOCAS Yoga significantly
improves circadian rhythm, anxiety, mood and sleep: a randomized,
controlled clinical trial among 410 cancer survivors. Support Care
Cancer. 2011;19(suppl 2):S317-S318.
Reid-Arndt SA, Matsuda S, Cox CR. Tai chi effects on neuropsychological, emotional, and physical functioning following cancer
treatment: a pilot study. Complement Ther Clin Pract. 2012;18:26-30.
Crevenna R, Schmidinger M, Keilani M, et al. Aerobic exercise for a
patient suffering from metastatic bone disease. Support Care Cancer.
2003;11:120-122.
Headley JA, Ownby KK, John LD. The effect of seated exercise on
fatigue and quality of life in women with advanced breast cancer.
Oncol Nurs Forum. 2004;31:977-983.
Porock D, Kristjanson LJ, Tinnelly K, et al. An exercise intervention for
advanced cancer patients experiencing fatigue: a pilot study. J Palliat
Care. 2000;16:30-36.
Oldervoll LM, Loge JH, Paltiel H, et al. The effect of a physical exercise
program in palliative care: a phase II study. J Pain Symptom Manage.
2006;31:421-430.
Adamsen L, Midtgaard J, Rorth M, et al. Feasibility, physical capacity,
and health benefits of a multidimensional exercise program for cancer
patients undergoing chemotherapy. Support Care Cancer. 2003;11:
707-716.
Oldervoll LM, Loge JH, Lydersen S, et al. Physical exercise for cancer
patients with advanced disease: a randomized controlled trial.
Oncologist. 2011;16:1649-1657.
249
PEDIATRIC ONCOLOGY
Controversies in Germline
Genetic Testing and Disclosure
in Pediatric Oncology
CHAIR
Kim E. Nichols, MD, PhD
Childrens Hospital of Philadelphia
Philadelphia, PA
SPEAKERS
Angela R. Bradbury, MD
Perelman School of Medicine at the University of Pennsylvania
Philadelphia, PA
Lainie Friedman Ross, MD, PhD
The University of Chicago Medicine
Chicago, IL
PEDIATRIC GENETIC TESTING FOR CANCER
The Advantages and Challenges of Testing Children for

Heritable Predisposition to Cancer
Chimene Kesserwan, MD, Lainie Friedman Ross, MD, PhD, Angela R. Bradbury, MD, and Kim E. Nichols, MD
OVERVIEW
The increased application of germline genetic testing is expanding our understanding of the risk factors associated with
childhood cancer development, and, in some cases, such testing is also informing clinical management. Nonetheless, the
incorporation of genetic testing into the pediatric oncology setting is complex and associated with many ethical and practical
challenges. The decision as to whether to pursue clinical genetic testing for hereditary cancer predisposition for children should
always be guided by the best interest of the child. Despite this fundamental ethical principle, patients, parents, and health care
providers may differ in their opinions. Clinical genetic testing to detect the presence of predisposition syndromes associated
with childhood-onset cancers, particularly those for which surveillance and preventive measures have proven to enhance
outcome, is currently well accepted. On the other hand, clinical genetic testing of children for syndromes associated with adultonset cancers has raised many concerns about the potential for psychological harm and disrespect of patient autonomy. As
a consequence, such testing is not encouraged. The challenges surrounding germline genetic testing are further complicated
when testing is done in the research setting and/or when it involves whole-exome or whole-genome sequencing approaches,
which can uncover genetic variants that may or may not be associated with the disease under study. Accordingly, there is great
debate around these processes and the most appropriate approaches regarding the return of test results. Future research is
needed to enhance knowledge about how best to incorporate genomic information into clinical practice.
he increased incorporation of genetic and genomic

testing in the pediatric oncology setting, including the
use of larger and more comprehensive multigene panels and
whole-exome sequencing, has greatly facilitated the identification of children with underlying predisposition syndromes. In this article, we discuss the advantages and
complexities related to germline genetic testing, with an
emphasis on the evaluation of children for heritable cancer
risk. We provide an ethical framework with which to approach clinical genetic testing in children and then discuss
the advantages and challenges related to testing children in
the clinical and research settings.
ETHICAL FRAMEWORK FOR APPROACHING

CLINICAL CANCER GENETIC TESTING IN
PEDIATRICS
Professional guidelines regarding the clinical genetic testing
of children all agree that the guiding principle should be this
standard: the best interest of the child.1-8 What this principle
means and how it should be applied in different clinical
settings are less clear. Although physicians may know what is
medically best for the child, the parents calculation considers

what is best for the child by incorporating and balancing many
types of needs and interests. That is, parents consider not only
the childs medical needs but also the childs social, cultural,
religious, and economic needs and interests. Ethical controversies related to clinical genetic testing of children include
the following: (1) whether to focus exclusively on the childs
self-regarding interests (e.g., concerns about the childs own
medical well-being) or to include other-regarding interests
(e.g., concerns about the needs and interests of family
members who may be affected by the information learned
from the childs genetic test results); (2) how to balance the
childs present-day needs and interests with his or her future
needs and interests; and (3) whether to include the childs
present and future autonomy interests (as well as present
and future privacy rights).8-10
In the context of clinical genetic testing of children, the
childs best interest will focus on the diagnosis of conditions
that require immediate attention. Traditionally, a physician
orders one or more genetic tests targeted to diagnose a
childs symptoms or phenotype. The increasing use of genomic sequencing has increased the number of successful
From the Department of Oncology, St. Jude Childrens Research Hospital, Memphis, TN; Departments of Pediatrics, Medicine, and Surgery, MacLean Center for Clinical Medical Ethics,
The University of Chicago, Chicago, IL; Department of Medicine, Department of Medical Ethics and Health Policy, Perelman School of Medicine of the University of Pennsylvania,
Philadelphia, PA.
Corresponding author: Kim E. Nichols, MD, 262 Danny Thomas Place, Memphis, TN 38105; email: kim.nichols@stjude.org.
251
KESSERWAN ET AL
diagnoses but has also resulted in discoveries of variants of

unknown significance and secondary and incidental findings. We provide an ethical framework for clinical genetic
and genomic testing in minors, examine scenarios that involve
predictive genetic testing in minors for cancer-predisposing
conditions by using single-gene testing, and then consider
scenarios in which genomic sequencing, rather than targeted
testing, is used.
Predictive Testing for Cancer Predisposition

Syndromes That Present in Childhood
In 2012, the American Academy of Pediatrics (AAP) and
the American College of Medical Genetics and Genomics
(ACMG) published a joint policy statement and technical
report on clinical genetic testing in children. 4,5 Earlier
statements made by these organizations objected to the
predictive testing in children1,3 but did not distinguish between testing for conditions that present in childhood and
those that present in adulthood. The timing of when a
condition is expected to present is crucial, because it affects
the relative importance of the childs role in the decisionmaking process, whether the child has a right to privacy from
his or her parents, and whether the parents have a right to
make such decisions on behalf of their child.
Of the cancer-predisposing conditions that present in childhood, some will be early-onset (e.g., hereditary retinoblastoma), and others will present in adolescence or young
adulthood (e.g., hereditary paraganglioma/pheochromocytoma
syndrome [HPPS]). Both of these are autosomal dominant,
therefore a child whose parent had one of these conditions is at
KEY POINTS
252
The guiding principle regarding clinical genetic and/or

genomic testing of children should always be the best
interest of the child.
Diagnostic genetic testing of a child with cancer can
guide disease management and allow for the initiation
of surveillance for second primary cancers.
Predictive genetic testing for an underlying
predisposition syndrome can identify children at
increased cancer risk and allow for patient and family
education, behavioral modification, and cancer
surveillance. Importantly, such testing can also
eliminate these procedures for children whose genetic
test results are negative.
Cancer genetic testing of children should take into
account the availability of evidence-based cancer
surveillance and risk-reduction strategies and the
probability of developing a malignancy during
childhood.
Whether to return genetic research results to research
participants is a subject of active debate. A clear plan for
return of genetic research results to research
participants should be formulated prior to recruitment
and explained during the informed consent process.
50% risk of inheriting a disease-causing mutation in RB1 or in one

of the genes associated with HPPS. Clinically, one needs to know
whether a child carries an RB1 mutation in infancy because the
cancer usually presents before age 5. In contrast, tumors associated with HPPS do not generally appear before adolescence,
so genetic testing is recommended at approximately age 10 or 5
to 10 years before the earliest HPPS-associated cancer identified
in the family.
Consider, then, the case of a child whose father developed a
paraganglioma at age 25 and was found to carry a causal
mutation in one of the HPPS genes. It is recommended to start
screening for HPPS-associated tumors at age 10 in mutationpositive children.11 Before the HPPS-associated genes were
discovered, it was advocated that all children whose parents had manifestations of HPPS begin lifelong biochemical
screening for abnormally high levels of catecholamines and
undergo annual physical examinations, starting at age 10, with
attention to the symptoms associated with paragangliomas
or pheochromocytomas, such as high blood pressure, heart
palpitations, anxiety, or headache.12 Today, only those individuals with germline HPPS-associated genetic mutations
need to undergo surveillance including MRI screening.11,12
What if the father wanted to test the child for the presence
of HPPS mutation when he was an infant? The arguments
to support waiting until the child is 10 years old are to avoid
demands for earlier initiation of annual surveillance screening; to reduce parental anxiety; and to allow the child to
participate in the decision-making process. However, all three
arguments can be refuted. First, the data find that parents rarely demand inappropriate treatments or initiation
of therapies earlier than necessary.13 Second, testing is only
done in children with a known family history, so parental
anxiety already exists, and testing would allow half of the
parents to be reassured.5,8,13 Furthermore, data show that
many individuals find uncertainty worse than a positive test
result.14,15 Third, although ethics supports the inclusion of
children to the extent that they are able to assent, the child in
this scenario does not really have a choice. Although mature
children can refuse to participate in many types of research,
refusal by a child of clinically indicated care is more restricted
when diagnosis, surveillance, and treatment can prevent
serious morbidity or mortality. This is not to ignore the mature
child, but respect can be shown in other ways (e.g., explanations of what at-risk status means and what the tests can
determine). Of note, given that the child in question is at 50%
risk of inheriting a syndrome in which tumors can be successfully treated if diagnosed early, his parents right to refuse is also restricted, given the states obligation, as parens
patriae, to protect children from abuse and neglect. If the
child and his parents both refuse genetic testing, they could
opt for regular screening (perhaps not the best approach, but
one that can be considered good enough).16
The arguments to allow parents to test a young child before
it is clinically indicated are to reduce parental anxiety and to
permit them to make other life plans for the child and the
family. The AAP/ACMG guidance gives parents wide leeway for
conditions that present at any point in childhood, because
parents are the presumptive decision makers for their minor

children.5 Genetic counseling can help parents decide when
this testing is appropriate for their child, and the decision to
test the child in this scenario at either age should be respected.
Predictive Testing for Cancer Predisposition

Syndromes That Present in Adulthood
Some parents seek genetic testing of minors for adult-onset
conditions, although surveillance is not needed until adulthood. The 2012 AAP/ACMG statement recommends against
predictive genetic testing for adult-onset disorders, because
the information is not necessary in childhood and because a
delay of testing allows the child to make this decision as an
adult for himself or herself (the concept of a childs right to an
open future17,18); respects the right to privacy that an adult
has, even against his or her own parents; and avoids unnecessary anxiety created by the time span between testing
and clinical relevance.5 The AAP/ACMG proscription is not
absolute but, rather, acknowledges that exceptional circumstances exist. It permits predictive genetic testing to
resolve disabling parental anxiety or to support life-planning
decisions that parents sincerely believe to be in the childs
best interest particularly when parents and mature adolescents jointly express interest in proceeding.5
Genomic Sequencing for Cancer Predisposition and

Other Conditions
The 2012 AAP/ACMG statement did not address the use of
genomic sequencing because this test was still considered a
research tool and the guidance focused on clinical care. One
month later, however, the ACMG unilaterally proposed that,
whenever clinical genomic sequencing was performed, the
sample should be interrogated opportunistically for 56 genes
that are highly penetrant in high-risk populations regardless
of the reason for sequencing, the age of the patient, or the
consent of the patient and physician.19
The Association of Molecular Pathologists objected because interrogating a sample for these 56 genes requires
additional time, resources, and expertise, and the ACMG
policy required no ones permission, so it was not clear who
was ordering the tests, who was willing to receive and
convey the results to patients and families, and who was
going to pay for this analysis.20 Others involved in pediatrics
and genetics objected to such testing because most of the 56
genes were associated with conditions that would not
present until adulthood,21-23 and the AAP/ACMG consensus
was that it was not in the childs best interest to routinely
test for adult-onset conditions.4,5 The ACMG responded that
such testing was indeed in a childs best interest because it
could identify disease-causing mutations that may be
present in their parents.24 One year later, the ACMG revised
the recommendations to allow patients (and surrogates) to
opt out of the screening of these 56 genes and to allow
health care professionals to limit the amount of information
requested.25
As genomic sequencing becomes cheaper, faster, and

more readily available, questions will continue to emerge
about its appropriate use. In 2015, the American Society of
Human Genetics published updated guidelines for genetic
testing of minors that examined single-gene and genomic
sequencing methodologies for both the clinical and research
settings.8 The American Society of Human Genetics concluded that targeted tests, or selective sequence analysis,
are usually preferable to less-discriminate data acquisition
when the clinical challenge can be addressed through a
targeted approach.8 However, even if the ACMG list of 56
genes is not intentionally interrogated when a childs blood
is sequenced, and even if sequencing is focused on analyzing
genes that are relevant to the childs condition, testing may
still identify incidental findings (such as variants in genes that
predispose to other health problems or variants that are only
relevant to reproductive decision making [e.g., carrier status]). When the sample comes from a child, some seek to
limit disclosure of findings to those conditions that may
manifest in childhood, whereas others seek to limit disclosure
to childhood conditions for which early surveillance or treatment can affect outcomes.23,26-30 Some support a policy of
strict nondisclosure of incidental findings to minors and their
parents, but others support pre- and post-test counseling and
disclosure of incidental findings as agreed upon by parents,
physicians, and when possible, the child.22,26-30 The clinical,
psychological, and economic costs and benefits of these
various approaches on families and on the health care system
remain uncertain and will need to be carefully evaluated.
CLINICAL FRAMEWORK FOR APPROACHING

CANCER GENETIC TESTING IN PEDIATRICS
The germline genetic testing of children with cancer is
primarily pursued for diagnostic or predictive purposes.
Diagnostic testing can explain tumor formation, provide
prognostic information, and guide decisions about cancer
treatment. It is recommended that patients with cancer
and Li-Fraumeni or Gorlin syndrome for example, who
harbor germline mutations in TP53 or PTCH1 respectively,
not be exposed to ionizing radiation because of the increased risk for secondary radiation-induced cancers.
Simiarly, patients with cancer and genetic syndromes
associated with defective DNA repair, such as Fanconi
anemia, should not be treated with conventional doses of
myelosuppressive chemotherapy and/or radiation because
of the excess toxicity associated with these approaches.
Genetic information can also guide the surgical approach
to cancer treatment, when no surgery or organ-sparing
surgery of paired at risk organs (e.g., the eyes in children
with hereditary retinoblastoma or the kidneys in children
with WT1-associated hereditary Wilms tumor) is preferred
over removal of an entire organ.
Predictive testing derives its benefit from the tenet that
early detection of tumors will lead to prevention or surveillance interventions that reduce overall morbidity and
mortality. In light of this issue, the American Society of
253
KESSERWAN ET AL
TABLE 1. Cancer Predisposition Syndromes and Associated Physical Findings

Syndrome
Physical Findings
Disorders Associated With Defects in DNA Repair

Ataxia Telangiectasia
Ataxia
Telangiectasia
Cafe-au-lait macules
Bloom Syndrome
Small size for age

Erythematous skin lesions on the nose and cheeks (butterfly rash)
Underdeveloped malar and lower mandibular areas
Fanconi Anemia
Radial ray anomalies: malformations of thumbs, forearms

Short stature
Pigmentary skin changes: cafe-au-lait macules, hypopigmentation
Other anomalies: small eyes, narrow ear canal and abnormal pinna, micropenis
Nijmegen Breakage Syndrome
Microcephaly
Growth retardation
Rothmund-Thomson Syndrome
Poikyloderma
Dysplastic nails
Rudimentary or hypoplastic teeth
Radial ray anomalies
Acral keratosis
Werner Syndrome
Short stature
Premature graying of the hair
Xeroderma Pigmentosum
Acute sun sensitivity

Marked freckle-like pigmentation
Severe keratitis
Photophobia
Squamous cell cancer of skin
Acquired microcephaly
Inherited Bone Marrow Failure Syndromes

Diamond-Blackfan Anemia
Microcephaly
Growth retardation
Ocular hypertelorism, ptosis
Congenital glaucoma, congenital cataract
Webbed neck, short neck
Klippel-Feil anomaly (short neck, low hair line and/or limited range of motion in the neck due to
fusion of cervical vertebrae)
Sprengel deformity (high scapula)
Flat thenar eminence
Triphalangeal, bifid, hypoplastic, or absent thumb
Dyskeratosis Congenita
Dystrophic nails
Dental abnormalities (hypodontia, early tooth loss)
Lacy reticular pigmentation of the upper chest and/or neck
Alopecia
Premature graying of the hair
GATA2 Deficiency
Cutaneous warts
Lymphedema of extremities
Shwachman-Diamond Syndrome
Short stature
Failure to thrive and poor growth
Continued
254
TABLE 1. Cancer Predisposition Syndromes and Associated Physical Findings (Contd)

Syndrome
Physical Findings
Neurocutaneous Syndromes
Neurofibromatosis Type 1
Freckling in the axillary and/or inguinal regions
Hypopigmented macules
Lisch nodules
Macrocephaly
Neurofibromas
Short stature
Tibial pseudarthrosis
Neurofibromatosis Type 2
Cutaneous schwannomas
Tuberous Sclerosis Complex
Hypomelanotic macules
Angiofibromas
Shagreen Patch
Confetti skin lesions
Dental enamel pits
Ungual fibromas
Neuroendocrine Syndromes
Carney Complex
Spotty skin pigmentation of lips, conjunctiva and inner or outer canthi, vaginal and penile mucosal
surfaces
Blue nevi
Cutaneous and mucosal myxomas
Acromegaly
Multiple Endocrine Neoplasia Type 1
Cutaneous collagenomas
Facial angiofibromas
Multiple Endocrine Neoplasia Type 2B
Marfanoid habitus
Neuromas of lips and tongue
Overgrowth Syndromes
AIP-Related Familial Isolated Pituitary
Adenomas
Gigantism or acromegaly
Beckwith-Wiedemann Syndrome
Macroglossia
Macrosomia
Ear lobe creases, helical pits
Hemihyperplasia
Omphalocele
Umbilical hernia
Diastasis recti
Nevoid Basal Cell Carcinoma Syndrome

(Gorlin Syndrome)
Coarse facial features, frontal bossing

Macrocephaly
Palmar and plantar pits
Preaxial or postaxial polydactyly
Basal cell carcinomas
PTEN Hamartoma Tumor Syndrome
Macrocephaly
Trichilemmomas
Lipomas
Oral mucosal papillomas
Acral keratosis
Hemangiomas
Penile freckling
Continued
255
KESSERWAN ET AL

Syndrome
Simpson-Golabi-Behmel Syndrome
Physical Findings
Macrocephaly
Macrosomia
Sotos Syndrome
Long narrow face with bitemporal narrowing

Long chin
Overgrowth
Sparse frontotemporal hair
Intestinal Polyposis Syndromes

Constitutional Mismatch Repair Defect
Hypopigmented macules
Familial Adenomatous Polyposis Syndrome
Dental abnormalities: missing or supernumerary teeth

Epidermoid cyst
Osteomas of the skull and/or mandible
Pilomatricomas
Peutz-Jeghers Syndrome
Melanotic macules around the mouth, eyes, nostrils, and perianal area
RASopathies
Cardiofaciocutaneous Syndrome
Coarse facial features

Bitemporal narrowing
Hypertelorism, downslanting palpebral fissures, and hypoplasia of the supraorbital ridges
Low set ears, ear lobe creases
Macrocephaly relative to body size
Short neck
Short stature
Keratosis pilaris
Lymphedema
Costello Syndrome

Curly or sparse fine hair
Deep palmar or plantar creases
Friendly, sociable personality
Full lips, large mouth
Papillomata of the face and anal region
Short stature
Ulnar deviation of wrists and fingers
Noonan Syndrome

Downslanted palpebral fissures
Vivid blue or blue-green irises
Fullness of upper eyelids and ptosis
Low set posteriorly rotated ears with fleshy helices
Broad or webbed neck
Short stature
Pectus deformity of chest
Widely set nipples
Cryptorchidism
Noonan Syndrome With Multiple Lentigines

Downslanted palpebral fissures
Widely spaced eyes and ptosis
Low set ears, posteriorly rotated with thick helices
Lentigines
Continued
256

Syndrome
Physical Findings
Short stature
Cryptorchidism
Disorders of Sex Chromosomes

Klinefelter Syndrome
Gynecomastia and/or small genitalia

Tall stature
Turner Syndrome
Short stature
Webbed neck
Low hair line
Shield-like chest
Wide-spaced nipples
Lymphedema of hands and feet
WT1-Related Disorders
Denys-Drash Syndrome
Ambiguous genitalia
Frasier Syndrome
Ambiguous genitalia
Wilms Tumor, Aniridia, and Growth

Retardation Syndrome
Aniridia
Ambiguous genitalia
Other
Rubinstein Taybi Syndrome
Convex nasal bridge with low hanging columella

Downslanting palpebral fissures
Broad radially deviated thumbs
Broad terminal phalanges
Grimacing smile
Pilomatricomas
Talon cusps (extra cusp-like projection of the primary or permanent anterior teeth)
Undescended testis
Clinical Oncology (ASCO) advises that cancer genetic testing of children take into account the availability of evidencebased risk-reduction strategies and the probability of
developing a malignancy during childhood. 31 Currently,
there are only a limited number of conditions for which effective
preventive measures are known to exist. One such example is
multiple endocrine neoplasia type 2 (MEN2), which is caused by
germline RET gene mutations. Predictive RET gene testing is the
clinical standard of care for all individuals with a positive family
history of MEN2 because of the very high risk to develop earlyonset medullary thyroid cancer in affected individuals, including
children. Prophylactic thyroidectomy during childhood is
therefore indicated for anyone who harbors a pathogenic gainof-function RET gene mutation because this procedure greatly
reduces or even eliminates the risk of developing thyroid
cancer. Risk stratification that is based on RET genotype
can inform decisions about the age to perform the prophylactic
thyroidectomy.32 The identification of a predisposing germline
mutation also guides strategies for cancer monitoring. Although
surveillance protocols are under investigation for many conditions, effective approaches do exist for certain conditions,
such as familial ademonatous polyposis, Beckwith Wiedeman
syndrome, Li-Fraumeni syndrome, and the WT1-associated
Wilms tumor syndromes, among others.33-38
Factors to Consider When Pursuing Clinical Cancer

Genetic Testing in Children
The factors most commonly used to prompt referral of a
child to a cancer genetics specialist include a personal
medical or family history consistent with, or suggestive of,
an underlying hereditary cancer syndrome. The personal
features that most support an underlying cancer predisposition in a child include the presence of: a clinical or
molecular diagnosis of a known predisposition syndrome;
very early onset or adult-type tumors (e.g., epithelial breast
or colon cancer); multifocal tumors; bilateral involvement of
paired organs (e.g., eyes, kidneys, adrenal glands, breasts);
specific cancer types and histologic features (e.g., medullary
thyroid cancer and MEN2); second primary neoplasms; and
specific physical findings (examples of these physical findings are provided as a reference and listed in Table 1).
The family history is also a key factor that helps guide
a cancer genetics evaluation. ASCO has recognized the
importance of an adequate family history in the identification of patients whose cancers may be associated with
inherited genetic factors. For each relative of a child with
cancer, it is important to record whether that relative developed cancer and, if so, to also record the type of cancer
and its histologic features; the age at cancer diagnosis; laterality;
maternal or paternal lineage relationships of affected relatives
257
KESSERWAN ET AL
TABLE 2. Family History Features Suggestive of an

Underlying Cancer Predisposition
Feature
Findings
Pattern of
Cancer
Occurrence
Three generations affected by cancer (on the same

side of the family)
Three or more family members with the same type of
cancer (on the same side of the family)
One or more first-degree relatives with cancer (parent,
child, sibling)
Age of Cancer
Onset
Earlier than expected age of onset for cancer type
Cancer Type/
Adult type cancers in children (e.g., epithelial cancers
Presentation
such as adenocarcinoma of the breast, ovary, and
colon)
Constellation of tumors consistent with a familial
syndrome (e.g., breast and ovarian cancer in
hereditary breast and ovarian cancer syndrome;
uterine and colon cancer in Lynch syndrome)
Multiple primary cancers
Bilateral or multifocal cancers
Ethnic
Background
Ashkenazi Jewish background

Consanguinity
Other Features Clinical or molecular diagnosis of a predisposition

syndrome in a close relative
with cancer; ethnicity; and results of prior genetic testing.39 A

minimum adequate family history should include cancer in firstand second-degree relatives of the proband and no less than a
three-generation pedigree. Table 2 lists the family history
features suggestive of an underlying predisposition.
Guidelines for the Referral of Children for Clinical

Cancer Genetic Testing
The ACMG and National Society of Genetic Counselors
(NSGC) have issued practice guidelines to facilitate identification and referral of adults with cancer and at-risk
individuals to a cancer genetics specialist. 40 Currently,
guidelines dedicated to the evaluation of children are
lacking. Tables 3, 4, and 5 provide practical information to aid
in the genetics referral of children with liquid, central
nervous system (CNS), and non-CNS solid tumors.
Challenges Associated With Cancer Genetic Testing in

Children
Despite the benefits of childhood cancer genetic testing,
several challenges remain. First, providers often lack the
time, tools, or expertise to collect and meaningfully interpret
family history information. Furthermore, it is important to
recognize that family histories may appear negative because
of inaccurate or inadequate medical information and because of incomplete penetrance or presence of de novo
germline mutations. Providers must not overlook the possibility of an underlying syndrome, even if the family history
appears negative. This is particularly true for children who
have cancers that are known to have a heritable basis in a
clinically relevant proportion of cases (e.g., retinoblastoma,
adrenocortical carcinoma, paraganglioma). Second, consensus
258
guidelines for cancer surveillance of many predisposition

syndromes that affect children do not yet exist. Multicenter,
prospective studies to assess the clinical utility of surveillance
protocols are critically needed. Third, there remains concern
that the testing of children for cancer predisposition will cause
adverse psychological outcomes for the child, the parents, and
other family members. Although it has been reported that
some children experience distress, discrimination, guilt, or
regret, most of the published literature indicates that there is
no increase in mean anxiety, depression, and distress scores
as a result of testing.41,42 Fourth, the increasing use of multigene panel and genomic testing has increased the difficulty
level and time required to adequately counsel patients and
parents. A 2015 ASCO policy statement highlights the elements of pretest counseling recommended for multigene
panel testing. In addition to including the general components of traditional counseling, the possibility of detecting
predisposing mutations not suggested by the personal or
family history factors, and the implications of positive results
in lesser-penetrance genes are key points to address.43 The
counseling process should include a discussion of variants of
unknown significance, which remain a source of confusion
for health care providers and patients. Providers should
ensure that families comprehend the concept of variants of
unknown significance and recognize that the presence of
such variants does not generally change clinical management. They should consult the literature and seek the assistance of experts for additional opinions, when necessary.
Given the challenges associated with genetic and genomic
testing, interpretation of test results, and the potential
impacts on clinical management, it is recommended that
children who are candidates for cancer genetic or genomic
testing be referred to physicians and genetic specialists with
clinical expertise and research interests in these subject
areas.
GUIDELINES AND CHALLENGES RELATED TO

RESEARCH GENETIC TESTING AMONG CHILDREN
As large, prospective, cohort studies with banked DNA increasingly have been used to evaluate the effects of genes,
the environment, and lifestyle, there has been increasing
debate about the obligations, if any, to share individual
genetic research results with research participants.44 The
traditional approach has been to not share individual research results with research participants or, in the setting of
pediatrics, with their parents or surrogates.45,46 This traditional approach is rooted in distinctions between the goals of
research and clinical care.27,45,47,48 Classically, the primary
goal of research is to create generalizable knowledge to
ultimately benefit a population as a whole. In contrast, the
goal of clinical care is to benefit an individual patient and
improve health outcomes. Arguments against returning
individual research results include blurring this important
distinction between research and clinical care,27,47-51 the
potential for misunderstanding or over-interpretation of the
clinical significance or limitations of genetic results identified
TABLE 3. Indications for Consideration of Genetics Evaluation in Children With Hematopoietic Malignancies
Type of
Hematopoietic
Malignancy
Consider Referring for Genetics Evaluation

When Any of the Following Is Present
Cancer Predisposition
Syndrome(s) to Consider
Gene(s)
ALL
Family history of LFS-associated tumors
LFS
TP53
Hypodiploid ALL
LFS
TP53
Cafe -au-lait macules
CMMRD
MLH1, MSH2, MSH6, PMS2, EPCAM
Pure red cell aplasia
DBA
RPL11, RPL26, RPL35A, RPL5, RPS7,

RPS10, RPS17, RPS19, RPS24,
RPS26, RPS29, GATA1
Pancreatic insufficiency
SDS
SBDS
Family history of ALL
Familial ALL caused by PAX5

mutations
PAX5
Familial ALL caused by ETV6

mutations
ETV6
Familial ALL caused by SH2B3

mutations
SH2B3
Family history of Lynch syndromeassociated tumors

Consanguinity
Growth failure
Bone marrow failure
Thrombocytopenia
AML/Myelodysplastic Family history of LFS-associated tumors
Syndrome
LFS
TP53
CMMRD
FA
BRCA2, BRIP1, ERCC4, FANCA, FANCB,

FANCC, FANCD2, FANCE, FANCF,
FANCG, FANCI, FANCL, PALB2,
RAD51C, SLX4, XRCC2
Pure red cell aplasia
DBA
RPL11, RPL26, RPL35A, RPL5, RPS7,

RPS10, RPS17, RPS19, RPS24,
RPS26, RPS29, GATA1
Pancreatic insufficiency
SDS
SBDS
Neutropenia
Severe congenital neutropenia/

Kostmann syndrome
CSF3R, ELANE, G6PC3, GF11, HAX1,

JAGN1
Pigmentary changes (neck/chest)
Dyskeratosis congenita
ACD, C16orf57, CTC1, DKC1, NHP2,

NOLA3, PARN, RTEL1, TERT, TERC,
TINF2, WRAP53
Family history of AML
Familial AML caused by CEBPA

mutations
CEBPA
Family history of AML
Familial platelet disorder with

predisposition to acute
myelogenous leukemia
RUNX1
MonoMac syndrome/Emberger
syndrome
GATA2
Rothmund-Thomson syndrome
RECQL4
Family history of Lynch syndromes-associated tumors

Consanguinity
Radial ray defect
Growth and developmental delay
Microcephaly
Skeletal anomalies
Growth failure
Bone marrow failure
Pulmonary fibrosis
Bone marrow failure
Dysplastic nails
Oral leukoplakia
Thrombocytopenia
Cytopenias
Immunodeficiency and increased risk for M. avium
complex infection
Lymphedema
Radial ray defect
Poikiloderma
Sparse hair/nail abnormalities
Continued
259
KESSERWAN ET AL
TABLE 3. Indications for Consideration of Genetics Evaluation in Children With Hematopoietic Malignancies
(Contd)
Type of
Hematopoietic
Malignancy
JMML
Consider Referring for Genetics Evaluation

When Any of the Following Is Present
Syndrome(s) to Consider
Family history and/or features of RASopathies or NF1 Noonan syndrome and related
RASopathies
NF1
Gene(s)
BRAF, CBL, HRAS, KRAS, MAP2K1,
MAP2K2, NRAS, PTPN11, RAF1,
RIT1, SHOC2, SOS1
NF1
Mosaic Monosomy 7
Family history of monosomy 7
Familial mosaic monosomy 7

syndrome
Unknown
Non-Hodgkin
Lymphoma
Males with B-cell lymphoma and a history of EBV

infection and/or family history of XLP features
X-linked lymphoproliferative
disease
SH2D1A
CMMRD
Ataxia-telangiectasia
ATM
Wiskott-Aldrich syndrome
WAS
X-linked agammaglobulinemia
BTK
Family history of Lynch syndromeassociated cancers

Consanguinity
Gait abnormalities
Cutaneous and/or ocular telangiectasia
Males with lymphoma, immunodeficiency, and/or
eczema
Thrombocytopenia, small platelets
Failure to thrive
Males with humoral immunodeficiency
Abbreviations: ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; CMMRD, constitutional mismatch repair deficiency; DBA, Diamond-Blackfan anemia; EBV, EpsteinBarr virus; FA, Fanconi anemia; JMML, juvenile myelomonocytic anemia; LFS, Li-Fraumeni syndrome; NF1, neurofibromatosis type 1; NF2, neurofibromatosis type 2; SDS, ShwachmanDiamond syndrome.
in the research setting,47,48,50 maintenance of privacy and

the right of the participant to not know,52 the prohibitive
costs for research team and biobanks,53-55 and the potential
negative impact on research progress if costs are shifted to
cover return of results at the expense of discovery.27,53,56
Yet, this traditional framework has been questioned over
time, given an increasing appreciation that some genetic
findings identified in the research setting could significantly
affect the health of a research participant.27,45,46,47,57-59
Others have argued that returning individual genetic research results should be considered on the basis of the
principles of beneficence, autonomy, reciprocity, and respect
for persons, and, in pediatrics, that the return of individual
genetic research results is in the best interest of the child.57-64
This debate has become more and more important with
the advent of next-generation sequencing, which allows
multiple genes or the entire genome to be sequenced rapidly
and at relatively accessible costs.48,65-67 Now, there are
numerous ways that potentially clinically significant genetic
findings can be identified in the research setting. First, many
studies designed to evaluate the genetic underpinnings of
disease evaluate a number of genes suspected to be related
to a specific condition. Even if these genes are not known to
have any relationship to the health of research participants
at the time they enroll in the study, the level of evidence may
accumulate over time, such that they become clinically
actionable in the future. Second, a research study may sequence large portions of the genome but only invest in the
interpretation of genetic variants related to the disease of
260
interest. Nonetheless, other potentially actionable mutations (e.g., the 56 highly penetrant genes outlined by the
ACMG) may have been sequenced, but the interpretation
and variant calling required to return individual results to
research participants would require additional resources
beyond the scope of the scientific aims. Third, somatic tumor
profiling (e.g., sequencing of tumors to identify acquired
genomic aberrations that can be used to identify targeted
therapy) can indirectly or directly reveal inherited variants of
clinical significance.66,67 Many somatic sequencing platforms use a normal sample for subtraction analysis to clarify
which variants are tumor specific. Dedicated analysis of the
normal or germline sample may identify inherited mutations. Tumor-only sequencing (those that do not include a
normal or germline sample) may also identify potential
germline mutations, because any variant found in the cancer
could, in theory, reflect variation in the normal DNA. Thus, a
germline susceptibility to cancer may be incidentally identified in the process of cataloging somatic mutations. Given
the increasing range of scenarios in which research can
identify potentially clinically significant genetic variants,
there is an increasing need to clarify what obligations research teams have to return genetic research results to
participants.
Guidelines Regarding Returning Research Results

In 1999, the National Bioethics Advisory Commission advised
against return of incidental research findings, except in rare
TABLE 4. Indications for Consideration of Genetics Evaluation for Children With Central Nervous System, Spinal, or
Intracranial Tumors
Type of Tumor
Astrocytoma
Consider Referring for Genetics Evaluation When Any

of the Following Is Present
Syndrome to Consider
Gene(s)
Anaplastic/high grade and family history of

LFS-associated tumors
LFS
TP53
Subependymal giant cell astrocytoma
Tuberous sclerosis complex
TSC1, TSC2
NF1
NF1
CMMRD
MLH1, MSH2, MSH6, PMS2,

EPCAM
Cortical tuber
NF1 features
Family history of NF1
Family history of Lynch syndrome-associated cancers
Consanguinity
Atypical Teratoid/
Rhabdoid Tumor
Refer all
RTPS
SMARCB1, SMARCA4
Cerebellar
Hemangioblastoma
Refer all
VHL
VHL
Choroid Plexus
Carcinoma
Refer all
LFS
TP53
Endolymphatic Sac
Tumor
Refer all
VHL
VHL
Ependymoma
Spinal ependymoma
NF2
NF2
Glioblastoma
CMMRD

EPCAM
LFS
TP53
Age , 3
NBCCS/Gorlin syndrome
PTCH1, SUFU
LFS
TP53
CMMRD

EPCAM
Family history of Lynch syndrome-associated cancers

Consanguinity
Medulloblastoma
Desmoplastic/extensive nodularity
Macrocephaly
Basal cell carcinoma
Palmar/plantar pitting
Jaw keratocysts
Calcification of the falx
Sonic Hedgehog subtype
Consanguinity
Family history of polyps, early-onset colon cancer
FAP
APC
Fanconi anemia
BRCA2, BRIP1, ERCC4, FANCA,

FANCB, FANCC, FANCD2,
FANCE, FANCF, FANCG, FANCI,
FANCL, PALB2, RAD51C, SLX4,
XRCC2
Clear cell subtype
Predisposition to clear cell

meningiomas
SMARCE1
Macrocephaly
NBCCS (Gorlin syndrome)
PTCH1, SUFU
NF2
NF2
Radial ray defect

Developmental delays
Microcephaly
Skeletal anomalies
Multiple congenital anomalies
Meningioma
Basal cell carcinoma

Palmar/plantar pitting
Jaw keratocysts
Calcification of the falx
Personal/family history of Schwannomas
Continued
261
KESSERWAN ET AL
TABLE 4. Indications for Consideration of Genetics Evaluation for Children With Central Nervous System, Spinal,
or Intracranial Tumors (Contd)
Type of Tumor
Consider Referring for Genetics Evaluation When Any

Gene(s)
Neurofibroma
Refer all
NF1
NF1
Personal/family history of Schwannomas
NF2
NF2
Optic Pathway Glioma
Family history or features of NF1
NF1
NF1
Pineoblastoma
Family history of DICER1-associated tumors
DICER1 syndrome
DICER1
Family history of retinoblastoma
Hereditary retinoblastoma
RB1
Pituitary Blastoma
Refer all
DICER1 syndrome
DICER1
Schwannoma
Refer all
NF2
NF2
Familial schwannomatosis
SMARCB1
Abbreviations: CMMRD, constitutional mismatch repair deficiency; FAP, familial adenomatous polyposis; LFS, Li-Fraumeni syndrome; NBCCS, nevoid basal cell carcinoma syndrome;
NF1, neurofibromatosis type 1; NF2, neurofibromatosis type 2; RTPS, rhabdoid tumor predisposition syndrome; VHL, Von Hippel-Lindau; yo, years old.
circumstances. Since this initial guidance, various organizations, advisory commissions, and scholars have provided
additional guidance statements, which reflect a growing
consensus that there may be some cases in which research participants should be offered the option of learning
about research results that could significantly affect their
health.45,68-72 Although the 2013 ACMG statement proposes
that laboratories have a fiduciary responsibility to seek and
report mutations in a minimum list of 56 genes (23 of which
are cancer related) regardless of test indication or age,
these recommendations have been controversial and were
designed specifically for sequencing in clinical care as opposed to research.21,22,66,73-81 There currently is no consensus list of actionable genes for obligatory disclosure in
the research setting.19,82-84 Later in 2013, the Presidential
Commission for Bioethical Issues published Anticipate and
Communicate: Ethical Management of Incidental and Secondary Findings in the Clinical, Research and Direct-toConsumer Contexts, which provides guiding principles
and general recommendations for the return of incidental
findings in the research setting.85 Key recommendations
included the following: (1) anticipating the potential for
incidental genetic findings; (2) communicating the potential and plans for returning individual results in advance
(e.g., at the time of consent to the study); and (3) honoring
the right of participants to refuse receipt of incidental
genetic research findings. Importantly, the Presidential
Commission rejected an obligation to opportunistically
seek actionable genetic variants (e.g., secondary findings)
in the research setting, highlighting the negative impact
this could have on the research enterprise, in which society
has a significant interest.85 In addition, the Commission
emphasized the need for additional research, deliberation,
and context-specific guidelines from professional organizations. These recommendations are consistent with
recommendations published by other expert groups and
represent a growing consensus that some analytically
valid genetic research findings should be considered for
return to participants and, in pediatrics, to parents and
surrogates.27,45,46,52
262
Remaining Areas of Controversy and Unique

Considerations for Children
Despite this emerging consensus, there remain many areas
of continued debate. There is no consensus on criteria or
settings in which there is an obligation to return actionable research findings to pediatric research participants,
their parents, or their surrogates45,51,52,59,86,87 Although the
clinical utility or actionability of results is often used as a
deciding factor, defining actionability is challenging, particularly as commercial tests become available prior to evidence of clinical utility.88 There are additional complexities
in pediatrics, given that some conditions (e.g., adult-onset
conditions such as adult cancer or Alzheimer disease) may
not be immediately relevant to the child but could be immediately relevant to his or her relatives.46 Although some
have argued that this is sufficient rationale to return results
in the pediatric setting, others have rejected this rationale
for justifying the return.19,27,46 Second, the duration of
obligations has been debated. Can research teams be held
responsible for returning results that may not become actionable until many years after the research was completed
and when they may not have remaining funding or resources
to support the return of results? This is particularly relevant in
pediatric research, because a child may not have the ability to
make an informed decision about receiving research results
at the time of enrollment but may be able to do so in the
future.46 Third, there is debate about whether genetic research results must be conducted or confirmed in a Clinical
Laboratory Improvement Amendments (CLIA)certified laboratory prior to return of results to research participants.
Some groups argue that only select results (e.g., those that are
clearly clinically actionable and have been confirmed in a CLIA
laboratory) should be returned.51,87,89 Yet, many research
investigations occur in non-CLIA certified laboratories, and
such requirements would significantly increase costs, detract
from the research at hand, and create barriers to returning
potentially actionable genetic research findings. Thus, others
argue that genetic research results should be offered to
participants, and, in pediatrics, to their parents or surrogates,
to let them make decisions about which results they would
TABLE 5. Indications for Consideration of Genetics Evaluation for Children With Solid Tumors
Type of Solid Tumor by
Anatomic Location
Consider Referring When Any

Gene(s)
Refer all
HBOC
BRCA1, BRCA2
Hereditary diffuse gastric cancer

syndrome
CDH1
PHTS
PTEN
LFS
TP53
PALB2-associated cancer
syndromes
PALB2
PJS
STK11
NF1
NF1
Breast
Breast Carcinoma
Cardiac
Fibroma
Refer all
PTCH1
Myxoma
Refer all
Carney complex
PRKAR1A
Rhabdomyoma
Refer all
TSC
TSC1/TSC2
Refer all
MEN1
MEN1
NF1
NF1
HDGC
CDH1
Lynch syndrome, CMMRD
FAP syndrome
APC
Gastrointestinal
Carcinoid
Gastric Cancer
GIST
Refer all
Refer all
Peutz-Jeghers syndrome
STK11
Predisposition to GIST
SDHA,SDHB, SDHC,SDHD
Carney-Stratakis syndrome
Pancreatic Cysts/Cancer
Small Bowel, Colon, Rectal Polyps

or Cancer
Refer all
Refer all
NF1
NF1
FAMMM
CDKN2A
VHL
VHL
LFS
TP53
FAP
APC
PJS
STK11
Lynch syndrome, CMMRD
FAP
APC
PJS
STK11
JPS
BMPR1A
JPS
SMAD4
HHT
PHTS
PTEN
MUTYH-associated polyposis
MUTYH
FAP
APC
General
Desmoid Tumor/Desmoid-Type
Fibromatosis
Refer all
Hemangioblastoma
Refer all
VHL
VHL
Myofibroma/Myofibromatosis
Refer all
Familial idiopathic basal ganglia

calcification
PDGFRB
Predisposition to myofibromas
NDRG4
RTPS
SMARCB1, SMARCA4
Rhabdoid Tumor
Refer all
Continued
263
KESSERWAN ET AL
TABLE 5. Indications for Consideration of Genetics Evaluation for Children With Solid Tumors (Contd)
Anatomic Location

Gene(s)
Gonadoblastoma
Refer all
WT1-related disorders
WT1
Gyrandroblastoma
Refer all
DICER1 syndrome
DICER1
Juvenile Granulosa Cell Tumor,

Testis
Refer all
Y chromosome structural
abnormality
Ovarian Sex Cord Tumor With

Annular Tubules
Refer all
PJS
STK11
RTPS
SMARCA4
Genitourinary
Juvenile Granulosa Cell Tumor,

Ovary
Small Cell Carcinoma of the Ovary, Refer all

Hypercalcemic Type
Sertoli-Leydig Cell Tumor, Ovary
Refer all
DICER1 syndrome
DICER1
Sertoli-Leydig Cell Tumor, Testis
Refer all
PJS
STK11
Carney complex
PRKAR1A
Head/Neck
Endolymphatic Sac Tumor
Refer all
VHL
VHL
Facial Angiofibroma
Refer all
TSC
TSC1/TSC2
Nasal Chondromesenchymal
Hamartoma
Refer all
DICER1 syndrome
DICER1
Paraganglioma
Refer all
Hereditary paraganglioma/
SDHA, SDHB, SDHC, SDHD, SDHF, MAX,
pheochromocytoma syndrome
TMEM127
Parathyroid Carcinoma
Thyroid Cancer
Refer all
MEN2
RET
VHL
VHL
NF1
NF1
MEN1
MEN1
CDC73-related disorders
CDC73
Medullary subtype
MEN2
RET
Papillary subtype
DICER1 syndrome
DICER1
PHTS
PTEN
Carney complex
PRKAR1A
Familial nonmedullary thyroid

cancer
HABP2
BWS/isolated hemihyperplasia
chr11p15 abnormalities
Family history of DICER-1

related tumors
Papillary, follicular subtype
Macrocephaly
Learning disorder/autism
spectrum disorder
Penile freckling
Features of Carney complex
Family history of thyroid cancer
Liver
Hepatoblastoma
Refer all
CDKN1C
FAP
APC
Lung
Carcinoid
Refer all
MEN1
MEN1
Mesothelioma
Refer all
DICER1 syndrome
DICER1
Refer all
FAP
APC
Pleuropulmonary Blastoma
Musculoskeletal
Desmoid Tumor/Desmoid-Type
Fibromatosis
Continued
264
Anatomic Location
Osteosarcoma

Gene(s)
Age at diagnosis , 10 years
LFS
TP53
Family history of LFS-associated

tumor
Rhabdomyosarcoma
Family history of retinoblastoma
RB1
Age at diagnosis , 3 years
LFS
TP53
DICER1 syndrome
DICER1
Anaplastic histology
tumors
Embryonal, botryoid subtype
Cervical/bladder
Features of BWS such as overgrowth, BWS/isolated hemihypeplasia
macroglossia, omphalocele,
and/or hemihyperplasia
CMMRD
NF1
NF1
Features of Noonan syndrome
Noonan syndrome
BRAF, CBL, HRAS, KRAS, MAP2K1,

MAP2K2, NRAS, PTPN11, RAF1, RIT1,
SHOC2, SOS1
Ciliary Body Medulloepithelioma
Refer all
DICER1 syndrome
DICER1
Ocular Melanoma
Refer all
FAMMM
CDKN2A/CDK4
CDKN1C

Consanguinity
NF1 features
Family history of NF1
Ocular
BAP1 syndrome
BAP1
Retinal Hemangioblastoma
Refer all
Von Hippel-Lindau syndrome
VHL
Retinoblastoma
Refer all
RB1
Malignant Nerve Sheath Tumors
Refer all
NF1
NF1
Neuroblastoma
Family history of ALK-related

tumors
Predisposition to ALK-related
tumors
ALK
Hypoventilation, Hirschprung
disease
CCHS
PHOX2B
Neurofibroma
Refer all (particularly plexiform

subtype)
NF1
NF1
Schwannoma
Refer all
NF2
NF2
Familial Schwannomatosis
SMARCB1
Peripheral Nervous System
Ganglioneuroblastoma
Ganglioneuroma
Neuroblastoma
Neuroblastoma
Renal/Adrenal
Adrenocortical Carcinoma
Refer all
LFS
TP53
CDKN1C
Angiomyolipoma
Refer all
TSC
TSC1/TSC2
Cystic Nephroma
Refer all
DICER1 syndrome
DICER1
Neuroblastoma (Adrenal)
Family history of ALK-related

tumors
Predisposition to ALK-related
tumors
ALK
Ganglioneuroma
Continued
265
KESSERWAN ET AL
Anatomic Location
Neuroblastoma (Adrenal)

Gene(s)
Hypoventilation, Hirshprung
disease
CCHS
PHOX2B
Refer all
Hereditary PGL/PCC syndrome
SDHA, SDHB, SDHC, SDHD, SDHF, MAX,

TMEM127
MEN2
RET
VHL
VHL
Ganglioneuroma
Pheochromocytoma
HLRCC
FH
Chr11p15 abnormalities
CDKN1C
Renal Cell Carcinoma
Refer all
VHL
VHL
PHTS
PTEN
TSC
TSC1/TSC2
HLRCC
FH
Rhabdoid Tumor
Refer all
Rhabdoid tumor predisposition

syndrome
SMARCB1, SMARCA4
Wilms Tumor
Bilateral
WT1
Hemihyperplasia
Family history of Wilms tumor
WT1
Family history of Wilms tumor
Familial Wilms tumor
REST
Unilateral multifocal
Hypospadias, undescended testis
CDKN1C
Skin
Basal Cell Carcinoma
Refer all
PTCH1
Melanoma
Family history of melanoma
FAMMM
CDKN2A/CDK4
BAP1 syndrome
BAP1

tumors
LFS
TP53
Family history of breast and

ovarian cancer
HBOC
BRCA2
Family history of pancreatic

cancer
Family history of melanoma
Family history of renal cell
carcinoma
Family history of mesothelioma
Abbreviations: BWS, Beckwith-Wiedemann syndrome; CCHS, congenital central hypoventilation syndrome; CMMRD, constitutional mismatch repair deficiency; FAMMM,
familial atypical multiple mole melanoma syndrome; FAP, familial adenomatous polyposis; GIST, gastrointestinal stromal tumor; HDGC, hereditary diffuse gastric cancer; HHT,
hereditary hemorrhagic telangiectasia; HLRCC, hereditary leiomyomatosis and renal cell carcinoma; LFS, Li-Fraumeni syndrome; HBOC, hereditary breast and ovarian cancer;
JPS, juvenile polyposis syndrome; MEN1: multiple endocrine neoplasia type 1; MEN2, multiple endocrine neoplasia type 2; NBCCS, nevoid basal cell carcinoma syndrome; NF1,
neurofibromatosis type 1; NF2, neurofibromatosis type 2; PCC, pheochromocytoma; PGL, paraganglioma; PHTS, PTEN hamartoma tumor syndrome; PJS, Peutz-Jeghers syndrome;
RTPS, rhabdoid tumor predisposition syndrome; TSC, tuberous sclerosis complex; VHL, Von Hippel-Lindau; yo, years old.
like to receive or decline,57,58,72,90,91 with recommendations

for CLIA confirmation of those results most likely to change
care. The costs of confirmatory testing and who should cover
these costs remain topics of active discussion.51,53-55,92
Equally important, little is known about how best to return
genetic research results.27,85,93 On the basis of principles of
autonomy and respect for persons,57-61,63,64,76,80 there is
consensus that research participants, parents, or surrogates should be informed of whether results will, or will not
be returned, what results will be returned, the plans for
266
returning results, and the opportunity to decline receipt of

results for themselves or their children.21,27,86 Given the
lack of consensus on how to best return actionable genetic
research results, and given the concerns about the complexity of genetic information, many research programs
have elected to use genetic providers, such as trained genetic counselors, to communicate individual research
results.27,94-97 However, the costs of this approach are
potentially prohibitory for many research programs, and
there are significant barriers to this approach in large
research cohorts. Alternatively, genetic research results can

be provided to a research participants primary medical
provider for disclosure. Although this may be more cost
effective, most providers do not have the expertise to
interpret or provide medical recommendations related
to genetic research findings, nor do they know how to
facilitate confirmation in a CLIA-certified laboratory if this
is needed. 27 Empirical research needs to develop best
practices for the return of genetic results identified in the
research setting.27,53,55,86,92,98
CONCLUSION
The application of germline genetic and genomic testing
has had a great impact on childhood cancer by providing
improved insights into tumor biology, diagnostic and
prognostic markers, and targets for therapeutic intervention

and/or cancer prevention. Nonetheless, much remains to be
learned about how to best incorporate this information into
clinical care in a meaningful way for children with cancer and
their families. Future research efforts will shed light onto the
important yet poorly understood domains surrounding
medical, psychosocial, and ethical impacts of pediatric genetic and genomic testing.
ACKNOWLEDGMENT
We thank Emily Berenson, Stacy Hines-Dowell, Jessica Valdez,
Rose McGee, Gina Nuccio, Emily Quinn, and Manish Kubal for
assistance in preparing Tables 35. We thank Rose McGee,
Gina Nuccio, and Emily Quinn for their critical review of this
manuscript.
References
1. American Society of Human Genetics (ASHG) Board of Directors. Points to
consider: ethical, legal, and psychosocial implications of genetic testing in
children and adolescents. Am J Hum Genet. 1995;57:1233-1241.
2. Canadian Paediatric Society. Guidelines for genetic testing of healthy
children. Paediatr Child Health. 2003;8:42-52.
3. American Academy of Pediatrics (AAP) Committee on Bioethics.
Ethical issues with genetic testing in pediatrics. Pediatrics. 2001;17:
1451-1455.
4. Committee on Bioethics, Committee on Genetics, American College of
Medical Genetics, et al. Ethical and policy issues in genetic testing and
screening of children. Pediatrics. 2013;131:620-622.
5. Ross LF, Saal HM, David KL, et al. Technical report: Ethical and policy
issues in genetic testing and screening of children. Genet Med. 2013;15:
234-245.
6. Clarke A. The genetic testing of children. J Med Genet. 1994;31:785-797.
7. British Society for Human Genetics. Report on the Genetic Testing of
Children. (Edgbaston, Birmingham, United Kingdom: British Society
of Human Genetics, 2010.) http://www.bsgm.org.uk/media/678741/
gtoc_booklet_final_new.pdf. Accessed December 23, 2015.
8. Botkin JR, Belmont JW, Berg JS, et al. Points to consider: Ethical, legal,
and psychosocial implications of genetic testing in children and adolescents. Am J Hum Genet. 2015;97:6-21.
9. Ross LF. Predictive genetic testing of children and the role of the best
interest standard. J Law Med Ethics. 2013;41:899-906.
10. Salter EK. Deciding for a child: A comprehensive analysis of the best
interest standard. Theor Med Bioeth. 2012;33:179-198.
11. Kirmani S, Young WF. Hereditary Paraganglioma-Pheochromocytoma
Syndromes. In Pagon RA, Adam MP, Ardinger HH, et al (eds). GeneReviews. Seattle, WA: University of WA, 2014.
12. St. Jude Childrens Research Hospital. Hereditary ParagangliomaPheochromocytoma Syndrome. https://www.stjude.org/disease/hereditaryparaganglioma-pheochromocytoma.html. Accessed December 26,
2015.
13. Wilfond B, Ross LF. From genetics to genomics: Ethics, policy, and
parental decision-making. J Pediatr Psychol. 2009;34:639-647.
14. McConkie-Rosell A, Spiridigliozzi GA, Melvin E, et al. Living with genetic
risk: Effect on adolescent self-concept. Am J Med Genet C Semin Med
Genet. 2008;148C:56-69.
15. Jones DL, Clayton EW. The role of distress in uptake and response to
predisposition genetic testing: The BMPR2 experience. Genet Test Mol
Biomarkers. 2012;16:203-209.
16. Buchanan A, Brock DW. Deciding for others. Milbank Q. 1986;64:17-94.
17. Feinberg J. The childs right to an open future. In: Freedom and Fulfillment: Philosophical Essays. Princeton, NJ: Princeton University Press,
1992;76-77, 79, 88, 89.
18. Davis DS. Genetic dilemmas and the childs right to an open future.
Hastings Cent Rep. 1997;27:7-15.
19. Green RC, Berg JS, Grody WW, et al. ACMG recommendations for
reporting of incidental findings in clinical exome and genome sequencing. Genet Med. 2013;15:565-574.
20. Hegde M, Bale S, Bayrak-Toydemir P, et al. Reporting incidental findings
in genomic scale clinical sequencinga clinical laboratory perspective:
a report of the Association for Molecular Pathology. J Mol Diagn. 2015;
17:107-117.
21. Burke W, Antommaria AH, Bennett R, et al. Recommendations for
returning genomic incidental findings? We need to talk! Genet Med.
2013;15:854-859.
22. Allyse M, Michie M. Not-so-incidental findings: the ACMG recommendations on the reporting of incidental findings in clinical whole
genome and whole exome sequencing. Trends Biotechnol. 2013;31:
439-441.
23. Knoppers BM, Avard D, Senecal K, et al. Return of whole-genome
sequencing results in pediatric research: a statement of the P3G international paediatrics platform. Eur J Hum Genet. 2014;22:3-5.
24. American College of Medical Genetics and Genomics. Incidental
findings in clinical genomics: a clarification. Genet Med. 2013;15:
664-666.
25. American College of Medical Genetics and Genomics Board of Directors.
ACMG policy statement: updated recommendations regarding analysis
and reporting of secondary findings in clinical genome-scale sequencing. Genet Med. 2015;17:68-69.
26. American Society of Clinical Oncology. American Society of Clinical
Oncology policy statement update: genetic testing for cancer susceptibility. J Clin Oncol. 2003;21:2397-2406.
27. Jarvik GP, Amendola LM, Berg JS, et al. Return of genomic results to
research participants: the floor, the ceiling, and the choices in between.
Am J Hum Genet. 2014;94:818-826.
267
KESSERWAN ET AL
28. Knoppers BM. Pediatric research and the communication of not-so

incidental findings. Paediatr Child Health. 2012;17:190-192.
29. Hallowell N, Hall A, Alberg C, et al. Revealing the results of wholegenome sequencing and whole-exome sequencing in research and
clinical investigations: some ethical issues. J Med Ethics. 2015;41:
317-321.
30. Borry P, Evers-Kiebooms G, Cornel MC, et al. Genetic testing in
asymptomatic minors: background considerations towards ESHG recommendations. Eur J Hum Genet. 2009;17:711-719.
31. Robson ME, Storm CD, Weitzel J, et al. American Society of Clinical
32. Marquard J, Eng C. Multiple Endocrine Neoplasia Type 2. In Pagon RA,
Adam MP, Ardinger HH, et al (eds). Gene Reviews. Seattle, WA: University of Washington, 2015.
33. Barnard J. Screening and surveillance recommendations for pediatric
gastrointestinal polyposis syndromes. J Pediatr Gastroenterol Nutr.
2009;48:S75-S78.
34. Weksberg R, Shuman C, Beckwith JB. Beckwith-Wiedemann syndrome.
Eur J Hum Genet. 2010;18:8-14.
35. Scott RH, Walker L, Olsen OE, et al. Surveillance for Wilms tumour in atrisk children: pragmatic recommendations for best practice. Arch Dis
Child. 2006;91:995-999.
36. Rao A, Rothman J, Nichols KE. Genetic testing and tumor surveillance for
children with cancer predisposition syndromes. Curr Opin Pediatr. 2008;
20:1-7.
37. Tan TY, Amor DJ. Tumour surveillance in Beckwith-Wiedemann syndrome and hemihyperplasia: a critical review of the evidence and
suggested guidelines for local practice. J Paediatr Child Health. 2006;42:
486-490.
38. Villani A, Tabori U, Schiffman J, et al. Biochemical and imaging surveillance in germline TP53 mutation carriers with Li-Fraumeni syndrome: a prospective observational study. Lancet Oncol. 2011;12:
559-567.
39. Lu KH, Wood ME, Daniels M, et al. American Society of Clinical Oncology
expert statement: collection and use of a cancer family history for
oncology providers. J Clin Oncol. 2014;32:833-840.
40. Hampel H, Bennett RL, Buchanan A, et al. A practice guideline from the
American College of Medical Genetics and Genomics and the National
Society of Genetic Counselors: referral indications for cancer predisposition assessment. Genet Med. 2015;17:70-87.
41. Malkin D. Surveillance for children at genetic risk for cancer: are we
ready? Pediatr Blood Cancer. 2014;61:1337-1338.
42. Wakefield CE, Hanlon LV, Tucker KM, et al. The psychological impact of
genetic information on children: a systematic review. Genet Med. Epub
2016 Jan 7.
44. Kaufman D, Murphy J, Scott J, et al. Subjects matter: a survey of public
opinions about a large genetic cohort study. Genet Med. 2008;10:
831-839.
45. Wolf SM, Crock BN, Van Ness B, et al. Managing incidental findings and
research results in genomic research involving biobanks and archived
data sets. Genet Med. 2012;14:361-384.
46. Abdul-Karim R, Berkman BE, Wendler D, et al. Disclosure of incidental
findings from next-generation sequencing in pediatric genomic research. Pediatrics. 2013;131:564-571.
47. Clayton EW, Ross LF. Implications of disclosing individual results of
clinical research. JAMA. 2006;295:37.
48. McGuire AL, Robinson JO, Ramoni RB, et al. Returning genetic research
results: study type matters. Per Med. 2013;10:27-34.
268
49. Beskow LM. Considering the nature of individual research results. Am J

Bioeth. 2006;6:38-40.
50. Meltzer LA. Undesirable implications of disclosing individual genetic
results to research participants. Am J Bioeth. 2006;6:28-30.
51. Clayton EW, McGuire AL. The legal risks of returning results of genomics
research. Genet Med. 2012;14:473-477.
52. Knoppers BM, Joly Y, Simard J, et al. The emergence of an ethical duty to
disclose genetic research results: international perspectives. Eur J Hum
Genet. 2006;14:1170-1178.
53. Dressler LG, Juengst ET. Thresholds and boundaries in the disclosure of
individual genetic research results. Am J Bioeth. 2006;6:18-20.
54. Bledsoe MJ, Clayton EW, McGuire AL, et al. Return of research results
from genomic biobanks: cost matters. Genet Med. 2013;15:103-105.
55. Bledsoe MJ, Grizzle WE, Clark BJ, et al. Practical implementation issues
and challenges for biobanks in the return of individual research results.
Genet Med. 2012;14:478-483.
56. Klitzman R. Questions, complexities, and limitations in disclosing individual genetic results. Am J Bioeth. 2006;6:34-36.
57. Fernandez CV, Santor D, Weijer C, et al. The return of research results to
participants: pilot questionnaire of adolescents and parents of children
with cancer. Pediatr Blood Cancer. 2007;48:441-446.
58. Ravitsky V, Wilfond BS. Disclosing individual genetic results to research
participants. Am J Bioeth. 2006;6:8-17.
59. Beskow LM, Burke W. Offering individual genetic research results:
context matters. Sci Transl Med. 2010;2:38cm20.
60. Manolio TA. Taking our obligations to research participants seriously:
disclosing individual results of genetic research. Am J Bioeth. 2006;
6:32-34.
61. Fernandez CV, Weijer C. Obligations in offering to disclose genetic
research results. Am J Bioeth. 2006;6:44-46.
62. Fryer-Edwards K, Fullerton SM. Relationships with test-tubes: wheres
the reciprocity? Am J Bioeth. 2006;6:36-38.
63. Sharp RR, Foster MW. Clinical utility and full disclosure of genetic results
to research participants. Am J Bioeth. 2006;6:42-44.
64. Richardson HS, Belsky L. The ancillary-care responsibilities of medical
researchers: an ethical framework for thinking about the clinical care
that researchers owe their subjects. Hastings Cent Rep. 2004;34:25-33.
65. Garraway LA, Verweij J, Ballman KV. Precision oncology: an overview.
J Clin Oncol. 2013;31:1803-1805.
66. Bombard Y, Robson M, Offit K. Revealing the incident alone when
targeting the tumor genome. JAMA. 2013;310:795-796.
67. Stadler ZK, Schrader KA, Vijai J, et al. Cancer genomics and inherited risk.
J Clin Oncol. 2014;32:687-698.
68. Fabsitz RR, McGuire A, Sharp RR, et al. Ethical and practical guidelines
for reporting genetic research results to study participants: updated
guidelines from a National Heart, Lung, and Blood Institute working
group. Circ Cardiovasc Genet. 2010;3:574-580.
69. Wolf SM, Lawrenz FP, Nelson CA, et al. Managing incidental findings in
human subjects research: analysis and recommendations. J Law Med
Ethics. 2008;36:219-248, 211.
70. Illes J, Kirschen MP, Edwards E, et al. Practical approaches to incidental
findings in brain imaging research. Neurology. 2008;70:384-390.
71. Caulfield T, McGuire AL, Cho M, et al. Research ethics recommendations
for whole-genome research: consensus statement. PLoS Biol. 2008;6:
e73.
72. Fullerton SM, Wolf WA, Brothers KB, et al. Return of individual research
results from genome-wide association studies: experience of the
Electronic Medical Records and Genomics (eMERGE) network. Genet
Med. 2012;14:424-431.
73. Ross LF, Rothstein MA, Clayton EW. Mandatory extended searches in all
genome sequencing: incidental findings, patient autonomy, and shared
decision making. JAMA. 2013;310:367-368.
74. Klitzman R, Appelbaum PS, Chung W. Return of secondary genomic

findings vs patient autonomy: implications for medical care. JAMA.
2013;310:369-370.
75. McGuire AL, Joffe S, Koenig BA, et al. Point-counterpoint: ethics and
genomic incidental findings. Science. 2013;340:1047-1048.
76. Wolf SM, Annas GJ, Elias S. Point-counterpoint: patient autonomy
and incidental findings in clinical genomics. Science. 2013;340:
1049-1050.
77. Green RC, Lupski JR, Biesecker LG. Reporting genomic sequencing results to ordering clinicians: incidental, but not exceptional. JAMA. 2013;
310:365-366.
78. Biesecker LG. The Nirvana fallacy and the return of results. Am J Bioeth.
2013;13:43-44.
79. Holtzman NA. ACMG recommendations on incidental findings are
flawed scientifically and ethically. Genet Med. 2013;15:750-751.
80. Townsend A, Adam S, Birch PH, et al. Paternalism and the ACMG
recommendations on genomic incidental findings: patients seen but not
heard. Genet Med. 2013;15:751-752.
81. Korf B. Response to Townsend et al. Genet Med. 2013;15:752-753.
82. Berg JS, Khoury MJ, Evans JP. Deploying whole genome sequencing in
clinical practice and public health: meeting the challenge one bin at a
time. Genet Med. 2011;13:499-504.
83. Berg JS, Amendola LM, Eng C, et al. Processes and preliminary outputs
for identification of actionable genes as incidental findings in genomic
sequence data in the Clinical Sequencing Exploratory Research Consortium. Genet Med. 2013;15:860-867.
84. Klitzman R, Appelbaum PS, Fyer A, et al. Researchers views on return of
incidental genomic research results: qualitative and quantitative
findings. Genet Med. 2013;15:888-895.
85. Presidential Commission for the Study of Bioethical Issues. Anticipate
and Communicate: Ethical Management of Incidental and Secondary
Findings in the Clinical Research and Direct-to-Consumer Contexts.
Washington, DC: Presidential Commission for the Study of Bioethical
Issues; 2013.
86. Wolf SM. Return of individual research results and incidental findings:
facing the challenges of translational science. Annu Rev Genomics Hum
Genet. 2013;14:557-577.
87. Bookman EB, Langehorne AA, Eckfeldt JH, et al. Reporting genetic
88.
89.
90.
91.
92.
93.
results in research studies: summary and recommendations of an NHLBI

working group. Am J Med Genet A. 2006;140:1033-1040.
Klitzman R, Buquez B, Appelbaum PS, et al. Processes and factors involved in decisions regarding return of incidental genomic findings in
research. Genet Med. 2014;16:311-317.
Greely HT. The uneasy ethical and legal underpinnings of large-scale
genomic biobanks. Annu Rev Genomics Hum Genet. 2007;8:343-364.
Rothstein MA. Tiered disclosure options promote the autonomy and
well-being of research subjects. Am J Bioeth. 2006;6:20-21.
Kohane IS, Mandl KD, Taylor PL, et al. Medicine: reestablishing the
researcher-patient compact. Science. 2007;316:836-837.
Clayton EW. Sharing individual research results with biospecimen
contributors: counterpoint. Cancer Epidemiol Biomarkers Prev. 2012;
21:260-261.
Appelbaum PS, Waldman CR, Fyer A, et al. Informed consent for return
of incidental findings in genomic research. Genet Med. 2014;16:
367-373.
94. Forrest LE, Young MA. Clinically significant germline mutations in
cancer-causing genes identified through research studies should be
offered to research participants by genetic counselors. J Clin Oncol.
2016;34:898-901.
95. Graves KD, Sinicrope PS, Esplen MJ, et al. Communication of genetic test
results to family and health-care providers following disclosure of research results. Genet Med. 2014;16:294-301.
96. Christensen KD, Roberts JS, Shalowitz DI, et al. Disclosing individual
CDKN2A research results to melanoma survivors: interest, impact, and
demands on researchers. Cancer Epidemiol Biomarkers Prev. 2011;20:
522-529.
97. Roberts JS, Shalowitz DI, Christensen KD, et al. Returning individual
research results: development of a cancer genetics education and risk
communication protocol. J Empir Res Hum Res Ethics. 2010;5:17-30.
98. Bledsoe MJ, Clayton EW, McGuire AL, et al. Return of research
results from genomic biobanks: a call for data. Genet Med. 2013;15:
159-160.
269
PEDIATRIC ONCOLOGY
Exploiting Laboratory Insights to

Improve Outcomes of Pediatric
Central Nervous System Tumors
CHAIR
Giles W. Robinson, MD
Memphis, TN
SPEAKERS
Adam Resnick, PhD
Childrens Hospital of Philadelphia
Philadelphia, PA
Hendrik Witt, PhD
University of Heidelberg
Heidelberg, Germany
ROBINSON, WITT, AND RESNICK
Exploiting Laboratory Insights to Improve Outcomes of

Pediatric Central Nervous System Tumors
Giles W. Robinson, MD, Hendrik Witt, MD, and Adam Resnick, PhD
OVERVIEW
Over a relatively short period of time, owing to improvements in biotechnology, our ability to identify the molecular
mechanisms within pediatric brain tumors has dramatically increased. These findings have reshaped the way that we
describe these diseases and have provided insights into how to better treat these often devastating diseases. Although still
far from reaching the full therapeutic potential these advancements hold, the impact of these findings is steadily taking hold
of pediatric brain tumor management. In this article, we summarize the major discoveries within three common pediatric
brain tumor categories; medulloblastoma, ependymoma, and low-grade glioma. We discuss the current impact of these
findings on treatment and the direction these findings may take the field of pediatric neuro-oncology.
n the topic of finding a cure for cancer, the American

Association for Cancer Research (AACR) president, Jose
Baselga, recently postulated, We have indeed reached an
inflection point, where the number of discoveries that are
being made at such an accelerated pace are saving lives and
bringing enormous hope for patients with cancer, even
those with advanced disease.1
This inflection point of new-found hope in recent discoveries resonates strongly within the field of pediatric
neuro-oncology where advances in molecular analyses of
brain tumors now appear to be on the cusp of making a huge
therapeutic impact. The past decades foray into molecular
discovery has brought about an improved understanding
of pediatric brain tumors origin, heterogeneity, molecular
makeup, and mutational landscape. This new knowledge is
fueling novel treatment approaches and opening pediatric
brain tumors to avenues of care not previously recognized.
In a field where mortality rates remain high, and even
successful therapy is often marred by high treatmentrelated morbidity rates, such advances are intensely
sought and enthusiastically welcomed. However, despite the
rapid pace of discovery, the principal challenge remains to
harness these discoveries into therapy while not detracting
from the current successes that conventional therapy has
brought.
Herein, we review three major pediatric neuro-oncologic
diagnoses: medulloblastoma, ependymoma, and low-grade
glioma. We review the breakthrough findings within each
tumor type and discuss the current and potential impact
these discoveries are imparting on the management of these

diseases.
MEDULLOBLASTOMA
The Discoveries
Medulloblastoma is one of the most common pediatric brain
tumors, accounting for approximately 20% of malignant
brain tumors among children. The combination of surgery,
radiation, and cytotoxic chemotherapy has been successful
at eradicating the disease for the majority of patients, and
current treatment protocols boast a 70% to 75% 5-year
overall survival (OS).2,3 However, despite this noteworthy
success, considerable challenges remain. Therapy frequently
leaves the surviving population facing cognitive deficits,
hearing loss, infertility, and endocrinopathies.4,5 Metastatic
spread of disease continues to confound and remains one of
the strongest predictors of treatment failure.6 Moreover,
the youngest children, who are also the most vulnerable to
treatment-related morbidities, continue to fare the worst.
Importantly, these challenges have remained relatively
constant despite attempts over the past 10 to 15 years to
optimize therapy.7 Dose escalations, stem cell rescue, advances in surgical techniques, and better quality supportive
care seem to only have had modest impacts on disease-free
survival. This suggests that a therapeutic plateau has been
reached, and, to improve outcome, treatment needs to
change.
Advanced molecular analysis of the disease provides the
means to exact this change because it dissociates disease
From the Division of Neuro-Oncology, Department of Oncology, St. Jude Childrens Research Hospital, Memphis, TN; Division of Pediatric Neurooncology, German Cancer Research
Center, Heidelberg, Germany; Center for Data Driven Discovery in Biomedicine, Division of Neurosurgery, The Childrens Hospital of Philadelphia, Philadelphia, PA.
Corresponding author: Giles W. Robinson, MD, St. Jude Childrens Research Hospital, 262 Danny Thomas Pl., M 260, Memphis, TN 38105; email: giles.robinson@stjude.org.
e540
LABORATORY INSIGHTS: PEDIATRIC CENTRAL NERVOUS SYSTEM TUMORS
into related entities that can be better managed. This approach has been championed in medulloblastoma because
of the diversity found within. In recent years, enhanced
molecular interrogation by transcriptomic, methylomic,
and mutational profiling has uniformly revealed medulloblastoma to be a highly heterogeneous disease best grouped
into four categories, or subgroups, called WNT, SHH, group 3,
and group 4.8-10
The WNT subgroup accounts for only 10% of medulloblastoma and is the most molecularly and clinically uniform
of all the subgroups.11 Molecularly, almost all of these tumors are found to have a gain-of-function mutation in the
CTNNB1 gene that constitutively activates the WNT pathway
as well as a heterozygous loss of chromosome 6.10 Clinically,
these tumors arise in midchildhood (average age of 10),
more commonly affect female patients, and are exclusively
described as classic by histology.12 Metastatic disease is rare
in this subgroup. The most striking feature of this group is
the excellent prognosis. Across multiple trials, which prescribe fundamentally similar but assorted treatment regimens, patients with these tumors have an excellent survival
rate (. 95%).2,12,13 This observation suggests a remarkable
sensitivity to therapy.
The SHH subgroup describes a heterogeneous mix of medulloblastoma, all of which aberrantly activate the SHH
pathway. Approximately 25% of medulloblastomas belong to
this subgroup.8,12 The incidence of these tumors has two
peaks, one in children younger than age 5 and another in
adolescents and adults older than age 16.12,14 All histologic
subtypes of medulloblastoma (classic, nodular-desmoplastic,
anaplastic, and large cell) are present, but nodular desmoplastic histology is only seen in this subgroup.13 Intriguingly,
although more common among young children and older
patients, those that occur during midchildhood (age 5 to 16)
are more aggressive. These aggressive tumors harbor large-cell
and anaplastic histology, MYCN amplifications, and TP53
mutations, all of which are associated with a very poor
KEY POINTS
Improvements in biotechnology have advanced our

understanding of pediatric brain tumors.
Medulloblastomas and ependymomas, two of the more
prevalent pediatric high-grade brain tumors, are now
biologically subcategorized into molecular subgroups
associated with clinical outcomes.
Similarly, pediatric low-grade gliomas are increasingly
subcategorized by the underlying gene fusions and
mutations identified within.
The immediate clinical impact of these findings is
improved prognostication, which affords opportunities
to risk stratify therapy and mitigate toxicity.
The future, and much anticipated, impact will be
tailoring therapy to target the specific molecular
aberrations that are identified within.
prognosis.15,16 Conversely, SHH medulloblastomas in young

children are associated with nodular desmoplastic histology
and a good prognosis.12,16,17
Group 3 medulloblastomas are almost always found in
young children (peak incidence is age 5) and are associated
with a poor prognosis.12 Approximately 25% of medulloblastomas belong to this subgroup. Patients with these
tumors commonly present with metastatic disease and
anaplastic histology. No association with a singular molecular pathway has been made, but these tumors frequently
harbor MYC overexpression and amplification.10 Only approximately 50% of patients diagnosed with this disease
survive despite aggressive therapy with high-dose radiation
and/or high-dose cytotoxic chemotherapy.12,18
Group 4 is the most common of all subgroups of medulloblastoma and accounts for approximately 40% of the
disease.8,12 Although molecularly quite similar to group 3,
patients with group 4 disease have a much better prognosis.
Many molecular characteristics of group 3 and group 4
overlap, making them difficult to completely segregate;
however, features like MYC overexpression and amplification are rare in group 4.19 The average age of patients with
this disease is higher than group 3, and metastatic presentation is less common. The OS among the group without
metastatic disease is greater than 80%, and, among patients
with metastatic disease, it is approximately 65% to 70%.18
Current Impact
The major and immediate clinical impact of these molecular
observations is improved prognostication, and with this comes
the opportunity to tailor therapy according to risk of relapse.
For patients with the lowest risk of relapse, judicious dose
reductions of conventional therapy are now being explored.
The proven excellent survival of patients with WNT medulloblastoma means that these patients are the perfect candidates for this approach. The latest St. Jude medulloblastoma
protocol, SJMB12 (NCT01878617), is investigating a lowered
dose of craniospinal radiation as well as dose reductions in the
infertility-causing agent cyclophosphamide for patients with
nonmetastatic WNT medulloblastoma. Similar dose reductions have been initiated on the newly released European
protocol, PNET5 (NCT02066220), and are anticipated on an
upcoming COG trial. One trial (NCT02212574) is evaluating
the feasibility of eliminating radiation completely from
therapy to this subgroup.
Patients with a predictably higher risk of relapse who were
previously classified as having lower-risk disease can have
their treatment escalated. On SJMB12, group 3 and group 4
patients with anaplastic histology, gains, or amplifications of
MYC or MYCN, which have been associated with an inferior
prognosis, are being prescribed more intensive regimens.
Analysis of the disease-free outcome of patients with tumors
that carry these characteristics will determine if these steps
are improving outcomes.
In this fashion, therapy is moving away from a uniform approach that exposes all patients to an equal risk of
e541
treatment-related side effects toward one where biologic

risk factors govern the amount of therapy needed to maintain survival and mitigate toxicity.
Future Directions
The biologic risk-based approach also brings an improved
ability to identify patients who are predicted to relapse even
after being prescribed maximal conventional therapy. Although troubling, because therapeutic options are limited
for these patients, this approach affords the opportunity to
identify and concentrate on select groups where sensible
and potentially targetable alternatives can be evaluated.
Of current interest are a number of novel agents: BRD4
inhibitors that downgrade the proliferative effects of MYC 20,21;
CDK inhibitors that can impair the continuous drive that
malignant cells place on the cell cycle; and epigenetic regulating drugs that can push malignant cells away from
mimicry of a stem celllike state.22 These are just some
classes of agents, which predictably would have been
overlooked without the advanced molecular profiling of
these tumors. Additionally, although early-phase clinical
trials of these agents will establish the safety profile of these
agents, the molecular characterization of these tumors will
arguably be more informative in defining efficacy and clinical
utility. In this fashion, early-phase trials can be streamlined
through use of biologic markers to small expansion cohorts
to identify an early efficacy signal without subjecting scores
of patients to irrelevant agents.
Even greater detail of molecular characterization of medulloblastoma will also help to identify subsets for which
therapy can be altered. Recent publications have connected
certain biomarkers found within subgroups to prognosis. For
example, whole loss of chromosome 11 or gain of chromosome 17 appears to be associated with improved survival
rates among patients with group 4 medulloblastomas and
GLI2 amplification with poor survival rates among patients
with SHH medulloblastoma.18 Findings such as these are
necessary to improve our understanding of the intricacies of
these diseases and to fuel even better stratification of patients on future clinical trials.
EPENDYMOMA
The Discoveries
Ependymoma is an aggressive pediatric brain tumor and
accounts for approximately 10% of all malignant neoplasms
in the pediatric brain. Particularly, young children and infants experience a very poor prognosis. The majority of cases
arise within the posterior fossa (70%), followed by the
supratentorial region (25%) and the spinal cord (5%). Several
independent studies have identified compartment-specific
molecular subtypes of ependymomas. Within the posterior
fossa, two molecular subtypes could be defined: group A
ependymomas (PF-EPN-A) found in young children and infants and group B (PF-EPN-B) tumors diagnosed in adolescents and adults.23 Two additional predominantly pediatric
ependymoma subtypes could be identified in the
e542
supratentorial region. One subgroup, ST-EPN-RELA, is characterized by fusions of the RELA gene, whereas the other
subgroup, ST-EPN-YAP1, is defined by fusions to the oncogene YAP1.24,25 Regarding clinical associations, the molecular subgroups outperform the current histopathological
classification.
Current Impact
Currently, major clinical studies are in process to integrate
the newly identified molecular subgroups to adjust a riskadapted treatment concept. The standard of care of pediatric ependymomas is maximal neurosurgical resection
followed by radiation therapy, and sometimes patients receive chemotherapy. In North America and across Europe,
patients with pediatric ependymoma are commonly enrolled into clinical trials: in North America, the Childrens
Oncology Group trial AVNS0831 (NCT01096368) and in
Europe, SIOP Ependymoma II trial (EudraCT No. 2013002766-39). On the basis of novel molecular insights
delivered scientifically during recent years, group A ependymomas (PF-EPN-A) of the posterior fossa and supratentorial tumors harboring the RELA fusion (ST-EPN-RELA)
can be considered as high-risk subtypes. In contrast, most
group B ependymomas (PF-EPN-B) of the posterior fossa can
be cured with the current treatment regimen. Group B
tumors are diagnosed in approximately 10% of childhood
posterior fossa ependymomas; the patients are adolescent
or young adults. This observation can lead to a de-escalation
treatment strategy for group B ependymomas (PF-EPN-B),
compared with the dose reduction in the treatment of WNT
medulloblastomas, respectively. However, the adjustment
of a reduced treatment has to be carefully evaluated in
clinical trials.
Future Directions
The vast majority of patients with high-risk ependymoma,
for whom effective therapeutic concepts are lacking, are
children who belong to the molecular subgroups PF-EPN-A
and ST-EPN-RELA. This points to the striking relevance of
molecular classification for the future clinical management
of ependymoma. The integration of the novel molecular
subgroups for more precise risk stratification within clinical
trials and the identification of novel biology-driven therapeutic concepts, as well as the development of adequate
preclinical models represent the paramount challenges for
the near future of ependymoma research and treatment.
Several new targets and pathways were identified during recent years, such as RELA-C11orf95 fusion, including
nuclear factor-kappa B activation in supratentorial ependymomas25 and the response of epigenetic modifiers,
e.g., histone deacetylase inhibitors and inhibitors of the
polycomb-repressive complex 2 polycomb in vitro and
in vivo.26,27 The therapeutic inhibition of the Hippo-signaling
pathway would be an option in YAP1 fusionpositive
ependymomas,24 which has to be evaluated in future
preclinical approaches. However, the translation from
preclinical experiments into clinical phase I or II trials is a

considerable process and a time-consuming challenge. On
the basis of the experiences of the comprehensive spectrum
of newly discovered genetic and epigenetic alteration in
pediatric ependymomas, combinational targeted treatment
approaches can be possibly promising strategies. In terms of
low numbers of this rare disease, multicenter studies should
be generated more effectively to increase the enrollment
numbers of patients with ependymoma. Future treatment
approaches for ependymomas should include up-front
stratification for molecular subtypes. Additional novel precision treatment ideas should be made on the basis of reproducible preclinical experiences of in vitro and in vivo
treatments of subtype-specific ependymoma models. We
hope that novel treatment concepts will broaden and
substitute current ependymoma trials in the foreseeable
future.
LOW-GRADE GLIOMA
The Discoveries
Pediatric low-grade gliomas (PLGGs) are the most common
type of brain tumor in children, comprising a histologically
heterogeneous group of World Health Organization (WHO)
grade I and II tumors.28,29 Classification of PLGGs has been
largely informed by the relevant constituting cell features,
including astrocytic, oligodendroglial, mixed oligoastrocytic,
neuronal, or mixed glioneuronal morphologies.28 However,
histology alone fails to predict the biologic course of PLGGs.
In contrast to adult low-grade gliomas, which often display
malignant transformation, PLGGs have very high OS rates
with a reported 87% 20-year OS.30 However, although
clinical symptoms may indeed take months to years to
progress, management of surgically intractable PLGGs or
recurrent and/or progressive tumors via traditional chemoradiation often proves challenging. Such treatments can
often lead to serious life-long neurocognitive and systemic
complications in surviving patients,31 including loss of vision
and/or hearing, endocrinopathies, and mood/behavior disorders.32 To harness the potential of new targeted precision
medicine approaches in addressing these morbidities,
a number of large-scale genomic characterization efforts have recently sought to further define the molecular underpinnings of PLGGs.33-35 These studies have
informed a consensus of an altered mitogen-activated protein kinase (MAPK) pathway as the unifying driver event in
PLGGs.
The genomic landscape of PLGGs is dominated by recurrent
structural rearrangements and gene fusion events. The characterization of KIAA1549-BRAF gene fusions as a common
structural alteration in cerebellar pilocytic astrocytomas defined the first and most frequent driver mutation underlying
the pathogenesis of PLGGs.36-39 Subsequent studies demonstrated the prevalence of additional BRAF alterations, including canonical BRAF V600E mutations as characteristic
of pleomorphic xanthoastrocytomas, gangliogliomas, and a
subset of extracerebellar pilocytic astrocytomas.40,41 BRAF
V600E mutations in PLGG recapitulate the most common
activation-loop amino acid substitution associated with

adult cancers including melanomas, papillary thyroid cancers, and colorectal malignancies.42 In contrast to adult lowgrade gliomas harboring IDH1 driver mutations and their
frequent clinical progression, secondary high-grade gliomas
arising from PLGGs are rare and define a distinct subgroup
harboring frequent BRAF V600E mutations co-occurring in
the context of CDKN2A deletion.43,44 In addition to
KIAA1549-BRAF, more than a half dozen other BRAF fusions
have been additionally identified in PLGGs occurring at lower
frequencies. Universally, the fusion events truncate the
BRAF protein, eliminating the RAS-binding N-terminus
regulatory region of the kinase and in turn engaging a
constitutively active kinase that functions in a dimeric state.
As such, distinct molecular mechanisms drive BRAF kinase
activation in the fusion versus the monomeric V600E point
mutant contexta biochemical novelty that imparts differential responsiveness to targeted therapeutics and an
associated clinical mandate for personalized, precision
medicine.
In addition to frequent BRAF alterations, a number of additional lower-prevalence recurrent mutational settings
have been identified in PLGGs involving gene fusion events
and structural rearrangements of RAF1 (CRAF, a BRAF kinase
family member), FGFR, NTRK, and MYB.33-35 How tumor
location and histology fully align with mutational status
continues to evolve in the context of our understanding of
PLGGs and has required increasingly larger cohorts. A recent
study using the comprehensive integration of published and
newly generated genomic data across 249 PLGGs identified MYB-QKI as diagnostic of angiocentric gliomas and
revealed a complex functional landscape of multiple oncogenic mechanisms being invoked.45 In the context of a
single fusion event a tripartite mechanism drives oncogenesis through a combination of oncogene activation,
tumor-suppressor loss function, and epigenetic dysregulation driving gene fusion expression.45 Whether such combinatorial mechanisms occur in other PLGG fusion settings
remains unknown.
Mitogen-activated protein kinase signaling alterations as
well as recurrent upstream kinase fusions that also impinge
on the PI3K/mTOR pathway are dominant features of the
PLGG genomic landscape and suggest a potential common,
targetable signaling hub for the two pathways in PLGGs.
Consistent with a biologic role for the mTOR pathway in
PLGG, neurofibromatosis type-1 (NF-1) and tuberous
sclerosis complex are two cancer-predisposition syndromes possessing aberrant mTOR activation that display
increased clinical incidence of PLGGs in affected patients.46
Cell-based studies have further confirmed mTOR pathway
activation as a shared, key growth-regulatory pathway
common to both sporadic and familial PLGGs.47 As such,
mTOR targeting and MAPK inhibition have formed the basis
of the rapidly expanding PLGG precision medicine clinical
trial landscape. However, despite more than 2 decades of
research and the recent development of highly specific
small-molecule inhibitors of the pathway, BRAF signaling
e543
and MAPK functional regulation have proven to be more

complex and often intractable in unexpected and clinically
relevant ways.
Current Impact
Current standards of care for PLGGs largely comprise surgical
resection and chemotherapy and/or radiation therapy independent of molecular classification. The prevalence of
MAPK alterations across adult cancers has provided strong
incentives for MAPK targeting and drug development with
successful clinical implementation in adult malignancies.
This includes the development of a number of potent and
selective inhibitors of RAF (e.g., vemurafenib, dabrafenib)
and MEK (e.g., trametinib, selumetinib, MEK162) and offers
novel precision medicine opportunities that could address
the critical need for new targeted therapies for patients with
PLGG.
Initial efforts to apply small-moleculemediated inhibition
in PLGGs included a phase II study of sorafenib. However,
patients experienced unexpected and unprecedented acceleration of tumor growth, leading to closure of the trial.48
This unexpected response was further informed by a remarkable development in small-molecule kinase targeting
with the recognition that selective inhibitors of BRAF, like
vemurafenib, could drive pathway activation in the context
of wild-type RAF through dimerization-mediated transactivation.49 Biochemical characterization of KIAA1549BRAF found it to signal as a constitutive homodimer.
Targeting with first-generation selective inhibitors like
vemurafenib induced robust paradoxical activation of the
MAPK pathway in cells and enhanced tumor growth. In
contrast, recently developed second-generation paradoxbreaking inhibitors are able to inhibit BRAF fusion signaling,
in part via the disruption of dimerization.50,51 Whether such
paradox-breaking inhibitors display similar efficacy across
all BRAF and CRAF fusions remains unknown and will likely
depend on the cognate RAF fusion partner and/or the
homodimerization versus heterodimerization of the mutant
RAF fusion with either another fusion or a wild-type RAF,
respectively. BRAF inhibition in PLGGs is currently restricted
to validated targeting of BRAF V600E tumors (PNOC-002).
However, MEK inhibition downstream of BRAF activation is
also currently undergoing clinical evaluation (NCT01089101,
NCT02124772) with reported responses in a recent phase I
study of AZD6244 suggesting single-agent efficacy. Additional single-agent clinical trials for PLGGs include mTOR
targeting with RAD001/everolimus. Indeed RAD001 has
recently been shown to have limited clinical efficacy, although molecular profiling of underlying mutations was
not performed.52,53 An additional on-going clinical trial
(NCT00831324) is further defining differential RAD001 responsiveness on the basis of tumor-associated S6
phosphorylation, a downstream mTOR target. In other nonPLGG contexts, single-agent MAPK targeting results in rapid,
emergent drug resistance suggest the necessity for combinatorial pathway targeting with MEK and mTOR inhibitors.
e544
However, recent efforts among adults to clinically combine

trametinib and everolimus resulted in frequent treatmentrelated adverse events (NCT00955773), suggesting differential dosing or alternative combinations may be necessary.
Future Directions
Understanding the contribution of the fusion context to
mutant RAF signaling across different BRAF and CRAF fusions
will be important given the strict dependency of drug
responsiveness on protein-protein interactions. Likewise,
emergent inhibitors that harness dimerization like
LY3009120 are likely to represent ideal RAF fusion inhibitors
that may offer additional MEK/RAF combinatorial targeting
opportunities in fusion-bearing PLGGs.54 For a comprehensive PLGG precision medicine platform, the characterization of small molecules targeting of NTRK and FGFR
structural variants is also required as well as novel targeting
approaches for MYB-altered tumors. In addition to
small-molecule inhibition, fusion kinases and the context of
fusion-specific epitopes also avail the use of emerging
immunotherapy platforms with at least one clinical trial
underway (NCT01130077). Remarkably, the prevalence of
fusion gene biology in cancer is only now beginning to be
appreciated and includes MYB-, RAF1-, BRAF, NTRK, and
FGFR fusions in a host of adult malignancies, suggesting the
molecular characterization and targeting of PLGGs may
provide for wider implementation across both adult and
pediatric malignancies.55
CONCLUSION
The molecular revolution for pediatric brain tumors has
begun, and one can envision an era of highly specialized
therapeutics will follow. Improved technology has enabled
us to interrogate the transcriptome, genome, epigenome,
and proteome of the tumors we study. Although not yet
perfect, these techniques have already improved our understanding of these diseases and are, importantly, providing us with new ideas for treatment. The ultimate goal is
to use this new-found knowledge to develop therapy that is
both efficacious and nontoxic. Although far from achieving
this goal, new molecular findings are already making inroads
into therapy.
For medulloblastoma and ependymoma, molecular subgrouping and biomarker identification has allowed for an
improvement in risk stratification. As such, groups at low risk
of relapse on conventional therapy are being evaluated for
dose reductions to mitigate toxicity and improve the quality
of life in the surviving population. High-risk groups are now
being recognized and earmarked for prospective therapeutic
interventions.
In the meantime, therapy targeted toward common molecular aberrations found in refractory low-grade gliomas
is already starting to show the benefit and power of
matching selective agents to their appropriate targets. Such
is the success of these novel targeted compounds that the
how and when to move these agents into front-line care for
newly diagnosed low-grade glioma is already being hotly

debated.
Although the complex world of cancer biology still remains
principally unfamiliar, the early insight afforded by recent
advancements is propelling the field forward at a rapid pace.

Now is an exciting time to be involved in pediatric neurooncology because the future, although demanding, appears
bright.
References
1. Bin Han Ong, M. Obama announces moonshot to cure cancer. The
Cancer Letter. 2016;42:1-4.
2. Gajjar A, Chintagumpala M, Ashley D, et al. Risk-adapted craniospinal
radiotherapy followed by high-dose chemotherapy and stem-cell rescue in children with newly diagnosed medulloblastoma (St Jude
Medulloblastoma-96): long-term results from a prospective, multicentre trial. Lancet Oncol. 2006;7:813-820.
3. Packer RJ, Gajjar A, Vezina G, et al. Phase III study of craniospinal radiation therapy followed by adjuvant chemotherapy for newly diagnosed average-risk medulloblastoma. J Clin Oncol. 2006;24:
4202-4208.
4. Mulhern RK, Palmer SL, Merchant TE, et al. Neurocognitive consequences of risk-adapted therapy for childhood medulloblastoma. J Clin
Oncol. 2005;23:5511-5519.
5. Laughton SJ, Merchant TE, Sklar CA, et al. Endocrine outcomes for
children with embryonal brain tumors after risk-adapted craniospinal
and conformal primary-site irradiation and high-dose chemotherapy
with stem-cell rescue on the SJMB-96 trial. J Clin Oncol. 2008;26:
1112-1118.
6. Ellison DW, Kocak M, Dalton J, et al. Definition of disease-risk stratification groups in childhood medulloblastoma using combined clinical,
pathologic, and molecular variables. J Clin Oncol. 2011;29:1400-1407.
7. Ostrom QT, de Blank PM, Kruchko C, et al. Alexs lemonade stand
foundation infant and childhood primary brain and central nervous
system tumors diagnosed in the United States in 2007-2011. Neurooncol. 2015;16:x1-x36.
8. Taylor MD, Northcott PA, Korshunov A, et al. Molecular subgroups of
medulloblastoma: the current consensus. Acta Neuropathol. 2012;123:
465-472.
9. Hovestadt V, Remke M, Kool M, et al. Robust molecular subgrouping
and copy-number profiling of medulloblastoma from small amounts of
archival tumour material using high-density DNA methylation arrays.
Acta Neuropathol. 2013;125:913-916.
10. Northcott PA, Jones DT, Kool M, et al. Medulloblastomics: the end of the
beginning. Nat Rev Cancer. 2012;12:818-834.
11. Gajjar AJ, Robinson GW. Medulloblastoma-translating discoveries from
the bench to the bedside. Nat Rev Clin Oncol. 2014;11:714-722.
12. Kool M, Korshunov A, Remke M, et al. Molecular subgroups of medulloblastoma: an international meta-analysis of transcriptome, genetic
aberrations, and clinical data of WNT, SHH, group 3, and group 4
medulloblastomas. Acta Neuropathol. 2012;123:473-484.
13. Ellison DW, Dalton J, Kocak M, et al. Medulloblastoma: clinicopathological correlates of SHH, WNT, and non-SHH/WNT molecular subgroups. Acta Neuropathol. 2011;121:381-396.
14. Northcott PA, Korshunov A, Witt H, et al. Medulloblastoma comprises
four distinct molecular variants. J Clin Oncol. 2011;29:1408-1414.
15. Zhukova N, Ramaswamy V, Remke M, et al. Subgroup-specific prognostic implications of TP53 mutation in medulloblastoma. J Clin Oncol.
2013;31:2927-2935.
16. Kool M, Jones DT, Jager N, et al; ICGC PedBrain Tumor Project. Genome
sequencing of SHH medulloblastoma predicts genotype-related response to smoothened inhibition. Cancer Cell. 2014;25:393-405.
17. Rutkowski S, von Hoff K, Emser A, et al. Survival and prognostic factors
of early childhood medulloblastoma: an international meta-analysis.
J Clin Oncol. 2010;28:4961-4968.
18. Shih DJ, Northcott PA, Remke M, et al. Cytogenetic prognostication
within medulloblastoma subgroups. J Clin Oncol. 2014;32:886-896.
19. Roussel MF, Robinson GW. Role of MYC in medulloblastoma. Cold
Spring Harb Perspect Med. 2013;3:a014308.
20. Bandopadhayay P, Bergthold G, Nguyen B, et al. BET bromodomain
inhibition of MYC-amplified medulloblastoma. Clin Cancer Res. 2014;
20:912-925.
21. Tang Y, Gholamin S, Schubert S, et al. Epigenetic targeting of Hedgehog
pathway transcriptional output through BET bromodomain inhibition.
Nat Med. 2014;20:732-740.
22. Robinson G, Parker M, Kranenburg TA, et al. Novel mutations target
distinct subgroups of medulloblastoma. Nature. 2012;488:43-48.
23. Witt H, Mack SC, Ryzhova M, et al. Delineation of two clinically and
molecularly distinct subgroups of posterior fossa ependymoma. Cancer
Cell. 2011;20:143-157.
24. Pajtler KW, Witt H, Sill M, et al. Molecular classification of ependymal
tumors across all CNS compartments, histopathological grades, and age
groups. Cancer Cell. 2015;27:728-743.
25. Parker M, Mohankumar KM, Punchihewa C, et al. C11orf95-RELA fusions drive oncogenic NF-kB signalling in ependymoma. Nature. 2014;
506:451-455.
26. Mack SC, Witt H, Wang X, et al. Emerging insights into the ependymoma
epigenome. Brain Pathol. 2013;23:206-209.
27. Milde T, Kleber S, Korshunov A, et al. A novel human high-risk ependymoma stem cell model reveals the differentiation-inducing potential
of the histone deacetylase inhibitor Vorinostat. Acta Neuropathol.
2011;122:637-650.
28. Louis DN, Ohgaki H, Wiestler OD, et al. The 2007 WHO classification of
tumours of the central nervous system. Acta Neuropathol. 2007;114:
97-109.
29. Dolecek TA, Propp JM, Stroup NE, et al. CBTRUS statistical report:
primary brain and central nervous system tumors diagnosed in the
United States in 2005-2009. Neuro-oncol. 2012;1:v1-v49.
30. Bandopadhayay P, Bergthold G, London WB, et al. Long-term outcome
of 4,040 children diagnosed with pediatric low-grade gliomas: an
analysis of the Surveillance Epidemiology and End Results (SEER) database. Pediatr Blood Cancer. 2014;61:1173-1179.
31. Gnekow AK, Falkenstein F, von Hornstein S, et al. Long-term follow-up of
the multicenter, multidisciplinary treatment study HIT-LGG-1996 for
low-grade glioma in children and adolescents of the German Speaking
Society of Pediatric Oncology and Hematology. Neuro-oncol. 2012;14:
1265-1284.
32. Armstrong GT, Conklin HM, Huang S, et al. Survival and long-term health
and cognitive outcomes after low-grade glioma. Neuro-oncol. 2011;13:
223-234.
33. Ramkissoon LA, Horowitz PM, Craig JM, et al. Genomic analysis of
diffuse pediatric low-grade gliomas identifies recurrent oncogenic
truncating rearrangements in the transcription factor MYBL1. Proc Natl
Acad Sci USA. 2013;110:8188-8193.
e545
34. Zhang J, Wu G, Miller CP, et al; St. Jude Childrens Research

HospitalWashington University Pediatric Cancer Genome Project.
Whole-genome sequencing identifies genetic alterations in pediatric
low-grade gliomas. Nat Genet. 2013;45:602-612.
35. Jones DT, Hutter B, Jager N, et al; International Cancer Genome Consortium PedBrain Tumor Project. Recurrent somatic alterations of
FGFR1 and NTRK2 in pilocytic astrocytoma. Nat Genet. 2013;45:
927-932.
36. Sievert AJ, Jackson EM, Gai X, et al. Duplication of 7q34 in pediatric lowgrade astrocytomas detected by high-density single-nucleotide
polymorphism-based genotype arrays results in a novel BRAF fusion
gene. Brain Pathol. 2009;19:449-458.
37. Cin H, Meyer C, Herr R, et al. Oncogenic FAM131B-BRAF fusion resulting
from 7q34 deletion comprises an alternative mechanism of MAPK
pathway activation in pilocytic astrocytoma. Acta Neuropathol. 2011;
121:763-774.
38. Jones DT, Kocialkowski S, Liu L, et al. Tandem duplication producing a
novel oncogenic BRAF fusion gene defines the majority of pilocytic
astrocytomas. Cancer Res. 2008;68:8673-8677.
39. Lin A, Rodriguez FJ, Karajannis MA, et al. BRAF alterations in primary
glial and glioneuronal neoplasms of the central nervous system with
identification of 2 novel KIAA1549:BRAF fusion variants. J Neuropathol
Exp Neurol. 2012;71:66-72.
40. Schindler G, Capper D, Meyer J, et al. Analysis of BRAF V600E mutation
in 1,320 nervous system tumors reveals high mutation frequencies in
pleomorphic xanthoastrocytoma, ganglioglioma and extra-cerebellar
pilocytic astrocytoma. Acta Neuropathol. 2011;121:397-405.
41. Dougherty MJ, Santi M, Brose MS, et al. Activating mutations in BRAF
characterize a spectrum of pediatric low-grade gliomas. Neuro-oncol.
2010;12:621-630.
42. Bollag G, Tsai J, Zhang J, et al. Vemurafenib: the first drug approved for
BRAF-mutant cancer. Nat Rev Drug Discov. 2012;11:873-886.
43. Korshunov A, Meyer J, Capper D, et al. Combined molecular analysis of
BRAF and IDH1 distinguishes pilocytic astrocytoma from diffuse astrocytoma. Acta Neuropathol. 2009;118:401-405.
e546
44. Mistry M, Zhukova N, Merico D, et al. BRAF mutation and CDKN2A

deletion define a clinically distinct subgroup of childhood secondary
high-grade glioma. J Clin Oncol. 2015;33:1015-1022.
45. Bandopadhayay P, Ramkissoon LA, Jain P, et al. MYB-QKI rearrangements in angiocentric glioma drive tumorigenicity through a tripartite
mechanism. Nat Genet. 2016;48:273-282.
46. Yunoue S, Tokuo H, Fukunaga K, et al. Neurofibromatosis type I tumor
suppressor neurofibromin regulates neuronal differentiation via its GTPaseactivating protein function toward Ras. J Biol Chem. 2003;278:26958-26969.
47. Kaul A, Chen YH, Emnett RJ, et al. Pediatric glioma-associated KIAA1549:
BRAF expression regulates neuroglial cell growth in a cell type-specific
and mTOR-dependent manner. Genes Dev. 2012;26:2561-2566.
48. Karajannis MA, Legault G, Fisher MJ, et al. Phase II study of sorafenib in
children with recurrent or progressive low-grade astrocytomas. Neurooncol. 2014;16:1408-1416.
49. Poulikakos PI, Zhang C, Bollag G, et al. RAF inhibitors transactivate RAF
dimers and ERK signalling in cells with wild-type BRAF. Nature. 2010;
464:427-430.
50. Sievert AJ, Lang SS, Boucher KL, et al. Paradoxical activation and RAF
inhibitor resistance of BRAF protein kinase fusions characterizing pediatric astrocytomas. Proc Natl Acad Sci USA. 2013;110:5957-5962.
51. Zhang C, Spevak W, Zhang Y, et al. RAF inhibitors that evade paradoxical
MAPK pathway activation. Nature. 2015;526:583-586.
52. Kieran MW, Yao X, Macy M, et al. Final results of a prospective multiinstitutional phase II study of everolimus (RAD001), an MTOR inhibitor,
in pediatric patients with recurrent or progressive low-grade glioma. A
Poetic Consortium Trial. Neuro-oncol. 2014;16:iii27.
53. Yalon M, Rood B, MacDonald TJ, et al. A feasibility and efficacy study of
rapamycin and erlotinib for recurrent pediatric low-grade glioma (LGG).
54. Peng SB, Henry JR, Kaufman MD, et al. Inhibition of RAF isoforms and
active dimers by LY3009120 leads to anti-tumor activities in RAS or BRAF
mutant cancers. Cancer Cell. 2015;28:384-398.
55. Stransky N, Cerami E, Schalm S, et al. The landscape of kinase fusions in
cancer. Nat Commun. 2014;5:4846.
PEDIATRIC ONCOLOGY
The State of the Art in

Neuroblastoma: From Biology to
Survivorship
CHAIR
Julie R. Park, MD
Seattle Childrens Hospital, University of Washington School of Medicine
Seattle, WA
SPEAKERS
Frank Speleman, PhD
Ghent University Hospital
Ghent, Belgium
Tara O. Henderson, MD, MPH
Chicago, IL
SPELEMAN, PARK, AND HENDERSON
Neuroblastoma: A Tough Nut to Crack

Frank Speleman, PhD, Julie R. Park, MD, and Tara O. Henderson, MD, MPH
OVERVIEW
Neuroblastoma, an embryonal tumor arising from neural crestderived progenitor cells, is the most common solid tumor in
childhood, with more than 700 cases diagnosed per year in the United States. In the past several decades, significant
advances have been made in the treatment of neuroblastoma. Treatment advances reflect improved understanding of the
biology of neuroblastoma. Although amplification of MYCN was discovered in the early 1980s, our understanding of
neuroblastoma oncogenesis has advanced in the last decade as a result of high-throughput genomic analysis, exome and
whole-genome sequencing, genome-wide association studies, and synthetic lethal drug screens. Our refined understanding
of neuroblastoma biology and genetics is reflected in improved prognostic stratification and appropriate tailoring of therapy
in recent clinical trials. Moreover, for high-risk neuroblastoma, a disease that was uniformly fatal 3 decades ago, recent
clinical trials incorporating autologous hematopoietic transplant and immunotherapy utilizing anti-GD2 antibody plus
cytokines have shown improved event-free and overall survival. These advances have resulted in a growing population of
long-term survivors of neuroblastoma. Examination of the late effects and second malignant neoplasms (SMNs) in both older
generations of survivors and more recently treated survivors will inform both design of future trials and surveillance
guidelines for long-term follow-up. As a consequence of advances in understanding of the biology of neuroblastoma,
successful clinical trials, and refined understanding of the late effects and SMNs of survivors, the promise of precision
medicine is becoming a reality for patients with neuroblastoma.
euroblastoma is undoubtedly one of the most enigmatic

tumors, with remarkable heterogeneous clinical behavior ranging from spontaneous regression to metastatic
disease that is refractory to therapy (see the therapy update
section). Progress in understanding the molecular basis of
the biologic and clinical behavior of neuroblastoma has
moved forward at a slower pace compared with many other
tumor entities, such as childhood leukemias. Although amplification of the MYCN oncogene was discovered as early
as 1983,1 the finding of further genes implicated in neuroblastoma oncogenesis took much longer. Low-resolution
whole-genome analyses revealed further highly recurrent
and large genomic DNA copy-number alterations, such as
1p and 11q deletions and 17q gain.2 In the past decade,
this picture has dramatically changed by virtue of highthroughput genomic analysis such as transcriptome profiling, exome and whole-genome sequencing, genome-wide
association studies, and synthetic lethal drug screens.
As a result, step by step, the promise of precision oncology is
now also approaching patients with neuroblastoma suffering from aggressive tumors at diagnosis or relapsed tumors with limited further treatment options.3 Interestingly,
many of the novel findings connect directly or indirectly to
regulatory pathways controlling MYCN activity, bringing the

old enemy into the spotlight of our current and future
targeted drug efforts. Fortunately, our increasing knowledge
of neuroblastoma biology and genetics, together with access
to a rapidly increasing number of actionable drugs, is now
offering new hope for significantly increased cure rates and
reduced toxicity.
NEUROBLASTOMA BIOLOGY UPDATE

Neuroblastoma Biology and Genetics: Progress in
Brief
In the recent past, several excellent reviews have described
progress in understanding neuroblastoma biology.3-8 Neuroblastoma, an embryonal tumor arising from neural
crestderived progenitor cells, has been considered as a
developmental disorder that develops through deregulation
of signaling pathways governing sympathetic nervous system development, such as neurothrophin receptor signaling, and several key transcription factors orchestrating
the complex developmental lineage commitment and
differentiation. 6,9-11
As for other cancers, a search for recurrent genetic alterations
was a major path toward understanding neuroblastoma
From the Center for Medical Genetics Ghent, Cancer Research Institute Ghent, Ghent, Belgium; Seattle Childrens Hospital, Seattle, WA; Department of Pediatrics, University of
Washington School of Medicine, Seattle, WA; University of Chicago Comer Childrens Hospital, Chicago, IL.
Corresponding author: Julie R. Park, MD, Seattle Childrens Hospital, 4800 Sandpoint Way NE, Seattle, WA 98105; email: julie.park@seattlechildrens.org.
e548
NEUROBLASTOMA BIOLOGY, THERAPY, AND SURVIVORSHIP
formation. Roughly three major subgroups were identified

using DNA copy-number analyses: favorable tumors among
children younger than 18 months without MYCN amplification with typically whole chromosome imbalances, and
high-risk tumors among older children with either MYCN
amplification and 1p deletions or 11q deletions in the
absence of MYCN amplifications, with both high-risk
groups exhibiting almost invariably 17q gain.2,12,13 In view
of the high frequency and large size of these segmental
aberrations, it has been proposed that dosage effects of
multiple genes located on the involved chromosomal
regions are implicated in tumor formation. Unfortunately,
the large size of these segments has precluded the identification of the culprit genes on 11q and 17q, whereas a few
genes are now considered to be involved for 1p, including
CHD5, CAMTA1, and CASZ1.14-16
In addition to MYCN, two other oncogenes, ALK and
LIN28B, were found to be amplified, albeit in very low
frequency.17,18 Furthermore, activating ALK mutations were
identified as the major cause of familial neuroblastomas,
although somatic mutations were discovered in less than
10% of neuroblastomas, opening the way for targeted
therapies.19-21 A remarkable productive approach was taken
by Maris et al using genome-wide association studies to
genetic variants predisposing to neuroblastoma and pointing
toward involvement of multiple loci, including LINC00340,
BARD1, LMO1, DUSP12, DDX4, LIN28B, HACE1, NEFL, and
TP53.22-28
KEY POINTS
Recent high-throughput genomics analyses have

provided a wealth of novel data that shed new light on
complex neuroblastoma tumor biology and clinical
behavior.
Novel insights into tumor biology are opening the way
for precision medicine among patients with high-risk
and relapsed neuroblastoma.
Tumor biology and clinical characteristics determine
prognosis and treatment recommendations.
Therapy reduction has proven possible for patients
without high-risk neuroblastoma; for patients with
high-risk neuroblastoma, dose-intensive multimodality
therapy, including dose-intensive induction
chemotherapy, high-dose chemotherapy with
autologous stem cell rescue, external beam
radiotherapy, and immunotherapy has improved
outcomes.
Understanding of the late effects and second
malignancies of survivors is essential for ensuring the
long-term health of survivors with neuroblastoma as
well as informing the design of future neuroblastoma
treatment trials aimed at improved survival along with a
concurrent reduction in late effects.
Untangling the MYCN Regulatory Network

MYCN biology has been intensively studied for many years
using in vitro model systems and a TH-MYCNdriven mouse
model, among others. More recently, new mouse and
zebrafish models were generated18,29-31 and novel prospects for drug MYCN activity have brought MYCN further
into the picture.32 MYCN controls multiple essential cellular
processes, including cell cycle, apoptosis, differentiation,
and metabolism, and it is involved in basic processes
connected to replication, transcription, and splicing.33-35
To further unravel the underlying molecular basis of MYCN
transcriptional control, chromatin immunoprecipitation
sequencing has been performed and many bona fide target
genes have been identified, although MYC(N) overall can
bind to thousands of promotors. The nature of these target
genes appears to be context dependent, and recent studies
provided evidence of a so-called amplifier role for MYC and
MYCN, boosting the transcription of already active genes, a
process that might be distinctly different between normal
and malignant cells. 36,37 MYCN both activates and represses certain gene sets through promotor binding and
association of activator or repressor protein complexes,
respectively. In this context, the binding of MYCN to EZH2,
which is overexpressed in neuroblastoma cells and is a core
component of the PRC2 repressive complex, may play a key
role in MYCN-controlled gene repression of neuronal differentiation genes. Further dissection of the regulatory
protein complexes that are acting in concert with MYCN in
oncogenic regulation of DNA replication, transcription,
splicing, and so forth will be pivotal.38,39 Moreover, MYCN
also controls expression of many microRNAs, most notably the miR-17-92 cluster, targeting transforming growth
factor-beta pathway components and DKK3, among
others.40-42
The Novel Neuroblastoma Oncogenes, ALK and

LIN28B, Are Wired Into the MYCN Regulatory
Network
Given the central role of MYCN in neuroblastoma biology,
understanding upstream regulators is also of key importance. LIN28B has been shown to positively regulate MYCN
levels through let-7 binding. In keeping with its important upstream regulator role for MYCN, LIN28B itself
has been recognized as a bona fide driver gene by mouse
modeling. 18 In addition, using an unbiased LIN28B-39
untranslated region reporter screen, we found that miR26a-5p and miR-26b-5p regulate LIN28B expression and
that MYCN indirectly affects the expression of miR-26a-5p,
hence regulating LIN28B. Furthermore, MYCN directly
regulates LIN28B expression through LIN28B promoter
binding. 43 Taken together, these data point to a complex regulatory relationship between both genes.44 Other
MYCN upstream regulators include SIRT1, SIRT2, and
PRMT5. 45-47
The finding of ALK mutations and rare amplifications in
neuroblastomas was exciting to neuroblastoma researchers
e549
and pediatric oncologists because this represented an important novel druggable target. Several small molecules
are now under clinical testing; although resistance to
such compounds has been reported, further in-depth investigation of the optimal design and potential powerful
combination therapies must be explored to exploit the full
potential of this molecular target.48,49 In this context, we
performed an in-depth analysis of downstream mutant ALK
signaling to gain a deeper understanding of the rewired
oncogenic signaling in ALK mutant neuroblastoma cells and
to provide a molecular basis to gain insights into possible
drug-resistance mechanisms and drug-induced compensatory pathways. This study provided a rich data set for
exploration and showed mutant ALK-driven upregulation
of mitogen-activated protein kinase negative feedback
regulators, among others, and upregulation of RET and RETdriven sympathetic neuronal markers of the cholinergic
lineage.50 Although this most likely connects ALK and RET
signaling in a developmental context (in keeping with
other observations),51,52 this observation also offers novel
venues for drug treatments that are currently under
investigation.
ALK has been connected to MYCN through multiple
mechanisms. ALK was shown to drive MYCN expression
by promotor activation53 and activation through ERK5.54
In addition, activated ALK downstream signaling through
phosphoinositide 3-kinase/AKT controls glycogen synthase
kinase 3 beta activity and MYCN protein stabilization. These
mechanisms explain the observed cooperative effect of
mutant ALK accelerated tumor formation in MYCN transgenic mice and zebrafish.29,31,55
The Whole Genome and Nothing But the Whole

Genome
In keeping with recent findings in other embryonal tumors,
the overall mutation burden was low, with an average
of fewer than 10 mutations. 56-59 No additional recurrent mutations with significant frequency, such as ALKactivating mutations, were found; however, Molenaar
et al56 collectively demonstrated that an increase in implication of low-frequency mutations of neuritogenesis
genes was observed overall. In addition, chromothripsis,
a complex pattern of interchromosomal and/or intrachromosomal rearrangements believed to arise through
a single time-point event, was also demonstrated in several cases. Perhaps the main power of whole-genome sequencing was the possibility to combine both mutation
and copy-number events at single base-pair resolution. By
combining such data, further recurrent events emerged,
including the ARID1A and ARID1B epigenetic components
of the SWI/SNF remodeling complex and, most notably,
ATRX inactivating events (mainly deletions) and hTERT
activating structural rearrangements. ATRX encodes a SWI/
SNF chromatin-remodeling ATP-dependent helicase; remarkably, ATRX mutations occur mainly in older children,
with a frequency rising to greater than 40% in children
e550
older than age 14. It was shown that most of the tumors
with ATRX loss showed evidence of alternative lengthening
of telomeres. Two recent articles further showed mutual
exclusivity for ATRX inactivating events versus hTERT activating rearrangements. hTERT is a known MYCN target
gene and previous lower-resolution DNA copy-number
analyses revealed recurrent gains and rearrangements
affecting the hTERT locus.60 This type of alteration was
studied in more detail using whole-genome sequencing
approaches and revealed that 5p15.33 rearrangements
juxtapose the TERT coding sequence to strong enhancer
elements, resulting in massive chromatin remodeling and
DNA methylation of the affected region and transcriptional
upregulation of hTERT.61,62 TERT rearrangements (23%),
ATRX deletions (11%), and MYCN amplifications (37%) identified three almost nonoverlapping groups of high-stage
neuroblastoma, each associated with very poor prognosis.62
Of further interest, sequencing whole genomes of diagnostic and relapsed tumors has provided both insight into
the nature of mutations driving therapy resistance and
molecular evidence for the existence of minor subclones
that emerge during therapy (e.g., ALK mutant subclones
in otherwise ALK wild-type tumors at diagnosis, among
others). 63,64
NEUROBLASTOMA THERAPY UPDATE

Because of the heterogeneous biology of neuroblastoma, its
clinical behavior and prognosis range from near uniform
survival to high risk for fatal demise. Significant advances in
our knowledge of neuroblastoma biology have led to improved prognostic stratification and appropriate tailoring of
therapy, although our evolving molecular understanding has
yet to translate fully into novel therapies. The International
Risk Group (INRG) classification system uses clinical and
biologic factors that have been shown to predict prognosis and risk of recurrence, including age, International
Neuroblastoma Staging System stage histopathology, DNA
index (ploidy), segmental chromosomal gains or losses, and
MYCN amplification,65 to classify patients as having low-risk,
intermediate-risk, and high-risk neuroblastoma. In addition, a
National Cancer Institutesponsored Clinical Trials Planning
Meeting has proposed revisions to the International Response
Criteria. These revised criteria will integrate tumor response
in the primary tumor, soft tissue and bone metastases, and
bone marrow. Primary and metastatic soft tissue sites will
be assessed using RECIST and 123I-metaiodobenzylguanidine
(MIBG) or 18F-fluorodeoxyglucose (FDG)PET scans if the
tumor is MIBG nonavid. 123I-MIBG or FDG-PET scans, for
MIBG nonavid disease, replace 99technetium bone scans for
osteomedullary metastases assessment. Bone marrow
will be assessed by histology/immunohistochemistry and
cytology/immunocytology. Urinary catecholamine levels
will not be included in response assessment. Overall response will be defined as complete response, partial response, minor response, stable disease, and progressive
disease. These international collaborations provide the
opportunity for comparison across international clinical

trials and have opened the door for international collaborative trials.
Neuroblastoma: When Less Is More

Low-risk disease is characterized by a localized primary
tumor (INRGSS L1) or INRGSS stage MS without the presence of MYCN amplification. Survival rates for patients
with low-risk neuroblastoma are excellent with surgery
alone and rare recurrences can often be cured with salvage chemotherapy, resulting in survival rates of greater
than 95%. 66-68 Chemotherapy is reserved for patients
with localized neuroblastoma who have life- or organthreatening symptoms or the minority of patients who
experience recurrence or progressive disease. Several
clinical trials have further established the potential for
spontaneous regression of localized neuroblastoma, especially among patients who are diagnosed at younger
than age 12 months.69-71 In addition, recent clinical trials
support a continued reduction of chemotherapy exposure and surgery for the majority of low-risk asymptomatic patients, and studies to examine strategies to
biologically identify the rare patient with localized disease who will have a poor outcome are underway.
Intermediate-risk classification encompasses a wide
spectrum of disease for which surgical resection and
moderate-dose multi-agent chemotherapy are the backbone of therapy. Intermediate risk includes subsets of
patients with locoregional disease (INRGSS L2) without
MYCN amplification and infants (younger than 18 months
at diagnosis) with metastatic disease and favorable tumor
biologic features including no MCYNA. Survival after
moderate-dose chemotherapy including carboplatin or
cisplatin, doxorubicin, etoposide, and cyclophosphamide is
greater than 90% for patients whose tumors exhibit favorable clinical and biologic characteristics, including infants with stage M who lack MYCN amplification. 72,73
Extent of surgical resection does not affect prognosis.72
Ongoing studies are investigating whether further reduction
(and, in some cases, elimination) of therapy for the majority
of patients with favorable histology and genomics can be
achieved.
Despite the overall excellent prognosis for the majority of
patients with intermediate-risk neuroblastoma, the prognosis is less favorable for patients older than 18 months with
locoregional disease (INRGSS L2) and unfavorable histology
or segmental chromosomal aberrations or those infants
with metastatic disease with unfavorable biologic characteristics. Event-free survival for this rare cohort of patients
ranges from 40% to 70%72,74-76 with a wide range of therapy
intensity used. Improved understanding of the underlying
tumor biology and international collaborative clinical trials
are needed to better define optimal therapy for this rare
cohort of patients.
High-Risk Neuroblastoma: Beyond Chemotherapy

Dose Intensity
High-risk neuroblastoma is defined by presence of tumor
MYCN amplification or children with metastatic disease
older than age 18 months at diagnosis. High-risk neuroblastoma is largely chemotherapy responsive, but approximately one-half of patients still experience disease relapse
despite improvements in overall response rates. Standard
therapy for patients with high-risk neuroblastoma has
evolved through the results of randomized clinical trials
performed by international cooperative groups and involves three components: multi-agent chemotherapy induction and surgical resection, consolidation using highdose chemotherapy with autologous hematopoietic stem
cell rescue, and postconsolidation/maintenance treatment
of minimal residual disease using a biologic differentiating
agent (isotretinoin) and immunotherapy with antiganglioside
(GD2) antibody plus cytokines.
Dose intensity of chemotherapy is an important factor
in successful treatment of high-risk neuroblastoma. Doseintensive multi-agent induction chemotherapy provided improved event-free survival compared with a less-intensive
chemotherapy regimen.77 Additional randomized trials
demonstrated that the use of myeloablative therapy and
autologous hematopoietic stem cell rescue further improved outcomes for high-risk neuroblastoma compared
with use of conventionally dosed chemotherapy or
observation.78-80 More recently, randomized clinical trials
demonstrated that the specific agents used for consolidation
chemotherapy (busulfan/melphalan)81 or the intensity of
consolidation therapy (tandem transplant; J. Park, personal
communication, April 2016) may further improve outcomes
for high-risk neuroblastoma.
Randomized clinical trials have also highlighted the role
of treating minimal residual disease to improve outcomes
for patients with high-risk neuroblastoma. Isotretinoin, a
noncytotoxic differentiating agent, led to improved eventfree survival from high-risk neuroblastoma. Furthermore,
the addition of immunotherapy utilizing anti-GD2 antibody
plus cytokines improved event-free and overall survival,
becoming the first example of the benefits of immunetargeted therapy for treatment of a solid tumor in
childhood.
Although outcomes for patients with high-risk neuroblastoma have markedly improved over the past 2 decades,
our current therapies for these patients remain suboptimal. Chemotherapy and radiotherapy resistance remain
the hallmark of failure. However, optimism remains as our
expanding knowledge of tumor biology provides ample
opportunities for the introduction of novel molecularly
targeted agents into an optimized backbone of chemotherapy and radiotherapy. Novel therapeutic approaches
may target more generalizable features of neuroblastoma,
including surface epitopes such as GD2 for immunotherapy 82 and targeted radionuclide delivery. Alternatively, the molecularly targeted therapy paradigm includes
e551
identification and pharmacologic antagonism of an oncogenic driver as a novel therapeutic.

Ongoing and planned clinical trials will exploit the radiosensitivity of neuroblastoma by incorporating 131I-MIBG
targeted radiotherapy. Neuroblastoma is one of the most
radiosensitive tumors of childhood, resulting in a marked
decreased in primary site-specific recurrence, yet 50% of
patients have recurrence in bone marrow and/or cortical
bone.83 Efforts to radiate all sites of metastatic disease
using conventional total body irradiation (TBI) have not
clearly benefited patients, in part because of toxicityassociated limitations of TBI beyond a dose of 1,000 to
1,200 cGy and associated second malignancies.84 MIBG,
a norepinephrine analog, is concentrated selectively in sympathetic tissues. Clinical trials have demonstrated the
safety of 131I-MIBG (1218 mCi/kg) administered with
chemotherapy, and objective response rates of up to 25%
have been documented among patients with recurrent and
refractory disease.85-87 131I-MIBG can be safely combined
with myeloablative chemotherapy for patients with a poor
response to induction chemotherapy with additional antitumor efficacy.88
Future therapies must also build on the success of antiGD2 monoclonal antibody immunotherapy and develop
novel approaches to enhance antibody efficacy, such as the
addition of lenolidomide,89 which increases cytokine levels
and activates natural killer cells, combination with cytotoxic
therapy that may alter the inhibitory tumor microenvironment (R. Mody, personal communication, April 2016) or
active immunization with anti-idiotype antibodies are being
studied among patients with relapsed disease.90 Promising
novel immunotherapies genetically modify a patients own
T cells to be redirected against tumor-associated antigens
(e.g., GD2, L1CAM). Autologous T cells engineered to overexpress chimeric antigen receptors are infused and have
been shown to persist and demonstrate antitumor activity
among patients with neuroblastoma.91,92
Translation of our evolving understanding of neuroblastoma biology and the increasing availability of active clinical
compounds has greatly enhanced the availability of molecularly targeted agents. However, these remain in earlyphase clinical trials for patients with recurrent high-risk
neuroblastoma, with the ultimate goal to bring those with
the most encouraging activity into front-line therapy. For
patients with MYCN-amplified neuroblastoma, preclinical
studies suggest that BET bromodomain inhibitors can induce
cell death by interfering with MYCN transcription.93 Other
drugs that have effects on MYCN stability (aurora kinase A
and mTOR inhibitors)5 as well as those that target MYCdependent metabolic changes (difluoromethylornithine)94
are being studied. ALK is a pharmacologically tractable target
and early experience suggests that its inhibition can induce
responses in pediatric tumors with activating ALK mutations.95,96 Emergent resistance to targeted therapeutics
is a concern.97,98 A major issue is that ALK mutations
confer differential sensitivity to available ALK inhibitors.98
The combination of an ALK inhibitor with cytotoxic
e552
chemotherapy may augment activity and will be studied in

up-front treatment of children whose neuroblastoma harbors ALK aberrations. Genetic studies of relapsed neuroblastoma demonstrate genetic evolution of tumors and
further support combination of molecularly targeted agents
for treatment of relapsed disease.64
NEUROBLASTOMA SURVIVORSHIP
As described, there has been tremendous success in the
treatment of neuroblastoma. Five-year relative survival has
increased from 54% among patients diagnosed between
1975 and 1984 to 78% for those diagnosed between 2005
and 2011.99 Subsequently, there has been a growing population of survivors of neuroblastoma. Consistent with other
childhood cancer survivors, these advances have not come
without a cost.
The long-term outcomes of 954 survivors of neuroblastoma diagnosed between 1970 and 1986 who are participants in the North American cohort, the Childhood Cancer
Survivor Study (CCSS), were reported in 2009.100 The 25-year
cumulative late mortality was 6%. Compared with the sexand age-matched general population, the risk of death was
almost sixfold higher (standardized mortality ratio [SMR],
5.6; 95% CI, 4.46.9). The most common cause of late
mortality was recurrence, followed by an SMN (SMR, 10.9;
95% CI, 5.818.7), pulmonary complications (SMR,
11.4; 95% CI, 3.129.3), and cardiac complications (SMR, 5.0;
95% CI, 1.014.5). The 20-year cumulative incidence of
a chronic health condition was 41.1%. Compared with
their siblings, survivors had an 8.3-fold risk of developing a
chronic health condition. Survivors of neuroblastoma were
at increased risk for at least one of the following health
complications: neurologic, musculoskeletal, endocrine, and
sensory complications, with 20-year cumulative incidences
of 30%, 8%, 8%, and 9%, respectively. In the CCSS cohort,
30 survivors of neuroblastoma developed an SMN 5 years
or later after the neuroblastoma diagnosis. Of the patients
with SMNs, eight had thyroid cancer, five had renal cell
carcinoma, three had soft tissue sarcomas, two had AML,
two had breast cancer, one had a brain tumor, one had
Hodgkin lymphoma, and seven had other SMNs. The risk of
developing an SMN was eightfold higher among survivors of
neuroblastoma than the general population (standardized
incidence ratio, 8.0; 95% CI, 5.411.4). Similar to other
childhood cancer survivors, risks of late effects and SMNs
were attributable to radiation exposure (e.g., SMN, thyroid
disease, or cardiac disease), cisplatin exposure (e.g., hearing
loss or renal disease), alkylator exposure (e.g., renal disease, infertility, or risk of treatment-related acute myeloid
leukemia/myelodysplastic syndrome), and other treatment
exposures.
Notably, the outcomes reported from the CCSS reflect
outcomes associated with the treatment and survival success of an earlier era for neuroblastoma. Although data on
stage and/or risk group were not available in the CCSS
analysis, only 20 survivors had undergone bone marrow
treatment with TBI.103,108 Survivors with growth failure

often exhibit suboptimal responses to growth hormone (GH)
therapy.109,110 The etiology of this striking growth failure
and inadequate GH response is unclear and may be attributable to compromised spine growth after radiation.
Recent data have suggested that it may also be a consequence of premature epiphyseal closure caused by 13-cisretinoic acid exposure.103,111 Other prevalent endocrine late
effects include hypothyroidism, abnormal glucose metabolism, and delayed/abnormal pubertal development and
maturation.103,111,112 The high prevalence of hypothyroidism in survivors of neuroblastoma reflects damage from
diagnostic exposure to radioactive iodine tracers (131I-MIBG)
and, in some cases, therapeutic radiation therapy that
involves the thyroid.113 A number of investigators reported
laboratory evidence of diabetes and metabolic syndrome
among survivors of high-risk neuroblastoma.112,114 Female
survivors may demonstrate abnormal pubertal progression
and premature ovarian failure; puberty and gonadal function has not been characterized in male individuals.103 In
addition to endocrine late effects, studies of survivors of
high-risk neuroblastoma suggest elevated rates of sensorineural hearing loss,104-108 cardiac dysfunction,103,104,108
hypertension/renal dysfunction,104-108 and neurocognitive
deficits including behavioral problems as well as speech
and/or learning disabilities.103,105 Importantly, these studies
were small institutional studies limiting investigators to
have the power to describe true incidence and risk of late
effects and identify the risk factors associated with their
development.
Recognizing the gap in knowledge regarding the combined effects on morbidity and quality of life of high-dose
chemotherapy, radiation, and non-TBIbased stem cell
transplantation in combination with cis-retinoic acid, immunocytokines, and immunotherapy in high-risk survivors
of neuroblastoma, researchers have identified a cohort
of 5-year survivors (. 700) of high-risk neuroblastoma
treated between 2000 and 2011 who were enrolled in the
Childrens Oncology Group (COG) Neuroblastoma Biology
transplantation, suggesting that most of the participants did

not have high-risk disease.100 Thus, these data do not reflect
the late effects or SMNs experienced by patients with recently treated neuroblastoma, such as high-risk patients
who received more modern therapies such as tandem hematopoietic stem cell transplantation, biologic therapies,
and immunotherapy. Currently, survivorship researchers are
focused on the questions of whether risk-adapted therapies
improve the late effect profiles of low- and intermediate-risk
survivors and, importantly, identifying the long-term outcomes of high-risk survivors exposed to therapies never
given alone or in combination to any other population of
childhood cancer survivors.
Regarding the effects of risk-based therapy on the longterm outcomes of survivors of neuroblastoma, a recent
study examined second cancers in the Surveillance, Epidemiology, and End Results database across treatment eras.
Among 2,801 patients with neuroblastoma treated between
1973 and 2006 (era 1, 1973 to 1989; era 2, 1990 to 1996;
and era 3, 1997 to 2006), 34 patients developed an SMN
(14 developed a carcinoma and 10 had a hematologic malignancy). Importantly, there was no difference in the incidence of
SMNs in era 1 compared with era 3. Consistent with other
institutional studies reporting a high incidence of SMNs in
high-risk survivors,101 the 30-year cumulative incidence of
SMNs among high-risk patients was 10.4% (95% CI, 4.020.5)
compared with 3.6% among lower-risk patients.102 A large,
comprehensive analysis of SMNs among these high-risk
survivors is necessary to examine the effect of modern
regimens and exposures (including the combination of radiotherapy, chemotherapy, cis-retinoic acid, immunotherapy, as well as 131I-MIBG) on risk for SMNs.
Current knowledge regarding the long-term outcomes of
high-risk survivors of neuroblastoma treated with more
modern approaches is limited to small institutional series.
These studies suggest a high burden of multiple late effects
for these survivors (summarized in Table 1).103-108 Endocrine
late effects are prominent, with most series identifying
growth failure and short stature even in the absence of
TABLE 1. Prevalence of Late Effects Among Survivors of High-Risk Neuroblastoma

Late Effect
Cohen, 2014103
Laverdiere,
`
2005104
Moreno, 2013105
Perwein, 2011106
Hobbie, 2008108
Trahair, 2007107
No. of Survivors
51
63
57
16
13
23
5 (8)
Late Effect
10 (20)
Pulmonary
12 (19)
4 (25)
7 (54)
Hypertension/Renal
6 (9)
8 (14)
10 (63)
8 (61)
10 of 21 (47)
Growth Abnormality
10 (20)
6 (9)
4 (7)
7 (44)
7 (54)
13 of 13 (100)
Gonadal Failure
1 (8)
Cardiac
9 (12)
13 (41)*
2 (4)
2 (13)
2 (15)
5 of 6 (83)
Hypothyroid
29 (49)
15 (24)
1 (2)
8 (50)
7 (54)
4 of 14 (29)
Hearing Loss
37 (73)
39 (62)
28 (49)
6 (38)
12 (92)
11 of 15 (73)
Neurocognitive
11 (38)
8 (13)
12 (21)
Data are given as numbers (percentages).

*Of 32 female survivors, 13 had ovarian failure.
e553
Study ANBL00B1.115 The COG LEAHRN (Late Effects After High

Risk Neuroblastoma) study (ALTE 15N2) plans to open in 2016
and will recruit almost 400 of these survivors for a crosssectional study to characterize the survivors late effects and
SMNs, assess for predictors of these late effects and SMNs,
and identify factors predictive of decreased health-related
quality of life. Furthermore, ALTE 15N2 will serve to establish a cohort of high-risk survivors of neuroblastoma, with
stored peripheral blood samples, as a resource for future
investigation.
Understanding of the late effects and SMNs of survivors is
essential for both ensuring the long-term health of survivors
of neuroblastoma as well as informing the design of future
neuroblastoma treatment trials aimed at improved survival
along with a concurrent reduction in late effects. While we
continue to refine our understanding of the late effects
of modern neuroblastoma treatments, all neuroblastoma
survivors should continue to participate in lifelong riskbased health care with surveillance and early intervention
of late effects and SMNs, as recommended by the Institute
of Medicine and the COG.116,117
EPILOGUE: NEW TECHNIQUES, NEW

DISCOVERIES, NEW IDEAS...IN THAT ORDER
Referencing the quote of Nobel prize winner Sydney Brenner
above, it is fair to say that the past decade of neuroblastoma
research has been tremendously fruitful and, most importantly, many novel insights in biology and emerging novel
drugs, not the least of those targeting MYCN or MYCN-driven
pathways, are opening new exciting possibilities.7,93,118-123
Now, we are facing a novel wave of technological developments that will further fuel discoveries and new ideas
and venues for drug treatment. These include the study of
long noncoding RNAs124,125; single cell profiling at the DNA,
RNA, and chromatin levels126; novel and very powerful
proteome and interactome methods; metabolome profiling;
and, importantly, further exploration of zebrafish and mouse
models using powerful novel genome-editing technologies.127,128 These efforts must continue to enhance our
understanding of neuroblastoma clinical behavior and inform the design of optimal therapy for children with neuroblastoma, therapy that will enhance survival while
minimizing late effects.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
e554
Schwab M, Alitalo K, Klempnauer KH, et al. Amplified DNA with limited

homology to myc cellular oncogene is shared by human neuroblastoma cell lines and a neuroblastoma tumour. Nature. 1983;305:
245-248.
Speleman F, De Preter K, Vandesompele J. Neuroblastoma genetics
and phenotype: a tale of heterogeneity. Semin Cancer Biol. 2011;21:
238-244.
Irwin MS, Park JR. Neuroblastoma: paradigm for precision medicine.
Pediatr Clin North Am. 2015;62:225-256.
Louis CU, Shohet JM. Neuroblastoma: molecular pathogenesis and
therapy. Annu Rev Med. 2015;66:49-63.
Barone G, Anderson J, Pearson AD, et al. New strategies in neuroblastoma: therapeutic targeting of MYCN and ALK. Clin Cancer Res.
2013;19:5814-5821.
Cheung NK, Dyer MA. Neuroblastoma: developmental biology,
cancer genomics and immunotherapy. Nat Rev Cancer. 2013;13:
397-411.
Bagatell R, Cohn SL. Genetic discoveries and treatment advances in
neuroblastoma. Curr Opin Pediatr. 2016;28:19-25.
Schulte JH, Eggert A. Neuroblastoma. Crit Rev Oncog. 2015;20:
245-270.
Jiang M, Stanke J, Lahti JM. The connections between neural crest
development and neuroblastoma. Curr Top Dev Biol. 2011;94:
77-127.
Nakagawara A. Trk receptor tyrosine kinases: a bridge between cancer
and neural development. Cancer Lett. 2001;169:107-114.
Nakagawara A, Ohira M. Comprehensive genomics linking between
neural development and cancer: neuroblastoma as a model. Cancer
Lett. 2004;204:213-224.
Schleiermacher G, Janoueix-Lerosey I, Delattre O. Recent insights into
the biology of neuroblastoma. Int J Cancer. 2014;135:2249-2261.
Domingo-Fernandez R, Watters K, Piskareva O, et al. The role of
genetic and epigenetic alterations in neuroblastoma disease pathogenesis. Pediatr Surg Int. 2013;29:101-119.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
Henrich KO, Bauer T, Schulte J, et al. CAMTA1, a 1p36 tumor suppressor candidate, inhibits growth and activates differentiation
programs in neuroblastoma cells. Cancer Res. 2011;71:3142-3151.
Liu Z, Rader J, He S, et al. CASZ1 inhibits cell cycle progression
in neuroblastoma by restoring pRb activity. Cell Cycle. 2013;12:
2210-2218.
Fujita T, Igarashi J, Okawa ER, et al. CHD5, a tumor suppressor gene
deleted from 1p36.31 in neuroblastomas. J Natl Cancer Inst. 2008;100:
940-949.
Mosse YP, Laudenslager M, Longo L, et al. Identification of ALK as a
major familial neuroblastoma predisposition gene. Nature. 2008;455:
930-935.
Molenaar JJ, Domingo-Fernandez R, Ebus ME, et al. LIN28B induces
neuroblastoma and enhances MYCN levels via let-7 suppression. Nat
Genet. 2012;44:1199-1206.
Janoueix-Lerosey I, Lequin D, Brugières L, et al. Somatic and germline
activating mutations of the ALK kinase receptor in neuroblastoma.
Nature. 2008;455:967-970.
Chen Y, Takita J, Choi YL, et al. Oncogenic mutations of ALK kinase in
neuroblastoma. Nature. 2008;455:971-974.
George RE, Sanda T, Hanna M, et al. Activating mutations in ALK
provide a therapeutic target in neuroblastoma. Nature. 2008;455:
975-978.
Capasso M, Diskin S, Cimmino F, et al. Common genetic variants in
NEFL influence gene expression and neuroblastoma risk. Cancer Res.
2014;74:6913-6924.
Diskin SJ, Capasso M, Diamond M, et al. Rare variants in TP53 and
susceptibility to neuroblastoma. J Natl Cancer Inst. 2014;106:dju047.
Maris JM, Mosse YP, Bradfield JP, et al. Chromosome 6p22 locus
associated with clinically aggressive neuroblastoma. N Engl J Med.
2008;358:2585-2593.
Diskin SJ, Capasso M, Schnepp RW, et al. Common variation at 6q16
within HACE1 and LIN28B influences susceptibility to neuroblastoma.
Nat Genet. 2012;44:1126-1130.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
Capasso M, Devoto M, Hou C, et al. Common variations in BARD1

influence susceptibility to high-risk neuroblastoma. Nat Genet. 2009;
41:718-723.
Diskin SJ, Hou C, Glessner JT, et al. Copy number variation at 1q21.1
associated with neuroblastoma. Nature. 2009;459:987-991.
Wang K, Diskin SJ, Zhang H, et al. Integrative genomics identifies LMO1
as a neuroblastoma oncogene. Nature. 2011;469:216-220.
Zhu S, Lee JS, Guo F, et al. Activated ALK collaborates with MYCN in
neuroblastoma pathogenesis. Cancer Cell. 2012;21:362-373.
Althoff K, Beckers A, Bell E, et al. A Cre-conditional MYCN-driven
neuroblastoma mouse model as an improved tool for preclinical
studies. Oncogene. 2015;34:3357-3368.
Heukamp LC, Thor T, Schramm A, et al. Targeted expression of mutated ALK induces neuroblastoma in transgenic mice. Sci Transl Med.
2012;4:141ra91.
Henssen A, Althoff K, Odersky A, et al. Targeting MYCN-driven
transcription by BET-bromodomain inhibition. Clin Cancer Res.
Epub 2015 Dec 2.
Huang M, Weiss WA. Neuroblastoma and MYCN. Cold Spring Harb
Perspect Med. 2013;3:a014415.
Seton-Rogers S. RNA splicing: MYC maintains high-fidelity splicing. Nat
Rev Cancer. 2015;15:385.
Zhang S, Wei JS, Li SQ, et al. MYCN controls an alternative RNA splicing
program in high-risk metastatic neuroblastoma. Cancer Lett. 2016;
371:214-224.
Walz S, Lorenzin F, Morton J, et al. Activation and repression by
oncogenic MYC shape tumour-specific gene expression profiles.
Nature. 2014;511:483-487.
Kress TR, Sabo` A, Amati B. MYC: connecting selective transcriptional control to global RNA production. Nat Rev Cancer. 2015;15:
593-607.
Conacci-Sorrell M, McFerrin L, Eisenman RN. An overview of MYC
and its interactome. Cold Spring Harb Perspect Med. 2014;4:
a014357.
Wang C, Liu Z, Woo CW, et al. EZH2 mediates epigenetic silencing of
neuroblastoma suppressor genes CASZ1, CLU, RUNX3, and NGFR.
Cancer Res. 2012;72:315-324.
Schulte JH, Horn S, Otto T, et al. MYCN regulates oncogenic microRNAs
in neuroblastoma. Int J Cancer. 2008;122:699-704.
De Brouwer S, Mestdagh P, Lambertz I, et al. Dickkopf-3 is regulated by
the MYCN-induced miR-17-92 cluster in neuroblastoma. Int J Cancer.
2012;130:2591-2598.
Mestdagh P, Bostrom AK, Impens F, et al. The miR-17-92 microRNA
cluster regulates multiple components of the TGF-b pathway in
neuroblastoma. Mol Cell. 2010;40:762-773.
Beckers A, Van Peer G, Carter DR, et al. MYCN-targeting miRNAs are
predominantly downregulated during MYCNdriven neuroblastoma
tumor formation. Oncotarget. 2015;6:5204-5216.
Cotterman R, Knoepfler PS. N-Myc regulates expression of pluripotency genes in neuroblastoma including lif, klf2, klf4, and lin28b. PLoS
One. 2009;4:e5799.
Marshall GM, Liu PY, Gherardi S, et al. SIRT1 promotes N-Myc oncogenesis through a positive feedback loop involving the effects of
MKP3 and ERK on N-Myc protein stability. PLoS Genet. 2011;7:
e1002135.
Liu PY, Xu N, Malyukova A, et al. The histone deacetylase SIRT2
stabilizes Myc oncoproteins. Cell Death Differ. 2013;20:503-514.
Park JH, Szemes M, Vieira GC, et al. Protein arginine methyltransferase
5 is a key regulator of the MYCN oncoprotein in neuroblastoma cells.
Mol Oncol. 2015;9:617-627.
Crescenzo R, Inghirami G. Anaplastic lymphoma kinase inhibitors. Curr
Opin Pharmacol. 2015;23:39-44.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
Bresler SC, Weiser DA, Huwe PJ, et al. ALK mutations confer differential oncogenic activation and sensitivity to ALK inhibition therapy in
neuroblastoma. Cancer Cell. 2014;26:682-694.
Lambertz I, Kumps C, Claeys S, et al. Upregulation of MAPK negative
feedback regulators and RET in mutant ALK neuroblastoma: implications for targeted treatment. Clin Cancer Res. 2015;21:3327-3339.
Cazes A, Lopez-Delisle L, Tsarovina K, et al. Activated Alk triggers
prolonged neurogenesis and Ret upregulation providing a therapeutic target in ALK-mutated neuroblastoma. Oncotarget. 2014;5:
2688-2702.
Reiff T, Huber L, Kramer M, et al. Midkine and Alk signaling in
sympathetic neuron proliferation and neuroblastoma predisposition.
Development. 2011;138:4699-4708.
Schonherr C, Ruuth K, Kamaraj S, et al. Anaplastic lymphoma kinase
(ALK) regulates initiation of transcription of MYCN in neuroblastoma
cells. Oncogene. 2012;31:5193-5200.
Umapathy G, El Wakil A, Witek B, et al. The kinase ALK stimulates the
kinase ERK5 to promote the expression of the oncogene MYCN in
neuroblastoma. Sci Signal. 2014;7:ra102.
Berry T, Luther W, Bhatnagar N, et al. The ALK(F1174L) mutation
potentiates the oncogenic activity of MYCN in neuroblastoma. Cancer
Cell. 2012;22:117-130.
Molenaar JJ, Koster J, Zwijnenburg DA, et al. Sequencing of neuroblastoma identifies chromothripsis and defects in neuritogenesis
genes. Nature. 2012;483:589-593.
Pugh TJ, Morozova O, Attiyeh EF, et al. The genetic landscape of highrisk neuroblastoma. Nat Genet. 2013;45:279-284.
Cheung NK, Zhang J, Lu C, et al; St Jude Childrens Research
HospitalWashington University Pediatric Cancer Genome Project.
Association of age at diagnosis and genetic mutations in patients with
neuroblastoma. JAMA. 2012;307:1062-1071.
Sausen M, Leary RJ, Jones S, et al. Integrated genomic analyses identify
ARID1A and ARID1B alterations in the childhood cancer neuroblastoma. Nat Genet. 2013;45:12-17.
Kumps C, Fieuw A, Mestdagh P, et al. Focal DNA copy number changes
in neuroblastoma target MYCN regulated genes. PLoS One. 2013;8:
e52321.
Peifer M, Hertwig F, Roels F, et al. Telomerase activation by genomic rearrangements in high-risk neuroblastoma. Nature. 2015;526:
700-704.
Valentijn LJ, Koster J, Zwijnenburg DA, et al. TERT rearrangements are
frequent in neuroblastoma and identify aggressive tumors. Nat Genet.
2015;47:1411-1414.
Schramm A, Koster J, Assenov Y, et al. Mutational dynamics between
primary and relapse neuroblastomas. Nat Genet. 2015;47:872-877.
Eleveld TF, Oldridge DA, Bernard V, et al. Relapsed neuroblastomas
show frequent RAS-MAPK pathway mutations. Nat Genet. 2015;47:
864-871.
Cohn SL, Pearson AD, London WB, et al; INRG Task Force. The International Neuroblastoma Risk Group (INRG) classification system:
an INRG Task Force report. J Clin Oncol. 2009;27:289-297.
Strother DR, London WB, Schmidt ML, et al. Outcome after surgery
alone or with restricted use of chemotherapy for patients with lowrisk neuroblastoma: results of Childrens Oncology Group study
P9641. J Clin Oncol. 2012;30:1842-1848.
Perez CA, Matthay KK, Atkinson JB, et al. Biologic variables in the
outcome of stages I and II neuroblastoma treated with surgery as
primary therapy: a childrens cancer group study. J Clin Oncol. 2000;
18:18-26.
Simon T, Spitz R, Faldum A, et al. New definition of low-risk neuroblastoma using stage, age, and 1p and MYCN status. J Pediatr Hematol
Oncol. 2004;26:791-796.
e555
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
e556
Nuchtern JG, London WB, Barnewolt CE, et al. A prospective study of

expectant observation as primary therapy for neuroblastoma in young
infants: a Childrens Oncology Group study. Ann Surg. 2012;256:
573-580.
Rubie H, De Bernardi B, Gerrard M, et al. Excellent outcome with
reduced treatment in infants with nonmetastatic and unresectable
neuroblastoma without MYCN amplification: results of the prospective INES 99.1. J Clin Oncol. 2011;29:449-455.
Simon T, Hero B, Benz-Bohm G, et al. Review of image defined risk
factors in localized neuroblastoma patients: results of the GPOH NB97
trial. Pediatr Blood Cancer. 2008;50:965-969.
Baker DL, Schmidt ML, Cohn SL, et al; Childrens Oncology Group.
Outcome after reduced chemotherapy for intermediate-risk neuroblastoma. N Engl J Med. 2010;363:1313-1323.
De Bernardi B, Gerrard M, Boni L, et al. Excellent outcome with
reduced treatment for infants with disseminated neuroblastoma
without MYCN gene amplification. J Clin Oncol. 2009;27:1034-1040.
Oberthuer A, Juraeva D, Hero B, et al. Revised risk estimation and
treatment stratification of low- and intermediate-risk neuroblastoma
patients by integrating clinical and molecular prognostic markers. Clin
Cancer Res. 2015;21:1904-1915.
Schleiermacher G, Mosseri V, London WB, et al. Segmental chromosomal alterations have prognostic impact in neuroblastoma:
a report from the INRG project. Br J Cancer. 2012;107:1418-1422.
Meany HJ, London WB, Ambros PF, et al. Significance of clinical and
biologic features in stage 3 neuroblastoma: a report from the International Neuroblastoma Risk Group project. Pediatr Blood Cancer.
2014;61:1932-1939.
Pearson AD, Pinkerton CR, Lewis IJ, et al; European Neuroblastoma
Study Group; Childrens Cancer and Leukaemia Group (CCLG formerly
United Kingdom Childrens Cancer Study Group). High-dose rapid and
standard induction chemotherapy for patients aged over 1 year with
stage 4 neuroblastoma: a randomised trial. Lancet Oncol. 2008;9:
247-256.
Matthay KK, Villablanca JG, Seeger RC, et al; Childrens Cancer Group.
Treatment of high-risk neuroblastoma with intensive chemotherapy,
radiotherapy, autologous bone marrow transplantation, and 13cis-retinoic acid. N Engl J Med. 1999;341:1165-1173.
Berthold F, Boos J, Burdach S, et al. Myeloablative megatherapy with
autologous stem-cell rescue versus oral maintenance chemotherapy
as consolidation treatment in patients with high-risk neuroblastoma:
a randomised controlled trial. Lancet Oncol. 2005;6:649-658.
Pritchard J, Cotterill SJ, Germond SM, et al. High dose melphalan in the
treatment of advanced neuroblastoma: results of a randomised trial
(ENSG-1) by the European Neuroblastoma Study Group. Pediatr Blood
Cancer. 2005;44:348-357.
Ladenstein RL, Poetschger U, Luksch R, et al. Busulphan-melphalan
as a myeloablative therapy (MAT) for high-risk neuroblastoma: results
from the HR-NBL1/SIOPEN trial. J Clin Oncol. 2011;29 (suppl; abstr 2).
Yu AL, Gilman AL, Ozkaynak MF, et al; Childrens Oncology Group. AntiGD2 antibody with GM-CSF, interleukin-2, and isotretinoin for neuroblastoma. N Engl J Med. 2010;363:1324-1334.
Matthay KK, Atkinson JB, Stram DO, et al. Patterns of relapse after
autologous purged bone marrow transplantation for neuroblastoma:
a Childrens Cancer Group pilot study. J Clin Oncol. 1993;11:2226-2233.
Flandin I, Hartmann O, Michon J, et al. Impact of TBI on late effects in
children treated by megatherapy for stage IV neuroblastoma. A study
of the French Society of Pediatric Oncology. Int J Radiat Oncol Biol
Phys. 2006;64:1424-1431.
DuBois SG, Allen S, Bent M, et al. Phase I/II study of (131)I-MIBG with
vincristine and 5 days of irinotecan for advanced neuroblastoma.
Br J Cancer. 2015;112:644-649.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
French S, DuBois SG, Horn B, et al. 131I-MIBG followed by consolidation with busulfan, melphalan and autologous stem cell transplantation for refractory neuroblastoma. Pediatr Blood Cancer.
2013;60:879-884.
Gaze MN, Chang YC, Flux GD, et al. Feasibility of dosimetry-based highdose 131I-meta-iodobenzylguanidine with topotecan as a radiosensitizer in children with metastatic neuroblastoma. Cancer Biother
Radiopharm. 2005;20:195-199.
Yanik GA, Villablanca JG, Maris JM, et al. 131I-metaiodobenzylguanidine
with intensive chemotherapy and autologous stem cell transplantation
for high-risk neuroblastoma. A new approaches to neuroblastoma
therapy (NANT) phase II study. Biol Blood Marrow Transplant. 2015;
21:673-681.
Xu Y, Sun J, Sheard MA, et al. Lenalidomide overcomes suppression of
human natural killer cell anti-tumor functions by neuroblastoma
microenvironment-associated IL-6 and TGFb1. Cancer Immunol
Immunother. 2013;62:1637-1648.
Koehn TA, Trimble LL, Alderson KL, et al. Increasing the clinical
efficacy of NK and antibody-mediated cancer immunotherapy:
potential predictors of successful clinical outcome based on observations in high-risk neuroblastoma. Front Pharmacol. 2012;3:91.
Park JR, Digiusto DL, Slovak M, et al. Adoptive transfer of chimeric
antigen receptor re-directed cytolytic T lymphocyte clones in patients
with neuroblastoma. Mol Ther. 2007;15:825-833.
Louis CU, Brenner MK. Cellular immunotherapy for neuroblastoma:
a review of current vaccine and adoptive T cell therapeutics. Curr
Pharm Des. 2009;15:424-429.
Puissant A, Frumm SM, Alexe G, et al. Targeting MYCN in neuroblastoma by BET bromodomain inhibition. Cancer Discov. 2013;3:
308-323.
Bassiri H, Benavides A, Haber M, et al. Translational development of
difluoromethylornithine (DFMO) for the treatment of neuroblastoma.
Transl Pediatr. 2015;4:226-238.
Mosse YP, Lim MS, Voss SD, et al. Safety and activity of crizotinib for
paediatric patients with refractory solid tumours or anaplastic largecell lymphoma: a Childrens Oncology Group phase 1 consortium
Mosse YP, Wood A, Maris JM. Inhibition of ALK signaling for cancer
therapy. Clin Cancer Res. 2009;15:5609-5614.
Carpenter EL, Haglund EA, Mace EM, et al. Antibody targeting of
anaplastic lymphoma kinase induces cytotoxicity of human neuroblastoma. Oncogene. 2012;31:4859-4867.
Bresler SC, Wood AC, Haglund EA, et al. Differential inhibitor sensitivity of anaplastic lymphoma kinase variants found in neuroblastoma.
Sci Transl Med. 2011;3:108ra114.
Altekruse S, Krapcho M, Neyman N, et al. (eds). SEER Cancer Statistics
Review 19752007. Bethesda, MD: National Cancer Institute; 2010.
http://seer.cancer.gov/csr/1975_2007/,/csr/1975_2007/., based on
November 2009 SEER data submission, posted to the SEER web site.
Accessed April 7, 2016.
Laverdiere
` C, Liu Q, Yasui Y, et al. Long-term outcomes in survivors of
neuroblastoma: a report from the Childhood Cancer Survivor Study.
Martin A, Schneiderman J, Helenowski IB, et al. Secondary malignant
neoplasms after high-dose chemotherapy and autologous stem cell
rescue for high-risk neuroblastoma. Pediatr Blood Cancer. 2014;61:
1350-1356.
Applebaum MA, Henderson TO, Lee SM, et al. Second malignancies in
patients with neuroblastoma: the effects of risk-based therapy.
Cohen LE, Gordon JH, Popovsky EY, et al. Late effects in children
treated with intensive multimodal therapy for high-risk neuroblastoma:
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
high incidence of endocrine and growth problems. Bone Marrow

Transplant. 2014;49:502-508.
Laverdière C, Cheung NK, Kushner BH, et al. Long-term complications
in survivors of advanced stage neuroblastoma. Pediatr Blood Cancer.
2005;45:324-332.
Moreno L, Vaidya SJ, Pinkerton CR, et al; European Neuroblastoma
Study Group; Childrens Cancer and Leukaemia Group (CCLG) (formerly
UKCCSG). Long-term follow-up of children with high-risk neuroblastoma:
the ENSG5 trial experience. Pediatr Blood Cancer. 2013;60:1135-1140.
Perwein T, Lackner H, Sovinz P, et al. Survival and late effects in children
with stage 4 neuroblastoma. Pediatr Blood Cancer. 2011;57:629-635.
Trahair TN, Vowels MR, Johnston K, et al. Long-term outcomes in
children with high-risk neuroblastoma treated with autologous stem
cell transplantation. Bone Marrow Transplant. 2007;40:741-746.
Hobbie WL, Moshang T, Carlson CA, et al. Late effects in survivors of
tandem peripheral blood stem cell transplant for high-risk neuroblastoma. Pediatr Blood Cancer. 2008;51:679-683.
Olshan JS, Gubernick J, Packer RJ, et al. The effects of adjuvant
chemotherapy on growth in children with medulloblastoma. Cancer.
1992;70:2013-2017.
Willi SM, Cooke K, Goldwein J, et al. Growth in children after bone
marrow transplantation for advanced neuroblastoma compared with
growth after transplantation for leukemia or aplastic anemia.
J Pediatr. 1992;120:726-732.
Hobbie WL, Mostoufi SM, Carlson CA, et al. Prevalence of advanced
bone age in a cohort of patients who received cis-retinoic acid for
high-risk neuroblastoma. Pediatr Blood Cancer. 2011;56:474-476.
van Waas M, Neggers SJ, Raat H, et al. Abdominal radiotherapy:
a major determinant of metabolic syndrome in nephroblastoma and
neuroblastoma survivors. PLoS One. 2012;7:e52237.
van Santen HM, de Kraker J, van Eck BL, et al. High incidence of
thyroid dysfunction despite prophylaxis with potassium iodide during
(131)I-meta-iodobenzylguanidine treatment in children with neuroblastoma. Cancer. 2002;94:2081-2089.
Meacham LR, Sklar CA, Li S, et al. Diabetes mellitus in long-term
survivors of childhood cancer. Increased risk associated with radiation
therapy: a report for the childhood cancer survivor study. Arch Intern
Med. 2009;169:1381-1388.
Mody R, London WB, Park JR, et al. Late recurrences, mortality and
second cancers in five-year survivors of high risk neuroblastoma.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
126.
127.
128.
Paper presented at: 46th Congress of the International Society of

Paediatric Oncology; October 2014; Toronto, ON, Canada.
Hewitt M, Weiner SL, Simone JV (eds); National Cancer Policy Board;
Institute of Medicine; National Research Council. Childhood Cancer
Survivorship: Improving Care and Quality of Life. Washington, DC:
National Academies Press; 2003.
Childrens Oncology Group. Long-Term Follow-up Guidelines for
Survivors of Childhood, Adolescent, and Young Adult Cancers, version
4.0. Monrovia, CA: Childrens Oncology Group; 2014.
Ham J, Costa C, Sano R, et al. Exploitation of the apoptosis-primed
state of MYCN-amplified neuroblastoma to develop a potent and
specific targeted therapy combination. Cancer Cell. 2016;29:159-172.
Chayka O, DAcunto CW, Middleton O, et al. Identification and
pharmacological inactivation of the MYCN gene network as a therapeutic strategy for neuroblastic tumor cells. J Biol Chem. 2015;290:
2198-2212.
Sottile F, Gnemmi I, Cantilena S, et al. A chemical screen identifies the
chemotherapeutic drug topotecan as a specific inhibitor of the
B-MYB/MYCN axis in neuroblastoma. Oncotarget. 2012;3:535-545.
Molenaar JJ, Ebus ME, Geerts D, et al. Inactivation of CDK2 is synthetically lethal to MYCN over-expressing cancer cells. Proc Natl Acad
Sci USA. 2009;106:12968-12973.
He S, Liu Z, Oh DY, et al. MYCN and the epigenome. Front Oncol. 2013;
3:1.
Moore NF, Azarova AM, Bhatnagar N, et al. Molecular rationale for
the use of PI3K/AKT/mTOR pathway inhibitors in combination with
crizotinib in ALK-mutated neuroblastoma. Oncotarget. 2014;5:
8737-8749.
Mestdagh P, Fredlund E, Pattyn F, et al. An integrative genomics
screen uncovers ncRNA T-UCR functions in neuroblastoma tumours.
Oncogene. 2010;29:3583-3592.
Pandey GK, Kanduri C. Long noncoding RNAs and neuroblastoma.
Oncotarget. 2015;6:18265-18275.
Gawad C, Koh W, Quake SR. Single-cell genome sequencing: current
state of the science. Nat Rev Genet. 2016;17:175-188.
Olivares O, Dabritz JH, King A, et al. Research into cancer metabolomics: towards a clinical metamorphosis. Semin Cell Dev Biol. 2015;
43:52-64.
White R, Rose K, Zon L. Zebrafish cancer: the state of the art and the
path forward. Nat Rev Cancer. 2013;13:624-636.
e557
Curing Burnout in Oncology:

Mindful Self-Compassion,
Communication, and Practice
CHAIR
Tait D. Shanafelt, MD
Mayo Clinic
Rochester, MN
SPEAKERS
Anthony L. Back, MD
Seattle Cancer Care Alliance
Seattle, WA
Fay J. Hlubocky, PhD, MA
The University of Chicago Medicine
Chicago, IL
ADDRESSING BURNOUT IN ONCOLOGY
Addressing Burnout in Oncology: Why Cancer Care Clinicians

Are At Risk, What Individuals Can Do, and How Organizations
Can Respond
Fay J. Hlubocky, PhD, MA, Anthony L. Back, MD, and Tait D. Shanafelt, MD
OVERVIEW
Despite their benevolent care of others, today, more than ever, the cancer care professional who experiences overwhelming
feelings of exhaustion, cynicism, and inefficacy is in grave jeopardy of developing burnout. Clinicians are repeatedly physically
and emotionally exposed to exceedingly long hours in direct care with seriously ill patients/families, limited autonomy over
daily responsibilities, endless electronic documentation, and a shifting medical landscape. The physical and emotional wellbeing of the cancer care clinician is critical to the impact on quality care, patient satisfaction, and overall success of their
organizations. The prevention of burnout as well as targeting established burnout need to be proactively addressed at the
individual level and organizational level. In fact, confronting burnout and promoting wellness are the shared responsibility of
both oncology clinicians and their organizations. From an individual perspective, oncology clinicians must be empowered to
play a crucial role in enhancing their own wellness by identification of burnout symptoms in both themselves and their
colleagues, learning resilience strategies (e.g., mindful self-compassion), and cultivating positive relationships with fellow
clinician colleagues. At the organizational level, leadership must recognize the importance of oncology clinician well-being;
engage leaders and physicians in collaborative action planning, improve overall practice environment, and provide institutional
wellness resources to physicians. These effective individual and organizational interventions are crucial for the prevention and
improvement of overall clinician wellness and must be widely and systematically integrated into oncology care.
CASE PRESENTATION
r. M is a medical oncologist who finished her fellowship 5

years ago and is now working in a cancer center for a
large integrated health system. She is well respected and has
worked hard to develop her practice. Now she has a large
patient panel and is on track to make partner next year. She
is active in the clinical trials group and spends time every
week participating in conference calls and meetings so that
her patients have access to the latest treatments. But lately
she is feeling stretched pretty thin. Her phone seems to ring
constantly, even into the evening. Her son, age 4, is starting
preschool, and she does not want to leave it all to the nanny.
Her husband, an attorney, is also working long hours to
become partner at his law firm. Yesterday, she sat in her
office at the end of the day feeling overwhelmed by the pile
of charts she had to take homeshe knew shed be staying
up late after putting her son to bedand thought, How
much longer can I work like this?
The demanding lifestyle of the present-day oncology clinician has become increasingly overwhelming and burdensome
because of the evolving landscape of clinical care and medicine. Dr. M is an exceptionally trained, dedicated oncologist
working at optimal performance professionally and attempting
to meaningfully meet the needs of her patients and practice;
however, she is feeling besieged at effectively addressing the
desires of her family. She finds her present work-life balance
much to her dissatisfaction. Dr. M is exhibiting signs of a
common syndrome universally experienced by oncology clinicians today referred to as burnout.
WHAT IS BURNOUT: SIGNS AND SYMPTOMS

Originally described in the mid-1970s by psychologist Herbert Freudenberger,1,2 burnout is a condition that occurs
when work coupled with additional life pressures exceed the
ability to cope, resulting in physical and mental distress.111
Although definitions of burnout have varied over the years,
in health care, and especially oncology, it has traditionally
been defined as an occupational-related syndrome characterized by physical and emotional exhaustion, cynicism
From the Department of Medicine, Section of Hematology Oncology, The University of Chicago Medicine, Chicago, IL; Division of Oncology, University of Washington, Fred Hutchinson
Cancer Research Center, Seattle, WA; Mayo Clinic, Rochester, MN.
Corresponding author: Fay J. Hlubocky, PhD, MA, Department of Medicine, Section of Hematology/Oncology, University of Chicago Medicine, 5841 South Maryland Ave., MC2115,
Chicago, IL 60637; email: fhlubock@medicine.bsd.uchicago.edu.
271
HLUBOCKY, BACK, AND SHANAFELT
and depersonalization (sense of detachment or disengagement), and low sense of professional accomplishment.111
These three-dimensional signs of burnout exist along a continuum characterized by distinctly unique symptoms as well as
an overlap of symptoms (Sidebar 1).13 For example, the
symptoms of physical and emotional exhaustion may include:
chronic fatigue, cardiovascular issues, cognitive dysfunction,
insomnia, gastrointestinal complaints, and affective and behavioral distress (anger, depression, and anxiety). Cynicism
and depersonalization may be characterized by pessimism/
depression, isolation, demoralization, and detachment. A low
sense of personal accomplishment may lead to feelings of
inefficacy, decreased productivity, and overall dissatisfaction
in work-life balance. Evidence reveals that the initial physical
and emotional symptoms of burnout may slowly develop over
the course of 1 year.13 Burnout is a physical and mental
response that manifests as chronic occupational and interpersonal stressors arise and persevere over an extended
period of time.117 Dr. M is feeling mentally and physically
exhausted at work due to equally rising demands as an oncologist, resulting in overall decreased productivity and
increased workload brought home. She feels overwhelmed
in her role as a clinician, which negatively impacts her personal dual roles as wife and mother. Increasingly, these signs
and symptoms may adversely affect Dr. M, leading to potentially long-term personal and professional consequences.
A Group of Related Concepts

Burnout is at the center of a group of related concepts with
overlapping features.1820 Compassion fatigue is a state of
tension and preoccupation with the individual or cumulative
trauma of ones clients as manifested in three major domains
that parallel symptoms of post-traumatic stress disorders: (1)
hyperarousal (irritability, hypervigilance), (2) avoidance of
Sidebar 1. Signs and Symptoms of Burnout

Syndrome
Physical and emotional exhaustion

Chronic fatigue
Cardiovascular issues
Cognitive dysfunction
Insomnia
Gastrointestinal complaints
Affective and behavioral distress: anger, irritability,
depression, anxiety
Cynicism and depersonalization
Pessimism
Isolation
Demoralization
Detachment
Low sense of personal accomplishment
Feelings of inefficacy
Decreased productivity
Overall dissatisfaction in work-life balance
stressful situations, and (3) re-experiencing difficult events

through persistent or intrusive thoughts or even dreams.18
Moral distress is the pain or anguish affecting the mind,
body, or relationships in response to a situation in which
the person is aware of a moral problem, acknowledges moral
responsibility, and makes a moral judgment about the correct
action; yet, as a result of real or perceived constraints, participates in perceived moral wrongdoing.19 Empathy fatigue
is a state of emotional, mental, physical, and occupational
exhaustion that occurs as the counselors own wounds are
continually revisited by the clients life stories of chronic illness, disability, trauma, grief, and loss.20 Although these
concepts all capture a different aspect of burnout, they have
not been clearly distinguished empirically. Thus, this review
will focus on burnout, even though clinicians should be aware
of the variety of concepts in the literature.
KEY POINTS
272
Burnout is defined as an occupational-related syndrome

characterized by: physical and emotional exhaustion,
cynicism/depersonalization, and low sense of
professional accomplishment.
Multiple oncology-specific risk factors associated with
an increased susceptibility for the development of
burnout may trigger personal and professional
consequences.
Addressing clinician burnout by promoting oncology
clinician well-being needs to be tackled at the individual
level and organizational level.
At the individual level, oncology clinicians have an
important role in identifying symptoms, acquiring
resilience skills, and building positive relationships with
colleagues.
At the organizational level, leadership must optimize the
clinical practice environment and institutional culture in
order to promote clinician well-being.
Risk Factors Associated With Burnout

Multiple individual and organizational risk factors have been
associated with an increased susceptibility to develop
burnout in oncology (Sidebar 2).2,517,2128 Individual risk
factors are internally based dispositional risk factors consisting of sociodemographic and personality characteristics.
Prior research revealed specific individual risk factors associated with burnout including: female gender, younger
age (# 55 years), junior physicians (residents, fellows, or
physicians # 5 years from training), years in practice, and
single, unmarried/nonpartnered physicians.417,2128 Personality characteristics identified as independent risk factors
for burnout include compulsiveness, neuroticism, extraversion, conscientiousness, alexithymia, psychologic hardiness,
and type A behavior.2,2128 Empirical study of neurobiological
characteristics including genetic factors or biomarkers (e.g.,
catecholamines) as potential risk factors for burnout have
been not been supported.2 Finally, externally based,
Sidebar 2. Individual and Organizational Risk

Factors: Why Cancer Care Clinicians Are At Risk
Female gender
Younger age ( 55 years)
Junior physicians: oncology fellows, physicians 5 years
from training completion
Single, unmarried/nonpartnered physicians
Personality characteristics: compulsiveness, extraversion,
type A behavior
Increased time in direct patient care
High occupational demands
Lack of control over daily tasks
Increased administrative responsibilities
Use of electronic medical record systems
Limited decision making
Unclear job expectations
Lack of social support
Changing health care system
Care of patients who are terminally ill with cancer
environmental, occupational, and organizational risk factors specific to oncology identified as contributing central
causes to burnout include increased time in direct patient
care, high occupational demands, lack of control over daily
tasks, increased administrative responsibilities, use of
electronic medical record systems, limited decision
making, unclear job expectations, lack of social support,
and the changing landscape in health care system.6,9,21,22
As a young, female, conscientious, compulsive junior physician working exceedingly long hours in direct patient care in a
busy community practice, Dr. M is at an increased risk of
developing burnout, resulting in significant consequences.
Personal and Professional Consequences of Burnout

Burnout is not formally diagnosed as a disorder given it is
primarily recognized as an occupational-related condition;
however, it is incorporated in the International Statistical
Classification of Diseases and Related Health Problems, 10th
revision, under the category Problems related to lifemanagement difficulty (Z73.0) due to its adverse effects
and influence on the individuals overall health.2,29 Psychiatric
disorders such as depression and post-traumatic stress disorders may be both precursors to the development of burnout
as well as consequences of burnout.2,617,21,22,3032 It has been
linked to and compared with both stress and depression given
similar shared symptomology and metabolic, physiologic
systems involved (e.g., systematic inflammation or autonomous nervous system).30,31,3335 However, unlike stress, which
tends to be fairly short-term and resolves completely once the
stressful situation has changed, burnout is a complex, insidious
process gradually and progressively developing over an extended period of time.2,3 Traditionally, the burnout process
occurs in 12 nonspecific, dynamic stages ranging from a
compulsion to establish oneself to a development of
multiple behavioral and mood changes, causing the final

burnout syndrome.2,3,32 In fact, it is due in part to this
gradual development in exhaustion symptoms that burnout
has been difficult to detect, identify, and intervene early on,
therefore resulting in long-term enduring health consequences for the individual.2,628,3035 Unlike stress and depression, symptoms of burnout may resolve once the
individual changes a job or resolves if the work environment is altered. Yet, long-term untreated and unaddressed
burnout may lead to personal consequences such as chronic
health conditions (heart disease, stroke, or obesity) or
mental health conditions (depression, anxiety, substance
use, and suicide).2,617,2128,3035 Professionally, long-term
burnout may lead to diminished quality care, reduced professional satisfaction, and accomplishment.21,22 At present,
no formal tailored clinical diagnostic assessment exists to
identify burnout symptoms in the individual; however, selfreported screening tools such as the Maslach Burnout
Inventory (MBI)36 and the Mayo Clinic Physician Well-Being
Index37 have been used in research, occupational, and clinical
settings to empirically study burnout and burnout-related signs
for various health and medical staff populations, including
oncology.
Prevalence in Oncology
The global incidence of burnout has drastically increased
over the past decade for oncologists in the United States,
Europe, and Australia.3841 In 2005, the survey study by
Allegra et al38 of over 1,700 oncologists revealed early on
that nearly 62% of oncologists in community practice in the
United States reported experiencing specific symptoms of
burnout, including the top three signs: frustration (78%),
emotional exhaustion (69%), and lack of work satisfaction
(50%). Today, 45% of American Society of Clinical Oncology
(ASCO) member medical oncologists have reported experiencing emotional exhaustion and/or depersonalization
symptoms related to burnout.39 In Europe and Australia,
burnout rates vary significantly, ranging from 52% to 78%
depending on medical oncology specialty, practice, health
care systems, and screening tools used.9,40,41 For example, in
France, a mailed survey study of 340 medical and radiation
oncology fellows using the MBI revealed that 44% believed
burnout was prevalent and associated with low perception
of health status and a desire to leave medicine.40 In Australia, 36% of gynecologic oncologists surveyed reported a
high degree of emotional exhaustion, with 43% reporting a
desire to leave their current position, 29% considering retirement, or 57% wishing to reduce work hours.41 These are
only a few of several notable studies uncovering the global
scale of the prevalence of burnout in medical oncology in
uniquely different health care systems. Large-scale studies
of the incidence and development of burnout remain underway, calling for not only the identification of risk factors
but also for the implementation of individual and institutional interventions to address this increasingly burdensome phenomenon.
273
WHAT ONCOLOGY CLINICIANS CAN DO TO

ADDRESS BURNOUT
Dr. Ms dedication to patient care was inspired by her
grandmother, who dealt with cancer while Dr. M was a
medical student. She feels that her patients must always
come first and prides herself on having a personal connection with every patient. She wants things to be perfect
for her patient because she remembers how frustrating it
was for her grandmother to wait for tests or treatments
when paperwork had not been finished. Her son is a rambunctious, happy 4-year-old, and she loves every minute she
has with him. Between her responsibilities as an oncologist
and a mother, she is on the run every minute and feels like
she never has a moment for herself.
The case of Dr. M illustrates a common scenario: she
realizes she is stressed but does not take the time to stop and
figure out what is happening, much less create an action
plan. Her belief, although she has not quite articulated it to
herself, is that although she realizes that she is at risk for
burnout, she does not know how to prevent it or address
symptoms of burnout. Her case represents a common situation for physicians, advanced practice providers, and nurses
in cancer care. Preventing burnout proactively and dealing
with established burnout must be addressed at the individual
and system level.3,42 Although system-level issues will be
covered in the following section, in this section, we will focus
on individual issues. At the individual level, clinicians have an
important role in recognizing symptoms in themselves and
their colleagues, learning skills that prevent burnout, and
rebuilding their local culture of clinicians (Sidebar 3).
Because many clinicians did not have burnout prevention
included in their training, it is worth naming common tacit
assumptions and strategies that do not work. Many clinicians assume that burnout occurs when a clinician has given
everything she has to her work and that when a fixed
quantity of energy is used up, that burnout becomes inevitable. In addition, many clinicians experienced during
training attitudes from role models that the way to deal
with stress is to tough it out and that clinicians need to deal
with these issues on their own time. The strategies that
result from these erroneous but common assumptions are
that burnout should be approached by working longer and
Sidebar 3. What Oncology Clinicians Can Do:

Individual Interventions
274
Recognize symptoms: irritability, impatience, exasperation, feeling burdened by work

Seek out professional advice
Develop an action plan
Acquire resilience strategies for wellness: fitness/sleep,
cognitive behavioral interventions, mindfulness, finding
meaning and purpose, connect with fellow clinicians
Rebuild local culture of oncology clinicians through
advocacy
harder, not admitting even to yourself that you are stressed,

and addressing colleagues mostly by venting behind their
backs. None of these strategies is effective; they simply
compound the problem.
Recognizing Symptoms
Perhaps the most important issue in recognizing symptoms
such as irritability, impatience, exasperation, or feeling burdened by workis that every clinician experiences these at
some point. The key to recognizing these symptoms as warning
signs is to track how often they occur. The most widely used
burnout instrument is based on frequency,4 and the question
clinicians should start to ask themselves is: how often is this
symptom happening? When symptoms are occurring weekly or
more, it is probably time to take action. In addition, clinicians
should recognize that their accuracy in self-reporting may not be
highmany clinicians minimize their own self-report of these
symptoms (for many of us, it is a habit formed in training)
and a trusted observer or friend may be the most accurate
reporter. This trusted observer could be a spouse, friend, colleague, or therapistbut it is worth asking them from time to
time the question: What are you observing about my stress
level these days? This is not a formally validated question,
but it is designed to ask for observations rather than
judgments or diagnoses, pointed toward stress, which is a
less loaded term than burnout, and gives an immediate time
frame for an implicit comparison. Part of the challenge in
starting to recognize burnout proactively, before it is established, is to identify language that is evocative but not
stigmatizing.
When a clinician feels that burnout is established, it is
worth seeking out a professional for an assessment and
creation of an action plan. Established burnout often restricts a
clinicians perspective, and because burnout can merge into a
more serious depression,28 it is not advised that clinicians
diagnose and treat themselves. Taking the time to begin this
process is an important step for clinicians who think they might
have a serious level of burnout.
Learning Resilience Skills

Research studies testing interventions for clinician burnout
have generally focused on a single modality, and there is a
developing evidence base identifying useful interventions. A
2015 Cochrane systematic review concluded that there is
some evidence that cognitive-behavioral interventions and
relaxation interventions have benefit, and whereas this
review classified interventions into just three categories
(cognitive-behavioral, relaxation, and organizational), it is
useful to consider a larger repertoire of resilience skills even
though they have not all been formally tested in cancer care
clinicians.
Fitness and sleep. A variety of studies in general populations point to the importance of physical activity and
sleep, both of which have well-established benefits in
cognitive function.43,44 Most Americans do not get enough
high-quality sleep, and, especially for clinicians who are on

call, this may be worth addressing. Some might feel that we
are stretching the point to discuss fitness and sleep skills, but
it is important to build habits that ensure good fitness levels
and adequate sleep benefit from focused attention, environmental cues, and setting goals.
Cognitive behavioral. A number of techniques derived from
cognitive-behavioral psychotherapy can be useful in proactively preparing for stress4548 and dealing with it when it
occurs. The specific techniques useful for cancer care clinicians involve tracking activation during the day (e.g.,
noticing when I am worked up and taking steps to calm
down), reframing cognitive distortions (e.g., noticing when I
am catastrophizing about a patient and taking steps to reel
back my negative predictions), and finding healthy boundaries (e.g., noticing when I am about to take on too much
work and taking steps to make it manageable). These techniques involve noticing patterns of thinking and addressing
the contents of ones thoughts. They appear to be simple,
but it takes practice to notice ones own patterns, catch
the pattern as it is appearing, and to remember to use the
specific steps.
Mindfulness. In contrast to cognitive-behavioral techniques that focus on the content of thoughts, mindfulness
involves placing attention on the process of thoughts. For
clinicians, the relevance of mindfulness is that it offers
training to change how we deal with multiple demands, the
constant pressure to multitask, and habits that keep us
focused on the cognitive skills we use clinically. Two studies
indicate that mindfulness training for practicing clinicians
can have an effect on burnout, empathy, and well-being for
clinicians (although not specifically in cancer care).49,50
Although a variety of approaches to mindfulness exist, the
most well-studied approaches involve secular forms of
meditation, such as sitting meditation, body scan, and yoga.
It is worth noting that mindfulness brings full attention to
the present moment, rather than distracting oneself, and
that not all forms of mindfulness have been equally studied.
Meaning and purpose. Reconnecting with ones purpose
as a physician, nurse, or other professional can provide
insight, inspiration, and motivation to do the work needed to
deal with stress. One widespread program that has touched
many clinicians was developed by Rachael Remen, author of
the best-seller Kitchen Table Wisdom, and incorporated in
many medical schools as The Healers Art.51 Another wellknown program was developed by Linda Clever, MD, called
Renew,52 which has been featured at previous ASCO
meetings. These interventions focus on reflective practices
in which clinicians set aside time to recall and savor meaningful experiences and to draw from them guidance for the
future. For some clinicians, this kind of reflection occurs in
their personal spiritual practices, and surveys have confirmed
the importance of personal spiritual practices in physician
well-being.28
Community and relationships. The ability for a clinician to

connect with other clinicians at a personal level, in a way
different from their typical clinical dutieseither with
family, friends, or colleagueshas also been identified in a
studies aimed at identifying components of well-being.53
Although taking time for relationships is often mentioned at
work, the involvement, commitment, and skill necessary to
sustain a close relationship is not something clinicians should
take for grantedthe high divorce rates of physicians suggest
inattention to this aspect of well-being. Notably, many interventions designed to prevent burnout have included a
component of community building, and one of the mindfulness interventions cited above is designed explicitly to
allow for informal clinician relationship building meant to
address the isolation that many clinicians feel at work.49
Rebuilding Local Culture

Especially for physicians, whose practices have changed
from self-run small businesses to large managed systems in
which they are employees, the issue of work culture represents an important transition.54 Physicians who are adjusting
to having their work defined and structured as employees,
dealing with payers in new ways, and mandates for quality and
documentation that are more complex than ever are dealing with change management, leadership challenges, and reimbursement changes that have changed significantly how
physicians regard each other as colleagues. In addition, the shift
of cancer care to a team-based model means that oncology
physicians have new kinds of collegial relationships with nurses
and other specialists. As cancer care is being restructured, the
importance of rebuilding a community of practice that connects clinicians as committed clinicians will require that individual clinicians think bigger than the organization roles they
are being assigned to the community of colleagues they want
to practice, learn, and care for patients with. A warning sign of
sorts appeared in the Journal of Oncology Practice recently in
the form of a study of internal medicine resident experiences
with a hematology/oncology inpatient unit at a respected New
York academic cancer centerthe study reported that residents finished their month-long rotation with less empathy,
more burnout, and less interest in oncology.55,56 As individual
clinicians, we may not possess the ability or power to change
a situation like that on our own, but we do each have a responsibility to advocate and participate in building the kind of
culture that ensures that we do our best work and offer the
best care for our patients.
THE ROLE OF THE HEALTH CARE ORGANIZATION

Dr. M works in a large integrated health system that has very
clear productivity expectations and established financial
incentives, and she typically sees 100120 outpatients per
week. Her colleagues are always willing to help provide their
advice when she sees a challenging case, but they are also
very busy, and she feels a bit disconnected from them. Her
practice operations are run by a manager that she has very
little contact with, and, recently, the nurses in the clinic were
275
all reassigned to make things more efficienteven though Dr. M

lost a working partner whom she felt had become a close collaborator and was someone she could talk things over with at
the end of the day. Two other female physician colleagues, also
with young childrena surgeon and an internistboth left
recently for other jobs because they could not manage their
hours in ways that allowed them enough time with their families.
The well-being of physicians is critical to the success of the
organizations and practices in which physicians work. Physician burnout has a powerful impact on quality of care,
safety, and patient satisfaction.5760 Burnout and professional
satisfaction are also drivers of physician turnover and have
implications for recruitment and retention.22,61,62 Thus, burnout and well-being have important repercussions for the
quality, sustainability, and financial health of hospitals and
practice groups.60,6365 It should also be noted that many of the
drivers of physician burnout are directly influenced by the
organization. For these reasons, addressing physician burnout
and promoting physician well-being are the shared responsibility of both the physicians and the organizations in
which they work (Sidebar 4).
As a first step to addressing this challenge, organizations
must recognize the key drivers of burnout and satisfaction in
physicians. Although numerous granular factors can influence physician well-being, these can typically be categorized
into one of six dimensions: workload, work efficiency, control/
flexibility, values alignment, meaning in work, and work-life
integration. In a simple sense, the organization is responsible
for having realistic workload/productivity expectations,
providing an efficient practice environment, giving oncologists input into the decisions that affect their practice, and
providing oncologists at least some flexibility/control over
their work.12 Providing opportunities for individual oncologists to focus at least some of their time on the aspects of work
that are most personally meaningful (e.g., a specific type of
cancer, end-of-life care, teaching, clinical trials, quality improvement, or administration), rather than treating them
as homogenous clinicians, is also critical to preserve engagement over the long term.66 Although work-life integration
is often considered the responsibility of the individual, practice
structure and organization (calls scheduled, hospital coverage
arrangements, office work hours, coverage for time away, etc.)
can either promote or hinder the likelihood individual oncologists will succeed in this domain.
Sidebar 4. How Organizations Can Respond:

Organizational Interventions
276
Recognize the importance of oncology physician wellbeing to achieving its mission

Assessment of burnout
Proactively engage organizational leaders and physicians
in collaborative action planning
Optimize the clinical practice environment and
institutional culture; and provide wellness resources
The introductory case illustrates a number of potential

system issues and problems with the environment. An expectation to see 100120 outpatients per week (2024/day)
is 30%60% more than the average seen by private practice
medical oncologists in the United States (15 outpatients per
day or 75 per week)39 and is likely unsustainable. The fact
that this physician provides a personal cell phone number
for patients to use for after-hours issues also implies an
inefficient cross-coverage system or an unhealthy culture
for the oncologists in the practice not to use the system that
is in place. This is unacceptable for a physician working
in a large, urban health system. The description that
Dr. M ends her day feeling overwhelmed by the pile of charts
she needs to take home implies excessive clerical work that
could be completed by other support staff, particularly in a
practice seeing 100120 patients/week (where tremendous
ancillary support staff should be the expectation). Her symptoms of feeling disconnected from colleagues and anxiety
about how she will integrate her professional duties with
being a wife and mother are also consistent with work-life
integration issues that have a high risk of resulting in this
physician moving to a new practice (as illustrated by her two
colleagues who recently left the practice).22,67
Evaluating and Improving the Practice Environment

Once the organization recognizes the importance of physician well-being to achieving its mission, it can begin to
proactively address this challenge. Before initiating changes,
organizations should consider anonymously assessing the
prevalence of burnout and professional satisfaction among
their physicians, and then track this variable over time in the
same way they evaluate other measures of organizational
health. For example, no leadership team or board of directors would neglect to monitor measures of financial
performance and productivity. If the well-being and morale
of physicians is deemed critical to the organizations success,
it must be monitored (anonymously) and considered in the
decision-making process as the organization sets priorities,
allocates resources, and develops its strategy. It is essential
that such assessments be anonymous, and they typically should
be performed by an outside consultant to allow physicians to
be honest and preserve the integrity of the process. A baseline
assessment can also serve to identify which driver dimensions
are most in need of attention and allow the organization to
assess whether the changes implemented have led to improvement. Short validated tools to assess physician wellbeing across a variety of dimensions are publically available
and could be used for this purpose.37,6870 These tools also
have national benchmarks for physicians by specialty and can
also allow a practice to understand how their physicians
compare with like specialists nationally.
Once baseline assessment has occurred, a process of collaborative action planning between physicians and organizational leaders is a constructive way to identify solutions and
engage physicians in the process.71,72 This approach gives
physicians the opportunity to provide input and share their
ideas regarding decisions affecting their work and helps

physicians and practice leaders work in partnership toward a
shared goal.72 Because few practices have the bandwidth to
make substantive changes in all six driver dimensions at the
same time, initially focusing on one dimension and applying
this process can lead to tangible and incremental improvements that help physicians recognize the organization is
committed to the issue and that things are getting better.
This can also build momentum to then tackle challenges in the
next dimension. In our experience, these efforts often initially
focus on ways to reduce clerical tasks and enhance efficiency in
the practice workflow or to identify changes to the scheduling/
coverage system to address perceptions of inequity and improve work-life integration. Notably, such solutions can result
in a large improvement in burnout and satisfaction at the
practice level and are often cost neutral.71 A number of innovative ways to reduce clerical burden, minimize the negative
aspects of the electronic environment, and improve efficiency
have been reported.7379 Effective leadership that engages
physicians and empowers them to help solve the problems
facing the organization is essential to this process.72
Organizational Resources for Individual Physicians

The primary focus of organizations should be to optimize the
practice environment and institutional culture as described
above. Given the prevalence of burnout and distress, however, it is also important for organizations to provide resources to individual physicians to prevent burnout, promote
well-being, and provide assistance to individual physicians
struggling with burnout.
With respect to prevention, evidence suggests that organizational efforts to foster community at work and help
physicians engage with colleagues is an effective way to
improve physician well-being.80 Two randomized trials have
evaluated the impact of the organization providing time
and/or a meal so that physicians can meet with colleagues
in a structured format to explore both the meaningful and
challenging aspects of their work. Both studies found that
this approach reduces burnout and improves physicians
perception of meaning in work.50,81 Providing residents,
fellows, and practicing physicians training in self-awareness
activities such as mindfulness and narrative medicine may
also be a useful approach to helping enhance resilience.49,8284
Although these strategies are an individual approach to wellbeing, organizations can provide physicians opportunities
to learn and develop these skills to enhance individual
resilience.49,8284 Among practicing physicians, however, if
such offerings are not coupled with meaningful efforts to

address unrealistic workload expectations and problems in
the practice environment, they are likely to be met with
cynicism (they are trying to imply this is my problem or
they only want to make me more resilient so they can add to
my workload). Accordingly, it is critical that such offerings
are part of a broader strategy that demonstrates the organization has skin in the game. Flexibility in scheduling
has also become an important dimension of job satisfaction
for physicians in recent years. This is particularly true as the
number of female physicians and physicians in a two-career
relationship has increased. Providing physicians the ability to
determine when they begin (early start or late start) and end
the work day can allow them to more easily meet their
personal life responsibilities and goals. Allowing oncologists
the option of working less than full time with a commensurate
reduction in pay can also help a practice retain talented
physicians who may otherwise leave the practice.8587 Large
medical centers may also be able to provide other resources
to assist with work-life integration (e.g., childcare resources),
self-care (exercise facilities or coaching), and resources for
professional development (leadership training).
Finally, individual resources must be provided for physicians experiencing severe burnout and/or distress in other
dimensions. The icon that physicians are super human and
concerns about stigma can be barriers to physicians seeking
help.28,88 It is critical that such resources be confidential and,
preferably, off-campus/outside the group. Ideally, these
resources should be tailored to physicians rather than generic employee assistance programs that do not appreciate
the unique challenges experienced by physicians or the
repercussions of their burnout for their medical practice.
External resources designed specifically to assist physicians
experiencing burnout are also provided by some large medical
centers such as the Mayo Clinic and available by self-referral
(www.mayoclinic.org/physicianhealthcenter).
CONCLUSION
Given their dedication to the care of the oncology patient, the
present day oncology clinician is at greatest risk for burnout
once work stressors coupled with life pressures exceed the
ability to cope. Both the prevention of burnout as well as
established burnout must be targeted using effective wellness
strategies and interventions at the individual level and organizational level. Such clinician-tailored and organizationalbased interventions need to be incorporated systematically
into routine oncology care.
References
1. Freudenberger HJ. Staff burn-out. J Soc Issues. 1974;30:159-165.
2. Bahrer-Kohler S (ed). Burnout for Experts: Prevention in the Context of
Living and Working. Berlin: Springer; 2014.
3. Schaufeli WB, Leiter MP, Maslach C. Burnout: 35 years of research and
practice. Career Dev Int. 2009;14:204-220.
4. Maslach C, Schaufeli WB, Leiter MP. Job burnout. Annu Rev Psychol.
2001;52:397-422.
5. Cohn KH, Panasuk DB, Holland JC. Workplace burnout. In Cohn KH (ed).
Better Communication for Better Care: Mastering Physician-Administration
Collaboration. Chicago, IL: Health Administration Press, 2005;56-62.
277
6. Kash KM, Holland JC, Breitbart W, et al. Stress and burnout in oncology.
Oncology (Williston Park). 2000;14:1621-1633, discussion 1633-1634,
1636-1637.
7. Holland JC, Brietbart WS, Jacobsen PB, et al. Psycho-Oncology. Oxford:
Oxford University Press; 2015.
8. Jacobsen PB, Holland JC, Steensma DP. Caring for the whole patient: the
science of psychosocial care. J Clin Oncol. 2012;30:1151-1153.
9. Trufelli DC, Bensi CG, Garcia JB, et al. Burnout in cancer professionals:
a systematic review and meta-analysis. Eur J Cancer Care (Engl). 2008;
17:524-531.
10. Lyckholm L. Dealing with stress, burnout, and grief in the practice of
oncology. Lancet Oncol. 2001;2:750-755.
11. Kearney MK, Weininger RB, Vachon ML, et al. Self-care of physicians
caring for patients at the end of life: Being connected... a key to my
survival. JAMA. 2009;301:1155-1164, E1.
12. Shanafelt TD, Sloan JA, Habermann TM. The well-being of physicians.
Am J Med. 2003;114:513-519.
13. Kakiashvili T, Leszek J, Rutkowski K. The medical perspective on burnout.
Int J Occup Med Environ Health. 2013;26:401-412.
14. Bourg Carter S. High Octane Women: How Superachievers Can Avoid
Burnout. Amherst, NY: Prometheus; 2011.
15. Center C, Davis M, Detre T, et al. Confronting depression and suicide in
physicians: a consensus statement. JAMA. 2003;289:3161-3166.
16. Ramirez AJ, Graham J, Richards MA, et al. Mental health of hospital
consultants: the effects of stress and satisfaction at work. Lancet. 1996;
347:724-728.
17. Ramirez AJ, Graham J, Richards MA, et al. Burnout and psychiatric
disorder among cancer clinicians. Br J Cancer. 1995;71:1263-1269.
18. Figley CR. Compassion Fatigue: Coping With Secondary Traumatic Stress
Disorder in Those Who Treat the Traumatized. New York: Brunner/
Mazel; 2005.
19. Rushton CH, Kaszniak AW, Halifax JS. A framework for understanding
moral distress among palliative care clinicians. J Palliat Med. 2013;16:
1074-1079.
20. Stebnicki MA. Empathy Fatigue: healing the mind, body, and spirit of
professional counselors. Am J Psych Rehab. 2007;10:317-338.
21. Shanafelt T, Dyrbye L. Oncologist burnout: causes, consequences, and
responses. J Clin Oncol. 2012;30:1235-1241.
22. Shanafelt TD, Raymond M, Kosty M, et al. Satisfaction with work-life
balance and the career and retirement plans of US oncologists. J Clin
Oncol. 2014;32:1127-1135.
23. McManus IC, Keeling A, Paice E. Stress, burnout and doctors attitudes
to work are determined by personality and learning style: a twelve year
longitudinal study of UK medical graduates. BMC Med. 2004;2:29.
24. Alarcon G, Escheleman KJ, Bowling NA. Relationships between personality variables and burnout: a meta-analysis. Work Stress. 2009;23:
244-263.
25. Shanafelt TD, Bradley KA, Wipf JE, et al. Burnout and self-reported
patient care in an internal medicine residency program. Ann Intern Med.
2002;136:358-367.
26. Oreskovich MR, Shanafelt T, Dyrbye LN, et al. The prevalence of substance use disorders in American physicians. Am J Addict. 2015;24:
30-38.
27. Tyssen R, Hem E, Vaglum P, et al. The process of suicidal planning among
medical doctors: predictors in a longitudinal Norwegian sample. J Affect
Disord. 2004;80:191-198.
28. Shanafelt TD, Balch CM, Dyrbye L, et al. Special report: suicidal ideation
among American surgeons. Arch Surg. 2011;146:54-62.
29. World Health Organization. The ICD-10 Classification of Mental and
Behavioural Disorders: Clinical Descriptions and Diagnostic Guidelines.
Geneva: World Health Organization; 1992.
278
30. Bianchi R, Schonfeld IS, Laurent E. Is burnout a depressive disorder? A

re-examination with special focus on atypical depression. Int J Stress
Manag. 2014;21:307-324.
31. Bianchi R, Boffy C, Hingray C, et al. Comparative symptomatology of
burnout and depression. J Health Psychol. 2013;18:782-787.
32. Kraft U. Burned Out: your job is extremely fulfilling. It is also extremely
demanding-and you feel overwhelmed. You are not alone. Sci Am Mind.
2006;28-33.
33. Toker S, Melamed S, Berliner S, et al. Burnout and risk of coronary heart
disease: a prospective study of 8838 employees. Psychosom Med. 2012;
74:840-847.
34. Honkonen T, Ahola K, Pertovaara M, et al. The association between
burnout and physical illness in the general populationresults from the
Finnish Health 2000 Study. J Psychosom Res. 2006;61:59-66.
35. Shirom A, Melamed S. Does burnout affect physical health? A review of
the evidence. In Alexander-Stamatios AG and Cooper CL (eds). Research
Companion to Organizational Health Psychology. Cheltenham: Edward
Elgar Publishing, 2005;599-622.
36. Maslach C, Jackson SE, Leiter MP. Maslach Burnout Inventory Manual,
3rd Ed. Palo Alto: Consulting Psycologist Press; 1996.
37. Dyrbye LN, Satele D, Sloan J, et al. Utility of a brief screening tool to
identify physicians in distress. J Gen Intern Med. 2013;28:421-427.
38. Allegra CJ, Hall R, Yothers G. Prevalence of burnout in the U.S. oncology
community: results of a 2003 survey. J Oncol Pract. 2005;1:140-147.
39. Shanafelt TD, Gradishar WJ, Kosty M, et al. Burnout and career satisfaction among US oncologists. J Clin Oncol. 2014;32:678-686.
40. Blanchard P, Truchot D, Albiges-Sauvin L, et al. Prevalence and causes of
burnout amongst oncology residents: a comprehensive nationwide
cross-sectional study. Eur J Cancer. 2010;46:2708-2715.
41. Stafford L, Judd F. Mental health and occupational wellbeing of Australian gynaecologic oncologists. Gynecol Oncol. 2010;116:526-532.
42. Back AL, Steinhauser K, Kamal A, et al. Building resilience for palliative
care clinicians: an approach to burnout prevention based on individual
skills and workplace factors. J Pain Symp Man. Epub 2016 Feb 24.
43. Killgore WDS. Effects of sleep deprivation on cognition. Prog Brain Res.
2010;185:105-129.
44. Barnes DE, Yaffe K, Satariano WA, et al. A longitudinal study of cardiorespiratory fitness and cognitive function in healthy older adults.
J Am Geriatr Soc. 2003;51:459-465.
45. Rowe MM. Four-year longitudinal study of behavioral changes in coping
with stress. Am J Health Behav. 2006;30:602-612.
46. Jones MC, Johnston DW. Evaluating the impact of a worksite stress
management programme for distressed student nurses: A randomised
controlled trial. Psychol Health. 2000;15:689-706.
47. Lokk CTJ, Arnetz BB. Impact of management change and an intervention
program on health care personnel. Psychother Psychosom. 2000;69:
79-85.
48. McCue JD, Sachs CL. A stress management workshop improves residents coping skills. Arch Intern Med. 1991;151:2273-2277.
49. Krasner MS, Epstein RM, Beckman H, et al. Association of an educational program in mindful communication with burnout, empathy,
and attitudes among primary care physicians. JAMA. 2009;302:
1284-1293.
50. West CP, Dyrbye LN, Rabatin JT, et al. Intervention to promote physician
well-being, job satisfaction, and professionalism: a randomized clinical
trial. JAMA Intern Med. 2014;174:527-533.
51. Rabow MW, Newman M, Remen RN. Teaching in relationship: the
impact on faculty of teaching the Healers Art. Teach Learn Med. 2014;
26:121-128.
52. Clever LH. A call to renew. Doctors who feel ground down can renew
their spirits and their values. BMJ. 1999;319:1587-1588.
53. Wallace JE, Lemaire J. On physician well being-youll get by with a little
help from your friends. Soc Sci Med. 2007;64:2565-2577.
54. Hoff TJ, McCaffrey DP. Adapting, resisting, and negotiating: How
physicians cope with organizational and economic change. Work Occup.
1996;23:165-189.
55. McFarland DC, Holland J, Holcombe RF. Inpatient hematology-oncology
rotation is associated with a decreased interest in pursuing an oncology
career among internal medicine residents. J Oncol Pract. 2015;11:
289-295.
56. Back AL, Safyan RA, Edwards KA. What residents learn from inpatient
hematology-oncology: a call to rebuild a community of practice. J Oncol
Pract. 2015;11:296-297.
57. Shanafelt TD, Balch CM, Bechamps G, et al. Burnout and medical errors
among American surgeons. Ann Surg. 2010;251:995-1000.
58. Wallace JE, Lemaire JB, Ghali WA. Physician wellness: a missing quality
indicator. Lancet. 2009;374:1714-1721.
59. West CP, Huschka MM, Novotny PJ, et al. Association of perceived
medical errors with resident distress and empathy: a prospective
longitudinal study. JAMA. 2006;296:1071-1078.
60. Jones JW, Barge BN, Steffy BD, et al. Stress and medical malpractice:
organizational risk assessment and intervention. J Appl Psychol. 1988;
73:727-735.
61. Shanafelt T, Sloan J, Satele D, et al. Why do surgeons consider leaving
practice? J Am Coll Surg. 2011;212:421-422.
62. Atkinson W, Misra-Hebert A, Stoller JK. The impact on revenue of
physician turnover: an assessment model and experience in a large
healthcare center. J Med Pract Manage. 2006;21:351-355.
63. Dewa CS, Loong D, Bonato S, et al. How does burnout affect physician
productivity? A systematic literature review. BMC Health Serv Res.
2014;14:325.
64. Dewa CS, Jacobs P, Thanh NX, et al. An estimate of the cost of burnout
on early retirement and reduction in clinical hours of practicing physicians in Canada. BMC Health Serv Res. 2014;14:254.
65. Buchbinder SB, Wilson M, Melick CF, et al. Estimates of costs of primary
care physician turnover. Am J Manag Care. 1999;5:1431-1438.
66. Shanafelt TD, West CP, Sloan JA, et al. Career fit and burnout among
academic faculty. Arch Intern Med. 2009;169:990-995.
67. Shanafelt TD, Boone SL, Dyrbye LN, et al. The medical marriage: a national survey of the spouses/partners of US physicians. Mayo Clin Proc.
2013;88:216-225.
68. Shanafelt TD, Kaups KL, Nelson H, et al. An interactive individualized
intervention to promote behavioral change to increase personal wellbeing in US surgeons. Ann Surg. 2014;259:82-88.
69. West CP, Dyrbye LN, Sloan JA, et al. Single item measures of emotional
exhaustion and depersonalization are useful for assessing burnout
in medical professionals. J Gen Intern Med. 2009;24:1318-1321.
70. Shanafelt TD, Hasan O, Dyrbye LN, et al. Changes in burnout and
satisfaction with work-life balance in physicians and the general US
working population between 2011 and 2014. Mayo Clin Proc. 2015;90:
1600-1613.
71. Swenson SJ, Kabcenell A, Shanafelt TD. Physician-organization collaboration to reduce physician burnout and promote engagement: the
Mayo Clinic experience. J Healthc Manag. In press.
72. Shanafelt TD, Gorringe G, Menaker R, et al. Impact of organizational

leadership on physician burnout and satisfaction. Mayo Clin Proc. 2015;
90:432-440.
73. The Joint Commission. Use of Unlicensed Persons Acting as Scribes.
http://www.jointcommission.org/standards_information/jcfaqdetails.
aspx?StandardsFaqId=426&ProgramId=47. Accessed February 27,
2016.
74. Reuben DB, Knudsen J, Senelick W, et al. The effect of a physician
partner program on physician efficiency and patient satisfaction. JAMA
Intern Med. 2014;174:1190-1193.
75. Shultz CG, Holmstrom HL. The use of medical scribes in health care
settings: a systematic review and future directions. J Am Board Fam
Med. 2015;28:371-381.
76. Hopkins K, Sinsky CA. Team-based care: saving time and improving
efficiency. Fam Pract Manag. 2014;21:23-29.
77. Anderson RJ. Optimizing the role of nursing staff to enhance physician
productivity: one physicians journey. Fam Pract Manag. 2013;20:
18-22.
78. Anderson P, Halley MD. A new approach to making your doctor-nurse
team more productive. Fam Pract Manag. 2008;15:35-40.
79. Sinsky CA, Willard-Grace R, Schutzbank AM, et al. In search of joy in
practice: a report of 23 high-functioning primary care practices. Ann
Fam Med. 2013;11:272-278.
80. Bruce SM, Conaglen HM, Conaglen JV. Burnout in physicians: a case for
peer-support. Intern Med J. 2005;35:272-278.
81. West CP, Dyrbye LN, Satele D, et al. A randomized controlled trial
evaluating the effect of COMPASS (COlleagues Meeting to Promote and Sustain Satisfaction) small group sessions on physician
well-being, meaning, and job satisfaction. J Gen Intern Med. 2015;
30:S89.
82. Beckman HB, Wendland M, Mooney C, et al. The impact of a program in
mindful communication on primary care physicians. Acad Med. 2012;
87:815-819.
83. Charon R. The patient-physician relationship. Narrative medicine:
a model for empathy, reflection, profession, and trust. JAMA. 2001;286:
1897-1902.
84. Shanafelt TD. Enhancing meaning in work: a prescription for preventing
physician burnout and promoting patient-centered care. JAMA. 2009;
302:1338-1340.
85. Mechaber HF, Levine RB, Manwell LB, et al; MEMO Investigators. Parttime physicians...prevalent, connected, and satisfied. J Gen Intern Med.
2008;23:300-303.
86. Harrison RA, Gregg JL. A time for change: an exploration of attitudes
toward part-time work in academia among women internists and their
division chiefs. Acad Med. 2009;84:80-86.
87. Levine RB, Harrison RA, Mechaber HF, et al. Professional characteristics and job satisfaction among SGIM members: a comparison of
part-time and full-time physician members. J Gen Intern Med. 2008;
23:1218-1221.
88. Balch CM, Shanafelt TS. Dynamic tension between success in a surgical
career and personal wellness: how can we succeed in a stressful environment and a culture of bravado? Ann Surg Oncol. 2011;18:
1213-1216.
279
SARCOMA
Is There a Future for

Immunotherapy in Sarcoma?
CHAIR
Seth M. Pollack, MD
Seattle, WA
SPEAKERS
Ronald P. DeMatteo, MD
New York, NY
Sandra P. DAngelo, MD
New York, NY
SARCOMA IMMUNOTHERAPY
Immunotherapy for Soft Tissue Sarcoma: Tomorrow Is Only a

Day Away
Alex Lee, MBChB, Paul Huang, PhD, Ronald P. DeMatteo, MD, and Seth M. Pollack, MD
OVERVIEW
Despite the advances taking place for patients with many types of cancer, to date there has been little success in meeting the
great need for novel treatments of advanced soft tissue sarcoma with effective immunologic therapies. Here, we review
recent clinical and preclinical data that indicate immune responses against sarcomas occur spontaneously and can also be
successfully provoked. Efforts to manipulate the sarcoma immune microenvironment have the potential to eradicate
disease and may also sensitize tumors to other tumor-targeted immunotherapeutic approaches. Other approaches, including vaccines and genetic engineering of T cells, offer a promising opportunity to actively direct cytotoxic lymphocytes
toward antigen-bearing sarcomas. Drawing parallels with recent advances made in other cancer types, we identify ways in
which sarcomas can be included in the ongoing immunotherapy revolution.
oft tissue sarcoma represents a rare and heterogeneous

group of malignancies, encompassing over 50 distinct
diagnoses that together account for around 1% of adult
cancers. Improved surgery delivered in specialist centers,
along with the additional benefit in some cases from preoperative or postoperative radiotherapy, has led to improved
outcomes from localized disease over the past 30 years.
However, metastatic disease is common, occurring among
25% to 50% of patients with sarcoma, depending on subtype
and initial stage. Advanced disease is fatal in the large majority
of patients, with median overall survival (OS) remaining
around 12 to 15 months from metastatic diagnosis.1
Cytotoxic chemotherapy remains the mainstay treatment
of advanced soft tissue sarcoma.2 However, these approaches
are highly toxic and generally only offer a limited period of
disease control. One of the main difficulties in the development of new therapies has been the differential efficacy of
particular drugs in specific subtypes. Randomized phase III
trials have too often been disappointing, in part because of
the logistical necessity of recruiting heterogeneous patient
populations. Other recently approved drugs have lacked
validated predictive biomarkers, with randomized trials underpowered to translate gains in progression-free survival to
clinically meaningful OS benefit.3-6
Although progress in most sarcoma subtypes has been slow,
one of the most successful examples of targeted molecular
therapy has been the application of the tyrosine kinase inhibitor imatinib mesylate to gastrointestinal stromal tumor
(GIST). These tumors contain an activating Kit mutation in

over 75% of cases.7-9 Imatinib mesylate inhibits both Kit and
PDGFRa, resulting in partial response or stable disease for over
80% of patients treated for advanced disease.10,11 This dramatic finding has led to an increase in median OS from 1 to
5 years since the introduction of the drug.12 However, the
disease does ultimately progress, in some cases because of a
secondary Kit mutation.13-15 Regardless, this demonstration
of such profound efficacy gives hope to patients with other
sarcoma subtypes that innovative new treatment strategies
may offer potential.
There remains an apparent and urgent need for effective
novel therapies for advanced soft tissue sarcoma. In this
article, we discuss ongoing clinical and preclinical work toward
treating patients with sarcoma using immunotherapy.
SARCOMA PLAYED AN IMPORTANT ROLE IN

CANCER IMMUNOTHERAPYS HISTORY
Since Coleys initial attempts over 100 years ago, the exploitation of the immune systems apparent ability to recognize and attack malignancy has represented an attractive
therapeutic mechanism.16 However, until recently, the biology underlying the antitumor immune responses was
mysterious and undefined. In the 1980s, Rosenberg performed pioneering trials with high doses of the key inflammatory cytokine interleukin 2 (IL-2), observing rare but
compelling instances of durable disease response in patients
with melanoma and renal cell carcinoma.17 The culture of
From the Department of Medical Oncology, Royal Marsden Hospital London, United Kingdom; Institute of Cancer Research, London, United Kingdom; Department of Surgery, Memorial
Sloan Kettering Cancer Center, New York, NY; Fred Hutchinson Cancer Research Center, Division of Oncology, University of Washington, Seattle, WA.
Corresponding author: Seth M. Pollack, MD, Fred Hutchinson Cancer Research Center, 1100 Fairview Ave., Suite D3-100, Seattle, WA 98109; email: spollack@fhcrc.org.
281
LEE ET AL
ex vivo expanded tumor-infiltrating lymphocytes (TILs) in

combination with lymphodepleting chemotherapy and
recombinant IL-2 was found to have striking evidence of
efficacy in melanoma, with objective response rates reported
at 20% to 70%, with durable responses and long-term survival,
suggesting that the lymphocytes in the tumor microenvironment target tumor-specific antigens.18-20 The potent antitumor effect of donor lymphocytes was highlighted by the
graft versus malignancy effect seen following allogeneic
bone marrow and hematopoietic stem cell transplantation for
the treatment of hematologic malignancy, ultimately demonstrating the presence of tumor-specific antigens recognized
by donor lymphocytes that had not previously provoked host
immunity.21,22 However, the expensive, challenging methodology and heavy burden of associated toxicity limited the
expansion of such immunotherapies from niche treatments
with occasional profound effect in melanoma and renal cell
carcinoma and prevented their arrival into the mainstream.
Recently, a new generation of immunotherapeutic agents
has revolutionized oncology. Inhibitory monoclonal antibodies targeting cell surface proteins that downregulate T-cell
activationthe so-called immune checkpointshave demonstrated unprecedented efficacy across a range of tumor
types with generally acceptable and manageable toxicity, and
they are now available for wide use in the treatment of
advanced melanoma and nonsmall cell lung cancer.23-27
T cells with genetically engineered antigen-specific T-cell
receptors (TCRs) have shown remarkable clinical activity as
have chimeric antigen receptors (CARs) in the targeting of
CD19-expressing hematologic malignancies, and they are also
under investigation in a range of solid tumors.28
Given William Coleys early foray into sarcoma immunotherapy in the early 1900s, it has been disappointing that
research into the immune biology of sarcoma has been slower
than in other disease groups in recent years.29 However, a flurry
of new research is attempting to correct this and bring immunotherapeutic approaches to sarcoma.
KEY POINTS
282
An increasing number of studies are characterizing the

sarcoma immune microenvironment and identifying
potential prognostic correlates.
Many soft tissue sarcomas express well-characterized
TAAs of variable type and are therefore excellent
potential targets for vaccines and antigen-specific T-cell
therapybased approaches.
GISTs are infiltrated with T cells and macrophages, and
checkpoint blockade has shown efficacy in an animal
model of GIST. Clinical trials of immunotherapy in
human GIST are now underway.
Selection and redirection of cytotoxic T-cell antigen
specificity toward TAAs through engineered T-cell
receptors has promising activity in a range of cancers,
including soft tissue sarcoma subtypes.
THE SARCOMA IMMUNE MICROENVIRONMENT

CONTAINS COMPLEX INFILTRATES
Given the capacity of TILs to lyse autologous tumor, multiple
investigators have examined and confirmed correlations
between the presence of lymphocytic infiltrates in the tumor
microenvironment and patient clinical outcome across a wide
range of tumor types.30,31 Although such studies characterizing the tumor immune microenvironment in sarcomas have
lagged behind other diseases, they are growing in number.
Positive association between the presence of TILs and improved clinical outcome have been reported in series of
patients with Ewing sarcoma, GIST, and cutaneous angiosarcoma, as well as cohorts of mixed subtypes including
leiomyosarcoma, synovial sarcoma, liposarcoma, and undifferentiated sarcomas, among others.32-38 As a consequence of the small numbers of patients, heterogeneous
populations, and varying methodologies between many of
these studies, a consistent picture has yet to emerge as to
which immune cells in which tumor compartment provide the
most useful prognostic information in sarcoma. However,
larger-scale immune-profiling projects are currently underway and promise to categorize sarcomas by the quality
and quantity of immune infiltrate and to highlight mechanisms used by tumor to evade eradication by a host immune
response. Such discoveries will guide the development and
application of existing and novel immunotherapeutics in this
disease type. Intriguingly, given the rich TIL infiltrates seen in
some sarcoma tumors, in vitro antitumor activity of TILs
isolated from patients with sarcoma has been investigated
in a small number of studies. Studies for patients with melanoma and renal cell carcinoma reported successful ex vivo
culture of TILs with IL-2 in a minority of patients, a proportion
that is even lower in other tumor types. Consequently, it is not
surprising that growing TILs extracted from sarcomas has
proven to be difficult in the subtypes studied so far.39,40
Compared with infiltrates from other bone sarcomas, only
TILs taken from osteosarcoma specimens demonstrated cytotoxicity against allogeneic tumor cells, whereas TILs isolated
from a rat osteosarcoma model demonstrated a similar
phenotypic and lytic profile to those derived from clinical
tumors. The fact that lymphocytes from osteosarcoma tumors may have similar ex vivo antitumor lytic capacity
compared with those from melanoma and renal cell carcinoma have led some to suggest that they may be susceptible
to the same sort of nonspecific immune activation approaches, such as high-dose IL-2 and TIL therapy. This is
supported by a German series of high-dose IL-2 in advanced
pediatric cancers in which two of four patients with osteosarcoma experienced durable complete remission.41
TUMOR-ASSOCIATED MACROPHAGES IN THE

SARCOMA TUMOR MICROENVIRONMENT
Macrophages are often classified as either classically (M1) or
alternatively (M2) activated, although they can shift phenotype and often exist on a spectrum between the two extremes.42 Tumor-associated macrophages (TAMs) may be
polarized toward M1 and be proinflammatory. These can

directly lyse tumor cells, secrete TNF and IL-12, present
antigen, and activate T cells. In contrast, M2 TAM are antiinflammatory and support tumor growth by producing IL-10
and promoting angiogenesis. Much of our understanding of
TAM has come from studies of mouse flank tumors, which
nearly always contain M2 TAM. Human TAMs frequently
have an M2 phenotype, based on limited immunohistochemistry (IHC) studies (often using CD163), but functional
analyses are lacking.43
In one of the first and most detailed analysis of isolated
TAMs, flow cytometry, transcriptome analysis, and in vitro
functional studies were performed on TAMs isolated from
mouse GIST and 57 fresh human GIST surgical specimens.44
Surprisingly, TAMs from mouse GIST were M1 at baseline.
Accordingly, TAM depletion with the colony-stimulating
factor-1 receptor (CSF1R) inhibitor PLX5622 increased tumor size. After imatinib therapy, TAMs became M2 through
phagocytosis of apoptotic tumor cells and upregulation of
CCAAT/enhancer binding protein b (C/EBPb), a known driver
of M2 macrophages, and its regulator cyclic AMP response
element-binding protein (CREB).45 Untreated human GISTs
had M1 TAMs that became M2 after imatinib therapy, but
reverted to M1 in tumors that became resistant to tyrosine kinase inhibitors. Transcriptome analysis of isolated
TAMs from 14 patients revealed no differentially expressed
(greater than twofold) genes between tumors from untreated and resistant patients, indicating that tumor cell
activity governs TAM polarity. These studies highlight the
critical importance of TAMs in the sarcoma microenvironment and suggest that research aimed at increasing favorable antitumor TAM activation should be given a high
priority.
Tumor-associated macrophages appear to play a significant role in the tumor immune microenvironment of sarcoma subtypes other than GIST. Through gene expression
profiling of 51 frozen tumor samples of various soft tissue
sarcoma subtypes, Lee et al46 identified high levels of expression of TAM-associated genes in all of eight cases of
leiomyosarcoma. Tumor-associated macrophage infiltration in 149 primary leiomyosarcoma tumors was then
assessed by IHC for the macrophage markers CD68 and
CD163 and correlated with patient outcomes. A high density
of CD68 and/or CD163-staining macrophages was associated
with a significantly worse disease-specific survival in nongynecological leiomyosarcoma, with a nonsignificant trend
seen in gynecologic leiomyosarcoma. Later studies from the
same group investigated the expression of CSF-1 (a major
macrophage chemoattractant) and related genes by similar
method, and they found a significant association between
expression of a TAM-associated four-gene signature and
poor prognosis in both uterine and nonuterine leiomyosarcoma in training and validation sets amounting to over
200 cases.47,48 An association between poor prognosis and
high levels of CD68-positive TAMs was reported in series of
78 patients with primary myxoid liposarcoma and 41 patients with primary Ewing sarcoma, further supporting a
potentially broad role of TAMs in sarcoma immunology.49,50

The depletion or manipulation of TAM responses as a
therapeutic target is currently under active clinical investigation in sarcoma.
The combination of PLX3397, a novel multitargeted kinase
inhibitor selective for the CSF1R kinase c-Fms as well as c-Kit,
demonstrated in vivo depletion of macrophages and significant tumor suppression when administered to mice
bearing malignant peripheral nerve sheath tumor xenografts
either as a single agent or in combination with the mTORC1
inhibitor rapamycin.51 A phase I study of the combination of
PLX3397 and sirolimus is currently recruiting patients with
malignant peripheral nerve sheath tumor and other unresectable sarcomas (NCT02584647).
Therapeutic modulation of TAM polarity toward a favorable antitumor phenotype has been demonstrated preclinically and is under current clinical investigation. A
humanized anti-CD47 monoclonal antibody (Hu5F9-G4) that
promotes macrophage phagocytosis of tumor cells in vivo
and in vitro is now in a phase I trial recruiting patients with
advanced solid tumors (NCT02216409).52 A synthetic agonist of Toll-like receptor 4 (TLR4), an activating receptor
expressed by innate immune cells including macrophages, is
currently in phase I development in metastatic sarcoma,
with the combination of radiotherapy included as a means of
increasing tumor antigen release (NCT02180698).
IMATINIB IS AN IMMUNE MODULATOR IN GIST

As shown in a genetically engineered mouse model of GIST,
imatinib works in part by altering the antitumor immune
response to increase the ratio of effector T cells to regulatory
T cells (T-regs).53 Transcriptome profiling of treated tumors
revealed the mechanism of immunomodulation depended
on imatinib inhibiting tumor cell production of the immunosuppressive enzyme indoleamine 2,3-dioxygenase (Ido).54
Ido is an intracellular enzyme that functions in the first and
rate-limiting step in the catabolism of the essential amino
acid L-tryptophan to 5-hydroxy-N-formyl-kynurenine. Local
deprivation of tryptophan and its downstream metabolites
(kynurenine, 3-OH anthranilic acid, and 3-OH kynurenine) is
thought to promote the development, stabilization, and
activation of T-regs while suppressing effector T cells.55 Ido is
made by tumor cells and dendritic cells and is tolerogenic in
the setting of pregnancy, transplantation, and cancer.56,57 Of
note, Ido2 is not expressed in GIST.
In a murine GIST model, imatinib decreased the transcription factor Etv4, which was shown to bind to the Ido
promoter. The related family member Etv1 has been
identified as a lineage survival factor in GIST. The intratumoral T-cell findings in mouse GIST were correlated to
freshly isolated intratumoral immune cells from 45 human
GIST samples.58 Human tumors that had responded to
imatinib also had a higher intratumoral effector T-cell to
T-reg ratio. Depletion of CD8 T cells blunted the effects of
imatinib. The importance of intratumoral T cells in human
GIST was later supported with IHC analysis of a different
283
LEE ET AL
cohort of localized GIST.35 Although imatinib therapy activated the immune system in mouse and human GIST, the
addition of antiCTLA-4 treatment enhanced T-cell cytokine
production and antitumor efficacy in the mouse tumor
model.54 This finding prompted an ongoing National Cancer
Institute (NCI) trial (NCT01643278) testing the combination
of ipilimumab (antiCTLA-4) and dasatanib in advanced,
refractory GIST.59
IMMUNE CHECKPOINTS IN SARCOMA

The interaction between PD-1 and PD-L1 is a normal homeostatic mechanism inhibiting T-cell activation and is now
recognized as an important mechanism of tumor immune
evasion that has been therapeutically targeted with significant success. The utility of tumor expression of PD-L1 as a
predictive biomarker for such therapies has been subject to
a great deal of recent discussion, not yet resolved into a
consensus conclusion, but nevertheless informing recent
approval decisions by licensing authorities. Although the
role of PD-L1 and PD-1 as prognostic biomarkers remains
similarly unresolved, the suggestion that there may be an
association with clinical outcome has led to a number of
groups examining the expression of these molecules in
sarcoma subtypes.60-64
The first report of PD-1 and PD-L1 expression in soft tissue
sarcoma included a cohort of 105 patients with multiple
subtypes, including leiomyosarcoma, synovial sarcoma,
undifferentiated pleomorphic sarcoma, and myxoid liposarcoma.65 PD-1expressing TILs and tumor PD-L1 expression was seen in 65% and 58% of tumors, respectively. Both
PD-1 and PD-L1 expression was associated with higher tumor
grade, more advanced stage, presence of metastasis, and
poor tumor differentiation, with both PD-1 and PD-L1
identified as independent prognostic indicators for OS
and progression-free survival, and PD-1/PD-L1positive
tumors associated with worst clinical outcome. A later study
of 50 patients with soft tissue sarcoma (14 GIST, four
leiomyosarcoma, five liposarcoma, three synovial sarcoma,
and 24 other subtypes) found lower levels of PD-1 expression on infiltrating leukocytes (22% of tumors) and PDL1 expression on tumor cells, infiltrating lymphocytes, and
macrophages (12%, 30%, and 58% of tumors, respectively).34
No association between PD-L1 expression, TILs, and survival
was seen, with authors proposing the small heterogeneous
cohort and the use of a different PD-L1 assay accounting for
the differing results to the earlier study.
There currently exists very limited available clinical data on
the efficacy of immune checkpoint inhibitors in the treatment of sarcoma. In view of the observation that treatment
with ipilimumab, an inhibitory monoclonal antibody targeting the immune checkpoint CTLA-4, enhanced the antitumor T-cell response toward the recurrent cancer/testis
antigen NY-ESO-1 in melanoma, Maki et al performed a small
pilot study in synovial sarcoma.66,67 Six patients with advanced disease of known NY-ESO-1 positivity received between one and three cycles of ipilimumab. All six patients
284
experienced disease progression during treatment, after

which the study was closed. Within the initial phase I cohort
of the antiPD-1 mAb pembrolizumab were two patients
with soft tissue sarcoma. One patient with leiomyosarcoma
achieved disease stability that endured at 30 weeks of
follow-up; one patient with malignant peripheral nerve
sheath tumor had stable disease at first assessment but then
progressed at 29 weeks.68 A much clearer picture of the
potential benefit of these agents in soft tissue sarcoma will
emerge over the coming few years, as the significant number of current trials investigating the use of PD-1targeting
drugs either as single agents or in combination with ipilimumab, kinase inhibitors, or chemotherapy report
their findings in cohorts open to a range of sarcoma
subtypes (NCT01643278; NCT02304458; NCT02428192;
NCT02301039; NCT02500797; NCT02636725; NCT02331251).
CANCER/TESTIS ANTIGENS IN SARCOMA

Cancer/testis antigens (CTAs) consist of over 40 identified
proteins that play a normal role in embryonic development.69 They are also found in some fetal organs, and some
have been reported to be expressed in placental tissue. They
are by definition expressed to a high degree in the primitive
spermatagonium of the testis, and, although they are highly
immunogenic, they provoke no natural immune response,
presumably because the testis is an immune-privileged site
protected by the bloodtestis barrier. A number of studies in
multiple malignancies have targeted CTAs without toxicity to
the testis or any other organs.70,71 At present, a small
number of studies have described expression of a range of
CTAs across soft tissue sarcoma subtypes.72,73
Synovial sarcoma is the soft tissue sarcoma subtype with
the best-characterized expression of CTAs. NY-ESO-1, a CTA
first identified through serological analysis of patients with
esophageal cancer, is expressed in the majority of synovial
sarcoma.74 Homogenous NY-ESO-1 expression is frequently
seen in tumors with monophasic histology, although biphasic variants also often express significant levels of this
CTA.75 Ubiquitous NY-ESO-1 expression was described in 25
myxoid/round-cell liposarcomas (MRCLs), with antigenspecific cytotoxic T cells (CTLs) derived from a patient
with synovial sarcoma able to lyse myxoid liposarcoma
cells in an antigen-specific manner. Although little native
NY-ESO-1 or other CTA expression is generally observed in
frozen chondrosarcoma samples, the treatment of chondrosarcoma cell lines with the demethylating agent decitabine induced NY-ESO-1 and PRAME expression, sensitizing
cells to lysis by antigen-specific CTLs.76 This result suggests
that the pool of patients eligible for tumor-associated antigen (TAA)-targeted immunotherapies could be widened
through the reversal of epigenetic silencing of tumor CTA
expression.
VACCINATION AGAINST SARCOMA

A number of small trials have used a variety of different
vaccine strategies in sarcoma with the aim of provoking
antitumor responses through exposure to a range of antigen

sources, ranging from whole tumor cells, tumor lysates, and
isolated or recombinant proteins and peptides. A variety
of vaccine adjuvants have been used, including nonspecific immunostimulants, such as granulocyte-macrophage
colony-stimulating factor (GM-CSF) or interferon-gamma
(IFN-g; Table 1). Immunologic correlates examined in
these studies indicate successful engagement of host immunity against antigen with some frequency, but it has been
difficult to conclusively demonstrate that these translate to
clinically meaningful benefit in terms of patient outcome.
This may relate to suboptimal adjuvant regimens or failure to
select antigens with corresponding cognate T-cells clones of
adequate affinity. Immune escape by downregulation of the
antigenic protein by the tumor or through co-option of
immunosuppressive T cell or myeloid populations also
poses a potential vulnerability of vaccine approaches.
A placebo-controlled multicenter phase II trial of a trivalent peptide vaccine against the ganglioside antigens GD2,
GD3, and GM2 has now completed treatment of 136 patients with stage IV sarcoma rendered free of demonstrable
disease following metastasectomy. Although serological
response toward GM2 and GD2 were seen in 98% and 21% of
patients in the vaccine and placebo groups, respectively,
there was no observed difference in progression-free survival between the trial arms. Follow-up for the secondary
endpoint of OS is ongoing.77 The future of this modality may
depend on the identification of recurrent, strongly antigenic
tumor-related proteins and the combination with potent
immunomodulators that skew the tumor immune microenvironment toward a proimmunity context.
A NY-ESO-1 lentiviral vaccine has been shown to be safe
when administered with an adjuvant TLR4 agonist, and there
is evidence that it can evoke humoral and cellular immune
responses. Investigation regarding its clinical efficacy in NYESO-1expressing sarcomas is ongoing.78 Because it is quite
possible that the PD-1 and PD-L1 axis has contributed to the
lack of success in other vaccine studies, this combination is
now also being evaluated with the PD-L1 inhibitor atezolizumab (NCT02609984).
SARCOMA-SPECIFIC T-CELL THERAPY

Targeted T-cell therapies have the potential to induce complete and durable responses in other solid tumors.18-20,71
T cells have the ability to not only kill tumor directly but
attract additional immune cells potentially eliciting an endogenous immune response.
Multiple groups have attempted to exploit the high NYESO-1 expression of synovial sarcoma for therapeutic benefit. Using NY-ESO-1 tetramer-based cell sorting, highly
purified NY-ESO-1specific T cells can be isolated from the
peripheral blood following peptide-pulsed dendritic cell
(DC)-based stimulation. 79 In this study of six patients
with NY-ESO-1expressing sarcomas, clinical grade levels of
antigen-specific CTLs were successfully generated in all
patients. These CTLs were oligoclonal, had effector memory
phenotype, and were capable of lysing NY-ESO-1expressing

tumor cells ex vivo. A phase I trial employing this method of
adoptive cell therapy (ACT) has recently completed recruitment of patients with NY-ESO-1expressing synovial
sarcoma and MRCL (NCT01477021). A further phase I trial
combining this ACT with palliative radiotherapy is currently
underway (NCT02319824).
Genetic modification of autologous T cells to express specific
TCRs can offer consistently high affinity directed toward TAAs,
as has been successfully used to direct clinically significant
antitumor immune responses across a range of tumor types
with well-characterized TAA expression. T-cell receptors with
effective antitumor activity can be designed to have enhanced
affinity through single- and dual-point mutations of the CDR3
region of the TCR as has been done for the high-affinity aLY
TCR used by the NCI and Adaptimmune, modified from the
already highly active 1G4 TCR.80
In a pilot study testing the activity of this TCR, patients with
pretreated advanced melanoma and synovial sarcoma with
proven NY-ESO-1 expression received an intravenous infusion of autologous engineered T cells preceded by lymphodepletion with cyclophosphamide and fludarabine.81 In
an initial cohort of 11 melanomas and six synovial sarcomas,
objective clinical responses were seen in five of 11 (45%) and
four of six (67%), respectively, without any incidences of
toxicity attributable to the cellular therapy. Updated results
were reported in 2015, with respective overall response
rates of 11 of 20 (55%) and 11 of 18 (61%) in patients with
melanoma and synovial sarcoma.82 In keeping with other
studies of ACT, total numbers of infused T cells and ex vivo
reactivity of T cells to antigenic peptide enzyme-linked
immunospot (ELISPOT) in terms of IFN-g secretion were
positively associated with response to therapy, although the
absence of an association between response and persistence of infused T cells in circulation was contrary to prior
studies by the same and other groups. Encouraging 3and 5-year survival rates of 38% and 14% in this chemorefractory synovial sarcoma cohort has led to further, currently ongoing trials of NY-ESO-1specific engineered
T cells in expanded NY-ESO-1positive sarcoma cohorts
(NCT01343043).
Although there are no published reports to date of severe
toxicity attributable to NY-ESO-1specific T-cell therapy, this
therapy is not entirely without risks. T-cell therapy trials
targeting other antigens have observed severe on-target
autoimmunity, resulting in hearing loss and uveitis in a trial
of MART-1targeted T cells and severe colitis following CEA
targeting.83-85 Furthermore, attempts to enhance TCR affinity through manipulation of the CDR2 TCR region that
influences HLA binding have resulted in off-target toxicities
as was demonstrated by the fatal cardiac complications in
the first two patients treated with affinity-enhanced HLAA*0201 MAGE A3specific engineered TCR cell therapy.86
Only after painstaking post-SAE peptide scanning were investigators able to identify sequence homology at amino
acid residues essential for epitope reactivity shared between
MAGE A3 and titin, a ubiquitously expressed muscle protein.
285
LEE ET AL
TABLE 1. Reported Vaccine Studies in Soft Tissue Sarcoma
Adjuvant
Disease Setting
No. of
Patients
Treated/
Enrolled Immune Correlates
Vaccine
Schedule
Autologous Tumor-Lysate Pulsed Autologous DC93
106 to 107 cells DCs cultured with GM-CSF Advanced pediatric 15/15
ID 3 3 2
and IL-4
solid tumors
weekly
Clinical Outcome
. 10-fold increase in
Partial response in
PBMC reactivity to tuone patient
mor lysate on ELISPOT
(fibrosarcoma)
in three of six evaluable
patients
DTH response in four of Stable disease in two
eight evaluable patients
patients, three
patients without
measurable disease progression
free at 1630
months
Autologous DC Pulsed 2 3 106 to 1.5 DCs cultured with GM-CSF Advanced pediatric
3 107 SC 12
and IL-4
solid tumors,
with Synthetic
weekly, six
refractory to
Translocation Fusion
to eight
autologous stem
Peptides or Tumor
94
doses
cell transplant
Lysate
5/5
Autologous T Cells and Single adminis- DCs cultured with GM-CSF, Newly diagnosed 30/52
DC Pulsed With
tration of DC
IL-4 (all patients), CD40L
metastatic or
Translocation Fusion
and T cells
(two of three patients)
recurrent aRMS
Peptides95
Influenza vaccine (all
or ESFT with t(2:
patients); IL-2 (moder13) or t(11;22).
ate, low or no dose)
Vaccine followed
completion of
induction
chemotherapy
Advanced sarcoma, various
subtypes
DTH to tumor material in Complete response

one of five
durable at 77
months in one
Increase in NK-cell activapatient (Ewing
tion (three of five)
sarcoma)
+
Increase in CD8 T cell
activation (two of five)
T-cell response to fusion 5-year OS: 31% for all
peptide by Cr51 assay in
enrolled patients,
nine of 23 evaluable
43% in patients
patients
who started
immunotherapy
Irradiated Autologous
Tumor Cells96
ID weekly for 3 GM-CSF or IFN-g

weeks, then
monthly
25/48
Converted to positive DTH Longer median OS in

to tumor cells in six of
DTH converters
18 evaluable patients
(16.6 vs. 8.2
months); no
objective responses
Synthetic NY-ESO-1
Peptide97
0.1 mg ID daily GM-CSF from third dose of Advanced NY-ESO- 19/20

for 5 days,
vaccine onward
1 or LAGEthen 3
expressing solid
weekly for 6
tumors (eight of
doses
12 sarcoma)
No cases of NY-ESO-1 se- Best response: 12 PD;

roconversion seven of
six SD; two
20 converted to DTH
not evaluable
positivity; three patients converted to
antigen-specific CTL
positivity
Synthetic Peptide Derived From SYT-SSX

Fusion Region (B
Peptide)98
0.1 or 1.0 mg Nil

SC 2 weekly
for six doses
No DTH responses
Best response at
CTLs specific for HLA;
completion of
peptide tetramer
treatment: five PD,
generated in four of six
one SD
Synthetic Peptide
Derived From SYTSSX Fusion Region
(B Peptide) + HLA1*2402-Anchor Substitute Variant (K91
Peptide)99
0.1 mg or
1.0 mg SC 2
weekly for
six doses
Advanced SS
6/6
Nil or incomplete Freunds HLA-A*2402-positive 21/21

adjuvant + IFN-a
advanced SS
No DTH responses twofold Best response at

or greater increase in
completion of
HLA; peptide tetramertreatment:
specific CTLs in seven
14 PD;
patients
seven SD (one no
adjuvant, six
received adjuvant)
Abbreviations: DC, dendritic cell; ID, intradermal; GM-CSF, granulocyte-macrophage colony-stimulating factor; IL-4, interleukin-4; PBMC, peripheral blood monocytic cell; DTH,
delayed-type hypersensitivity; SC, subcutaneously; NK, natural killer; aRMS, alveolar rhabdomyosarcoma; ESFT, Ewing sarcoma family of tumors; OS, overall survival; IFN, interferon;
PD, progressive disease; SD, stable disease; CTL, cytotoxic T cell; SS, synovial sarcoma.
Chimeric antigen receptors provide an alternative means

of redirecting immune effector cell antigen specificity.
Chimeric antigen receptors consist of an extracellular antigen recognition domain usually consisting of a single-chain
variable fragment coupled via transmembrane domains to
the CD3z chain of the TCR, in addition to either the CD28 or
286
41BB costimulatory domains that enhance lymphocyte activation following antigen recognition. When transduced
into T cells, CAR allows the TAA-specific direction of ACT
toward cell surface peptide, carbohydrate, and lipid antigens
in a high-affinity manner without the MHC restriction that
imposes haplotype limitations and vulnerability to loss of
tumor MHC molecule expression upon engineered TCR.

Chimeric antigen receptors have incredible clinical activity
and are completely revolutionizing the care for CD19positive hematologic malignancies. In fact, these T cells
sometimes are so effective that patients can suffer from
cytokine storm as a result of their rapid tumor recognition
and expansion.87-92 The efficacy of CAR T cells requires a
highly cancer-specific cell surface protein, and many groups
have been searching for these for many years with limited
success. However, there are trials using these exciting receptors in solid tumors, including the use of GD-2targeting
CAR in ganglioside-positive sarcomas (NCT02107963;
NCT01953900).
CONCLUSION
The diverse and rare biologies that underlie soft tissue
sarcoma have historically contributed to the often painstakingly slow and inefficient development of effective new
therapies, and outcomes for patients with these diseases
remain poor. However, an accumulating understanding of
the molecular pathology of some soft tissue sarcoma subtypes has yielded important therapeutic breakthroughs. The
dramatic and increasingly far-reaching impact that modern
immunotherapeutic techniques have had on oncology
presents a tremendous opportunity for further progress in
sarcoma treatment, while the mounting evidence of the
potentially idiosyncratic immune biology of these diseases is
of particular interest to the immuno-oncologist. As seen with
the effect of imatinib on Ido expression in GIST, further

research may uncover novel roles for the immune system in
contributing to the success or failure of existing treatment
modalities. Ongoing characterization of the immune microenvironment and antigenic signatures of other sarcoma
types will reveal promising new immunotherapeutic targets
for vaccine and T-cellbased therapies, as exemplified by the
identification and targeting of NY-ESO-1 in synovial sarcoma
and MRCL.
The progress of immunotherapy over the past 5 years has
been unrivaled, as has been the accompanying enthusiasm
within the oncology community. As the field becomes more
and more ensconced in the day-to-day practice of all cancer
physicians, the current wave of immunotherapy trials
promises to add further agents to the immuno-oncologists
toolbox. The therapeutic manipulation of the tumor immune microenvironment, the dismantling of tumor immune
evasion and activation of tumor-infiltrating T cells, the
generation of antigen-specific immune responses, and the
enrichment and redirection of T cells toward specific tumor
targets all represent exciting avenues of immunotherapy
that are under active investigation in soft tissue sarcoma.
However, it is also important to keep a historical perspective
and to bear in mind the decades of research that has
underpinned recent successes in immunotherapy. Only
through continued careful analysis and translational research over the coming years will the full promise of
immuno-oncology be realized for patients with sarcoma.
References
1. Karavasilis V, Seddon BM, Ashley S, et al. Significant clinical benefit of
first-line palliative chemotherapy in advanced soft-tissue sarcoma:
retrospective analysis and identification of prognostic factors in 488
patients. Cancer. 2008;112:1585-1591.
2. Linch M, Miah AB, Thway K, et al. Systemic treatment of soft-tissue
sarcoma-gold standard and novel therapies. Nat Rev Clin Oncol. 2014;
11:187-202.
3. van der Graaf WT, Blay JY, Chawla SP, et al; EORTC Soft Tissue and Bone
Sarcoma Group; PALETTE Study Group. Pazopanib for metastatic softtissue sarcoma (PALETTE): a randomised, double-blind, placebocontrolled phase 3 trial. Lancet. 2012;379:1879-1886.
4. Kawai A, Araki N, Sugiura H, et al. Trabectedin monotherapy after
standard chemotherapy versus best supportive care in patients with
advanced, translocation-related sarcoma: a randomised, open-label,
phase 2 study. Lancet Oncol. 2015;16:406-416.
5. Bui-Nguyen B, Butrynski JE, Penel N, et al; European Organisation for Research and Treatment of Cancer Soft Tissue and Bone Sarcoma Group
(EORTC/STBSG) and the Sarcoma Alliance for Research Through Collaboration (SARC). A phase IIb multicentre study comparing the efficacy of
trabectedin to doxorubicin in patients with advanced or metastatic untreated
soft tissue sarcoma: the TRUSTS trial. Eur J Cancer. 2015;51:1312-1320.
6. Schoffski P, Ray-Coquard IL, Cioffi A, et al; European Organisation for
Research and Treatment of Cancer (EORTC) Soft Tissue and Bone
Sarcoma Group (STBSG). Activity of eribulin mesylate in patients with
soft-tissue sarcoma: a phase 2 study in four independent histological
subtypes. Lancet Oncol. 2011;12:1045-1052.
7. Ducimetière F, Lurkin A, Ranchère-Vince D, et al. Incidence of sarcoma

histotypes and molecular subtypes in a prospective epidemiological
study with central pathology review and molecular testing. PLoS One.
2011;6:e20294.
8. Hirota S, Isozaki K, Moriyama Y, et al. Gain-of-function mutations of c-kit
in human gastrointestinal stromal tumors. Science. 1998;279:577-580.
9. Heinrich MC, Corless CL, Duensing A, et al. PDGFRA activating mutations
in gastrointestinal stromal tumors. Science. 2003;299:708-710.
10. Verweij J, Casali PG, Zalcberg J, et al. Progression-free survival in
gastrointestinal stromal tumours with high-dose imatinib: randomised
trial. Lancet. 2004;364:1127-1134.
11. Blanke CD, Rankin C, Demetri GD, et al. Phase III randomized, intergroup
trial assessing imatinib mesylate at two dose levels in patients with
unresectable or metastatic gastrointestinal stromal tumors expressing
the kit receptor tyrosine kinase: S0033. J Clin Oncol. 2008;26:626-632.
12. Corless CL, Ballman KV, Antonescu CR, et al. Pathologic and molecular
features correlate with long-term outcome after adjuvant therapy of
resected primary GI stromal tumor: the ACOSOG Z9001 trial. J Clin
Oncol. 2014;32:1563-1570.
13. Chen LL, Trent JC, Wu EF, et al. A missense mutation in KIT kinase
domain 1 correlates with imatinib resistance in gastrointestinal stromal
tumors. Cancer Res. 2004;64:5913-5919.
14. Debiec-Rychter M, Cools J, Dumez H, et al. Mechanisms of resistance to
imatinib mesylate in gastrointestinal stromal tumors and activity of the
PKC412 inhibitor against imatinib-resistant mutants. Gastroenterology.
2005;128:270-279.
287
LEE ET AL
15. Antonescu CR, Besmer P, Guo T, et al. Acquired resistance to imatinib in

gastrointestinal stromal tumor occurs through secondary gene mutation. Clin Cancer Res. 2005;11:4182-4190.
16. McCarthy EF. The toxins of William B. Coley and the treatment of bone
and soft-tissue sarcomas. Iowa Orthop J. 2006;26:154-158.
17. Rosenberg SA, Lotze MT, Yang JC, et al. Experience with the use of highdose interleukin-2 in the treatment of 652 cancer patients. Ann Surg.
1989;210:474-484, discussion 484-485.
18. Topalian SL, Solomon D, Avis FP, et al. Immunotherapy of patients with
advanced cancer using tumor-infiltrating lymphocytes and recombinant
interleukin-2: a pilot study. J Clin Oncol. 1988;6:839-853.
19. Geukes Foppen MH, Donia M, Svane IM, et al. Tumor-infiltrating lymphocytes
for the treatment of metastatic cancer. Mol Oncol. 2015;9:1918-1935.
20. Rosenberg SA, Yang JC, Sherry RM, et al. Durable complete responses in
heavily pretreated patients with metastatic melanoma using T-cell
21. Weiden PL, Sullivan KM, Flournoy N, et al. Antileukemic effect of chronic
graft-versus-host disease: contribution to improved survival after allogeneic marrow transplantation. N Engl J Med. 1981;304:1529-1533.
22. Warren EH, Gavin M, Greenberg PD, et al. Minor histocompatibility
antigens as targets for T-cell therapy after bone marrow transplantation. Curr Opin Hematol. 1998;5:429-433.
711-723.
24. Robert C, Schachter J, Long GV, et al; KEYNOTE-006 Investigators.
Pembrolizumab versus ipilimumab in advanced melanoma. N Engl J
Med. 2015;372:2521-2532.
advanced squamous-cell non-small-cell lung cancer. N Engl J Med. 2015;
373:123-135.
advanced nonsquamous non-small-cell lung cancer. N Engl J Med. 2015;
373:1627-1639.
28. Maus MV, Grupp SA, Porter DL, et al. Antibody-modified T cells: CARs
take the front seat for hematologic malignancies. Blood. 2014;123:
2625-2635.
29. Pollack SM, Loggers ET, Rodler ET, et al. Immune-based therapies for
sarcoma. Sarcoma. 2011;2011:438940.
30. Fridman WH, Pages
` F, Sautes-Fridman
`
C, et al. The immune contexture
in human tumours: impact on clinical outcome. Nat Rev Cancer. 2012;
12:298-306.
31. Gooden MJM, de Bock GH, Leffers N, et al. The prognostic influence of
tumour-infiltrating lymphocytes in cancer: a systematic review with
meta-analysis. Br J Cancer. 2011;105:93-103.
32. Berghuis D, Santos SJ, Baelde HJ, et al. Pro-inflammatory chemokinechemokine receptor interactions within the Ewing sarcoma microenvironment determine CD8+ T-lymphocyte infiltration and affect tumour
progression. J Pathol. 2011;223:347-357.
33. Feng Y, Shen J, Gao Y, et al. Expression of programmed cell death ligand
1 (PD-L1) and prevalence of tumor-infiltrating lymphocytes (TILs) in
chordoma. Oncotarget. 2015;6:11139-11149.
34. DAngelo SP, Shoushtari AN, Agaram NP, et al. Prevalence of tumorinfiltrating lymphocytes and PD-L1 expression in the soft tissue sarcoma
microenvironment. Hum Pathol. 2015;46:357-365.
35. Rusakiewicz S, Semeraro M, Sarabi M, et al. Immune infiltrates are
prognostic factors in localized gastrointestinal stromal tumors. Cancer
Res. 2013;73:3499-3510.
288
36. Fujii H, Arakawa A, Utsumi D, et al. CD8 tumor-infiltrating lymphocytes

at primary sites as a possible prognostic factor of cutaneous angiosarcoma. Int J Cancer. 2014;134:2393-2402.
37. Sorbye SW, Kilvaer T, Valkov A, et al. Prognostic impact of lymphocytes
in soft tissue sarcomas. PLoS One. 2011;6:e14611.
38. Sorbye SW, Kilvaer TK, Valkov A, et al. Prognostic impact of
peritumoral lymphocyte infiltration in soft tissue sarcomas. BMC Clin
Pathol. 2012;12:5.
39. Theoleyre S, Mori K, Cherrier B, et al. Phenotypic and functional analysis
of lymphocytes infiltrating osteolytic tumors: use as a possible therapeutic approach of osteosarcoma. BMC Cancer. 2005;5:123.
40. Rivoltini L, Arienti F, Orazi A, et al. Phenotypic and functional analysis
of lymphocytes infiltrating paediatric tumours, with a characterization
of the tumour phenotype. Cancer Immunol Immunother. 1992;34:
241-251.
41. Schwinger W, Klass V, Benesch M, et al. Feasibility of high-dose
interleukin-2 in heavily pretreated pediatric cancer patients. Ann
Oncol. 2005;16:1199-1206.
42. Biswas SK, Mantovani A. Macrophage plasticity and interaction
with lymphocyte subsets: cancer as a paradigm. Nat Immunol. 2010;11:
889-896.
43. Heusinkveld M, van der Burg SH. Identification and manipulation of
tumor associated macrophages in human cancers. J Transl Med. 2011;9:
216.
44. Cavnar MJ, Zeng S, Kim TS, et al. KIT oncogene inhibition drives
intratumoral macrophage M2 polarization. J Exp Med. 2013;210:
2873-2886.
45. Ruffell D, Mourkioti F, Gambardella A, et al. A CREB-C/EBPbeta cascade
induces M2 macrophage-specific gene expression and promotes muscle
injury repair. Proc Natl Acad Sci USA. 2009;106:17475-17480.
46. Lee C-H, Espinosa I, Vrijaldenhoven S, et al. Prognostic significance of
macrophage infiltration in leiomyosarcomas. Clin Cancer Res. 2008;14:
1423-1430.
47. Espinosa I, Beck AH, Lee CH, et al. Coordinate expression of colonystimulating factor-1 and colony-stimulating factor-1-related proteins is
associated with poor prognosis in gynecological and nongynecological
leiomyosarcoma. Am J Pathol. 2009;174:2347-2356.
48. Ganjoo KN, Witten D, Patel M, et al. The prognostic value of tumorassociated macrophages in leiomyosarcoma: a single institution study.
Am J Clin Oncol. 2011;34:82-86.
49. Nabeshima A, Matsumoto Y, Fukushi J, et al. Tumour-associated
macrophages correlate with poor prognosis in myxoid liposarcoma
and promote cell motility and invasion via the HB-EGF-EGFR-PI3K/Akt
pathways. Br J Cancer. 2015;112:547-555.
50. Fujiwara T, Fukushi J, Yamamoto S, et al. Macrophage infiltration
predicts a poor prognosis for human Ewing sarcoma. Am J Pathol. 2011;
179:1157-1170.
51. Patwardhan PP, Surriga O, Beckman MJ, et al. Sustained inhibition of
receptor tyrosine kinases and macrophage depletion by PLX3397 and
rapamycin as a potential new approach for the treatment of MPNSTs.
Clin Cancer Res. 2014;20:3146-3158.
52. Liu J, Wang L, Zhao F, et al. Pre-clinical development of a humanized
anti-CD47 antibody with anti-cancer therapeutic potential. PLoS One.
2015;10:e0137345.
53. Sommer G, Agosti V, Ehlers I, et al. Gastrointestinal stromal tumors in a
mouse model by targeted mutation of the Kit receptor tyrosine kinase.
Proc Natl Acad Sci USA. 2003;100:6706-6711.
54. Balachandran VP, Cavnar MJ, Zeng S, et al. Imatinib potentiates antitumor T cell responses in gastrointestinal stromal tumor through the
inhibition of Ido. Nat Med. 2011;17:1094-1100.
55. Munn DH, Mellor AL. Indoleamine 2,3 dioxygenase and metabolic
control of immune responses. Trends Immunol. 2013;34:137-143.
56. Uyttenhove C, Pilotte L, Theate I, et al. Evidence for a tumoral immune

resistance mechanism based on tryptophan degradation by indoleamine 2,3-dioxygenase. Nat Med. 2003;9:1269-1274.
57. Mellor AL, Munn DH. IDO expression by dendritic cells: tolerance and
tryptophan catabolism. Nat Rev Immunol. 2004;4:762-774.
58. Chi P, Chen Y, Zhang L, et al. ETV1 is a lineage survival factor that
cooperates with KIT in gastrointestinal stromal tumours. Nature. 2010;
467:849-853.
59. Shoushtari AN, DAngelo SP, Keohan ML, et al. Combined KIT and CTLA4 blockade in patients with refractory GIST and other advanced sarcomas. J Clin Oncol. 2014;32 (suppl; abstr 10521).
60. Xu F, Xu L, Wang Q, et al. Clinicopathological and prognostic value of
programmed death ligand-1 (PD-L1) in renal cell carcinoma: a metaanalysis. Int J Clin Exp Med. 2015;8:14595-14603.
61. Tamura T, Ohira M, Tanaka H, et al. Programmed death-1 ligand-1
(PDL1) expression is associated with the prognosis of patients with
stage II/III gastric cancer. Anticancer Res. 2015;35:5369-5376.
62. Clemente CG, Mihm MC Jr, Bufalino R, et al. Prognostic value of tumor
infiltrating lymphocytes in the vertical growth phase of primary cutaneous melanoma. Cancer. 1996;77:1303-1310.
63. Velcheti V, Schalper KA, Carvajal DE, et al. Programmed death
ligand-1 expression in non-small cell lung cancer. Lab Invest. 2014;
94:107-116.
64. Cierna Z, Mego M, Miskovska V, et al. Prognostic value of programmeddeath-1 receptor (PD-1) and its ligand 1 (PD-L1) in testicular germ cell
tumors. Ann Oncol. 2016;27:300-305.
65. Kim JR, Moon YJ, Kwon KS, et al. Tumor infiltrating PD1positive lymphocytes and the expression of PD-L1 predict poor prognosis of soft tissue sarcomas. PLoS One. 2013;8:e82870.
66. Yuan J, Gnjatic S, Li H, et al. CTLA-4 blockade enhances polyfunctional
NY-ESO-1 specific T cell responses in metastatic melanoma patients
with clinical benefit. Proc Natl Acad Sci USA. 2008;105:20410-20415.
67. Maki RG, Jungbluth AA, Gnjatic S, et al. A pilot study of anti-CTLA4
antibody ipilimumab in patients with synovial sarcoma. Sarcoma. 2013;
2013:168145.
68. Patnaik A, Kang SP, Rasco D, et al. Phase I study of pembrolizumab (MK3475; anti-PD-1 monoclonal antibody) in patients with advanced solid
tumors. Clin Cancer Res. 2015;21:4286-4293.
69. Scanlan MJ, Simpson AJG, Old LJ. The cancer/testis genes: review,
standardization, and commentary. Cancer Immun. 2004;4:1.
70. Odunsi K, Matsuzaki J, Karbach J, et al. Efficacy of vaccination with
recombinant vaccinia and fowlpox vectors expressing NY-ESO-1 antigen
in ovarian cancer and melanoma patients. Proc Natl Acad Sci USA. 2012;
109:5797-5802.
71. Hunder NN, Wallen H, Cao J, et al. Treatment of metastatic melanoma
with autologous CD4+ T cells against NY-ESO-1. N Engl J Med. 2008;358:
2698-2703.
72. Ayyoub M, Taub RN, Keohan ML, et al. The frequent expression of
cancer/testis antigens provides opportunities for immunotherapeutic
targeting of sarcoma. Cancer Immun. 2004;4:7.
73. Hemminger J, Toland AE, Scharschmidt TJ, et al. Expression of cancertestis antigens MAGEA1, MAGEA3, ACRBP, PRAME, SSX2, and CTAG2 in
myxoid and round cell liposarcoma. Mod Pathol. 2014;27:1238-1245.
74. Pollack SM, Jungbluth AA, Hoch BL, et al. NY-ESO-1 is a ubiquitous
immunotherapeutic target antigen for patients with myxoid/round cell
liposarcoma. Cancer. 2012;118:4564-4570.
75. Jungbluth AA, Antonescu CR, Busam KJ, et al. Monophasic and biphasic
synovial sarcomas abundantly express cancer/testis antigen NY-ESO-1
but not MAGE-A1 or CT7. Int J Cancer. 2001;94:252-256.
76. Pollack SM, Li Y, Blaisdell MJ, et al. NYESO-1/LAGE-1s and PRAME are
targets for antigen specific T cells in chondrosarcoma following
treatment with 5-Aza-2-deoxycitabine. PLoS One. 2012;7:e32165.
77. Carvajal RD, Agulnik M, Ryan CW, et al. Trivalent ganglioside vaccine and
immunologic adjuvant versus adjuvant alone in metastatic sarcoma
patients rendered disease-free by surgery: a randomized phase 2 trial.
78. Mahipal A, Odunsi K, Gnjatic S, et al. A first-in-human phase 1 doseescalating trial of G305 in patients with solid tumors expressing NYESO-1. J Clin Oncol. 2015;33 (suppl; abstr 3073).
79. Pollack SM, Jones RL, Farrar EA, et al. Tetramer guided, cell sorter
assisted production of clinical grade autologous NY-ESO-1 specific CD8+
T cells. J Immunother Cancer. 2014;2:36.
80. Robbins PF, Li YF, El-Gamil M, et al. Single and dual amino acid substitutions in TCR CDRs can enhance antigen-specific T cell functions.
J Immunol. 2008;180:6116-6131.
81. Robbins PF, Morgan RA, Feldman SA, et al. Tumor regression in patients
with metastatic synovial cell sarcoma and melanoma using genetically
engineered lymphocytes reactive with NY-ESO-1. J Clin Oncol. 2011;29:
917-924.
82. Robbins PF, Kassim SH, Tran TL, et al. A pilot trial using lymphocytes
genetically engineered with an NY-ESO-1-reactive T-cell receptor: longterm follow-up and correlates with response. Clin Cancer Res. 2015;21:
1019-1027.
83. Johnson LA, Morgan RA, Dudley ME, et al. Gene therapy with
human and mouse T-cell receptors mediates cancer regression and
targets normal tissues expressing cognate antigen. Blood. 2009;
114:535-546.
84. Parkhurst MR, Yang JC, Langan RC, et al. T cells targeting carcinoembryonic antigen can mediate regression of metastatic colorectal cancer
but induce severe transient colitis. Mol Ther. 2011;19:620-626.
85. Morgan RA, Chinnasamy N, Abate-Daga D, et al. Cancer regression and
neurological toxicity following anti-MAGE-A3 TCR gene therapy.
J Immunother. 2013;36:133-151.
86. Cameron BJ, Gerry AB, Dukes J, et al. Identification of a Titin-derived
HLA-A1-presented peptide as a cross-reactive target for engineered
MAGE A3-directed T cells. Sci Transl Med. 2013;5:197ra103.
87. Bonini C, Ferrari G, Verzeletti S, et al. HSV-TK gene transfer into
donor lymphocytes for control of allogeneic graft-versus-leukemia.
Science. 1997;276:1719-1724.
88. Straathof KC, Pulè MA, Yotnda P, et al. An inducible caspase 9 safety
switch for T-cell therapy. Blood. 2005;105:4247-4254.
89. Lanitis E, Poussin M, Klattenhoff AW, et al. Chimeric antigen receptor
T cells with dissociated signaling domains exhibit focused antitumor
activity with reduced potential for toxicity in vivo. Cancer Immunol Res.
2013;1:43-53.
90. Wilkie S, van Schalkwyk MC, Hobbs S, et al. Dual targeting of ErbB2 and
MUC1 in breast cancer using chimeric antigen receptors engineered to
provide complementary signaling. J Clin Immunol. 2012;32:1059-1070.
91. Fedorov VD, Themeli M, Sadelain M. PD-1- and CTLA-4-based inhibitory
chimeric antigen receptors (iCARs) divert off-target immunotherapy
responses. Sci Transl Med. 2013;5:215ra172.
92. Wang X, Chang WC, Wong CW, et al. A transgene-encoded cell surface
polypeptide for selection, in vivo tracking, and ablation of engineered
cells. Blood. 2011;118:1255-1263.
93. Geiger JD, Hutchinson RJ, Hohenkirk LF, et al. Vaccination of pediatric
solid tumor patients with tumor lysate-pulsed dendritic cells can expand specific T cells and mediate tumor regression. Cancer Res. 2001;
61:8513-8519.
94. Suminoe A, Matsuzaki A, Hattori H, et al. Immunotherapy with autologous dendritic cells and tumor antigens for children with refractory
malignant solid tumors. Pediatr Transplant. 2009;13:746-753.
95. Mackall CL, Rhee EH, Read EJ, et al. A pilot study of consolidative
immunotherapy in patients with high-risk pediatric sarcomas. Clin
Cancer Res. 2008;14:4850-4858.
289
LEE ET AL
96. Dillman R, Barth N, Selvan S, et al. Phase I/II trial of autologous tumor
cell line-derived vaccines for recurrent or metastatic sarcomas. Cancer
Biother Radiopharm. 2004;19:581-588.
97. Bender A, Karbach J, Neumann A, et al. LUD 00-009: phase 1 study of
intensive course immunization with NY-ESO-1 peptides in HLA-A2 positive
patients with NY-ESO-1-expressing cancer. Cancer Immun. 2007;7:16.
290
98. Kawaguchi S, Wada T, Tsukahara T, et al. A quest for therapeutic antigens in bone and soft tissue sarcoma. J Transl Med. 2005;3:31.
99. Kawaguchi S, Tsukahara T, Ida K, et al. SYT-SSX breakpoint peptide
vaccines in patients with synovial sarcoma: a study from the
Japanese Musculoskeletal Oncology Group. Cancer Sci. 2012;103:
1625-1630.
SANDRA P. DANGELO
Manipulating the Immune System With Checkpoint Inhibitors

for Patients With Metastatic Sarcoma
Sandra P. DAngelo, MD
OVERVIEW
Sarcomas are a rare group of malignant tumors of mesenchymal origin that comprise 1% of all adult cancers. Despite initial
surgery, distant metastatic disease will develop in approximately 25% of patients, and standard chemotherapy has limited
durable efficacy. There is a dire need for more effective and less toxic therapies for the treatment of metastatic sarcoma. The
immune system plays a major role in cancer control and progression. There have been tremendous breakthroughs in other
malignancies by manipulating the immune system with checkpoint inhibitors. These agents, either alone or in combination
with other approaches such as radiation, chemotherapy, targeted agents, or immunotherapeutics, have generally led to
improved efficacy in selected malignancies thus far. Although promising, these drugs can cause specific immune-related
adverse events that require prompt recognition and treatment. In addition, characterizing response and progression radiographically has become somewhat more challenging. Identifying predictive biomarkers of benefit will be essential. There
remains optimism and hope that the strides made in other cancers will be emulated in sarcoma.
arcomas are a rare group of malignant tumors of mesenchymal origin that comprise 1% of all adult cancers.
There are over 50 histologic subtypes. Most patients undergo curative surgical resection with or without adjuvant
radiation and chemotherapy.1 Adjuvant chemotherapy is
generally reserved for specific histologic subtypes.2 Despite
initial surgery, distant metastatic disease will develop in
approximately 25% of patients.3,4 Complete responses to
chemotherapy for metastatic sarcoma are infrequent and
the median survival remains dismal.5,6 There is a dire need
for the development of more effective, less toxic therapies
for the treatment of metastatic sarcoma. It has become
more apparent that the immune system plays a major role in
cancer control and progression. There have been tremendous breakthroughs in other malignancies by manipulating
the immune system with checkpoint inhibitors. Therefore,
there remains optimism for the development of checkpoint
inhibitors in sarcoma.
Immune checkpoints are necessary to prevent autoimmunity and protect tissue from damage when the immune
system responds to foreign pathogens.7 Alternatively, tumors often overexpress immune checkpoint proteins as a
mechanism to avoid death.7 As a result, dysregulation of
these immune checkpoint pathways lead to tumor immune
resistance. Immune checkpoints are typically ligand-receptor
interactions, which are amenable to blockade by antibodies.
CTLA-4 BLOCKADE
The CTLA-4 receptor was one of the first receptors to be
targeted.8-10 CTLA-4 binds to CD28 to downregulate T-cell
function (Fig. 1).9,11-14 There are two human monoclonal
antibodies, ipilimumab and tremelimumab, that have been
used to target CTLA-4. Ultimately, ipilimumab was approved
by the U.S. Food and Drug Administration (FDA) for metastatic melanoma, and recent data have demonstrated durable long-term survival with 20% of patients surviving
beyond 3 years.15
Limited clinical data are available regarding the efficacy of
CTLA-4 blockade in sarcoma. In a small phase II study, six
patients with synovial sarcoma received 3 mg/kg of ipilimumab every 3 weeks.16 Four patients completed three
doses of ipilimumab, whereas two patients each received
one and two doses because of clinical or radiologic progression. There were no documented responses. The patients had rapidly progressive disease and their median
survival was only 8.75 months, which is much worse than the
average patient with metastatic synovial sarcoma. Given
their advanced state, there may not have been sufficient
time to see a potential benefit from the ipilimumab, which
can typically take time. Further, radiologic progression after
just one dose of ipilimumab is not necessarily consistent with
true progression in the setting of immunotherapy, and lack
of immediate response may not translate to lack of benefit.
From the Memorial Sloan Kettering Cancer Center, Weill Cornell Medical College, New York, NY.
Corresponding author: Sandra P. DAngelo, MD, Memorial Sloan Kettering Cancer Center, 300 East 66th St., New York, NY 10065; email: dangelos@mskcc.org.
e558
CHECKPOINT INHIBITORS IN METASTATIC SARCOMA
FIGURE 1. Mechanism of Action of CTLA-4 and PDL1 Blockade
(A) Interactions between the tumor antigen on the dendritic cell with the TCR, along with interaction of the costimulatory molecules B7 and CD28, are necessary for T-cell
activation. As a negative regulator of the immune response, CTLA-4 competes with CD28 to bind with B7. Ipilimumab and tremelimumab are monoclonal antibodies that
bind to CTLA-4 and inhibit negative regulation and activate T cells. (B) The role of the PD-1 receptor is generally more important in the tumor microenvironment during Tcell activation. Antigen exposure leads to expression of PD-1 on the T cells. Interaction of the PD-1 receptor with its ligands PD-L1 and PD-L2 results in negative regulation of
T cells in the tumor microenvironment. Antibodies that bind to PD-1 or PD-L1 lead to activation of T cells.
Abbreviations: MHC, major histocompatibility complex; TCR, T-cell receptor.
Adapted from DAngelo SP, Tap WD, Schwartz GK, Carvajal RD. Sarcoma immunotherapy: past approaches and future directions. Sarcoma. 2014;2014:Article ID 391967.
PD-1 BLOCKADE
PD-1 and its ligand PD-L1 is another immunologic checkpoint
pathway that serves as a negative regulator of the immune
response. There are two ligands, PD-L1 and PD-L2, that
are specific for PD-1 and, upon binding, downregulate
T-cell activation (Fig. 1).17,18 There are multiple antibodies
that target PD-1 (e.g., nivolumab, pembrolizumab, and
pidilizumab) or PD-L1 (e.g., atezolizumab, durvalumab, and
avelumab). Thus far, nivolumab was FDA approved for
metastatic melanoma, nonsmall cell lung cancer (NSCLC),
and renal cell carcinoma. Pembrolizumab was FDA approved
for metastatic melanoma that progressed during treatment
with ipilimumab or a BRAF-inhibitor, as well as for NSCLC
that is PD-L1positive as confirmed by the PD-L1 immunohistochemistry (IHC) 22C3 pharmDx test.19,20 Since then,
pembrolizumab is now approved in the front-line setting for
KEY POINTS
The CTLA-4 receptor was one of the first checkpoint

inhibitors to be targeted with limited exploration and
efficacy in sarcoma.
The PD-1/PD-L1 immunologic checkpoint pathway has
demonstrated efficacy in many malignancies; studies
are ongoing in sarcoma.
There is substantial interest in combining checkpoint
blockade antibodies with multiple approaches including
chemotherapy, targeted therapy, radiation therapy, and
other immunotherapeutics.
Understanding, recognizing, and promptly treating
immune-related adverse events is essential as these
immunotherapeutics continue clinical development.
Identifying biomarkers predictive of benefit is essential
to continue to move the field forward.
patients with metastatic melanoma. Atezolizumab, durvalumab, and avelumab have been studied in multiple malignancies including bladder cancer, head and neck
cancer, gastrointestinal (GI) malignancies, and Merkel cell
carcinoma.21-24 Most recently, the SARC028 phase II study of
the antiPD-1 antibody pembrolizumab for patients with
advanced sarcomas was completed (NCT02301039).
COMBINATION APPROACHES
To improve clinical efficacy, there has been much interest in
combining checkpoint blockade antibodies with systemic
approaches such as chemotherapy, targeted therapy, radiation therapy, and other immunotherapeutics (Fig. 2).
Chemotherapy can alter the immune milieu by reducing
myeloid-derived suppressor cells,25 increasing the ratio of
M1 to M2 macrophages, and increasing CD8+ T cells and
natural killer (NK) cells.26,27 Trabectedin has been shown to
synergize with an antiPD-1 antibody in a mouse ovarian
cancer model.28 As trabectedin is now FDA approved for
selected sarcomas, there may be rationale to explore this
combination. There is an ongoing study of pembrolizumab
with chemotherapy (pembroplus; NCT02331251). Regardless, it is essential to be mindful of the increased risk of
adverse events. For example, in melanoma and NSCLC, some
studies that combined chemotherapy with ipilimumab
demonstrated an increased risk of transaminitis.29
Targeted therapies such as tyrosine kinase inhibitors (TKIs)
can stimulate immune cells, including CD4+ and CD8+
T cells,30,31 NK cells,32 and dendritic cells.33 Combined
blockade of CTLA-4 plus KIT via imatinib or dasatinib led to an
improved antitumor effect in a murine model of GI stromal
tumor (GIST).34 This provided the scientific rationale to
launch a phase I study of dasatinib plus ipilimumab in patients with advanced GIST and other sarcomas.35 The
maximum tolerated dose was identified as 70 mg of
e559
SANDRA P. DANGELO
FIGURE 2. Strategies to Enhance Antitumor Immunity
Immunotherapy with checkpoint inhibitors for sarcoma can be approached in many different ways. Various aspects of the immune response can be targeted. (A) T-cell
activation requires interaction of TCR with MHC and interaction of costimulatory molecules, such as CD28 and CD80/86. (B) Enhanced T-cell activation can occur with
costimulatory agonists such as antibodies that target 4-1BB, ICOS, OX40, or GITR. There are many pathways that can lead to downregulation of the immune, as noted on the
left side of the figure. (C) Inhibition of the immune response can be mitigated by blocking checkpoint inhibitors beyond PD-1 and CTLA-4, such as LAG-3 or TIM-4, (D) or by
decreasing regulatory T-cells with antibodies that target CD25 or CCR-4. (E) Chemotherapy can alter the ratio of M1 to M2 macrophages and potentially deplete MDSC.
Antibodies that bind colony stimulating factor-1R may also deplete M2 macrophages. (F) IDO is a negative regulator of the immune response found in the tumor
microenvironment that can be targeted with anti-IDO antibodies. (G) Increased exposure of tumor-associated antigens with radiation, as well as chemotherapy, can ultimate
enhance the immune response.
Abbreviations: MHC, major histocompatibility complex; TCR, T-cell receptor; APC, antigen-presenting cell; ICOS, inducible T-cell costimulator; GITR, glucocorticoid-induced
tumor necrosis factor receptorrelated protein; TIM-3, T-cell immunoglobulin and mucin domaincontaining; LAG-3, lymphocyte-activation gene; T reg, regulatory T cell; CCR4,
chemokine C-C Motif receptor 4; MDSC, myeloid derived suppressor cells; CSF-1R, colony stimulating factor 1 receptor; IDO, indoleamine 2,3-dioxygenase; TAA, tumor-associated
antigen.
Adapted from Zamarin D, Postow MA. Immune checkpoint modulation: rational design of combination strategies. Pharmacol Ther. 2015;150:23-32.
dasatinib per day administered concurrently with 10 mg/kg

of ipilimumab. In this treated cohort, three of eight patients
with heavily treated GIST and one of five patients with soft
tissue sarcoma achieved durable disease control. There is an
ongoing expansion trial using the maximum tolerated dose
(NCT01643278). In renal cell carcinoma, nivolumab with
sunitinib or pazopanib was found to be more efficacious,
albeit with increased hepatic and renal toxicity.36 Therefore,
TKIs must be cautiously combined with PD-1 inhibitors.
There is an ongoing clinical trial of pembrolizumab and
axitinib for selected sarcomas (NCT02636725).
Radiotherapy has been shown to increase antigenic expression and increase specific immune responses. In reported case series of patients with melanoma, the so-called
abscopal effect has resulted in decreased metastatic disease
sites after local radiation.37 There are ongoing clinical trials
investigating the safety and efficacy of checkpoint blockade
with radiation therapy.
e560
Most recently, combined CTLA-4 and PD-1 blockade with

ipilimumab and nivolumab was FDA approved for metastatic
melanoma. In the trial that led to its approval, the combination approach resulted in an overall response rate of 61%
compared with 11% in the ipilimumab alone group.38 These
exciting findings have led to an interest in exploring these
agents in multiple disease types. The phase II study of
nivolumab with or without ipilimumab for metastatic sarcoma has nearly completed accrual and data analysis is
pending at this time (NCT02500797). Beyond combining PD-1
with CTLA-4 blockade, there are ongoing efforts of finetuning the immune response with various costimulatory
molecules (OX-40, glucocorticoid-induced tumor necrosis
factor receptorrelated protein [GITR], CD28, 4-1BB, CD27)
or co-inhibitory molecules (T cell immunoglobulin, mucin
domaincontaining [TIM-3], and lymphocyte-activation
gene [LAG-3]). 39-42 There are several immune inhibitory mechanisms that can be targeted as well, including
tumor-infiltrating regulatory T cells via inhibition of chemokine receptor 4 (CCR-4) or CD25, tumor-associated macrophages, and myeloid-derived suppressor cells via anticolony
stimulating factor-1 receptor (CSF-1R). 43,44 Indoleamine
2,3-dioxygenase (IDO) is expressed on tumor cells and is
upregulated in many malignancies. IDO is an essential enzyme that depletes metabolites necessary for immune
function. There are inhibitors of IDO being evaluated in
clinical trials.45,46 There is rationale for targeting these
mechanisms in combination with PD-1/PD-L1 blockade, and
studies are currently underway to evaluate these strategies
in different tumor types. As these clinical trials begin to
unfold, being cognizant of the potential additive immunemediated adverse events will be critical.
response criteria (irRC) define progressive disease as total

disease growth up to 25% from baseline or total disease burden
(new lesions plus the target lesion) greater than 25%.51 These
criteria have correlated with overall survival for patients with
melanoma who were treated with antiCTLA-4 blockade.51
Much of the experience with irRC has been developed specifically in patients who were treated with CTLA-4 therapy
(Table 1). One would anticipate similar findings with PD-1/PD-L1
blockade. However, generally the median time to response
tends to be more rapid and higher with PD-1/PD-L1 blockade.
Nonetheless, these criteria must be prospectively validated to
ultimately serve as a surrogate for efficacy with these agents.
BIOMARKERS ASSOCIATED WITH OUTCOME

UNDERSTANDING AND MONITORING
IMMUNE-RELATED SIDE EFFECTS
Unique patterns of adverse events have been observed in patients treated with checkpoint blockade inhibitors. As a result of
nonspecific immune activation, these drugs can result in various
adverse events referred to as immune-related adverse events.
These events can affect multiple organ systems, including, but
not limited to, dermatologic, pulmonary, hepatic, renal, neurologic, endocrine, and GI.47,48 Grade 3 or 4 events have been
reported in 10% to 20% of patients treated with CTLA-4 or
PD-1/PD-L1 blockade.47,48 It seems as though the incidence of
adverse events may be somewhat lower with antiPD-1/PD-L1
antibodies because their effects tend to occur peripheral to the
tumor, whereas CTLA-4 blockade effects tend to occur more
centrally. Generally, the approach in treating these adverse
events is early recognition. Depending on severity, withholding
treatment and immediate initiation of immunosuppression,
typically with corticosteroids, is essential. For patients whose
disease is refractory to corticosteroids treatment, tumor necrosis factor a antagonists (infliximab) is commonly used for
diarrhea/colitis or mycophenolate mofetil for transaminitis.
Following established algorithms is essential to ensure these
adverse events are addressed appropriately.47,49
MEASURING CLINICAL RESPONSE:

IMMUNE-RELATED RESPONSE CRITERIA
When patients pursue cytotoxic chemotherapy, RECIST can be
used to effectively determine changes in tumor burden.50
When treated with immunotherapy agents, patients can
sometimes develop an increase in tumor burden or new lesions
followed by subsequent decrease or disease stabilization. Per
RECIST or World Health Organization (WHO) criteria, this would
qualify as disease progression. Interestingly, of 123 patients
with metastatic melanoma treated with ipilimumab who
progressed at week 12, almost 20% went on to have a partial
response or stable disease.51 In another analysis of patients
who were long-term survivors of melanoma who had received
ipilimumab, nearly 25% had progressive disease per WHO
criteria, yet had survived 4 years after completion of therapy.52
This prompted an interest in re-evaluating response criteria for
patients treated with immunotherapy. The immune-related
It is clear that predictive biomarkers of benefit with checkpoint

blockade are required for ultimate optimal patient selection.
There was initial suggestion that PD-L1 expression may correlate
with clinical activity of PD-1 blockade.53 The definition of positive
PD-L1 expression by IHC has varied based on the assay and
disease type. Expression of PD-L1 in sarcoma has been performed with at least three different antibodies, and the results
have varied. One series, using the rabbit monoclonal antihuman
PD-L1 antibody (clone 28-8) and an automated assay developed
by Dako identified greater than 1% PD-L1 expression in six out of
50 (12%) samples.54 Another group used the CD274/B7-1H
antibody to determine PD-L1 expression by IHC in 161 samples of
osteosarcoma, 46 samples of leiomyosarcoma, and 33 samples
of Ewing sarcoma.55 Positive PD-L1 expression was noted in 36%
of osteosarcoma samples, 97% of leiomyosarcomas, and 39% of
Ewing sarcomas. Finally, a third group tested PD-L1 expression in
105 cases of soft tissue sarcoma with the Santa Cruz antibody. In
their study, 58% of soft tissue sarcoma had intratumoral infiltration of PD-1positive lymphocytes and 65% of soft tissue
sarcoma expressed PD-L1. Lymphocyte and macrophage PD-L1
expression was noted in 30% and 58%, respectively, and lymphocyte and macrophage infiltration was present in 98%
and 90%, respectively. There was variability and a wide range of
types of lymphocytes present (e.g., CD3, CD4, CD8, FoxP3.) It is
challenging to understand the significance of these particular
values and what defines high or low in sarcoma. In this study,
lymphocyte and macrophage infiltration was noted to be fairly
common in sarcoma, while PD-L1 tumor expression is uncommon in sarcoma, with the highest frequency observed in
GIST.
Although the initial phase I trial of nivolumab in solid
tumors suggested that tumors with PD-L1 expression by IHC
may benefit more frequently from PD-1 blockade,53 more
recent data from the combination ipilimumab plus nivolumab trial in metastatic melanoma have demonstrated
that lack of PD-L1 expression did not preclude benefit.56
Since then, there have been data presented in multiple
malignancies that have demonstrated benefit of PD-1 blockade,
regardless of tumor PD-L1 expression.22,23,56,57 The role of PD-L1
expression as a predictive biomarker of benefit with antiPD-1
therapy continues to evolve. Better biomarkers are necessary to
help identify which patients are most likely to benefit.
e561
SANDRA P. DANGELO
TABLE 1. Comparison of RECIST 1.1, World Health Organization, and Immune-Related Response Criteria51
RECIST 1.1
WHO
irRC
New Measurable
Lesions ( 5 x 5 mm)
PD
PD
Added into tumor burden
New Nonmeasurable
Lesions ( 5 x 5 mm)
PD
PD
Does not define PD,

but will not allow irCR
Nonindex Lesions
Contribute to BOR
Contribute to BOR
Contribute to irRC
Complete Response
Disappearance of all lesions
Disappearance of all lesions*
Disappearance of all lesions*
Partial Response
At least a 30% decrease in sum of diameters

of target lesions, no new lesions, or
unequivocal progression in nonindex lesions
A $ 50% decrease in SPD, no new

lesions, or unequivocal
progression in nonindex lesions*
A $ 50% decrease in tumor

burden compared
with baseline*
Stable Disease
Does not meet criteria for PR or PD, no

new lesions or unequivocal progression
of nonindex lesions
Does not meet criteria for PR or PD
Does not meet

criteria for PR or PD
Progressive Disease
At least a 20% increase in sum of the diameters

of target lesions, smallest sum from study is
used as reference. Sum must demonstrate an
increase of at least 5 mm. One or more new
lesions considered PD.
At least 25% increase in SPD

compared with nadir and/or
unequivocal progression of
nonindex lesions and/or
new lesions
At least 25% increase in tumor

burden compared with nadir*
*Confirmation necessary at least 4 weeks apart.

Abbreviations: WHO, World Health Organization; irRC, immune-related response criteria; PD, progressive disease; BOR, best overall response; SPD, sum of the product of the greatest
diameter; PR, partial response.
The impact of mutational burden and novel tumor antigen

expression (neoantigens) with response to ipilimumab was
evaluated in 25 patients with metastatic melanoma who were
treated with ipilimumab.58 Of these patients, 11 had evidence
of long-term benefit as noted by their median overall survival
of 4.4 years (range 26.9 years) and median duration of
response of 59 weeks (range 42361 weeks). Alternatively,
there were 14 patients who had minimal or no benefit; their
median overall survival was 0.9 years (range 0.42.7 years)
without any evidence of radiographic response. In this
analysis, high mutational burden was statistically associated with benefit distinguishing long-term responders from
nonresponders. However, there were patients with high
mutational burden who did not appear to benefit. Using
a bioinformatics pipeline, immunogenic mutations were
identified that led to the production of specific peptide sequences, which resulted in candidate neoantigens that ultimately correlated with benefit to ipilimumab. Similar findings
were noted in NSCLC whereby higher mutational burden in
the tumor was associated with improved response, clinical
benefit, and progression-free survival.59 It is unknown if
mutational burden and/or neoantigen formation will correlate with potential benefit to checkpoint inhibitors in sarcoma. Radiation-associated sarcomas presumably do have
substantial DNA damage, but most sarcomas are not
characterized by DNA damage caused by ultraviolet exposure or tobacco use. Nonetheless, although some sarcomas are characterized by simple translocations, many do
have a fair amount of genomic instability. Exploring
mutational burden and possible neoantigen production
remains an area of research interest.
CONCLUSION
The field of sarcoma immunotherapy is in its infancy. New
successful therapies, such as the checkpoint inhibitors, have
begun to transform the care of many patients with metastatic
melanoma, lung, and renal cell carcinoma. The success and efficacy of these therapies have led to better understanding of
cancer immunology. Moving forward in the sarcoma field, there
are many questions that remain unanswered. Continuing to
explore the activity of immunomodulatory agents alone or in
combinatorial approaches is essential. Being conscious of the
potential of immune-mediated adverse events will minimize risk
to patients. It will be necessary to continue to optimize the
treatment of patients who experience immune-mediated adverse events. Prospective clinical trials will be necessary to further
investigate the role of irRC to determine response and progression. Finally, determining and identifying predictive biomarkers will be essential to continue to move the field forward.
References
1. Brennan MF, Antonescu CR, Maki RG. Management of Soft Tissue
Sarcoma. New York: Springer; 2012.
2. DeVita VT, Lawrence TS, Rosenberg SA, et al (eds). Cancer: Principles and
Practice of Oncology. Philadelphia: Lippincott, Williams and Wilkins; 2008.
3. Lindberg RD, Martin RG, Romsdahl MM, et al. Conservative surgery and
postoperative radiotherapy in 300 adults with soft-tissue sarcomas.
Cancer. 1981;47:2391-2397.
e562
4. Weitz J, Antonescu CR, Brennan MF. Localized extremity soft tissue

sarcoma: improved knowledge with unchanged survival over time. J Clin
Oncol. 2003;21:2719-2725.
5. Billingsley KG, Burt ME, Jara E, et al. Pulmonary metastases from
soft tissue sarcoma: analysis of patterns of diseases and postmetastasis survival. Ann Surg. 1999;229:602-610, discussion 610612.
6. Van Glabbeke M, van Oosterom AT, Oosterhuis JW, et al. Prognostic

factors for the outcome of chemotherapy in advanced soft tissue
sarcoma: an analysis of 2,185 patients treated with anthracyclinecontaining first-line regimensa European Organization for Research
and Treatment of Cancer Soft Tissue and Bone Sarcoma Group Study.
J Clin Oncol. 1999;17:150-157.
8. Allison JP, Krummel MF. The Yin and Yang of T cell costimulation.
Science. 1995;270:932-933.
9. Schneider H, Downey J, Smith A, et al. Reversal of the TCR stop signal by
CTLA-4. Science. 2006;313:1972-1975.
10. Walunas TL, Lenschow DJ, Bakker CY, et al. CTLA-4 can function as a
negative regulator of T cell activation. Immunity. 1994;1:405-413.
11. Krummel MF, Allison JP. CTLA-4 engagement inhibits IL-2 accumulation
and cell cycle progression upon activation of resting T cells. J Exp Med.
1996;183:2533-2540.
12. Linsley PS, Brady W, Urnes M, et al. CTLA-4 is a second receptor for the
B cell activation antigen B7. J Exp Med. 1991;174:561-569.
13. Qureshi OS, Zheng Y, Nakamura K, et al. Trans-endocytosis of CD80 and
CD86: a molecular basis for the cell-extrinsic function of CTLA-4. Science. 2011;332:600-603.
14. Riley JL, Mao M, Kobayashi S, et al. Modulation of TCR-induced transcriptional profiles by ligation of CD28, ICOS, and CTLA-4 receptors. Proc
Natl Acad Sci USA. 2002;99:11790-11795.
15. Schadendorf D, Hodi FS, Robert C, et al. Pooled analysis of long-term
survival data from phase II and phase III trials of ipilimumab in
unresectable or metastatic melanoma. J Clin Oncol. 2015;33:
1889-1894.
16. Maki RG, Jungbluth AA, Gnjatic S, et al. A pilot study of anti-CTLA4
antibody ipilimumab in patients with synovial sarcoma. Sarcoma. 2013;
2013:168145.
17. Carter L, Fouser LA, Jussif J, et al. PD-1:PD-L inhibitory pathway affects
both CD4(+) and CD8(+) T cells and is overcome by IL-2. Eur J Immunol.
2002;32:634-643.
18. Latchman Y, Wood CR, Chernova T, et al. PD-L2 is a second ligand for PD1 and inhibits T cell activation. Nat Immunol. 2001;2:261-268.
19. Herbst RS, Baas P, Kim DW, et al. Pembrolizumab versus docetaxel for
previously treated, PD-L1-positive, advanced non-small-cell lung cancer
(KEYNOTE-010): a randomised controlled trial. Lancet. Epub 2015 Dec
18.
20. Robert C, Ribas A, Wolchok JD, et al. Anti-programmed-death-receptor1 treatment with pembrolizumab in ipilimumab-refractory advanced
melanoma: a randomised dose-comparison cohort of a phase 1 trial.
Lancet. 2014;384:1109-1117.
21. Heery CR, OSullivan Coyne GH, Madan RA, et al. Phase I open-label,
multiple ascending dose trial of MSB0010718C, an anti-PD-L1 monoclonal antibody, in advanced solid malignancies. J Clin Oncol. 2014;32:5s
22. Herbst RS, Gordon MS, Fine GD, et al. A study of MPDL3280A, an
engineered PD-L1 antibody in patients with locally advanced or metastatic tumors. J Clin Oncol. 2013;31 (suppl; abstr 3000).
23. Powles T, Vogelzang NJ, Fine GD, et al. Inhibition of PD-L1 by
MPDL3280A and clinical activity in pts with metastatic urothelial
bladder cancer (UBC). J Clin Oncol. 2014;32:5s (suppl; abstr
5011).
24. Segal NH, Antonia SJ, Brahmer JR, et al. Preliminary data from a multiarm expansion study of MEDI4736, an anti-PD-L1 antibody. J Clin Oncol.
2014;32:5s (suppl; abstr 3002).
25. Ko HJ, Kim YJ, Kim YS, et al. A combination of chemoimmunotherapies
can efficiently break self-tolerance and induce antitumor immunity in a
tolerogenic murine tumor model. Cancer Res. 2007;67:7477-7486.
26. Fridlender ZG, Sun J, Singhal S, et al. Chemotherapy delivered after viral
immunogene therapy augments antitumor efficacy via multiple
immune-mediated mechanisms. Mol Ther. 2010;18:1947-1959.
27. Suzuki E, Kapoor V, Jassar AS, et al. Gemcitabine selectively eliminates
splenic Gr-1+/CD11b+ myeloid suppressor cells in tumor-bearing animals and enhances antitumor immune activity. Clin Cancer Res. 2005;
11:6713-6721.
28. Guo Z, Wang H, Meng F, et al. Combined Trabectedin and anti-PD1
antibody produces a synergistic antitumor effect in a murine model of
ovarian cancer. J Transl Med. 2015;13:247.
29. Robert C, Thomas L, Bondarenko I, et al. Ipilimumab plus dacarbazine
for previously untreated metastatic melanoma. N Engl J Med. 2011;364:
2517-2526.
30. Lee KC, Ouwehand I, Giannini AL, et al. Lck is a key target of imatinib and
dasatinib in T-cell activation. Leukemia. 2010;24:896-900.
31. Ozao-Choy J, Ma G, Kao J, et al. The novel role of tyrosine kinase inhibitor in the reversal of immune suppression and modulation of tumor
microenvironment for immune-based cancer therapies. Cancer Res.
2009;69:2514-2522.
32. Kreutzman A, Juvonen V, Kairisto V, et al. Mono/oligoclonal T and NK
cells are common in chronic myeloid leukemia patients at diagnosis and
expand during dasatinib therapy. Blood. 2010;116:772-782.
33. Ray P, Krishnamoorthy N, Oriss TB, et al. Signaling of c-kit in dendritic
cells influences adaptive immunity. Ann N Y Acad Sci. 2010;1183:104122.
34. Balachandran VP, Cavnar MJ, Zeng S, et al. Imatinib potentiates antitumor T cell responses in gastrointestinal stromal tumor through the
inhibition of Ido. Nat Med. 2011;17:1094-1100.
35. Shoushtari AN, DAngelo SP, Keohan ML, et al. Combined KIT and CTLA-4
blockade in patients with refractory GIST and other advanced sarcomas.
36. Amin A, Plimack ER, Infante JR, et al. Nivolumab (anti-PD-1; BMS936558, ONO-4538) in combination with sunitinib or pazopanib in
patients (pts) with metastatic renal cell carcinoma (mRCC). J Clin Oncol.
2014;32:5s (suppl; abstr 5010).
37. Barker CA, Postow MA. Combinations of radiation therapy and immunotherapy for melanoma: a review of clinical outcomes. Int J Radiat
Oncol Biol Phys. 2014;88:986-997.
2006-2017.
39. Curran MA, Kim M, Montalvo W, et al. Combination CTLA-4 blockade
and 4-1BB activation enhances tumor rejection by increasing T-cell
infiltration, proliferation, and cytokine production. PLoS One. 2011;6:
e19499.
40. Kohrt HE, Houot R, Goldstein MJ, et al. CD137 stimulation enhances the
antilymphoma activity of anti-CD20 antibodies. Blood. 2011;117:
2423-2432.
41. Kohrt HE, Houot R, Weiskopf K, et al. Stimulation of natural killer cells
with a CD137-specific antibody enhances trastuzumab efficacy in
xenotransplant models of breast cancer. J Clin Invest. 2012;122:
1066-1075.
42. Marabelle A, Kohrt H, Sagiv-Barfi I, et al. Depleting tumor-specific Tregs
at a single site eradicates disseminated tumors. J Clin Invest. 2013;123:
2447-2463.
43. Curiel TJ, Coukos G, Zou L, et al. Specific recruitment of regulatory T cells
in ovarian carcinoma fosters immune privilege and predicts reduced
survival. Nat Med. 2004;10:942-949.
44. Reinartz S, Schumann T, Finkernagel F, et al. Mixed-polarization phenotype of ascites-associated macrophages in human ovarian carcinoma:
correlation of CD163 expression, cytokine levels and early relapse. Int J
Cancer. 2014;134:32-42.
e563
SANDRA P. DANGELO
45. Munn DH, Mellor AL. IDO and tolerance to tumors. Trends Mol Med.
2004;10:15-18.
46. Whiteside TL. Tumor-induced death of immune cells: its mechanisms
and consequences. Semin Cancer Biol. 2002;12:43-50.
48. Weber JS, Kahler KC, Hauschild A. Management of immune-related
adverse events and kinetics of response with ipilimumab. J Clin Oncol.
2012;30:2691-2697.
49. Bristol-Meyers Squibb. Immune-Mediated Adverse Reactions Management Guide: Ipilimumab. http://www.hcp.yervoy.com/servlet/
servlet.FileDownload?file=00Pi000000PI1ZVEA1. Accessed March 5,
2016.
50. Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation
criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J
Cancer. 2009;45:228-247.
51. Wolchok JD, Hoos A, ODay S, et al. Guidelines for the evaluation of
e564
52. Wolchok JD, Weber JS, Maio M, et al. Four-year survival rates for
patients with metastatic melanoma who received ipilimumab in phase II
clinical trials. Ann Oncol. 2013;24:2174-2180.
53. Topalian SL, Hodi FS, Brahmer JR, et al. Safety, activity, and immune correlates of anti-PD-1 antibody in cancer. N Engl J Med. 2012;366:2443-2454.
54. DAngelo SP, Shoushtari AN, Agaram NP, et al. Prevalence of tumorinfiltrating lymphocytes and PD-L1 expression in the soft tissue sarcoma
microenvironment. Hum Pathol. 2015;46:357-365.
55. Raj S, Bui M, Gonzales R, et al. Impact of PD-L1 expression on clinical
outcomes in subtypes of sarcoma. Ann Oncol. 2014;25 (suppl 4).
56. Wolchok JD, Kluger H, Callahan MK, et al. Nivolumab plus ipilimumab in
57. Powderly JD, Koeppen H, Hodi SF, et al. Biomarkers and associations
with the clinical activity of PD-L1 blockade in a MPDL3280A study. J Clin
Oncol. 2013:31 (suppl; abstr 3001).
58. Snyder Charen A, Makorov V, Merghoub T, et al. The neoantigen
landscape underlying clinical response to ipilimumab. J Clin Oncol. 2014;
32:5s (suppl; abstr 3003).
cancer. Science. 2015;348:124-128.
SARCOMA
Noncytotoxic Approaches to the

Treatment of Metastatic Soft
Tissue Sarcoma
CHAIR
Alessandro Gronchi, MD
Fondazione IRCCS Istituto Nazionale dei Tumori
Milan, Italy
SPEAKERS
B. Ashleigh Guadagnolo, MD, MPH
Houston, TX
Joseph Patrick Erinjeri, MD, PhD
New York, NY
GRONCHI, GUADAGNOLO, AND ERINJERI
Local Ablative Therapies to Metastatic Soft Tissue Sarcoma

Alessandro Gronchi, MD, B. Ashleigh Guadagnolo, MD, MPH, and Joseph Patrick Erinjeri, MD, PhD
OVERVIEW
The approach to metastatic soft tissue sarcoma is complex and depends upon several factors, such as the extent of the
disease, the histologic subtype of the primary tumor, the disease-free interval, patient status and comorbidities, and
previous treatments. The effect of systemic chemotherapy is suboptimal, therefore local ablative therapies are often
considered when the disease is limited, especially if confined to a single site/organ. Historically, surgery has been considered
the treatment of choice for isolated lung metastases. This approach also has been extended to metastases in the liver,
although a formal demonstration of its benefit has never been provided. Radiation therapy instead has been mainly used to
obtain pain control and to reduce the risk of bone fracture and cord compression. Advances in techniques, such as the
development of more precise conformational modalities and the employment of particles, may change the role of this
modality in the strategic approach to metastatic soft tissue sarcoma. Recently, the use of interventional radiology in this
scenario has expanded. Ablative approaches, such as radiofrequency ablation and cryoablation, have shown durable
eradication of tumors. Catheter-directed therapies, such as hepatic artery embolization, are potential techniques for
treating the patient who has multiple unresectable liver metastases. Understanding the timing and role of these three
different modalities in the multidisciplinary approach to metastatic soft tissue sarcoma is critical to provide better care and
to personalize the approach to the single patient.
he primary approach to metastatic soft tissue sarcoma

may vary broadly according to the extent of the disease
and the number of metastatic sites, the interval between the
primary tumor and the development of distant metastases,
the histologic subtype of the primary tumor, the site of
metastasis, or the patient condition (including comorbidities) and will. In general, the likelihood of long-term disease
control (. 5 years) and possibly of cure for metastatic soft
tissue sarcoma is in the 5% range,1 although selected subgroups of patients who have been successfully treated by
local ablative therapies may live longer than 5 years and may
possibly be cured.2-7
In patients affected by oligometastatic disease, surgical
resection of all metastatic sites is in fact considered the
primary treatment because complete remission is critical for
cure.8,9 Radiation therapy and interventional radiology instead have historically been used predominantly for palliation and are not considered in this setting, whereas
systemic chemotherapy is an option that may complement
the use of surgery in selected subtypes or when the interval
between the resection of primary tumor and the development of distant metastases is short (, 1 year). Recent
advances in radiation techniques may change the role of
radiation therapy in the metastatic setting in the near

future.
In patients affected by widespread metastatic disease,
anthracycline-based systemic chemotherapy is considered the standard front-line treatment, although the
response rate is only in the 25% range and the median
overall survival (OS) is in the 1-year range.10 Surgery is
reserved for patients who have obtained major disease
responses, whereas other locoregional modalities and, at
times, surgery itself are used for palliation whenever it is
clinically indicated.
The use of locoregional therapies may be repeated
at metastatic recurrence/progression, alone or in combination with systemic agents, to maximize disease
control and quality of life. The decision making is complex, depends upon diverse presentations and histologies, and should always be taken in a multidisciplinary
fashon.8,9
We detail herein the recent advances in the use of
the three main local ablative techniquessurgery, radiation
therapy, and interventional radiologyto update their
potential and to discuss their specific roles in the treatment
of metastatic soft tissue sarcoma.
From the Department of Surgery, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy; Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center,
Houston, TX; Department of Radiology, Memorial Sloan Kettering Cancer Center, New York, NY.
Corresponding author: Alessandro Gronchi, MD, Department of Surgery, Fondazione IRCCS Istituto Nazionale dei Tumori, Via Venezian 1, 20133 Milan, Italy;
email: alessandro.gronchi@istitutotumori.mi.it.
e566
MANAGING OLIGOMETASTATIC STS BY LOCOREGIONAL THERAPIES
ROLE OF SURGICAL RESECTION OF METASTATIC

DISEASE IN SOFT TISSUE SARCOMA
In general, extremity and trunk wall soft tissue sarcoma
metastasizes to the lung before spreading to any other
organ, and a significant number of patients do not develop
extrapulmonary disease. As an example, in a series of 2,003
patients treated for primary soft tissue sarcoma of the extremities or trunk wall at Fondazione IRCCS Istituto Nazionale dei Tumori in Milan, Italy, for primary disease over a
20-year time span (1993 to 2013), 441 patients developed
distant metastases. In 350 patients (79%), the lung was
the first metastatic site. Two hundred twenty-eight of the
350 patients (65%) did not develop metastases at any other
site (final analysis ongoing; manuscript in preparation).
Selected histotypes, though, may prefer other metastatic
paths that mainly involve soft tissues. In the aforementioned series, 91 patients (21%) developed extrapulmonary
metastases only. The most common histotypes in this subgroup
were myxoid liposarcoma (30 occurrences; 33%), myxofibrosarcoma (nine occurrences; 10%), leiomyosarcoma (nine
occurrences; 10%), undifferentiated pleomorphic sarcoma
(eight occurrences; 9%), epithelioid sarcoma (six occurrences;
7%), and clear cell sarcoma (five occurrences; 5%). Moreover,
liver and peritoneum may be the most common first metastatic site in patients affected by intra-abdominal gynecologic/
gastrointestinal or retroperitoneal leiomyosarcoma as well as
retroperitoneal dedifferentiated liposarcoma.
The role of surgery with curative intent in metastatic soft
tissue sarcoma has been studied predominantly in retrospective series of patients affected by lung metastases
only2-7 and in a few small, retrospective series of patients
affected by liver metastases only.11-15 Nonetheless, surgery
KEY POINTS
The initial metastatic spread of soft tissue sarcoma is

often confined to a single organ (predominantly the
lung, but also the liver and soft tissues), making the use
of locoregional therapies attractive.
Surgery may be the treatment of choice or may
complement systemic therapy in patients affected by
isolated oligometastatic disease with favorable
prognostic factors (long disease-free interval, limited
disease, site of metastasis).
Radiation therapy can be an effective palliative
intervention in metastatic sarcoma. More research is
warranted into whether highly conformal, ablative
radiation therapy can prolong disease-specific survival
in select patients with metastatic sarcoma.
Interventional radiology therapies like ablation play a
role to control oligometastatic disease in patients with
sarcoma who cannot tolerate surgery.
Liver-directed interventional radiology therapies, like
hepatic artery embolization, chemoembolization, and
radioembolization, may prolong life in patients with
liver-dominant metastatic disease.
also is used on an individualized basis in patients affected by

extrapulmonary metastases, as long as the disease extent is
limited. A disease-free interval longer than 1 year and/or
response to chemotherapy, disease control at the primary
site (i.e., limited risk of locoregional recurrence), and limited
disease are known to predict a better outcome in patients
selected for surgery.2-7,11-15 Nevertheless, decisions are
made on a case-by-case basis, and surgery may well be
used in less favorable presentations. Also, surgery may be
considered for palliation. This is typically the case in bone
metastases for pain control and function preservation,
especially if the metastases lead to a risk of cord compression or fracture.16
Isolated Pulmonary Metastases

Patients who present with isolated pulmonary metastases
from soft tissue sarcoma should have the disease properly
staged to rule out any other possible disease site. Pulmonary
metastasectomy is a well-established component of the
management of metastatic soft tissue sarcoma, although a
formal demonstration of its benefit is lacking. The surgical
approach may be videothoracoscopy or thoracotomy, and
surgery may be through staged lateral thoracotomies
or bilateral through an anterior approach. Pulmonary resections are usually individualized, depending on the location, size, and number of metastases, with the intent to
spare as much lung parenchyma as possible. This approach is
associated to 5-year OS rates ranging from 15% to 20%
(Table 1). Several prognostic factorssuch as sex and age of
the patient, site, malignancy grade and histology subtype of
the primary tumor, disease-free interval (defined as time
between surgery of the primary tumor and appearance of
distant metastases), number of nodules, bilateral lung involvement, completeness of surgical resection, and repeated metastasectomieshave been investigated over the
years. Only disease-free interval and completeness of surgical resection have been consistently found to affect
prognosis (Table 1). Therefore, metachronous (disease-free
interval $ 1 year) resectable lung metastases without
extrapulmonary disease are generally managed with surgery, if complete excision of all lesions is feasible (Fig. 1A), as
an acceptable standard treatment.2-7
Nonetheless, the 5-year OS after pulmonary metastasectomy is not necessarily attributable to the surgical
procedure. Whether, in fact, this is a result of a careful
selection of better-suited patients or it is a direct effect of
the surgical procedure is far from being demonstrated. In
theory, whether pulmonary metastasectomy is effective in
prolonging life or not would require a proof that it is associated with an extension in survival beyond that which
occurs without pulmonary metastasectomy. This proof is
lacking completely, so we rely on retrospective evidence
provided by both prospectively maintained institutional
databases and the very few observational studies. These
studies suggest that long-term disease control in metastatic
soft tissue sarcoma can be achieved only in patients who
obtain complete remission.
e567
TABLE 1. Main Reported Series About Pulmonary Metastasectomy

Study
Period
No. of
Patients
Median DFI
(months)
5-Year OS
(%)
5-Year RFS
(%)
van Geel et al
1993
255
26
38
35
Billingsley
et al3
1992
138
12
14
Rehders et al4
19912002
61
21
25
N/A
Smith et al5
19762000
94
25
18
Blackmon
et al6
19982006
234*
N/A
N/A
N/A
DFI, $ 3 repeated metastasectomies
Schur et al7
20032013
46
12
32
N/A
DFI, repeated metastasectomies
Author
2
Prognostic Factors
DFI, completeness of surgical resection, malignancy
grade, age
DFI, completeness of surgical resection
Repeated metastasectomies, no predisposing cancer
condition
*Including bone sarcomas.

Abbreviations, DFI, disease-free interval; N/A, not available; OS, overall survival; RFS, relapse-free survival.
An additional level of complexity is that soft tissue sarcoma

is a family of more than 50 different histologic subtypes with
different natural histories. Surgery of pulmonary metastases
may well be indicated in some and not in others or may
simply have a different impact among the subtypes. The task
of studying this question by means of prospective trials
is challenging, especially when subgroups must be considered. Retrospective and observational clinical studies in
selected patient subgroups are as important as prospective
controlled studies because they can allow us to gain vital

information on the natural history and clinical characteristics
of different sarcoma subtypes. This is why, despite all of the
uncertainties mentioned above, surgery of resectable isolated lung metastases with a disease-free interval longer
than 1 year is considered a standard treatment option.8,9
Moreover, the 5-year OS does not equate with a cure;
several patients alive at 5 years have undergone repeated metastasectomies. This approach, when feasible, is
FIGURE 1. CT and MRI Scans
(A) Axial view CT scan showing a single right lung metastasis detected 2 years after the resection of a monophasic synovial sarcoma of the right foot in a 21-year-old man.
(B) Portal phase of a contrast-enhanced CT scan, axial view, showing a single right hepatic metastasis detected 18 months after the resection of a leiomyosarcoma of the left
retroperitoneum in a 56-year-old woman. (C) Portal phase of a contrast-enhanced CT scan, axial view, showing a large single intra-abdominal metastasis detected 3 years
after the resection of a myxoid liposarcoma of the leg in a 31-year-old man. (D) Contrast-enhanced T1-weighted MRI of the root of the right thigh showing a soft tissue
metastases detected 2 years after the resection of a myxofibrosarcoma of the knee in a 55-year-old woman.
e568
associated with better survival and quality of life. Reported

series suggest the number of redo thoracotomies and the
interval between them is associated with the outcome. That
is, only few of the long-term survivors have undergone a
single surgical procedure and can be possibly considered
cured.6,17 Criteria for a repeat surgery are rather similar to
criteria for the initial surgery; disease-free interval and resectability of all visible macroscopic disease are the two most
important factors.
Chemotherapy may be added to surgery as an option,
taking into account adverse prognostic factors (e.g., a short
previous disease-free interval and a high number of lesions
are adverse factors, which encourage the addition of chemotherapy), although there is a complete lack of formal
evidence that this approach improves outcome.18 Chemotherapy is preferably given before surgery to assess tumor
response and, thus, to modulate treatment. Response to
chemotherapy may encourage a surgical resection, even in
patients whose initial prognostic factors were less favorable,
whereas tumor progression during chemotherapy should
definitely discourage the use of salvage surgery.18,19 Several
drugs are now available and, other than the anthracyclinebased first-line regimens, are usually tailored to the different
histotypes.20
In cases for which lung metastases are synchronous in the
absence of extrapulmonary disease, standard treatment is
chemotherapy or best supportive care. Surgery of completely resectable residual lung metastases along with
surgery of the primary tumor may be offered as an option,
especially when a tumor response is achieved.8
(GIST). The treatment of this disease has been revolutionized

during the past 15 years by the introduction of tyrosine
kinase inhibitors21 and goes beyond the scope of the present
review. GIST is worth mentioning, though, because many of
the few reported series of hepatic metastasectomy in sarcoma do include GIST, which confounds the interpretation of
the results.22 Long-term outcome (i.e., 5-year OS) figures
should be calculated separately for GIST and other sarcomas,
because a similar medical therapy for non-GIST sarcoma is
lacking.
Other than GIST, the liver can be the first metastatic site of
intra-abdominal /retroperitoneal leiomyosarcoma and few
other sarcoma subtypes that originate outside of the abdominal cavity. Surgery can play a role when the disease is
confined to the liver (Fig. 1B). The few reported series show
results similar to those of lung metastasectomy, with a
5-year OS in the 20% range (Table 2).11-15 Disease-free interval and completeness of surgical resection are the main
prognostic factors. Surgery should be aimed at obtaining a
macroscopic clearance of the tumor, sparing as much liver
parenchyma as possible. To do so in selected patients,
surgery also can be combined with intraoperative radiofrequency ablation, although better results seem to be
obtained when the metastatic tumor is cleared by surgery
alone. Whether the better results associated with liver
metastasectomy compared with those associated with
radiofrequency ablation or medical therapies alone are
attributable to the surgical procedure or simply reflect
a selection of the patients with better prognoses is far
from being demonstrated. Moreover, repeated liver metastasectomy may be occasionally considered.
Chemotherapy is often associated with the surgical procedures. In the study by Brudvik et al it is intriguing to see the
reported difference between post-liver metastasectomy OS
and relapse-free survival in leiomyosarcoma (Table 2).15 The
postmetastasectomy outcome may well be attributed to
the availability of several potentially active drugs in advanced leiomyosarcoma, such as anthracyclines, dacarbazine,
gemcitabine, trabectedin, and, more recently, pazopanib,20
rather than to the surgical procedure. The relative contribution of the different treatments, including surgery, to
a prolonged survival in the metastatic setting is then difficult
to understand. Considerations similar to the one made
for lung metastasectomy apply here. Nonetheless, the more
Isolated Extrapulmonary Metastases

Although most soft tissue sarcoma metastasizes to the lung
before spreading to other organs, myxoid liposarcoma,
myxofibrosarcoma, leiomyosarcoma, and few other rarer
subtypes may skip the lung and spread initially to soft tissues, liver, lymph nodes, bone oroccasionallybrain
(i.e., alveolar soft part sarcoma). When these metastases are
isolated, principles similar to isolated lung metastases do
apply.
Liver metastases. The commonest sarcoma subtype that
metastasizes to the liver is gastrointestinal stromal tumor
TABLE 2. Main Reported Series About Liver Metastasectomy

Author
Study Period No. of Patients Median DFI (months) 5-Year OS (%) 5-Year RFS (%) Prognostic Factors
DeMatteo et al11 19822000
56*
38
30
20
Pawlik et al12
66*
N/A
27
N/A
Completeness of surgical resection
13
19962005
Adam et al
19832004
125
38
30
N/A
DFI, completeness of surgical resection, age
Groeschl et al14
19902009
98
43
32
N/A
None
Brudvik et al15
19982013
N/A
48.4, LMS;
44.9, other
3.4, LMS;
21.4, other
46, LMS;
50, other**
*Including gastrointestinal stromal tumor.

**Other = other sarcoma.
Abbreviations: DFI, disease-free interval; LMS, leiomyosarcoma; N/A, not available; OS, overall survival; RFS, relapse-free survival.
e569
limited availability of retrospective data and the rarer

presentation of this scenario do not allow the same conclusions as those made for lung metastasectomy. Surgery of
liver metastasis in soft tissue sarcoma is not a standard
approach in any of the available guidelines,8,9 although it is
considered an option to be discussed in a multidisciplinary
setting when favorable prognostic factors are present.
Soft tissue metastases. Some histology subtypes, predominantly myxoid liposarcoma23 and myxofibrosarcoma,24
and occasionally others, may spread initially to a single
soft tissue site virtually anywhere in the body. Myxoid
liposarcoma typically prefers either the abdomen/pelvis
(Fig. 1C) or the soft tissue of the extremities/trunk wall.
Myxofibrosarcoma largely prefers the soft tissues at the
root of the limb from which the primary tumor originated
(Fig. 1D). Epithelioid sarcoma,25 clear cell sarcoma,26 and,
less frequently, angiosarcoma may initially give rise to intransit metastases and/or lymph node metastases, such
as in melanoma. Extraskeletal myxoid chondrosarcoma27
can also spread to locoregional lymph nodes/soft tissue at
the root of the affected limb.
Some of these tumors are exceedingly rare. They all have
different natural histories and biologic backgrounds. Sensitivity to medical agents broadly varies among them. The
indication to resect the isolated metastatic site should be
weighed against the possible morbidities and discussed
within the multidisciplinary approach to each disease. There
are no specific reported data about these different scenarios. Decisions are usually based on the overall experience
with the approach to metastatic disease. Surgery is discussed
when the disease-free interval is long andpossiblythe
tumor has responded to medical therapy, although it may be
considered a salvage procedure even when the tumor
progresses during medical therapy while remaining confined
to a single site.
Bone metastases. Bones can occasionally be the primary
and isolated metastatic site of a few histologic subtypes,
such as myxoid liposarcoma and solitary fibrous tumor.28
Surgery can be considered when a single long bone or the
pelvis is affected, especially when there is a risk of fracture.
Similarly, when the metastatic tumor is located in the spine,
consideration of a formal spondylectomy rather than of a
simple cord decompression should be made. Disease-free
interval and expected morbidity are the main factors that
drive the decision.16
Surgery is also considered in the setting of pure palliation
when the risk of fracture or cord compression is significant.
Few data are reported about these scenarios. The decision
for surgery is usually part of a multidisciplinary approach to
this disease.
Brain metastases. Isolated brain metastases predominantly
occur in patients affected by alveolar soft part sarcoma.29
Surgery may be considered as an option, although other
modalities such as radiation therapy can be considered
e570
because the natural history of this disease is always characterized by the development of other metastatic sites.
Surgery should be aimed at obtaining symptom relief and
improving function preservation. Similar to bone metastases, very few reported data are available about the use of
surgery in these scenarios. The decision for surgery is part
of a multidisciplinary approach to the disease.
Multiple Sites
Multiple sites of metastatic soft tissue sarcoma are treated
with chemotherapy as standard treatment.8,9 In highly selected cases, surgery of responding metastases may be
offered as an option after a multidisciplinary evaluation,
taking into consideration the metastatic sites and the natural
history of the disease in the individual patient.
THE ROLE OF RADIATION THERAPY IN

METASTATIC SOFT TISSUE SARCOMA
Consideration of the use of radiation therapy in the management of metastatic sarcoma most often arises when a
patient presents with need for palliation from a symptomatic lesion. However, with the development of highly
conformal radiotherapeutic techniques, such as intensitymodulated radiation therapy and stereotactic body
radiotherapy (SBRT), there is growing interest in whether
outcomes for patients with oligometastases can be improved upon with the use of radical doses of radiation
therapy to limited sites of metastatic involvement. Among
patients who develop sarcoma metastases, performance of
metastasectomy is associated with better survival.30 Interest
has grown about whether advances in radiation therapy
techniques that allow delivery of ablative doses of radiation
may extend this potential survival benefit to those patients
in whom surgery is not recommended. The French Sarcoma
Group31 recently published a comparative study of outcomes for 281 patients with soft tissue sarcoma and oligometastases treated with or without ablative therapy for
their metastatic lesions. Ablative therapy in this study
consisted of surgery, radiofrequency ablation, or radiation
therapy. In the analysis, patients undergoing ablative
therapy for their metastases had a significantly decreased
risk of death (adjusted hazard ratio, 0.47; 95% CI, 0.290.78).
The subgroup of patients undergoing radiation therapy as
the ablative modality was small in this retrospective study.
Most of the series examining the use of ablative doses of
radiation therapy for metastatic cancer include patients with
nonsarcoma histologies. In general, in those patients undergoing SBRT for metastatic cancer, control rates for targeted lesions are approximately 80%.32 However, for larger
lesions (e.g., . 3 cm), the local control outcomes are
poorer.33,34 Patients with longer disease-free intervals
between primary tumor presentation and metastatic
recurrence30,35,36 and those with three or fewer metastatic
lesions37 are more likely to experience a progression-free
survival benefit from the use of ablative radiation therapy
to sites of metastases. Toxicity rates from SBRT are low;
most studies report rates of grade 3 toxicities of 10% or less

regardless of anatomic site treated or doses delivered.32
Because the lung is the most common and often the first
site of metastases in patients with sarcoma, the discussion
of whether SBRT can improve cancer-control outcomes for
patients with metastatic sarcoma is likely most relevant to
patients with pulmonary oligometastases. In 2015, two
studies were published that specifically reported results of
ablative radiotherapy techniques in patients with oligometastatic sarcoma to the lung. Navarria and colleagues35
analyzed outcomes for 28 patients treated for 51 lesions
with SBRT to pulmonary metastases from soft-tissue sarcomas. Administered radiation doses were as follows: 30 Gy
in one fraction for peripheral lesions of 10 mm or smaller,
60 Gy in three fractions for peripheral lesions of 10 mm to
20 mm, 48 Gy in four fractions for peripheral lesions greater
than 20 mm, and 60 Gy in eight fractions for lesions located
centrally in the thorax. With a median follow-up time of
39 months from discovery of lung disease, both the 2- and
5-year local control rates of treated lesions were 96%. The
2- and 5-year OS rates after evidence of lung metastases
were 85% and 54%, respectively. The only factor prognostic
for survival was disease-free interval between diagnosis and
development of lung disease. No patient experienced severe
(grades 3 to 4) pulmonary toxicity. Similarly, Frakulli and
colleagues38 published their experience using SBRT to treat
68 lung metastases in 24 patients with metastatic bone
sarcoma or soft tissue sarcoma. They delivered 30 to 60 Gy in
three to eight fractions. With a median follow-up time of
17 months, the 2-year local control rate for treated lesions
was 86%, and the 2-year OS was 66%. The investigators did
not find any clinical factor to be prognostic for survival.
Larger tumor volume (. 5 mL) was associated with poorer
local control. No patient experienced grade 3 or greater
pulmonary toxicity, and one patient presented with a grade
1 rib fracture 19 months after SBRT. These results are
promising and will hopefully stimulate additional investigation through prospective clinical trials of this therapeutic
strategy specifically in patients with sarcoma. Additionally, no
published data currently address whether SBRT to liver oligometastases may also improve outcomes in patients with
limited-burden sarcoma metastases to the liver (e.g., as may
occur with some retroperitoneal sarcoma histologies, GIST).
Although advances in radiation technologies have bolstered the use of ablative radiation therapy as a new potential approach for improving progression-free survival in
metastatic sarcoma management, the most common indication for radiation therapy will be one of symptom control
and quality of life improvement for patients with symptomatic sarcoma metastases. General principles of palliative
radiotherapy apply, with goals of delivering meaningful
radiation therapy dose to achieve symptom improvement
while minimizing risk of potential toxicity. Although sarcomas are generally less responsive to radiation therapy than
some tumor types, there are no data that support dose
escalation beyond standard multifraction palliative doses
of radiation therapy for sarcomas in the palliative setting.
A recent investigation of multifraction radiation dosing using

conventional radiation techniques for spinal cord compression included patients with sarcoma as part of the study
cohort and showed response rates of 87% to 89% in patients
treated with 30 Gy in 10 fractions or 20 Gy in five fractions.39
Another randomized study that supported single fractions of
8 Gy of conventional radiation therapy compared with
multifraction regimens for bone metastases did not include
patients with sarcoma40, so whether conventional dosing
and techniques for single-fraction radiotherapy provides
durable palliation for sarcoma bone metastases remains
questionable. However, single-fraction therapy might be
desired for patient convenience or to expedite transition to
systemic management. Folkert and colleagues41 reported
that image-guided spine stereotactic radiosurgery for patients with metastatic sarcoma to the spine resulted in
excellent local control, 88%, at 18 months. Patients in this
study received a median of 24 Gy (range, 18 to 24 Gy) in one
fraction (Fig. 2), and acute and chronic toxicity (grade 3 or
greater) rates were low at 1% and 5%, respectively. Occurrence of brain metastases is a rare event among patients
with sarcoma, but data suggest that a radiotherapeutic
approach to sarcoma brain metastases should not differ
from the approaches for other tumor types. Yaeh and
colleagues42 demonstrated that local control for brain
metastases treated with stereotactic radiosurgery was
similar among patients with metastatic sarcoma and those
with typically nonradiation-resistant tumors. Similarly,
Ahmed and colleagues43 demonstrated that postoperative
FIGURE 2. Cross-sectional Image of a Radiation

Therapy Plan for Spinal Stereotactic Radiosurgery in
a Patient With Metastatic Osteosarcoma to T11
The patient received 24 Gy in one fraction to a T11 lesion (shaded red lesion/
tumor on the left side of the vertebral body). The plan was delivered by using
a 9-beam intensity-modulated stereotactic radiotherapy technique; both CT onrails and megavoltage imaging were used for set-up verification. The anterior
vertebral body is encompassed by the 16-Gy isodose (blue) line.
e571
fractionated stereotactic radiotherapy to a resection cavity

after surgery for brain metastases affected local control
rates similarly among radiation therapyresponsive and
radiation therapyresistant histologies, including sarcomas.
For patients in whom urgent whole brain radiation therapy is
indicated, standard whole brain dosing and fractionation
schedules are acceptable. The decision about the use of
radiation therapy and about aggressiveness of radiation
therapy dosing and technique in all patients with metastatic
sarcoma should be made in multidisciplinary consultation,
and consultation with palliative care specialists should be
also considered, especially in cases of advanced or symptomatic disease burden.
THE ROLE OF INTERVENTIONAL RADIOLOGY IN

METASTATIC SOFT TISSUE SARCOMA
Interventional radiologists have an expanding role in the
treatment of metastases from sarcoma. Increasing data on
the ability to treat metastases with locoregional therapies
have solidified this position. Ablative approaches, such as
radiofrequency ablation and cryoablation, have shown durable eradication of tumors. Catheter-directed therapies,
such as hepatic artery embolization, are potential techniques for treating the patient who has multiple unresectable liver metastases. Understanding the timing and role of
these techniques in the multidisciplinary care of the patient
who has sarcoma with metastases is critical. Implementation
of the interventional radiology clinic for consultation has
enabled a more robust multidisciplinary approach to integration of these image-guided techniques into the patients
overall care and has facilitated improved opportunities
for clinical studies. Given the broad range of techniques
available to the interventional radiologist, it will become
important for these techniques to be investigated in clinical
trials to optimally integrate interventional radiology into
the multidisciplinary care of the patient who has sarcoma
with metastases.
Ablation of Sarcoma Metastasis

Patients undergoing liver or lung resection of their sarcoma
metastases have prolonged survival.3,44 Ablation technologies cause local destruction of malignant tissue without
open resection using minimally invasive percutaneous
techniques. When coupled with imaging guidance, this
targeted destruction can be aimed at a particular metastasis.
Similar to surgical resection, the goal of the ablation is to
create a margin of destruction around the targeted tumor to
prevent recurrence. The main advantage of ablation techniques compared with resection is that they are less invasive
than surgical resection, so they provide a lower-morbidity
treatment option. Patients can generally be treated as
outpatients and have rapid recovery to normal activities.
Although chemotherapy routines are frequently interrupted
for 6 weeks by surgical resection, this same requirement is
not present with percutaneous techniques.45 Prospective
randomized controlled studies of image-guided ablation
e572
against the gold standard of surgical resection are difficult

to perform because of the heterogeneity of sarcoma as a
disease and the institutional preferences in locoregional
treatment.
There are a number of ablative tools available, including
radiofrequency ablation (Fig. 3), microwave ablation, cryoablation, laser ablation, and focused ultrasound ablation.
All of these modalities rely on extreme temperature conditions of heat or cold to destroy malignant tissue.46 A new
nonthermal technique called irreversible electroporation
uses electric fields to cause cell death without harming the
tissue protein architecture that makes up structures such as
bile ducts and blood vessels. Irreversible electroporation
may open up new ablative opportunities near critical structures that were previously risky when thermal ablation tools
that could potentially damage these critical structures were
used.47 More research is needed to better understand this
modality. The technical differences among all of these techniques is beyond the scope of this review, but radiofrequency
ablation has been the most commonly used technique with
the most extensive literature for metastases of the liver and
lung.
Just as surgical metastasectomy is associated to improved
survival in patients with sarcoma metastases, the addition of
percutaneous ablation or stereotactic radiotherapy to surgical
locoregional control strategies is associated to improved
survival. The French Sarcoma Group recently retrospectively
reviewed patients who underwent local regional therapy
versus no locoregional therapy for sarcoma with one to five
hepatic, pulmonary, or other sites of metastasis.31 Locoregional therapies included surgery, radiofrequency ablation, and stereotactic radiotherapy. The median OS was
45.3 months in the locoregional therapy group, and it was
12.6 months in the group without locoregional therapy. For
nonsurgical candidates with pulmonary metastasis from
sarcoma, Gillams reported a mean survival of 51 months
(median not reached) after radiofrequency ablation and a
median progression-free survival of 12 months.
The test-of-time approach by using percutaneous ablation
was proposed by Livraghi et al48 as an approach to maintain
quality of life in patients with liver metastases. The concept
is that many patients may be able to avoid unnecessary
surgery by undergoing ablation of metastases in the interval
from the time of diagnosis of metastases to the time of
metastasectomy. The theory is that radiofrequency ablation
can completely treat many metastases with small local recurrence rates. The patients without recurrence after ablation will have avoided the potentially higher risks and
morbidity of resection. The patients with development of
innumerable metastases after ablation would also be spared
unnecessary surgery that would not have been beneficial
because the original metastasis was just the tip of the
iceberg. Livraghi et al48 tested this theory by treating
88 patients who had surgically resectable liver metastasis with radiofrequency ablation instead of surgery; he
then observed the patients to see if they ultimately
required surgery. Sixty percent of the patients treated
FIGURE 3. Imaging of Uterine Leiomyosarcoma in a 50-Year-Old Woman
Two years after resection, the patient developed a solitary pulmonary metastasis, which was treated with radiofrequency ablation. A ground-glass halo is seen around the
lesion 10 minutes following radiofrequency ablation, indicating a compete margin. Over the next 18 months, the ablation site contracted and formed a stable scar.
with radiofrequency ablation never had surgery (25% were

free of disease after radiofrequency ablation, and 35% developed new extensive metastatic disease that precluded
surgery). Therefore, the test of time approach by Livraghi
et al48 would allow both the patient with a successful ablation and the patient for whom surgery would not have
helped because of an early explosion of metastases to
benefit by avoiding the morbidity of resection.
Arterial Therapies for Sarcoma Liver Metastases

Image-guided intra-arterial therapies rely on the fact that
liver cancers predominantly derive their blood supply from
hepatic arteries, whereas normal liver parenchyma predominantly has a portal vein source of blood supply.49
Hepatic artery embolization refers to the delivery of the
microparticles to liver arteries to kill tumors by depriving
them of oxygen and glucose (bland embolization). Conventional transarterial chemoembolization delivers a slurry
of chemotherapy (usually doxorubicin), ethiodized oil, and
particles (gelfoam or spherical embolics) to the hepatic
artery. Drug-eluting bead transarterial chemoembolization

has been developed to allow chemotherapy release after
the bead has been embolized into the tumor microcirculation. Radioembolization relies on intra-arterial delivery
of 90Y permanently bound to microspheres, enabling selective intra-arterial delivery of doses greater than 120 Gy to the
targeted tumor without reaching the liver toxicity threshold.
Arterial therapies for sarcoma liver metastases can be performed to complement or salvage the effects of systemic
therapy.
Arterial therapies are indicated in patients with unresectable sarcoma who have large-volume liver metastatic
disease with little or no extrahepatic disease. This is a fairly
uncommon clinical scenario, so there are few papers on
arterial therapies for sarcoma, and all are retrospective
studies. Patients with sarcoma metastases to the liver who
underwent hepatic artery embolization with bland embolic
particles showed a median OS of 24 months in a series of
24 patients.50 In patients with GIST who experienced progression despite imatinib therapy, the patients treated with
e573
hepatic artery embolization as second-line therapy had a

median OS of 14.9 months, and those treated with hepatic
artery embolization as third-line therapy had a median OS
of 23.8 months.51 In another study, a regimen of cisplatin,
doxorubicin, and mitomycin in transarterial chemoembolization was used to treat 32 patients with sarcoma
metastatic to the liver; the median OS was 21.2 months, and
the progression-free survival was 6.3 months.52 No studies
review drug-eluting bead chemoembolization or radioembolization in patients with sarcoma metastasis. However,
insight can be gained from a randomized study by Van Hazel
et al53 that compared radioembolization plus systemic
chemotherapy with chemotherapy alone as first-line therapy for colorectal liver metastases. The authors found that
the addition of radioembolization to systemic therapy
improved the median survival compared with systemic

therapy alone (29.4 months compared with 11.8 months).
CONCLUSION
In patients affected by metastatic soft tissue sarcoma,
complete remission is critical for long-term disease control
andpossiblyfor cure. Surgery still remains the preferred
modality in oligometastatic disease, especially if isolated to a
single organ, but emerging techniques both in the radiation
therapy and interventional radiology fields may expand their
uses. Prospectively maintained databases and observational
studies will be instrumental to study the specific role of
these three modalities in the year to come and to help
expand their roles in the difficult multidisciplinary approach
to metastatic disease.
References
1. Blay JY, van Glabbeke M, Verweij J, et al. Advanced soft-tissue sarcoma:
a disease that is potentially curable for a subset of patients treated with
chemotherapy. Eur J Cancer. 2003;39:64-69.
2. van Geel AN, Pastorino U, Jauch KW, et al. Surgical treatment of lung
metastases: the European Organization for Research and Treatment of
Cancer Soft Tissue and Bone Sarcoma Group study of 255 patients.
Cancer. 1996;77:675-682.
3. Billingsley KG, Burt ME, Jara E, et al. Pulmonary metastases from soft
tissue sarcoma: analysis of patterns of diseases and postmetastasis
survival. Ann Surg. 1999;229:602-610.
4. Rehders A, Hosch SB, Scheunemann P, et al. Benefit of surgical
treatment of lung metastasis in soft tissue sarcoma. Arch Surg. 2007;
142:70-75, 76.
5. Smith R, Pak Y, Kraybill W, et al. Factors associated with actual long-term
survival following soft tissue sarcoma pulmonary metastasectomy. Eur
J Surg Oncol. 2009;35:356-361.
6. Blackmon SH, Shah N, Roth JA, et al. Resection of pulmonary and
extrapulmonary sarcomatous metastases is associated with long-term
survival. Ann Thorac Surg. 2009;88:877-884.
7. Schur S, Hoetzenecker K, Lamm W, et al. Pulmonary metastasectomy for
soft tissue sarcoma: report from a dual institution experience at the
Medical University of Vienna. Eur J Cancer. 2014;50:2289-2297.
8. ESMO/European Sarcoma Network Working Group. Soft tissue and
visceral sarcomas: ESMO clinical practice guidelines for diagnosis,
treatment and follow-up. Ann Oncol. 2014;25:iii102-iii112 (suppl 3).
9. von Mehren M, Randall RL, Benjamin RS, et al; National Comprehensive
Cancer Network. Soft tissue sarcoma, version 2.2014. J Natl Compr
Canc Netw. 2014;12:473-483.
10. Judson I, Verweij J, Gelderblom H, et al; European Organisation and
Treatment of Cancer Soft Tissue and Bone Sarcoma Group. Doxorubicin
alone versus intensified doxorubicin plus ifosfamide for first-line
treatment of advanced or metastatic soft-tissue sarcoma: a randomised controlled phase 3 trial. Lancet Oncol. 2014;15:415-423.
11. DeMatteo RP, Shah A, Fong Y, et al. Results of hepatic resection for
sarcoma metastatic to liver. Ann Surg. 2001;234:540-547.
12. Pawlik TM, Vauthey JN, Abdalla EK, et al. Results of a single-center
experience with resection and ablation for sarcoma metastatic to the
liver. Arch Surg. 2006;141:537-543, discussion 543-544.
13. Adam R, Chiche L, Aloia T, et al; Association Franaise de Chirurgie.
Hepatic resection for noncolorectal nonendocrine liver metastases:
e574
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
analysis of 1,452 patients and development of a prognostic model. Ann

Surg. 2006;244:524-535.
Groeschl RT, Nachmany I, Steel JL, et al. Hepatectomy for noncolorectal
non-neuroendocrine metastatic cancer: a multi-institutional analysis.
J Am Coll Surg. 2012;214:769-777.
Brudvik KW, Patel SH, Roland CL, et al. Survival after resection of
gastrointestinal stromal tumor and sarcoma liver metastases in
146 patients. J Gastrointest Surg. 2015;19:1476-1483.
Fehlings MG, Nater A, Tetreault L, et al. Survival and clinical outcomes in
surgically treated patients with metastatic epidural spinal cord compression: results of the prospective multicenter AOSpine study. J Clin
Oncol. 2016;34:268-276.
Weiser MR, Downey RJ, Leung DH, et al. Repeat resection of pulmonary
metastases in patients with soft-tissue sarcoma. J Am Coll Surg. 2000;
191:184-190, discussion 190-191.
Canter RJ, Qin LX, Downey RJ, et al. Perioperative chemotherapy in patients
undergoing pulmonary resection for metastatic soft tissue sarcoma of the
extremity: a retrospective analysis. Cancer. 2007;110:2050-2060.
Stephens EH, Blackmon SH, Correa AM, et al. Progression after chemotherapy is a novel predictor of poor outcomes after pulmonary
metastasectomy in sarcoma patients. J Am Coll Surg. 2011;212:821-826.
Radaelli S, Stacchiotti S, Casali PG, et al. Emerging therapies for adult
soft tissue sarcoma. Expert Rev Anticancer Ther. 2014;14:689-704.
Verweij J, Casali PG, Zalcberg J, et al. Progression-free survival in
gastrointestinal stromal tumours with high-dose imatinib: randomised
trial. Lancet. 2004;364:1127-1134.
Page AJ, Weiss MJ, Pawlik TM. Surgical management of noncolorectal
cancer liver metastases. Cancer. 2014;120:3111-3121.
Fiore M, Grosso F, Lo Vullo S, et al. Myxoid/round cell and pleomorphic
liposarcomas: prognostic factors and survival in a series of patients
treated at a single institution. Cancer. 2007;109:2522-2531.
Sanfilippo R, Miceli R, Grosso F, et al. Myxofibrosarcoma: prognostic
factors and survival in a series of patients treated at a single institution.
Ann Surg Oncol. 2011;18:720-725.
Baratti D, Pennacchioli E, Casali PG, et al. Epithelioid sarcoma: prognostic factors and survival in a series of patients treated at a single
institution. Ann Surg Oncol. 2007;14:3542-3551.
Bianchi G, Charoenlap C, Cocchi S, et al. Clear cell sarcoma of soft tissue:
a retrospective review and analysis of 31 cases treated at Istituto
Ortopedico Rizzoli. Eur J Surg Oncol. 2014;40:505-510.
27. Paioli A, Gronchi A, Palmerini A, et al. Extraskeletal myxoid chondrosarcoma: A retrospective analysis of 69 patients with localized
disease and molecularly confirmed diagnosis. J Clin Oncol. 2014; 32:5s
28. Baldi GG, Stacchiotti S, Mauro V, et al. Solitary fibrous tumor of all
sites: outcome of late recurrences in 14 patients. Clin Sarcoma Res.
2013;3:4.
29. Pennacchioli E, Fiore M, Collini P, et al. Alveolar soft part sarcoma:
clinical presentation, treatment, and outcome in a series of 33 patients
at a single institution. Ann Surg Oncol. 2010;17:3229-3233.
30. Kang S, Kim H-S, Kim S, et al. Post-metastasis survival in extremity soft
tissue sarcoma: a recursive partitioning analysis of prognostic factors.
Eur J Cancer. 2014;50:1649-1656.
31. Falk AT, Moureau-Zabotto L, Ouali M, et al; Groupe Sarcome FrancaisGroupe Detude Des Tumeurs Osseuses. Effect on survival of local
ablative treatment of metastases from sarcomas: a study of the French
sarcoma group. Clin Oncol (R Coll Radiol). 2015;27:48-55.
32. Tree AC, Khoo VS, Eeles RA, et al. Stereotactic body radiotherapy for
oligometastases. Lancet Oncol. 2013;14:e28-e37.
33. Rusthoven KE, Kavanagh BD, Burri SH, et al. Multi-institutional phase
I/II trial of stereotactic body radiation therapy for lung metastases.
J Clin Oncol. 2009;27:1579-1584.
34. Milano MT, Katz AW, Zhang H, et al. Oligometastases treated with
stereotactic body radiotherapy: long-term follow-up of prospective
study. Int J Radiat Oncol Biol Phys. 2012;83:878-886.
35. Navarria P, Ascolese AM, Cozzi L, et al. Stereotactic body radiation
therapy for lung metastases from soft tissue sarcoma. Eur J Cancer.
2015;51:668-674.
36. Inoue T, Katoh N, Onimaru R, et al. Clinical outcomes of stereotactic
body radiotherapy for patients with lung tumors in the state of oligorecurrence. Pulm Med. 2012;2012:369820.
37. Salama JK, Hasselle MD, Chmura SJ, et al. Stereotactic body radiotherapy for multisite extracranial oligometastases: final report of a dose
escalation trial in patients with 1 to 5 sites of metastatic disease. Cancer.
2012;118:2962-2970.
38. Frakulli R, Salvi F, Balestrini D, et al. Stereotactic radiotherapy in the
treatment of lung metastases from bone and soft tissue sarcomas.
Anticancer Res. 2015;35:5581-5586.
39. Rades D, S egedin B, Conde-Moreno AJ, et al. Radiotherapy with 4 Gy 3 5
versus 3 Gy 3 10 for metastatic epidural spinal cord compression: final
results of the SCORE-2 trial (ARO 2009/01). J Clin Oncol. 2016;34:
597-602.
40. Hartsell WF, Scott CB, Bruner DW, et al. Randomized trial of shortversus long-course radiotherapy for palliation of painful bone metastases. J Natl Cancer Inst. 2005;97:798-804.
41. Folkert MR, Bilsky MH, Tom AK, et al. Outcomes and toxicity for
hypofractionated and single-fraction image-guided stereotactic radiosurgery for sarcomas metastasizing to the spine. Int J Radiat Oncol
Biol Phys. 2014;88:1085-1091.
42. Yaeh A, Nanda T, Jani A, et al. Control of brain metastases from radioresistant tumors treated by stereotactic radiosurgery. J Neurooncol. 2015;
124:507-514.
43. Ahmed KA, Freilich JM, Abuodeh Y, et al. Fractionated stereotactic
radiotherapy to the post-operative cavity for radioresistant and radiosensitive brain metastases. J Neurooncol. 2014;118:179-186.
44. Jaques DP, Coit DG, Casper ES, et al. Hepatic metastases from soft
tissue sarcoma. Ann Surg. 1995;221:392-397.
45. Erinjeri JP, Fong AJ, Kemeny NE, et al. Timing of administration of
bevacizumab chemotherapy affects wound healing after chest wall port
placement. Cancer. 2011;117:1296-1301.
46. Chu KF, Dupuy DE. Thermal ablation of tumours: biological mechanisms and advances in therapy. Nat Rev Cancer. 2014;14:199-208.
47. Scheffer HJ, Nielsen K, de Jong MC, et al. Irreversible electroporation
for nonthermal tumor ablation in the clinical setting: a systematic
review of safety and efficacy. J Vasc Interv Radiol. 2014;25:997-1011,
quiz 1011.
48. Livraghi T, Solbiati L, Meloni F, et al. Percutaneous radiofrequency
ablation of liver metastases in potential candidates for resection: the
test-of-time approach. Cancer. 2003;97:3027-3035.
49. de Baere T, Deschamps F. Arterial therapies of colorectal cancer metastases to the liver. Abdom Imaging. 2011;36:661-670.
50. Maluccio MA, Covey AM, Schubert J, et al. Treatment of metastatic
sarcoma to the liver with bland embolization. Cancer. 2006;107:
1617-1623.
51. Takaki H, Litchman T, Covey A, et al. Hepatic artery embolization for liver
metastasis of gastrointestinal stromal tumor following imatinib and
sunitinib therapy. J Gastrointest Cancer. 2014;45:494-499.
52. Chapiro J, Duran R, Lin M, et al. Transarterial chemoembolization in softtissue sarcoma metastases to the liver: the use of imaging biomarkers as
predictors of patient survival. Eur J Radiol. 2015;84:424-430.
53. Van hazel G, Blackwell A, Anderson J, et al. Randomised phase 2 trial
of SIR-Spheres plus fluorouracil/leucovorin chemotherapy versus
fluorouracil/leucovorin chemotherapy alone in advanced colorectal
cancer. J Surg Oncol. 2004;88:78-85.
e575
TUMOR BIOLOGY
Expanding the Clinically

Actionable Cancer Genome
CHAIR
Elizabeth M. Swisher, MD
University of Washington School of Medicine
Seattle, WA
SPEAKERS
Alan D. DAndrea, MD
Boston, MA
Johann S. De Bono, MBChB, MSc, FRCP, PhD, FMedSci
The Royal Marsden NHS Foundation Trust and The Institute of Cancer Research
INTERROGATING THE CANCER GENOME
Interrogating the Cancer Genome to Deliver More Precise

Cancer Care
Joaquin Mateo, MD, and Johann S. de Bono, MBChB, MSc, FRCP, PhD, FMedSci
OVERVIEW
The aim of precision medicine is to select the best treatment option for each patient at the appropriate time in the natural
history of the disease, based on understanding the molecular makeup of the tumor, with the ultimate objective of improving
patient survival and quality of life. To achieve it, we must identify functionally distinct subtypes of cancers and, critically,
have multiple therapy options available to match to these functional subtypes. As a result of the development of better and
less costly next-generation sequencing assays, we can now interrogate the cancer genome, enabling us to use the DNA
sequence itself for biomarker studies in drug development. The success of DNA-based biomarkers requires analytical
validation and careful clinical qualification in prospective clinical trials. In this article, we review some of the challenges the
scientific community is facing as a consequence of this sequencing revolution: reclassifying cancers based on biologic/
phenotypic clusters relevant to clinical decision making; adapting how we conduct clinical trials; and adjusting our
frameworks for regulatory approvals of biomarker technologies and drugs. Ultimately, we must ensure that this revolution
can be safely implemented into routine clinical practice and benefit patients.
ancer is a result of the accumulation of numerous events

in the genome of a cell. These damaging events originate
from both endogenous and exogenous causes; although a
normal cell will continuously be exposed to mutagens, only
some genomic events will confer a survival and proliferative
advantage to tumor cells in their microenvironment and
an uncontrolled, expansive phenotype. Indeed, cancer is
reported to be the consequence of a stochastic process,
Darwinian-like with regard to clonal evolution that leads to
the provision of advantages to the cancer cell. Sequencing
technologies allow us to identify these genomic aberrations
in order to apply our understanding of tumor biology to
support the development of better, more precise, therapeutic anticancer strategies.1
WHAT THE GENOME CAN AND CANNOT TELL US

Almost 4 decades ago, Frederick Sanger reported on a
capillary-based sequencing assay capable of interrogating
the DNA sequence.2 Although this approach is still used
today, the advent of massive parallel sequence is what has
dramatically transformed our understanding of the cancer
genome, permitting the incorporation of genomic tests into
clinical decision making and drug development in oncology.
Briefly, these next-generation sequencing (NGS) assays allow for millions of DNA molecules to be captured at the
same time from tumor samples, supporting the reading in

parallel of the sequence of billions of base pairs in a single
run.3,4
Reading the DNA of the cancer genome can potentially
lead to identifying millions of sequence variants (understanding normal and variant as epidemiologic rather
than functional concepts). We can group these variants in
DNA sequence as:
Mutations: These are substitutions of one or more base
pairs by an alternative sequence, which could lead to
changes in the amino acid chain. Sometimes, small numbers of base pairs can be inserted or deleted, resulting
in the addition or loss of codons or inducing a change
in the reading pattern (frameshift mutations). These
mutations can activate a kinase or transcription factor
or inactivate a tumor suppression gene.5-8 Mutations
leading to cancer promotion occur primarily in the
somatic (tumor) DNA, although inherited mutations
in the germline DNA have been shown to increase
cancer risk, affecting prognosis or sensitivity to specific therapies.9,10
Copy number (CN) variations: These are the emergence
of focal deletions or amplifications of DNA sequence,
involving one or more genes. Two main types of CN
aberrations can be found in the somatic cancer genome:
From The Institute of Cancer Research, Sutton, United Kingdom; The Royal Marsden NHS Foundation Trust, Sutton, United Kingdom.
Corresponding author: Johann S. de Bono, MB ChB, MSc, FRCP, PhD, FMedSci, Drug Development Unit, The Royal Marsden NHS Foundation Trust, Downs Rd., Sutton, Surrey SM2 5PT,
United Kingdom; email: johann.de-bono@icr.ac.uk.
e577
MATEO AND DE BONO
those occurring close to a centromere, usually

affecting a small region and including one or several
genes, and those occurring closer to telomeres, usually
involving bigger fragments of DNA. It is now clear that
somatic CN aberrations in cancer tend to occur in
particular regions of the genome, giving an advantage
to the cancer cell because of the presence in these
regions of oncogenes and tumor suppressor genes.11,12
Rearrangements/Translocations: These are breakages
in the DNA chain with aberrant rejoining in a different
area of the same or another chromosome. As a result,
two otherwise separate genes end up fused. In solid
tumors, the 59 end of the resulting fusion is usually a
tumor type-specific gene, while the 39 end of the recurrent fusions observed is commonly a kinase or a
transcription factor. This is the case of the ETS family
fusions observed in up to half of advanced prostate
cancer cases13,14 and the clinically relevant EML4-ALK
fusion, which is present in lung cancer and associates
with sensitivity to crizotinib (but has also been described
in breast and colorectal carcinomas).15-17
Furthermore, the complexity of the cancer genome extends
beyond the DNA level. We now know that aberrations occur
at the RNA or protein levels without impacting the DNA of
the gene encoding for that particular protein but still altering
functioneither through DNA methylation, transcription
regulation, or other mechanisms.18,19 For example, long
noncoding RNA structures are reported to regulate different
processes in cancerous cells similarly to how miRNAs do.20,21
UNDERSTANDING CLINICAL RELEVANCE

The vast majority of genomic variants do not have a large
effect on the normal transcription of genes or on the function
of the associated proteins. Even when a genomic aberration
KEY POINTS
e578
Interrogating the cancer genome can advance precision

medicine for cancer care.
Although a cancer genome can harbor a million variants,
most will not have any functional impact and very few
(< 20) will be true drivers of cancer growth; it is critical to
identify them and their impact in the disease course.
Clinical trial design in the era of personalized medicine
should incorporate innovative designs to evaluate drugs
and biomarkers to maximize the chances of patients
deriving benefit.
Data sharing, standardization of clinical qualification
processes, simplification of -omics data analyses, and
education of patients and physicians in interpreting
genomic data remain key challenges.
Whole-genome studies remain difficult to implement
routinely in clinical practice until better bioinformatic
tools are available; targeted sequencing approaches
represent a short-term solution.
affects protein function, not all aberrations will contribute

equally to cancer cell survival and progression. Although a few
genomic switches will be true drivers of cancer progression,
the unstable cancer genome can lead to the emergence of
many other genomic aberrations that are not critical for cancer
cell survival. We refer to these as passenger mutations.
Probably no more than five to 20 driver events are necessary for carcinogenesis in many solid tumors, conferring a
growth advantage to these cancer cells.22,23 Identifying true
drivers of cancer survival and progression is critical for
planning targeted treatment in cancer medicine, as targeting
noncritical genomic events may not result in cancer cell
death. Still, the concept of driver aberrations may be dynamic during the course of the disease, as nondriver events
may become critical if they lead to resistance to treatment.
Once a genetic variant has been identified, and an analytically validated assay is developed to detect it, clinical qualification is required to determine if the biomarker will
distinguish clinically relevant subtypes of disease.24,25 Some
genomic aberrations inform on prognosis; some are relevant
to stratifying patient risk and, for example, can help inform on
the need for adjuvant treatment. Other biomarkers have
predictive value (i.e., are associated with response or lack of
sensitivity to a certain treatment) and are critical to delivering
personalized cancer care and selecting the right treatment for
the individual patient. A third type of biomarker describes
genomic variants informing the emergence of therapy resistance. Repeating tumor biopsies over the course of therapy
is logistically and ethically challenging; the advent of new
technologies to study tumor DNA from minimally invasive
sources, such as plasma DNA or circulating tumor cells, represents an opportunity to monitor genetic changes resulting
from treatment-selective pressures. These new technologies
can also be used as multipurpose biomarkers for the study of
predictive, response, and resistance biomarkers.26,27
CLINICAL TRIAL DESIGN IN THE ERA OF

PERSONALIZED MEDICINE
Drug development through clinical trials is a key tool for
improving cancer care for our patients. However, two main
factors are limiting the ability to bring new and better drugs to
clinical practice: the tremendous cost of developing new
compounds and their low yield of success, which is especially
concerning when these failures happen in late-stage trials.28,29
There is an urgent and unmet medical, ethical, and financial
need to make our clinical trials more efficient, to increase the
odds of benefit for patients, and to enable the early identification of compounds that do not merit development. Interestingly, although the costs of new drugs and clinical trials
keep increasing, the costs of NGS is plummeting; therefore,
identifying subgroups more likely to benefit from a therapy will
be positive for patients, researchers, and health payers.30,31
Using Genomics to Drive Clinical Trial Design

To optimally design drug registration strategies, the putative
patient population for that treatment has to be identified
and characterized, qualitatively and quantitatively. In the

past 5 years, joint efforts from academia, philanthropic
organizations, and industry have led to several landscape
studies describing the genomic map of different tumor
types.32-36 These Rosetta stonesproviding insights on the
molecular underpinning of different cancershave permitted us to identify putative drivers and targetable genomic
events and determine their frequency at different stages of
the disease. For example, analysis of the mutational landscape of 150 patients with advanced and lethal prostate
cancer has clarified the role of the androgen receptor
pathway, DNA repair machinery, and the PI3K/Akt/mTOR
signaling pathway in this common disease37; indeed, this
study reported that more than 20% of patients with advanced prostate cancer harbor somatic mutations, germline
mutations, or deletions in tumor suppressor genes involved
in DNA repair, with such tumors arguably being more
sensitive to treatment with DNA-damaging agents (e.g.,
platinum-based chemotherapy) or PARP inhibition.
The reporting of exceptional responder cases in case reports
or patient series may also be very helpful in guiding a therapys
optimal development. These small and usually retrospective
studies can lead to hypothesis generation by revealing novel
putative response and resistance biomarkers.38-40 Sequencing
of outlier responders is particularly critical for rare tumor
types, where it may be difficult to compile large sample sizes
for more comprehensive studies. Still, the main challenge
remains generating data on exceptional responders, which
requires the prospective collection of preferably fresh tumor
material for genomic analyses. Following the example above
of the mutational landscape of prostate cancer, small series of
patients with advanced prostate cancer who have deep and
long-standing responses to platinum-based chemotherapy
have been reported with retrospective identification of
function loss in key DNA repair genes.41,42 Such reports,
however, often describe responses to drug combinations,
which can make it difficult to elucidate which of the drugs
involved is the main reason for the response.
Lastly, design of clinical trials must also take into consideration emerging preclinical data on the mechanism of action
of compound and suggested biomarkers of response and
resistance in the laboratory.43,44 This implies a need for designing adaptive clinical trials to allow rigorous clinical testing
with, preferably, prospective test and validation cohorts.
Clinical Trial Design and Genomics: Yesterday, Today,

and Tomorrow
Clinical trials can be classified as (1) proof-of-concept trials
(including phase I-IIA studies), with the main objectives of
selecting the right dose and schedule of administration of a
compound, as well as assessing biologic and antitumor activity
and characterizing the target population and (2) confirmatory
late-stage trials (phase III clinical trials), where the main
endpoint is the demonstration that the observed effect
warrants bringing a therapeutic strategy into clinical practice.
The pharmacological audit trail (PhAT), initially described by
Workman and colleagues, represents a science-driven

framework for optimal drug development, where companion biomarker discovery and validation is key.25,45 Historically,
clinical trials considered the tissue of origin and histologic
differentiation as the sole basis for classifying patients,
without any further prospective stratification of patients
based on molecular features. Under this approach, therapeutic regimens that did not achieve a statistically significant
benefit in late-stage studies for an all-comers population were
discarded, even if some patients benefited.46-48
In the past 15 years, and with the advent of targeted
therapies, we have progressively transitioned to a one
gene, one biomarker, one drug model, which, despite its
limitations, has successfully led to approvals of several new
drugs for different tumor types only for biomarker-positive
cancers, typically in the presence of a single genetic event or
the expression of a protein. Examples of this successful
strategy include the approval of imatinib in 2001 for chronic
myeloid leukemia with BCR-ABL translocations49; EGFR inhibitors (gefitinib, erlotinib, afatinib, osimertinib) for EGFRmutant nonsmall cell lung cancers,50 anti-HER2 therapies
for breast cancer (trastuzumab, lapatinib, pertuzumab,
trastuzumab emtansine)51-54; and the successful development of BRAF and MEK inhibitors for melanomas with
BRAF-activating mutations.55-57 More recently, in 2014,
olaparib was the first PARP inhibitor approved by the U.S.
Food and Drug Administration (FDA) for ovarian cancer
associated with pathogenic somatic or germline loss of
BRCA1 or BRCA2 function.58,59
We envision that the dissection of the cancer genome will
represent a big change in the way we design clinical trials,
necessitating novel therapeutic approaches and biomarker
strategies. Trial designs that include the prospective molecular characterization of patients with a particular tumor
type are already ongoing for many tumor types.60,61 In fact,
these umbrella trials serve as multiple, simultaneous, openlabel trials for different targeted therapies. Both the
American Society of Clinical Oncology and the National
Cancer Institute (NCI) are actively supporting and sponsoring
umbrella trials of precision medicine (TAPUR and NCIMATCH), prospectively matching patients with different
tumor types to targeted therapies.62 These sorts of trials are
key when a biomarker has low prevalence; however, umbrella studies demand a deductively defined match and
therefore are more intended for validation rather than
biomarker discovery (Fig. 1).
Another example of novel trial design based on genomic
studies are basket trials, where patients with different tumor
types can be recruited together onto a trial based on a
common biomarker of interest presumed to be associated to
sensitivity to a particular therapy. Successful examples of the
feasibility of basket trial designs are the recently reported
phase II studies of the PARP inhibitor olaparib for BRCA1/2mutated tumors,63 regardless of tumor type, and a phase II
study of the FDA-approved vemurafenib for BRAF-mutated
nonmelanoma cancers.64 This second trial raised questions
on how single mutations are interpreted in their complex
e579
MATEO AND DE BONO
FIGURE 1. Clinical Trials in the Era of Genomics and Personalized Medicine
The diagram represents how different trial designs can complement each other during the development of a therapy. To optimize efforts and maximize benefit, drug
development should incorporate biomarker identification, validation, and qualification starting in the early stages of clinical testing.
genomic context, as indicated by the fact that patients with

colorectal cancer with BRAF mutations did not benefit from
vemurafenib. Moreover, these designs are at risk for biasing
outcome and assuming that all is known about which biomarkers sensitize to a drug or drug combination.
More flexible biomarker-driven adaptive trials designs,
incorporating test and validation sets, may represent a more
efficient way to conduct phase II trials incorporating genomics. Based on emerging data on DNA repair gene mutations
in advanced prostate cancer, we recently reported on the
first stage of a multistep adaptive clinical trial of the PARP
inhibitor olaparib in heavily pretreated metastatic prostate
cancer.65 We used a prospectively planned retrospective
biomarker analysis in the test set, recruiting all-comers first
and pursuing retrospective exome and transcriptome sequencing studies to associate putative predictive biomarkers
with response or resistance to the drug. Then, once we
identified that patients with not only BRCA2 mutations but
also deleterious aberrations of other genes involved in the
DNA repair machinery responded to olaparib, the study
progressed to a second validation stage where patients are
prospectively selected through genomic studies to further
clinically qualify the biomarker suite reported in the test set.
Measuring How Precision Medicine Improves Patient

Care
It is necessary to develop tools quantifying the benefit
derived from the implementation of genomics into cancer
care to provide evidence for this strategys efficiency. Jardim
et al recently reported on a retrospective review of 112 trials
conducted between 1998 and 2013 involving 58 drugs that
e580
achieved FDA approval for cancer treatment66; their analysis

shows that those clinical trials incorporating genomic biomarker selection or enrichment into their design resulted in
higher response rates and longer progression-free and
overall survival when compared with those trials testing the
same drugs without using any genomic stratification. Although this report may be biased by selecting drugs that
were finally approved, it supports incorporating biomarkerdriven strategies into clinical trials to optimize drug development and improve efficiency.
Health economic studies assessing the impact of precision
medicine are clearly urgently needed; arguably, if precision
medicine improves patient outcome, the costs of NGS will be
compensated by a reduction of morbidity-associated costs
for the health care system. Such studies, though, are in need
of novel biostatistic designs to allow data from different
studies to be comprehensively compared and compiled.
DEFINING THE REGULATORY FRAMEWORK FOR

GENOMIC TESTS AND TARGETED THERAPIES
Regulatory agencies are also adapting to this new genomic
landscape, with more precise drug therapeutic strategies
resulting in broader therapeutic indices and greater benefits
leading to more rapid drug approvals. Despite this, a large
challenge for the community and regulators will be oversight
of the analytic validation of NGS assays. Traditional approaches for approving medical devicessuch as an electrocardiogram reader or a machine capable of enumerating
circulating tumor cells from a blood samplehave been
based on the accuracy of such technology to identify the true
result for a patient from a range of prespecified values.
Genomic testing challenges this concept with new variants

being regularly identified. However, some of these aberrations are of low prevalence, making it difficult to define
deductively all the possible outcomes from a test because of
the complexity of human cancers. In a recent workshop on
the regulatory framework for NGS, the FDA stated that we
may need new approaches, such as accepting orthogonal
testing to prospectively compare the performance of different tests between themselves rather than against a gold
standard.67-69 Moreover, the analytic validity of these tests
may depend on not only the device in question but also a
number of other variables. Indeed, there should be a preanalytic procedures validation, including controlling for the
samples and testing the patient population, and a later
postanalytic procedures validation, including the processes
adapted at the local laboratory or the workflow for results
interpretation.24
Regulatory oversight of the clinical qualification of a
biomarker also creates other challenges. Historically, all
qualification has been conducted in premarketing studies;
however, the number of genomic variants of unknown
clinical significance is large and continuously growing. Although widely used public genomic databases, such as
ClinVar, have entries for less than 200,000 variants, sequencing of one cancer genome can detect millions of
variants. To implement NGS in clinical practice, we need
systems in place to allow for revision of data in the postmarketing space. In fact, the use of public databases is
another matter of discussion for regulatory agencies
specifically, can data not controlled or overseen by the FDA
or EMEA be acceptable for regulatory purposes? How should
data generated by laboratories not being bound to Clinical
Laboratory Improvement Amendment requirements or data
not deposited in public databases be used?
STEPS AND CHALLENGES TO IMPLEMENT

GENOMICS INTO ROUTINE CLINICAL PRACTICE
A major bottleneck to the wider implementation of genomic
studies in clinical practice remains the precise and rapid
interpretation of multidimensional data. Current genomic
studies still demand intensive bioinformatics resources,
which slow the turnaround time for treating physicians
when more comprehensive genome coverage is pursued.
Moreover, even if the costs of the wet laboratory aspects
of NGS keep decreasing, performing somatic whole-exome
or whole-genome sequencing studies at high coverage may
still be prohibitively expensive in light of the intense data
processing required. In this context, targeted sequencing
panels appear to be a very appealing interim approach, as
they can provide multiplex biomarker testing at a reasonable
cost, while requiring limited bioinformatic effort. Several
academic centers and commercial institutions have developed targeted sequencing panels for rapid tumor profiling of actionable aberrations to select patients for targeted
treatments, with these biomarkers being rapidly incorporated into clinical decision making.70-72 Nonetheless,
reimbursement for genomic profiling is currently limited to

specific cases (e.g., breast, lung, melanoma, colon, ovarian,
and thyroid cancers).
Success of these targeted sequencing approaches may also
be limited by the quality of the analyzed tumor tissue.70-72
Improved methods for tissue fixation, optimized for nucleic
acid preservation, are now urgently needed to minimize the
risk of patients missing out on treatments because of poor
quality assurance in pathology laboratories.
Although these targeted approaches are an important
strategy for the short term, we envision that in middle and
long term, whole-genome (or at least whole-exome) and
transcriptome studies will become easier to access and may
become the preferred approach, as they will provide more
comprehensive analyses with costs continuously decreasing.73 A major challenge to the incorporation of multidimensional genomic data into routine clinical practice is
ensuring that treating physicians are able to correctly interpret genomic data. A number of academic institutions
leading the development of personalized medicine for
cancer have opted for developing multidisciplinary sequencing boards where genomic results are analyzed and
the treating physician receives a report with a summary of
the most relevant and actionable findings, along with advice
on how these could affect sensitivity to different therapies.74,75 This may not be feasible for all institutions,
however, because not all institutions have experienced
personnel to interpret complex genomic data. Collaboration
between academic, industry, and regulatory bodies is critically important to ensure that genomic data in clinical
practice is optimized and becomes accessible to the majority
of the population.
Lastly, implementing genomic analyses into clinical care
needs to also convey to patients the effect of their
resultsparticularly when this involves germline testing
while warranting their privacy. Importantly, a proportion of
druggable genomic events may involve germline inherited
aberrations and may, therefore, affect not only the patients
management but also the patients risk for other diseases
and the risk of cancer for the patients relatives with the
associated fiscal and insurance policy implications. Balancing
patients expectations, their privacy rights, implications for
relatives, and the need for continuously increasing knowledge sharing, make the clinical consent process for genomic
studies particularly criticalperhaps especially in special
situations such as treating pediatric cancers or cancers in
adolescents.
CONCLUSION
Embedding genomic studies in clinical trials can transform
the development of novel anticancer therapies and is
changing patient care. Academics, the pharmaceutical industry, physicians, patients, and regulatory agencies now
must adapt their frameworks and processes to the new
opportunities arising in the care of patients with cancer.
These new processes will require wider accessibility to
e581
MATEO AND DE BONO
improved, inexpensive, high-throughput, and precise sequencing technologies, with improved bioinformatics
pipelines. Thiscombined with the sharing of sequencing
data linked to patient outcome and treatment response, the
standardization of clinical qualification processes, and the

simplification of omics data analysesremain the key
challenges to our translation of the cancer genome to make a
difference for our patients.
References
1. Van Allen EM, Wagle N, Stojanov P, et al. Whole-exome sequencing
and clinical interpretation of formalin-fixed, paraffin-embedded tumor samples to guide precision cancer medicine. Nat Med. 2014;20:
682-688.
2. Sanger F, Coulson AR. A rapid method for determining sequences in
DNA by primed synthesis with DNA polymerase. J Mol Biol. 1975;94:
441-448.
3. Margulies M, Egholm M, Altman WE, et al. Genome sequencing in
microfabricated high-density picolitre reactors. Nature. 2005;437:
376-380.
4. Rogers Y-H, Venter JC. Genomics: massively parallel sequencing. Nature. 2005;437:326-327.
5. Davies H, Bignell GR, Cox C, et al. Mutations of the BRAF gene in human
cancer. Nature. 2002;417:949-954.
6. Samuels Y, Wang Z, Bardelli A, et al. High frequency of mutations of the
PIK3CA gene in human cancers. Science. 2004;304:554.
7. Nigro JM, Baker SJ, Preisinger AC, et al. Mutations in the p53 gene occur
in diverse human tumour types. Nature. 1989;342:705-708.
8. Rhei E, Kang L, Bogomolniy F, et al. Mutation analysis of the putative
tumor suppressor gene PTEN/MMAC1 in primary breast carcinomas.
Cancer Res. 1997;57:3657-3659.
9. Gayther SA, Mangion J, Russell P, et al. Variation of risks of breast and
ovarian cancer associated with different germline mutations of the
BRCA2 gene. Nat Genet. 1997;15:103-105.
10. Fong PC, Boss DS, Yap TA, et al. Inhibition of poly(ADP-ribose) polymerase in tumors from BRCA mutation carriers. N Engl J Med. 2009;361:
123-134.
11. Zack TI, Schumacher SE, Carter SL, et al. Pan-cancer patterns of somatic
copy number alteration. Nat Genet. 2013;45:1134-1140.
12. Liu W, Laitinen S, Khan S, et al. Copy number analysis indicates
monoclonal origin of lethal metastatic prostate cancer. Nat Med. 2009;
15:559-565.
13. Tomlins SA, Rhodes DR, Perner S, et al. Recurrent fusion of TMPRSS2
and ETS transcription factor genes in prostate cancer. Science. 2005;
310:644-648.
14. Hermans KG, van Marion R, van Dekken H, et al. TMPRSS2:ERG fusion by
translocation or interstitial deletion is highly relevant in androgendependent prostate cancer, but is bypassed in late-stage androgen
receptor-negative prostate cancer. Cancer Res. 2006;66:10658-10663.
15. Soda M, Choi YL, Enomoto M, et al. Identification of the transforming
EML4-ALK fusion gene in non-small-cell lung cancer. Nature. 2007;448:
561-566.
16. Lin E, Li L, Guan Y, et al. Exon array profiling detects EML4-ALK fusion in
breast, colorectal, and non-small cell lung cancers. Mol Cancer Res.
2009;7:1466-1476.
17. Solomon BJ, Mok T, Kim D-W, et al. First-line crizotinib versus chemotherapy in ALK-positive lung cancer. N Engl J Med. 2014;371:
2167-2177.
18. Esteller M. Cancer epigenomics: DNA methylomes and histonemodification maps. Nat Rev Genet. 2007;8:286-298.
19. LaTulippe E, Satagopan J, Smith A, et al. Comprehensive gene expression analysis of prostate cancer reveals distinct transcriptional
e582
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
programs associated with metastatic disease. Cancer Res. 2002;62:

4499-4506.
Spizzo R, Almeida MI, Colombatti A, et al. Long non-coding RNAs and
cancer: a new frontier of translational research? Oncogene. 2012;31:
4577-4587.
Mattick JS, Gagen MJ. The evolution of controlled multitasked gene
networks: the role of introns and other noncoding RNAs in the development of complex organisms. Mol Biol Evol. 2001;18:1611-1630.
Greenman C, Stephens P, Smith R, et al. Patterns of somatic mutation in
human cancer genomes. Nature. 2007;446:153-158.
Beerenwinkel N, Antal T, Dingli D, et al. Genetic progression and the
waiting time to cancer. PLOS Comput Biol. 2007;3:e225.
de Gramont A, Watson S, Ellis LM, et al. Pragmatic issues in biomarker
evaluation for targeted therapies in cancer. Nat Rev Clin Oncol. 2015;
2014:197-212.
Yap TA, Sandhu SK, Workman P, et al. Envisioning the future of early
anticancer drug development. Nat Rev Cancer. 2010;10:514-523.
Dawson S-J, Tsui DWY, Murtaza M, et al. Analysis of circulating tumor
DNA to monitor metastatic breast cancer. N Engl J Med. 2013;368:
1199-1209.
Lohr JG, Adalsteinsson VA, Cibulskis K, et al. Whole-exome sequencing
of circulating tumor cells provides a window into metastatic prostate
cancer. Nat Biotechnol. 2014;32:479-484.
Kola I, Landis J. Can the pharmaceutical industry reduce attrition rates?
Nat Rev Drug Discov. 2004;3:711-715.
Mateo J, Ong M, Tan DSP, et al. Appraising iniparib, the PARP inhibitor
that never waswhat must we learn? Nat Rev Clin Oncol. 2013;10:
688-696.
Sullivan R, Peppercorn J, Sikora K, et al. Delivering affordable cancer
care in high-income countries. Lancet Oncol. 2011;12:933-980.
Collier R. Rapidly rising clinical trial costs worry researchers. CMAJ.
2009;180:277-278.
Waddell N, Pajic M, Patch AM, et al. Whole genomes redefine the
mutational landscape of pancreatic cancer. Nature. 2015;518:495-501.
Patch A-M, Christie EL, Etemadmoghadam D, et al. Whole-genome
characterization of chemoresistant ovarian cancer. Nature. 2015;521:
489-494.
Stransky N, Egloff AM, Tward AD, et al. The mutational landscape of
head and neck squamous cell carcinoma. Science. 2011;333:1157-1160.
Lawrence MS, Stojanov P, Mermel CH, et al. Discovery and saturation
analysis of cancer genes across 21 tumour types. Nature. 2014;505:495-501.
Robinson D, Van Allen EM, Wu Y-M, et al. Integrative clinical genomics
of advanced prostate cancer. Cell. 2015;161:1215-1228.
Mullard A. Learning from exceptional drug responders. Nat Rev Drug
Discov. 2014;13:401-402.
Takebe N, McShane L, Conley B. Biomarkers: exceptional
respondersdiscovering predictive biomarkers. Nat Rev Clin Oncol.
2015;12:1-2.
Iyer G, Hanrahan AJ, Milowsky MI, et al. Genome sequencing identifies a
basis for everolimus sensitivity. Science. 2012;338:221.
41. Cheng HH, Pritchard CC, Boyd T, et al. Biallelic inactivation of BRCA2 in
platinum-sensitive metastatic castration-resistant prostate cancer. Eur
Urol. Epub 2015 Dec 24.
42. Beltran H, Eng K, Mosquera JM, et al. Whole-exome sequencing of
metastatic cancer and biomarkers of treatment response. JAMA Oncol.
2015;1:466-474.
43. Begley CG, Ellis LM. Drug development: raise standards for preclinical
cancer research. Nature. 2012;483:531-533.
44. de Bono JS, Ashworth A. Translating cancer research into targeted
therapeutics. Nature. 2010;467:543-549.
45. Workman P. Auditing the pharmacological accounts for Hsp90 molecular chaperone inhibitors: unfolding the relationship between
pharmacokinetics and pharmacodynamics. Mol Cancer Ther. 2003;2:
131-138.
46. Amiri-Kordestani L, Fojo T. Why do phase III clinical trials in oncology fail
so often? J Natl Cancer Inst. 2012;104:568-569.
47. Sternberg CN, Petrylak DP, Sartor O, et al. Multinational, double-blind,
phase III study of prednisone and either satraplatin or placebo in patients with castrate-refractory prostate cancer progressing after prior
chemotherapy: the SPARC trial. J Clin Oncol. 2009;27:5431-5438.
48. OShaughnessy J, Schwartzberg L, Danso MA, et al. Phase III study of
iniparib plus gemcitabine and carboplatin versus gemcitabine and
carboplatin in patients with metastatic triple-negative breast cancer.
J Clin Oncol. 2014;32:3840-3847.
49. Druker BJ, Talpaz M, Resta DJ, et al. Efficacy and safety of a specific
inhibitor of the BCR-ABL tyrosine kinase in chronic myeloid leukemia.
N Engl J Med. 2001;344:1031-1037.
50. Mok TS, Wu Y-L, Thongprasert S, et al. Gefitinib or carboplatin-paclitaxel
in pulmonary adenocarcinoma. N Engl J Med. 2009;361:947-957.
52. Romond EH, Perez EA, Bryant J, et al. Trastuzumab plus adjuvant
chemotherapy for operable HER2-positive breast cancer. N Engl J Med.
2005;353:1673-1684.
53. Baselga J, Gelmon KA, Verma S, et al. Phase II trial of pertuzumab and
trastuzumab in patients with human epidermal growth factor receptor
2-positive metastatic breast cancer that progressed during prior trastuzumab therapy. J Clin Oncol. 2010;28:1138-1144.
54. Baselga J, Bradbury I, Eidtmann H, et al. Lapatinib with trastuzumab for
HER2-positive early breast cancer (NeoALTTO): a randomised, openlabel, multicentre, phase 3 trial. Lancet. 2012;379:633-640.
55. Chapman PB, Hauschild A, Robert C, et al. Improved survival with
vemurafenib in melanoma with BRAF V600E mutation. N Engl J Med.
2011;364:2507-2516.
56. Hauschild A, Grob J-J, Demidov LV, et al. Dabrafenib in BRAF-mutated
57. Flaherty KT, Robert C, Hersey P, et al. Improved survival with MEK
inhibition in BRAF-mutated melanoma. N Engl J Med. 2012;367:
107-114.
in platinum-sensitive relapsed ovarian cancer. N Engl J Med. 2012;366:
1382-1392.
in patients with platinum-sensitive relapsed serous ovarian cancer:
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
a preplanned retrospective analysis of outcomes by BRCA status in a

Kim ES, Herbst RS, Wistuba II, et al. The BATTLE trial: personalizing
therapy for lung cancer. Cancer Discov. 2011;1:44-53.
Herbst RS, Gandara DR, Hirsch FR, et al. Lung master protocol (LungMAP)-a biomarker-driven protocol for accelerating development of
therapies for squamous cell lung cancer: SWOG S1400. Clin Cancer Res.
2015;21:1514-1524.
Conley BA, Doroshow JH. Molecular analysis for therapy choice: NCI
MATCH. Semin Oncol. 2014;41:297-299.
Kaufman B, Shapira-Frommer R, Schmutzler RK, et al. Olaparib monotherapy in patients with advanced cancer and a germline BRCA1/2
Hyman DM, Puzanov I, Subbiah V, et al. Vemurafenib in multiple
nonmelanoma cancers with BRAF V600 mutations. N Engl J Med. 2015;
373:726-736.
Fontes Jardim DL, Schwaederle M, Wei C, et al. Impact of a biomarkerbased strategy on oncology drug development: a meta-analysis of
clinical trials leading to FDA approval. J Natl Cancer Inst. 2015;107:
djv253.
U.S. Department of Health and Human Services, Food and Drug
Administration, Center for Devices and Radiological Health, and Office
of In Vitro Diagnostics and Radiological Health. FDA Notification and
Medical Device Reporting for Laboratory Developed Tests (LDTs),
Draft Guidance. http://www.fda.gov/downloads/medicaldevices/
deviceregulationandguidance/guidancedocuments/ucm416684.pdf.
Evans BJ, Burke W, Jarvik GP. The FDA and Genomic Tests Getting
Regulation Right. N Engl J Med. 2015;372:2258-2264.
U.S. Department of Health and Human Services, U.S. Food and Drug
Administration, Center for Devices and Radiological Health, and Office of
In Vitro Diagnostics and Radiological Health. Framework for Regulatory
Oversight of Laboratory Developed Tests (LDTs), Draft Guidance. http://
www.fda.gov/downloads/medicaldevices/deviceregulationandguidance/
guidancedocuments/ucm416685.pdf. Accessed February 12, 2016.
Ong M, Carreira S, Goodall J, et al. Validation and utilisation of highcoverage next-generation sequencing to deliver the pharmacological
audit trail. Br J Cancer. 2014;111:828-836.
Cheng DT, Mitchell TN, Zehir A, et al. Memorial Sloan KetteringIntegrated Mutation Profiling of Actionable Cancer Targets (MSKIMPACT): a hybridization capture-based next-generation sequencing
clinical assay for solid tumor molecular bncology. J Mol Diagn. 2015;17:
251-264.
Tsimberidou A-M, Iskander NG, Hong DS, et al. Personalized medicine
in a phase I clinical trials program: the MD Anderson Cancer Center
initiative. Clin Cancer Res. 2012;18:6373-6383.
Maggi E, Montagna C. AACR precision medicine series: highlights of the
integrating clinical genomics and cancer therapy meeting. Mutat Res.
2015;782:44-51.
Simon R, Roychowdhury S. Implementing personalized cancer genomics
in clinical trials. Nat Rev Drug Discov. 2013;12:358-369.
Roychowdhury S, Iyer MK, Robinson DR, et al. Personalized oncology
through integrative high-throughput sequencing: a pilot study. Sci
Transl Med. 2011;3:111ra121.
e583
TUMOR BIOLOGY
Molecular Diagnostics in Cancer

CHAIR
Reena Philip, PhD
Bethesda, MD
SPEAKERS
Ignacio Ivan Wistuba, MD
Houston, TX
Daniel F. Hayes, MD, FASCO
University of Michigan Comprehensive Cancer Center
Ann Arbor, MI
DANIEL F. HAYES
Considerations for Implementation of Cancer Molecular

Diagnostics Into Clinical Care
Daniel F. Hayes, MD, FASCO
OVERVIEW
Physicians have provided personalized care with as much precision as possible for several centuries. However, increasingly
sophisticated understanding of the human genome and of cancer biology has permitted identification of genetic and
phenotypic distinctions that might permit development of new tumor biomarker tests for risk categorization, screening,
differential diagnosis, prognosis, prediction, and monitoring. Both commercial and academic laboratories are offering tests
for single analytes, panels of tests of single analytes, multiparameter assays coalesced into a signature, and total genomic,
transcriptomic, or proteomic analyses. However, the absence of a consistent regulatory environment has led to marketing of
assays without proven analytic validity or clinical utility. U.S. Food and Drug Administration (FDA) approval or clearance does
not necessarily imply that use of the test will improve patient outcomes, and FDA discretion to permit laboratory-developed
tests results in unknown benefit, or harm, of others. In this regard, a bad tumor marker is as bad as a bad drug. Caveat
emptor is not a satisfactory approach to delivering high-quality care. Rather, adoption of tumor biomarker tests should be
based on high levels of evidence generated in scientifically rigorous studies that demonstrate both analytical validity and
clinical utility. Doing so will ensure that clinicians and patients are confident that a tumor biomarker test is likely to improve
their outcomes.
s noted by the President Barack Obama in his State of

the Union address in January 2014, we are now entering
the era of personalized or precision medicine. The implication of this statement is that we will be able to provide the
right therapy to the right patient at the right time, schedule,
and dosea laudable goal. Of course, current hype notwithstanding, physicians have been trying to provide
personalized medicine for centuries. Certainly, public health
measures, such as sewers, vaccines, and smoking cessation
have led to major improvements in quality and duration of
life. However, individual caregivers are taught the importance of careful analysis of diagnostic data to develop a
differential diagnosis, and then to act accordingly, balancing
the benefits and risks of available treatment options for the
presumed final diagnosis. In this regard, clinicians have tried
to do so with as much precision as possible, given the
available tools and technologies at their disposal.
Cloning of the human genome, which took approximately
10 years and $3 billion, has opened technical, diagnostic,
and therapeutic possibilities not previously imagined.1
Indeed, The Cancer Genome Atlas has exploited rapid
technical advances that have made total genomic, transcriptomic, and even proteomic evaluations much quicker
and less expensive.2 Coupled with advances in molecular

biology, the knowledge gained from this technical progress
offers an unprecedented opportunity to develop new diagnostic tests that will provide unprecedented opportunities
for clinicians to provide truly precise medicine.
Cancer is ideally suited for development of more precise
diagnostic tests. We have ample evidence that germline
single-nucleotide polymorphisms in inherited DNA serve to
increase an individuals susceptibility to cancer, and tests for
BRCA1 and BRCA2, microsatellite instability, and others are
now widely used to help in decision making regarding
screening and prophylaxis that are otherwise unacceptable
for unaffected patients.3 Furthermore, the availability of
somatic cancer-associated changes, which can be measured
in cancer tissue, blood, or secreted fluids, and other specimens, is ideal for development of diagnostic tests.
However, the rapid introduction of these tests to direct
patient care has occurred with a paucity of regulatory
oversight, and often in the absence of rigorous scientific
support or high levels of evidence that the test is either
well performed or that its use improves patient outcomes
compared with not using it. Would we permit introduction of a new therapeutic based on assumption and weak
From the Breast Oncology Program, University of Michigan Comprehensive Cancer Center, Ann Arbor, MI.
Corresponding author: Daniel F. Hayes, MD, FASCO, Breast Oncology Program, University of Michigan Comprehensive Cancer Center, 1500 East Medical Center Dr., Ann Arbor,
MI 48109; email: hayesdf@umich.edu.
292
IMPLEMENTATION OF CANCER MOLECULAR DIAGNOSTICS
science? The answer is no. A bad tumor marker is as bad

as a bad drug. As the fascinating advances in omics provide
more and more information, development of tumor biomarker tests and introduction of them into the clinic must be
carefully considered.
WHAT IS A TUMOR BIOMARKER TEST?

It is important to note the distinction between a tumor
biomarker and a test, or assay, for that biomarker. Whereas
the former represents an indication that the tumor differs
from normal tissue in some way, many assays, in many
different formats, may identify the biomarker in many different ways. For example, no one doubts that HER2 is a
useful biomarker in breast cancer.4 However, what are we
testing when we assay for HER2? To begin with, there are
at least three separate names for this biomarker: HER2,
ERBB2, and c-neu.5 Regardless, we might test for amplification6 or, more recently, activating mutations in the gene
that encodes for HER2,7 ERBB2, or for overexpression
(compared with normal cells) at the RNA or protein level.8
We usually interrogate the cancer tissue itself for these
abnormalities, but elevated levels of the extracellular domain of HER2 can be detected in blood,9 as can ERBB2
mutations and perhaps even amplification, in cell-free DNA.
Furthermore, both amplification and overexpression of
HER2 can be detected in circulating tumor cells. Multiple
assays may exist for each of these conditions. For example,
different formats such as Southern blots, dot blots, comparative chromosomal hybridization, or fluorescent or
chromogenic in situ hybridization have all been used to
identify and quantify ERBB2 amplification. Measuring HER2
protein has been reported using enzyme-linked immunoassays,10 quantitative immunofluorescence,11 and immunohistochemistry,4 and the latter has been reported with at
least 15 different antibodies, all of which differ in their
sensitivity and specificity.12 These considerations are not
just academic, semantic debates. When we use an assay to
direct care, the assay becomes as important as the drug we
elect to give or withhold. Therefore, a sophisticated understanding of not only the biomarker but also the test to
measure it, is essential for high-quality clinical activity.
Categories of Tumor Biomarker Tests

Physicians are used to ordering tests, of course. We may
order either a test for a single analyte (such as a serum
calcium level) or, more frequently, a panel of multiple
analytes, such as a complete blood count. The latter provides
us with a dashboard picture of bone marrow function and
the associated hematologic status: hemoglobin, hematocrit,
red blood cell indices, platelet count, white blood cell count,
and differential and absolute leukocyte counts. In cancer,
both single analyte and panels of analytes have been part of
the tumor biomarker test armamentarium for years. Again,
turning to breast cancer as an example, for at least a decade,
if not longer, standard of care has dictated that all breast
tissues should be sent to the pathologist for measurement of
estrogen receptor (ER), progesterone receptor (PgR), and
HER2 analysis.4,13,14
More recently, both academic and commercial laboratories are offering panel testing for a number of tumor biomarkers that are not necessarily associated with one
disease, but rather cut across histologic origins of cancers,
providing doctors with results that may be relevant to one
type of cancer, but not necessarily to another. Taken a step
further, assays that now provide total genomic and/or
transcriptomic evaluation, or representative evaluation
using huge numbers of candidate genes, have been available, both commercially and within research arenas. The
National Cancer Institute is currently sponsoring several
clinical trials (Lung MAP, MATCH, etc.) in which such
technology is applied to metastatic biopsies to try to identify
genetic abnormalities that could serve as targets for therapeutic agents in associated phase II clinical trials.15
Other types of assays have been developed and are
available for clinical use.16 One strategy has been to measure
many analytes in one specimen and combine the results
into a weighted, multiparameter algorithmic index, or
signature. In this case, the physician is usually provided a
single value, or score, which is then placed into categories of
good, intermediate, or bad. In this regard, pathologists have
done exactly this for decades by performing histologic
grading, but more recently scientists have used genomic
expression array technologies to develop even more sophisticated assays. For example, again in breast cancer, the
21-gene Recurrence Score (OnctypeDx; Genomics Health,
Inc.) is one such test.17-19
KEY POINTS
WHAT MAKES A GOOD TUMOR MARKER ASSAY?
A bad tumor biomarker test is as bad as a bad drug.

There are key elements that must be met before one
uses a tumor biomarker test to direct care.
The intended use context must be delineated for a
tumor biomarker test, e.g., risk categorization,
prognosis, prediction, or monitoring.
A tumor biomarker test must have analytical validity to
be used to guide patient care.
A tumor biomarker test must have clinical utility for the
intended use context to be used for patient care.
Distinguishing a Test for a Tumor Biomarker From the

Tumor Biomarker Itself
We now return to our original thesis, a bad tumor marker
test is as bad as a bad drug. In 2009, the Evaluation of
Genomic Applications in Practice and Prevention (EGAPP)
Initiative coined three important terms that articulated the
evolution of a tumor marker to a clinically valuable tumor
marker test20: (1) analytical validity, meaning that the
preanalytical and analytical aspects of the test for the
tumor marker condition are well defined, practical, accurate,
293
DANIEL F. HAYES
TABLE 1. Criteria for Analysis of Biomarker Tests for

Application in the Clinic
Criteria
Analysis
Analytical Validity
Is the assay accurate, reproducible and

reliable? Have the preanalytical issues
been evaluated to be certain the assay has
analytical validity in the type of specimen
and conditions under which it will be
performed?
Clinical/Biologic
Validity
Does the assay divide one population into

two or more separate groups with
statistical significance?
Clinical Utility
Do results of the assay lead to a clinical

decision that has been shown with high
level of evidence to improve outcomes?
Adapted from Teutsch et al20 with permission from the publisher.
and reproducible; (2) clinical validity, meaning that the results of the tumor biomarker test divides one population into
two or more based on either biologic or, more importantly,
clinical outcomes differences; and (3) clinical utility, meaning
that high levels of evidence are available that demonstrate
that clinical outcomes are improved when the assay is applied compared with if it were not applied for a given clinical
use context (Table 1). Whereas clinical validity is usually as
step toward applying a tumor biomarker test to clinical care,
analytical validity and demonstration of clinical utility are
essential. Clinical use contexts may include risk categorization for unaffected individuals, application or screening to
individuals at risk, or determination of prognosis, prediction
of therapeutic effect, or monitoring to determine disease
course.21 The high levels of evidence required to achieve
clinical utility can either be produced within prospective
clinical trials designed specifically to address the value of
the tumor marker,22,23 or from prospective retrospective
studies using specimens collected and archived from previously conducted prospective therapeutic trials that have
been designed to permit investigation of the clinical use
context of interest for the biomarker test.24 In a review
of the application of omics-based tests to clinical practice,
the Institute of Medicine has endorsed these terms and
outlined a strategy for progression of a tumor marker from
first observation to development of an analytically validated
test with clinical validity to clinical utility (Fig. 1).1
market for clinical use: (1) application for FDA approval

or clearance under premarket approval or substantial
equivalence (510K) mechanisms, respectively; and (2)
development of a laboratory-developed test within an
individual laboratory following practices described under
the Clinical Laboratory Improvement Amendments (CLIA)
of 1988.
These disparate pathways mean that FDA clearance or
approval of a tumor biomarker test does not imply or
require demonstration of clinical utility. FDA has not
adopted the precise EGAPP or Clinical Assay Development
Program terminology but does require analytical validation of a new test and evidence that the test performance is aligned with the claims, or intended use, of
the manufacturer.20 Furthermore, tumor biomarker tests
offered as laboratory-developed tests never have been
scrutinized for analytical validity or clinical utility, because CLIA has no requirement that, or even a review
process to determine if, individual tests have either. Indeed,
FDA is considering dramatic changes to the current regulatory
FIGURE 1. A Road Map for Development of

a Molecular-Based Tumor Biomarker Test
CURRENT TUMOR BIOMARKER CLINICAL STATUS

Sadly, very few tumor biomarker tests have undergone
the scientific rigor to demonstrate analytical validity and
clinical utility for a given clinical context use. There are
many reasons for this gap in our knowledge, stemming
from an overall historical lack of value for tumor biomarkers, relative to therapeutic agents.25 First, the regulatory environment for diagnostics in general, and tumor
biomarkers in particular, is inconsistent in the United
States and elsewhere. There are two regulatory pathways
in the United States for bringing a tumor biomarker test
(classified as an in vitro diagnostic device) onto the
294
Modified from the Institute of Medicine1 with permission from the publisher.
IMPLEMENTATION OF CANCER MOLECULAR DIAGNOSTICS
environment, and the American Society of Clinical Oncology

(ASCO) has supported doing so.
Second, the reimbursement environment for tumor biomarker tests is also quite inconsistent. Although the enormous
cost of cancer therapeutics has recently come under scrutiny,26 reimbursement for tumor biomarker tests, especially
those with proven high clinical utility, has in general been
insufficient for their value.25 We have argued that a good
tumor biomarker test that both directs patients to life-saving
and improving therapies, while directing patients away from
expensive and toxic therapies that they do not need or from
which they will not benefit, are equally as valuable as effective
drugs, and should be reimbursed commensurately.25
SO WHAT IS A CLINICIAN AND PATIENT TO DO?

Currently, the tumor biomarker test situation is chaotic, and
frankly getting worse. Clinicians are being bombarded daily
with marketing for germline and somatic tests, either from
tissue or blood, that provide information about single
analytes, panels of analytes that may or may not pertain to
the type of cancer the patient has, or signature scores that
may or may not have been proven to have analytical validity
or clinical utility, or even omics-based tests that provide
the entire, or major parts of, the genetic, transcriptomic, or
proteomic (and more recently metabolomics) construct of a

patients germline or tumor content.
Each of these explicitly, or implicitly, claims remarkable
benefits for our patients. However, many of these are
backed up with questionable, or even no, scientific evidence. Some of them have undergone FDA scrutiny and
achieved approval or clearance, and yet that does not
mean they should be used to care for patients. Others are
laboratory-developed tests, some of which may be remarkably effective in improving patient outcomes, and
yet the clinician cannot be certain that they are even
analytically valid.
ASCO, the National Comprehensive Cancer Network,
and other societies have made efforts to provide clear and
evidence-based guidelines regarding the use of tumor
biomarker tests in daily oncologic practice.27-29 However,
with the proliferation of broad analyte panels for both
germline and somatic testing, a large number of multiparameter signature assays for a variety of diseases and
use contexts, and assays for circulating plasma tumor DNA
mutations on the horizon, the choices are bewildering and
difficult to evaluate. The clinician is advised to keep as well
informed as possible, but perhaps more importantly to
avoid hype and maintain an evidence-based, scientifically
rigorous approach to patient care.
References
1. Institute of Medicine. Evolution of Translational Omics: Lessons Learned and
the Path Forward. Washington, DC: The National Academies Press; 2012.
2. Cancer Genome Atlas Network. Comprehensive molecular portraits of
human breast tumours. Nature. 2012;490:61-70.
Oncology Policy Statement Update: Genetic and Genomic Testing for
Cancer Susceptibility. J Clin Oncol. 2015;33:3660-3667.
4. Wolff AC, Hammond ME, Hicks DG, et al; American Society of Clinical
Oncology; College of American Pathologists. Recommendations for
human epidermal growth factor receptor 2 testing in breast cancer:
American Society of Clinical Oncology/College of American Pathologists
clinical practice guideline update. J Clin Oncol. 2013;31:3997-4013.
5. King CR, Kraus MH, Aaronson SA. Amplification of a novel v-erbB-related
gene in a human mammary carcinoma. Science. 1985;229:974-976.
6. Slamon DJ, Clark GM, Wong SG, et al. Human breast cancer: correlation
of relapse and survival with amplification of the HER-2/neu oncogene.
Science. 1987;235:177-182.
7. Bose R, Kavuri SM, Searleman AC, et al. Activating HER2 mutations in
HER2 gene amplification negative breast cancer. Cancer Discov. 2013;3:
224-237.
8. Fehm T, Becker S, Duerr-Stoerzer S, et al. Determination of HER2 status
using both serum HER2 levels and circulating tumor cells in patients
with recurrent breast cancer whose primary tumor was HER2 negative
or of unknown HER2 status. Breast Cancer Res. 2007;9:R74.
9. Yamauchi H, ONeill A, Gelman R, et al. Prediction of response to antiestrogen therapy in advanced breast cancer patients by pretreatment
circulating levels of extracellular domain of the HER-2/c-neu protein.
J Clin Oncol. 1997;15:2518-2525.
10. Regidor PA, Callies R, Schindler AE. Level of c-erbB-2 oncoprotein
in the homogenate of malignant and benign breast tumor samples.
Eur J Gynaecol Oncol. 1995;16:130-137.
11. Harigopal M, Barlow WE, Tedeschi G, et al. Multiplexed assessment

of the Southwest Oncology Group-directed Intergroup Breast
Cancer Trial S9313 by AQUA shows that both high and low levels of
HER2 are associated with poor outcome. Am J Pathol. 2010;176:
1639-1647.
12. Press MF, Hung G, Godolphin W, et al. Sensitivity of HER-2/neu antibodies in archival tissue samples: potential source of error in immunohistochemical studies of oncogene expression. Cancer Res. 1994;54:
2771-2777.
13. Harris L, Fritsche H, Mennel R, et al; American Society of Clinical Oncology. American Society of Clinical Oncology 2007 update of recommendations for the use of tumor markers in breast cancer. J Clin Oncol.
2007;25:5287-5312.
14. Hammond ME, Hayes DF, Dowsett M, et al; American Society of Clinical
Oncology; College of American Pathologists. American Society of
Clinical Oncology/College of American Pathologists guideline recommendations for immunohistochemical testing of estrogen and progesterone receptors in breast cancer (unabridged version). Arch Pathol
Lab Med. 2010;134:e48-e72.
15. Mullard A. NCI-MATCH trial pushes cancer umbrella trial paradigm. Nat
Rev Drug Discov. 2015;14:513-515.
16. Schott AF, Perou CM, Hayes DF. Genome medicine in cancer: whats in a
name? Cancer Res. 2015;75:1930-1935.
351:2817-2826.
18. Albain KS, Barlow WE, Shak S, et al; Breast Cancer Intergroup of North
America. Prognostic and predictive value of the 21-gene recurrence
score assay in postmenopausal women with node-positive, oestrogenreceptor-positive breast cancer on chemotherapy: a retrospective
analysis of a randomised trial. Lancet Oncol. 2010;11:55-65.
295
DANIEL F. HAYES
19. Sparano JA, Gray RJ, Makower DF, et al. Prospective validation of a
21-gene expression assay in breast cancer. N Engl J Med. 2015;373:
2005-2014.
20. Teutsch SM, Bradley LA, Palomaki GE, et al; EGAPP Working Group. The
Evaluation of Genomic Applications in Practice and Prevention (EGAPP)
Initiative: methods of the EGAPP Working Group. Genet Med. 2009;11:3-14.
21. Henry NL, Hayes DF. Uses and abuses of tumor markers in the diagnosis,
monitoring, and treatment of primary and metastatic breast cancer.
Oncologist. 2006;11:541-552.
22. Sargent DJ, Conley BA, Allegra C, et al. Clinical trial designs for predictive marker
validation in cancer treatment trials. J Clin Oncol. 2005;23:2020-2027.
23. Freidlin B, McShane LM, Polley MY, et al. Randomized phase II trial
designs with biomarkers. J Clin Oncol. 2012;30:3304-3309.
24. Simon RM, Paik S, Hayes DF. Use of archived specimens in evaluation of
prognostic and predictive biomarkers. J Natl Cancer Inst. 2009;101:
1446-1452.
296
25. Hayes DF, Allen J, Compton C, et al. Breaking a vicious cycle. Sci Transl
Med. 2013;5:196cm6.
26. Tefferi A, Kantarjian H, Rajkumar SV, et al. In support of a patient-driven
initiative and petition to lower the high price of cancer drugs. Mayo Clin
Proc. 2015;90:996-1000.
27. Van Poznak C, Somerfield MR, Bast RC, et al. Use of biomarkers to guide
decisions on systemic therapy for women with metastatic breast
cancer: American Society of Clinical Oncology clinical practice guideline.
J Clin Oncol. 2015;33:2695-2704.
28. Harris L, Ismaila N, McShane L, et al. Use of biomarkers to guide decisions on adjuvant systemic therapy for women with early-stage invasive breast cancer: American Society of Clinical Oncology clinical
practice guideline. J Clin Oncol. Epub 2016 Feb 8.
29. Febbo PG, Ladanyi M, Aldape KD, et al. NCCN Task Force report:
evaluating the clinical utility of tumor markers in oncology. J Natl Compr
Canc Netw. 2011;9 Suppl 5:S1-S32; quiz S33.
TUMOR BIOLOGY
Tumor Heterogeneity and

Therapeutic Resistance
CHAIR
Ravi Salgia, MD, PhD
Chicago, IL
SPEAKERS
April K.S. Salama, MD
Duke University
Durham, NC
Christine M. Lovly, MD, PhD
Nashville, TN
TUMOR HETEROGENEITY AND RESISTANCE

Christine M. Lovly, MD, PhD, April K.S. Salama, MD, and Ravi Salgia, MD, PhD
OVERVIEW
The rapidly changing landscape of oncology has brought new light, and with it, new challenges to optimizing therapeutic
strategies for patients. Although the concept of patient heterogeneity is well known to any practicing clinician, a more
detailed understanding of the biologic changes that underscore the clinical picture is beginning to emerge. Thus, tumor
heterogeneity has come to encompass more than just the clinical picture and can represent both intratumor and intertumor
differences. Within the fields of thoracic oncology and melanoma, the discovery of key molecular drivers has resulted in
landmark breakthroughs in therapy. However, the complexities of tumor genetics and the interaction within the environment continue to drive the search for better therapies. Ongoing challenges include the accurate and timely assessment of
genetic changes as well as the development of resistance and the resultant compensatory mechanisms. Novel technologies,
including commercially available next-generation sequencing, have allowed for a greater breadth and depth of information
to be gained from a single pathologic specimen, and it is now being incorporated into routine clinical practice. Translational
advances have subsequently provided valuable insight into mechanisms of resistance, with the development of novel
treatment strategies. Future work will focus on novel diagnostic techniques and adaptive mechanisms that can ultimately
drive the development of the next generation of cancer therapy.
n an address to the New Haven Medical Association in

1903, Sir William Osler, an icon of modern medicine, declared, Variability is the law of life no two individuals react
alike and behave alike under the abnormal conditions which
we know as disease.1 Certainly, all clinicians involved in the
care of patients with cancer have a deep understanding of
this sentiment. In our day-to-day practices, we are accustomed to the variability, or heterogeneity, we observe
among different patients diagnosed with the same type of
cancer. This heterogeneity comes from differences in age,
gender, and comorbid conditions as well as numerous other
factors that influence disease course and treatment decisions. Although this demographic heterogeneity is partand-parcel to the daily practice of oncologists, there is
increasing necessity to incorporate tumor heterogeneity into
our daily practice and decision-making plans.
Tumor heterogeneity is not one distinct term but rather
encompasses several facets that render tumors unique.2,3 At
the highest level, there is variability both within an individual tumor (intratumor heterogeneity) and across several different tumors (intertumor heterogeneity; Fig. 1).
Just as no two patients are the same, no two tumors are the
same, even if they carry the identical histopathologic diagnosis. Tumors arising from the same cell type in different
patients may share some common attributes (for example,
expression of a particular cell surface protein) but are not

identical. This intertumor heterogeneity is also referred to as
population heterogeneity.4
Conversely, intratumor heterogeneity refers to the presence of different cell subpopulations within a given tumor
sample. These subpopulations, which are often referred to
as tumor clones, may differ in cell morphology, genetic
makeup, metabolism, proliferation rate, and metastatic
potential.2,3,5 Indeed, intratumor heterogeneity exists at
several levels. Most commonly, intratumor heterogeneity is
equated with genetic and epigenetic variability. Recent
advances in sophisticated -omic-based technologies have
resulted in detailed analyses of tumor samples, even to the
single cell level. These types of analyses have characterized
the molecular variations between different tumor samples
(intertumor heterogeneity) and between different cells in the
same tumor, overall leading to an increasingly detailed and
complex view of tumor cell biology. For example, deep sequencing of tumor cells has revealed differences in present
mutations and genes expressed within a given tumor sample,
which have therapeutic implications for the patient.
Molecular variabilities, or tumor intrinsic properties, are
not the only source of tumor heterogeneity. Extrinsic
properties of the tumor cellincluding interactions with
surrounding cells and proximity to blood vesselscan
From the Department of Medicine, Department of Cancer Biology, Vanderbilt Ingram Cancer Center, Vanderbilt University School of Medicine, Nashville, TN; Department of Internal
Medicine, Duke University Medical Center, Durham, NC; Department of Medical Oncology and Therapeutics Research, City of Hope, Duarte, CA.
Corresponding author: Ravi Salgia, MD, PhD, City of Hope, 1500 East Duarte Rd., Duarte, CA 91010; email: rsalgia@coh.org.
e585
LOVLY, SALAMA, AND SALGIA
influence the tumor cell phenotype, leading to what is

termed positional heterogeneity.4 For example, tumor
cells can interact with other tumor cells, the extracellular
matrix, surrounding stromal cells, and immune cells.
These cell-to-cell communications affect the signals the
tumor cell receives, thereby modulating properties such
as metabolism, motility, and drug responsiveness. The
location of the tumor can also influence the availability of
nutrients, oxygen, and growth factors, all of which can
modulate tumor cell growth.2,3,5 Furthermore, temporal
heterogeneity refers to the concept that tumors are
constantly changing under various selective pressures.
Patient-specific factors, such as medications, concurrent
illnesses, nutrition, and hormonal status, as well as the
cancer therapies themselves, may influence the evolution
of a tumor.4
What is the origin of intratumor heterogeneity? Two
models have been postulated to explain this phenomenon.3,4
First, the cancer stem cell model proposes that a small
subpopulation of cells with self-renewal properties drive
tumor progression. The differentiation of these cancer stem
cells may generate the variability observed within a tumor
and may also be responsible for the repopulation of a tumor
after treatment. Second, the clonal evolution model proposes that premalignant or malignant cells accumulate
KEY POINTS
e586
Tumor heterogeneity is not one distinct term but rather

encompasses several facets that make tumors unique.
Intra-tumor heterogeneity results from variability
within an individual tumor. Intertumor heterogeneity
results from variability across different tumors from
different individuals, even within the same
histopathological diagnosis.
Intratumor heterogeneity may result from genetic and
epigenetic variability, interactions of tumor cells with
surrounding cells, proximity to blood vessels, and
presence of tumor-infiltrating lymphocytes. Each of
these properties can influence the tumor cell phenotype
and therefore lead heterogeneity.
Two models have been proposed to explain the origin of
heterogeneity, the cancer stem cell model and the clonal
evolution model. These models are not mutually
exclusive.
Advances in sophisticated omic-based technologies
have resulted in a greater understanding of intra- and
intertumor heterogeneity.
Acquired therapeutic resistance refers to genotypic
and/or phenotypic changes within the tumor that alter
drug sensitivity, the natural selection of drug tolerant
clones. Intensive research efforts are underway to
better understand acquired resistance and to develop
strategies for this important clinical problem.
Representative patient cases are presented herein to
highlight the complexity of heterogeneity in clinical
decision making.
genetic changes over time due to inherent genomic instability. These changes may confer selective advantages or
disadvantages for the cell, and, over time, the cells that
acquire advantageous genetic changes are selected in a
Darwinian-like evolutionary process. The cells continue to
accumulate genetic changes, thereby driving the diversification of the tumor and leading to the phenotypes observed
in an advanced cancerabnormal proliferation, invasion,
evasion of apoptosis, and drug resistance.
It is important to note that these models are not mutually
exclusive. Several studies have now demonstrated that clonal
evolution may occur within the cancer stem cell compartment. For example, Notta et al showed that samples obtained
at the time of diagnosis from patients with BCR-ABL lymphoblastic leukemia contain multiple genetically distinct
leukemia-initiating stem cell subclones.6
INTEGRATING THE HETEROGENEITY OF CANCER

IN CLINICAL DECISION MAKING
Intertumor and intratumor heterogeneity are important
obstacles to overcome when designing the most effective
therapeutic strategies for patients with cancer. The genotypic
and phenotypic variability of tumors can have important
consequences for diagnosis, prognosis, and treatment. However, our ability to interrogate the underlying heterogeneity
present within a given tumor sample is limited by several
factors. First, analysis of a single core biopsy sample or fine
needle aspiratetechniques that are commonly used for the
diagnosis of advanced cancersmay not be an accurate
representation of a given tumor site or of the tumor as a
whole for an individual patient. The full spectrum of distinct
clones may not be captured within one small region from a
tumor biopsy specimen.5,7 Therefore, the treatment selected
based on a single biopsy site may not result in response in all
areas of the tumor. In addition, a biopsy is only reflective of a
single moment in time within the tumors evolution, and
there is no standard way to clinically predict how a tumor will
evolve over time. A repeat biopsy of the tumor at the time of
disease progression during treatment with a given therapy to
assess changes in genetic and phenotypic makeup has not
been routine practice until recently, and, even now, this
practice is not the standard of care for all tumor types.
As diagnostic platforms advance and clinical treatment
paradigms progress, our ability to effectively tackle the problem of tumor heterogeneity increases. This clinical relevance
of this evolution is highlighted here using examples from the
fields of lung cancer and melanoma. In lung cancer, initial
trials of the EGFR tyrosine kinase inhibitors (TKIs), erlotinib
and gefitinib, in an unselected patient population yielded less
than promising results, with only a small subset of patients
who experienced a benefit, albeit dramatically. Subsequent
research efforts focused on identifying molecular markers of
response leading to the identification of activating mutations
within the EGFR tyrosine kinase domain in the tumor samples
from those patients who derived substantial antitumor
benefit.8,9 This finding took a large heterogeneous group of
FIGURE 1. Intertumor and Intratumor Heterogeneity
Intertumor heterogeneity results from variability across different tumors from different individuals, even with the same histopathologic diagnosis. Intratumor heterogeneity
results from variability within an individual tumor. Subpopulations exist within a given tumor, as represented by the different colored cells (red, blue, and gray) shown within
the tumor. These subpopulations, often referred to as tumor clones, may differ in cell morphology, genetic makeup, metabolism, proliferation rate, and, ultimately, response
to therapy.
patients with lung cancer, encompassing variability in clinical

factors and tumor histology, and narrowed down drug selection to a specific tumor genotype. Conversely, in melanoma, a relevant genetic target in BRAF was identified in
2002; approximately half of all melanomas carry an activating
mutation at amino acid position 600 (V600E),10 yet therapeutic agents with appropriate selectivity did not exist. Sorafenib, initially developed as a RAF-1 inhibitor, failed to benefit
patients with BRAF-mutated advanced melanoma.11,12 The
subsequent development of the highly selective BRAF inhibitors vemurafenib and dabrafenib resulted in meaningful
clinical benefit in large randomized studies, completely altering the treatment landscape for this subset of patients.13,14
However, even BRAF, a validated oncogenic target that has
been successfully targeted in melanoma, is clearly not onesize-fits-all. The BRAF gene is mutated in 1% to 3% of non
small cell lung cancers (NSCLCs), 10% of colorectal cancers,
50% of papillary thyroid cancers, and in varying degrees
in other malignancies.15-18 However, although early-phase
studies in lung and papillary thyroid cancers yielded promising clinical activity, responses are rarely seen in BRAFmutated colorectal cancer.19-21
ADVANCES IN GENETIC TESTING

Next-generation sequencing (NGS) has facilitated remarkable advances in terms of the breadth and depth of potential
information that can now be discerned from a single tumor biopsy. The molecular profile of lung cancer is highly
heterogeneous, and research continues to discover new

markers to characterize and target. One study evaluated the
molecular profile of patients with a broad number of cancer
histologies.22 Although this particular study characterized the
molecular profile of this cohort with great detail, only 7% of
the cohort had lung cancer. Vigneswaran et al collected and
characterized the genomic profiles of 364 patients with NSCLC
whose tumor biopsy samples were analyzed by NGS over
5 years at a single institution.23 Out of those 364 patients, 289
(75%) patients were diagnosed with adenocarcinoma. Most
of the patients (99.4%) had genomic alterations with an
average of 10.8 alterations per sample. In the molecular
profile of the adenocarcinoma cohort, 23.3% had EGFR alterations, 32.5% had KRAS alterations, 7.5% had ALK alterations, 10.8% had RET alterations, 5.0% had MET alterations,
and 0.8% had ROS1 alterations. Of the samples with KRAS
mutations, 90% were from smokers, which included current
and former smokers. Furthermore, 46% of EGFR and 61% of
ALK alterations were detected in never smokers.
Kris et al evaluated the genomic profiles of 733 patients
with lung cancer as part of the multi-institutional Lung
Cancer Mutation Consortium (LCMC) study. An oncogenic
driver mutation was found in 466 of 733 patients (64%).24
Among the 733 tumors, the distribution of known driver
mutations was 182 (25%) in KRAS, 122 (17%) TKI-sensitizing
in EGFR, 57 (8%) ALK rearrangements, 29 (4%) other EGFR
mutations, 24 (3%) with two or more concurrent alterations,
19 (3%) in ERBB2, 16 (2%) in BRAF, six (less than 1%) in
PIK3CA, five (less than 1%) MET amplifications, five (less than
e587
1%) in NRAS, one (less than 1%) in MEK1, and none in AKT1.
Results were used to select a targeted therapy or trial among
275 of 1,007 patients (28%).
Despite the numerous advantages, however, traditional
platforms still require the availability of tumor tissue, and thus
an invasive procedure, which limits the ability to obtain
multiple samples in real time. Most recently, the development and clinical implementation of sophisticated NGS
technologies that can interrogate for genomic alterations in
plasma-derived cell-free tumor DNA are being used with
greater frequency. These so-called liquid biopsies have been
shown to track the evolution of resistance mutations, such as
EGFR T790M and EGFR C797S, during therapy over a patients
disease course, thereby giving clinicians a way to monitor
tumor clones and assess genetic heterogeneity.25,26 For patients with melanoma, detection of BRAF mutations in circulating tumor DNA has been shown to be feasible and may
have some predictive value.27,28 Importantly, additional
limited and defined mutations in RAF, as well as mutations in
RAS, have been described as resistance mechanisms to
RAF/MEK inhibitors; these alterations can also be assessed
through sensitive measures of circulating cell-free tumor DNA.
Most importantly, the ability to understand the heterogeneity leading to drug resistance and the ability to monitor
this heterogeneity over time has led to the development of
more refined treatment strategies that account for the
heterogeneity, and, hopefully, it will result in better outcomes for patients.
ACQUIRED RESISTANCE
Unfortunately, even patients who derive substantial initial
benefit from targeted therapy will eventually experience
disease progression, a concept referred to as acquired
resistance. Acquired resistance simply refers to genotypic
and/or phenotypic changes within the tumor that alter drug
sensitivity, the natural selection of drug-tolerant clones.
Understanding that tumors evolve over time under selective
pressures such as drug therapy, the study of acquired resistance to EGFR TKI therapy necessitated the acquisition
of a fresh biopsy specimen at the time of disease progression. Analysis of these tumors revealed the presence of a
second site mutation within EGFR, T790M, which is attributed to the development of approximately 50% of cases of
acquired resistance to EGFR TKI therapy.29 Other pathways
also play a role in the development of resistance, including
MET, in which amplification has been shown to cause
resistance in approximately 5% of patients.30,31 Tumor
rebiopsy at the time of acquired resistance is now the accepted standard of care for patients with EGFR-mutant lung
cancer. Tumor rebiospy allows for evaluation of T790M and
several other potential resistance mechanisms, such as
transition to small cell lung cancer histology, all of which
have important therapeutic implications.
Acquired resistance to the EGFR TKIs, erlotinib and gefitinib,
can also be driven by activation of several receptor tyrosine
kinases, including MET,32,33 HER2,34 HER3,35 IGF-1R,36 and
e588
AXL.37 In addition, activation of the MAPK pathway can

drive EGFR TKI resistance in vitro and in vivo.29,30
In targeting BRAF, the picture of acquired resistance is
similarly complex, and a deeper understanding has led to
important clinical developments. Early on, the recognition
that MAPK pathway reactivation was a major cause of resistance to BRAF inhibitor therapy led to the development of
combined BRAF/MEK inhibition. However, the heterogeneity of resistance mechanisms presents a challenge in
creating a rationally designed combination strategy that will
be applicable for all patients. In one of the largest series to
date, 132 samples obtained at disease progression during
treatment with a BRAF inhibitor were analyzed using wholeexome sequencing (WES). A defined mechanism of resistance was identified in only 58%, with the most common
being NRAS or KRAS mutations. Other alterations included
BRAF splice variants, BRAF amplification, MEK1 and MEK2
mutations, and non-MAPK mechanisms. Most striking was
the marked heterogeneity in patients who had more than
one biopsy at progression; 18 of 19 patients had distinct
mechanisms identified in each sample.38 Earlier work has
identified other pathways, including PDGFR, IGF-1R, PI3K,
and EGFR, among others.12 Although dual BRAF/MEK inhibition has resulted in improved response rates and, in
some studies, improved survival for patients with BRAFmutated melanoma, progression remains a concern.39-41
Early analyses to assess for mechanisms of resistance suggest that many overlapping pathways may still play a
role.32,42 The collection of tissue in future studies will remain
critical to further elucidate additional mechanisms of resistance and to define optimal therapeutic combinations.
Assessing the importance of compensatory pathways
within the kinome may also provide additional insight into
the degree of heterogeneity seen in tumors driven by the
RAS/RAF/MAPK pathway.23,43 Inhibition of the RAS/RAF/MAPK
pathway has been shown to upregulate many receptor tyrosine kinase pathways, particularly the PDGFR, VEGF, and
EGFR pathways. Profiling of compensatory pathways has been
studied in only a limited fashion to date among patients
with cancer.44-46 Relatedly, RAS/RAF/MEK inhibition has been
shown to modulate the host immune response, including
alteration of T-cell responses and downregulation of mediators such as PD-L1, IL-1, and IL-8.47,48 These factors can directly
modulate tumor heterogeneity, in part by modulating the
micro-environment and immune checkpoint expression.47,48
In one of the few reports on this topic, changes in the inflammatory mediators IGFBP3 and PDGF-BB were correlated
with benefit from trametinib in a randomized phase II study of
this agent in advanced pancreatic cancer.49
REPORTS
Here, we highlight representative examples from our own
clinical practice in which patients either have well-documented
genomic alterations and are receiving straightforward treatment, or uncharacterized alterations, which demand a multifaceted approach.
Case 1
A 68-year-old white male who was a never-smoker initially
presented with early-stage T1N0MX NSCLC and underwent right lower lobe resection with no adjuvant
therapy (Fig. 2). Thereafter, he was monitored as an
outpatient with chest x-rays performed every 6 months.
He did well for 5 years with surveillance, but then suddenly
developed chest pain and dyspnea. A CT scan of the chest
showed multiple enlarged mediastinal lymph nodes and a
5-mm right lower lobe lung nodule. A subsequent bronchoscopy with biopsy was positive for NSCLC. A PET scan
showed metastatic disease in the neck, chest, abdomen,
and pelvis. Next-generation sequencing analysis of the
tumor biopsy sample revealed a PRKAR1A-ALK fusion. As a
result, the patient was treated with the ALK TKI, crizotinib
(250 mg orally), and he had a partial response (PR) by
Response Evaluation Criteria in Solid Tumors (RECIST)
criteria.
Issues for consideration. As NGS has become common in
our analysis of metastatic lung cancer, we can identify
known and novel genetic alterations. This patient responded
well to a known TKI (crizotinib) that targets the ALK translocation. Of note for this case, the patients tumor did not
harbor an EML4-ALK translocation, which is typically the
ALK translocation found in lung cancer. Other ALK fusions
partners, such as PRKAR1A, have been described, albeit with
decreased frequency compared with EML4-ALK translocations in lung cancer. Nonetheless, the PRKAR1A-ALK
translocation was therapeutically targetable with the

ALK-targeted TKI, crizotinib.
Case 2
A 67-year-old black male who was a current smoker was
diagnosed with lung adenocarcinoma (Fig. 3). The patient
was lost for follow-up until more than a 1 year later.
Thirteen months after his initial diagnosis, the patient
underwent an MRI, which showed laminar necrosis and
encephalomacia. A PET scan performed on the same day
also showed hypermetabolic tumor in the anterior mediastinum, right supraclavicular, and the right axillary lymph
node. A CT of the head was done a week later and showed
no metastatic lesions in the brain. The patient then began
to receive treatment with carboplatin and paclitaxel for
4 weeks. Next-generation sequencing analysis of the
patients tumor sample revealed 24 genetic alterations
and 79 variants of unknown significance, one of which was
the EGFR R1068* mutation. There was no data to suggest that the EGFR R1068* mutation was sensitive to
EGFR TKI therapy. The patient was initially treated with
60 Gy of radiation to the right lung, chest wall, and mediastinum over the course of 6 weeks. During that time, results
received from the VeriStrat test (which is a proteomic test
designed to predict benefit from EGFR TKI therapy) demonstrated a VeriStrat good score, indicating a favorable
tumor signature for response to EGFR TKI therapy. Shortly
thereafter, the patient began erlotinib therapy.
FIGURE 2. Overview of the Timeline From Diagnosis to the Development of Metastatic Disease for a Patient With
Lung Cancer Harboring an ALK Translocation
e589
Lung Cancer With a Number of Genetic Alterations
Issues for consideration. Even with the most available tools

for precision medicine, we have still a long way to go for
therapeutic decision making. Our traditional therapies are
still of value for better outcomes for complex patients. As
our patients respond to therapy and therefore survive
longer, we have to consistently consider repeat analysis of
tumor burden, the nature of genetic/proteomic changes,
and further therapies.
Case 3
A 58-year-old woman presented to discuss treatment
options for metastatic melanoma (Fig. 4). She had initially
been diagnosed with a melanoma on the left shoulder
(Breslow 3.2 mm, Clark level IV with ulceration) 4 years
prior. After her initial wide local excision and sentinel
lymph node biopsy, which was negative, she was observed clinically with routine dermatologic examinations.
She was in her usual state of health until she presented
with dyspepsia and abdominal pain. A CT scan of the
chest, abdomen, and pelvis performed at that time
showed multiple hepatic masses, with no other sites of
disease. An MRI of the brain was negative for intracranial
metastases. An ultrasound-guided liver biopsy was positive
for metastatic melanoma. She was initially treated with
combined ipilimumab (CTLA-4 inhibitor) and nivolumab
(PD-1 inhibitor). After three cycles of therapy, she developed worsening symptoms, including abdominal pain
and decreased appetite. A CT scan indicated disease progression with multiple new and enlarging hepatic metastases. Next-generation sequencing was also performed
e590
and was negative for BRAF, KIT, and NRAS mutations;

however, an MEK K57N mutation was found. Because there
was no agent approved by the U.S. Food and Drug Administration (FDA) that targeted this specific gene mutation, she
elected to undergo screening for an early-phase clinical trial.
Issues for consideration. This case highlights several important factors regarding the novelty and complexity of
information obtained from tumor NGS assays for the
practicing clinician. In melanoma, selective inhibitors are
readily available for more common mutations such as
BRAF, and clinical trials show evidence of clinical activity
with agents targeted to KIT or NRAS.43,44 However, we
have little data to guide us regarding less common variants. MEK K57N is a rare mutation that has previously
been identified in melanomas.45 It has also been identified in
other tumor types including NSCLC and neuroblastoma.46-48
The K57N mutation occurs in the nonkinase portion of the
protein, and preclinical data suggest that this variant can
induce MAPK pathway upregulation that may be potentially sensitive to MEK inhibition.46 Pursuing off-label
treatment with a commercially available MEK inhibitor
would be one option, although in our experience this
can result in substantial delays in the administration of
therapy, primarily related to coverage given the high cost
of these agents. Newer trial designs, centered around
specific genetic abnormalities (as opposed to a specific
disease subtype), such as the NCI-Molecular Analysis for
Therapy Choice (NCI-MATCH, NCT02465060) may allow
for more therapeutic options for patients with rare
alterations.
Metastatic Melanoma Harboring a MEKK57N Mutation
FUTURE RESEARCH DIRECTIONS FOR ASSESSING

HETEROGENEITY: FRACTALS AND CHAOS
Fractals are mathematical constructs that show selfsimilarity over a range of scales and noninteger fractal dimensions. The important feature of fractals within
the context of lung cancer heterogeneity is that small
differences in initial conditions can yield widely diverging
outcomes by which the cancer cells grow and mutate in
seemingly random patterns. This process of deterministic
nature and unpredictability can be studied mathematically
through the use of chaos theory, where behavior of a
system can be predicted in principle. One such method is
through the use of fractal dimensions, which aim to calculate the changes in shape and surface area, and offer
another alternative to liquid biopsies by observing the
irregular patterns present in lung cancer growth over
space and time. By measuring the changes in fractal dimension and lacunarity (texture) of lung cancer growth
and regression, we hope to create a prognostic model that
predicts the rate of growth and the change in the shape of
the tumor. In one analysis, Lennon et al (also a good
reference article for the topic presented here) evaluated
the fractal analysis of lung cancer histology by analyzing
representative hematoxylin and eosinstained tissue
slides for normal lung and common lung cancer histologies
(adenocarcinoma, large-cell carcinoma, and small cell
carcinoma).49 The fractal dimension and lacunarity of
the lung tissue samples was then calculated using imaging
software. All of the lung tumor tissues exhibited an increase in fractal dimension and a decrease in lacunarity
showing marked clusters. In contrast, normal lung tissue
showed stable levels of fractal dimension and lacunarity. This

type of image analysis, together with chromatin, cellular, and
radiologic analyses, could be useful in clinical diagnosis and
classification of patients with lung cancer. Thus, fractal dimension and lacunarity of lung cancer, as measured in radiologic images or even pathologic slides, could someday
potentially be used as biomarkers. Ultimately, heterogeneity
can be measured at the genomic, proteomic, cellular,
organoid, and metastatic levels. Unique biomarkers can be
developed, as well as potential for therapeutics.
CONCLUSION
An increased understanding of the genetic underpinnings
of cancer over the past several decades has led to remarkable advances in the field of oncology, affecting everything from the way cancers are classified to the way they
are treated. A view that one gene, even a powerful oncogenic driver, would have the same effect across diseases or
that one type of treatment will produce predictable results
has long been a vestige of the past. Tumor heterogeneity has
many facetswithin a disease, within a patient, even within
the tumor itself. With this evolution, however, new questions have emerged regarding the complex interactions of
these systems and the development of resistance to therapy. Careful collaboration will be required to further elucidate these processes, with the hope that this will translate
into better outcomes for patients.
ACKNOWLEDGMENT
Dr. Lovly and Dr. Salama contributed equally to this article.
e591
References
1. Osler W. On the educational value of the medical society. Yale Medical
Journal. 1903;9:325.
2. Holzel M, Bovier A, Tuting T. Plasticity of tumour and immune cells:
a source of heterogeneity and a cause for therapy resistance? Nat Rev
Cancer. 2013;13:365-376.
3. Michor F, Polyak K. The origins and implications of intratumor heterogeneity. Cancer Prev Res (Phila). 2010;3:1361-1364.
4. Welch DR. Tumor heterogeneity-A contemporary concept founded on
historical insights and predictions. Cancer Res. 2016;76:4-6.
5. Zellmer VR, Zhang S. Evolving concepts of tumor heterogeneity. Cell
Biosci. 2014;4:69.
6. Notta F, Mullighan CG, Wang JC, et al. Evolution of human BCR-ABL1
lymphoblastic leukaemia-initiating cells. Nature. 2001;469:362-367.
7. Andor N, Graham TA, Jansen M, et al. Pan-cancer analysis of the extent and
consequences of intratumor heterogeneity. Nat Med. 2016;22:105-113.
8. Lynch TJ, Bell DW, Sordella R, et al. Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-smallcell lung cancer to gefitinib. N Engl J Med. 2004;350:2129-2139.
9. Hanna N, Shepherd FA, Fossella FV, et al. Randomized phase III trial of
pemetrexed versus docetaxel in patients with non-small-cell lung cancer
previously treated with chemotherapy. J Clin Oncol. 2004;22:1589-1597.
10. Davies H, Bignell GR, Cox C, et al. Mutations of the BRAF gene in human
cancer. Nature. 2002;417:949-954.
11. Hauschild A, Agarwala SS, Trefzer U, et al. Results of a phase III, randomized, placebo-controlled study of sorafenib in combination with
carboplatin and paclitaxel as second-line treatment in patients with
unresectable stage III or stage IV melanoma. J Clin Oncol. 2009;27:
2823-2830.
12. Salama AK, Flaherty KT. BRAF in melanoma: current strategies and
future directions. Clin Cancer Res. 2013;19:4326-4334.
13. Hauschild A, Grob JJ, Demidov LV, et al. Dabrafenib in BRAF-mutated
14. Chapman PB, Hauschild A, Robert C, et al. Updated overall survival (OS)
results for BRIM-3, a phase III randomized, open-label, multicenter trial
comparing BRAF inhibitor vemurafenib (vem) with dacarbazine (DTIC)
in previously untreated patients with BRAFV600E-mutated melanoma.
15. Xing M. BRAF mutation in thyroid cancer. Endocr Relat Cancer. 2005;12:
245-262.
16. Nagasaka T, Sasamoto H, Notohara K, et al. Colorectal cancer with
mutation in BRAF, KRAS, and wild-type with respect to both oncogenes
showing different patterns of DNA methylation. J Clin Oncol. 2004;22:
4584-4594.
17. Tiacci E, Trifonov V, Schiavoni G, et al. BRAF mutations in hairy-cell
leukemia. N Engl J Med. 2011;364:2305-2315.
18. Chen D, Zhang LQ, Huang JF, et al. BRAF mutations in patients with nonsmall cell lung cancer: a systematic review and meta-analysis. PLoS One.
2014;9:e101354.
19. Planchard D, Kim TM, Mazieres J, et al. Dabrafenib in patients with BRAF
V600E-mutant advanced non-small cell lung cancer (NSCLC): a multicenter, open-label, phase II trial (BRF113928). Ann Oncol. 2014;25
(suppl 4; abstr LBA38_PR).
20. Falchook GS, Millward M, Hong D, et al. BRAF inhibitor dabrafenib in
patients with metastatic BRAF-mutant thyroid cancer. Thyroid. 2015;
25:71-77.
21. Kopetz S, Desai J, Chan E, et al. Phase II pilot study of vemurafenib in
patients with metastatic BRAF-mutated colorectal cancer. J Clin Oncol.
2015;33:4032-4038.
e592
22. Johnson DB, Dahlman KH, Knol J, et al. Enabling a genetically informed
approach to cancer medicine: a retrospective evaluation of the impact
of comprehensive tumor profiling using a targeted next-generation
sequencing panel. Oncologist. 2014;19:616-622.
23. Vigneswaran J, Tan Y, Murgu S, et al. Comprehensive genetic testing
identifies targetable genomic alterations in most patients with nonsmall cell lung cancer, specifically adenocarcinoma, single institute
investigation. Oncotarget. Epub 2016 Feb 26.
24. Kris MG, Johnson BE, Berry LD, et al. Using multiplexed assays of
oncogenic drivers in lung cancers to select targeted drugs. JAMA.
2014;311:1998-2006.
25. Piotrowska Z, Niederst MJ, Karlovich CA, et al. Heterogeneity underlies
the emergence of EGFRT790 wild-type clones following treatment of
T790M-positive cancers with a third-generation EGFR inhibitor. Cancer
Discov. 2015;5:713-722.
26. Thress KS, Paweletz CP, Felip E, et al. Acquired EGFR C797S mutation
mediates resistance to AZD9291 in non-small cell lung cancer harboring
EGFR T790M. Nat Med. 2015;21:560-562.
27. Shinozaki M, ODay SJ, Kitago M, et al. Utility of circulating B-RAF DNA
mutation in serum for monitoring melanoma patients receiving biochemotherapy. Clin Cancer Res. 2007;13:2068-2074.
28. Board RE, Ellison G, Orr MC, et al. Detection of BRAF mutations in
the tumour and serum of patients enrolled in the AZD6244 (ARRY142886) advanced melanoma phase II study. Br J Cancer. 2009;101:
1724-1730.
29. Pao W, Miller VA, Politi KA, et al. Acquired resistance of lung adenocarcinomas to gefitinib or erlotinib is associated with a second mutation
in the EGFR kinase domain. PLoS Med. 2005;2:e73.
30. Sequist LV, Waltman BA, Dias-Santagata D, et al. Genotypic and
histological evolution of lung cancers acquiring resistance to EGFR
inhibitors. Sci Transl Med. 2011;3:75ra26.
31. Yu HA, Arcila ME, Rekhtman N, et al. Analysis of tumor specimens at the
time of acquired resistance to EGFR-TKI therapy in 155 patients with
EGFR-mutant lung cancers. Clin Cancer Res. 2013;19:2240-2247.
32. Long GV, Fung C, Menzies AM, et al. Increased MAPK reactivation in
early resistance to dabrafenib/trametinib combination therapy of
BRAF-mutant metastatic melanoma. Nat Commun. 2014;5:5694.
33. Nazarian R, Shi H, Wang Q, et al. Melanomas acquire resistance to B-RAF
(V600E) inhibition by RTK or N-RAS upregulation. Nature. 2010;468:
973-977.
34. Wagle N, Emery C, Berger MF, et al. Dissecting therapeutic resistance to
RAF inhibition in melanoma by tumor genomic profiling. J Clin Oncol.
2011;29:3085-3096.
35. Liu L, Mayes PA, Eastman S, et al. The BRAF and MEK inhibitors dabrafenib and trametinib: effects on immune function and in combination
with immunomodulatory antibodies targeting PD-1, PD-L1, and CTLA-4.
Clin Cancer Res. 2015;21:1639-1651.
36. Wilmott JS, Long GV, Howle JR, et al. Selective BRAF inhibitors induce
marked T-cell infiltration into human metastatic melanoma. Clin Cancer
Res. 2012;18:1386-1394.
37. Heymach J, Tran HT, Nixon AB, et al. Circulating baseline plasma cytokines
and angiogenic factors (CAF) as markers of clinical response in the study
of trametinib (T) plus gemcitabine (G) vs placebo (P) plus gemcitabine for
patients (pts) with untreated metastatic adenocarcinoma of the pancreas
(MEK113487). J Clin Oncol. 2013;31 (suppl; abstr 4042).
38. Johnson DB, Menzies AM, Zimmer L, et al. Acquired BRAF inhibitor
resistance: A multicenter meta-analysis of the spectrum and frequencies, clinical behaviour, and phenotypic associations of resistance
mechanisms. Eur J Cancer. 2015;51:2792-2799.
39. Larkin J, Ascierto PA, Dreno B, et al. Combined vemurafenib and

cobimetinib in BRAF-mutated melanoma. N Engl J Med. 2014;371:
1867-1876.
40. Long GV, Stroyakovskiy D, Gogas H, et al. Dabrafenib and trametinib
versus dabrafenib and placebo for Val600 BRAF-mutant melanoma:
a multicentre, double-blind, phase 3 randomised controlled trial.
Lancet. 2015;386:444-451.
41. Robert C, Karaszewska B, Schachter J, et al. Improved overall survival in
melanoma with combined dabrafenib and trametinib. N Engl J Med.
2015;372:30-39.
42. Wagle N, Van Allen EM, Treacy DJ, et al. MAP kinase pathway alterations
in BRAF-mutant melanoma patients with acquired resistance to
combined RAF/MEK inhibition. Cancer Discov. 2014;4:61-68.
43. Carvajal RD, Antonescu CR, Wolchok JD, et al. KIT as a therapeutic target
in metastatic melanoma. JAMA. 2011;305:2327-2334.
44. Ascierto PA, Schadendorf D, Berking C, et al. MEK162 for patients with
advanced melanoma harbouring NRAS or Val600 BRAF mutations:
45.
46.
47.
48.
49.
a non-randomised, open-label phase 2 study. Lancet Oncol. 2013;14:

249-256.
Murugan AK, Dong J, Xie J, et al. MEK1 mutations, but not ERK2 mutations, occur in melanomas and colon carcinomas, but none in thyroid
carcinomas. Cell Cycle. 2009;8:2122-2124.
Marks JL, Gong Y, Chitale D, et al. Novel MEK1 mutation identified
by mutational analysis of epidermal growth factor receptor signaling
pathway genes in lung adenocarcinoma. Cancer Res. 2008;68:5524-5528.
Arcila ME, Drilon A, Sylvester BE, et al. MAP2K1 (MEK1) mutations
define a distinct subset of lung adenocarcinoma associated with
smoking. Clin Cancer Res. 2015;21:1935-1943.
Shukla N, Ameur N, Yilmaz I, et al. Oncogene mutation profiling of
pediatric solid tumors reveals significant subsets of embryonal rhabdomyosarcoma and neuroblastoma with mutated genes in growth
signaling pathways. Clin Cancer Res. 2012;18:748-757.
Lennon FE, Cianci GC, Cipriani NA, et al. Lung cancer-a fractal viewpoint.
Nat Rev Clin Oncol. 2015;12:664-675.
e593
TUMOR BIOLOGY
Whats Next in Cancer

Immunotherapy?
CHAIR
Vamsidhar Velcheti, MD
Cleveland Clinic
Cleveland, OH
SPEAKERS
Jamie E. Chaft, MD
Memorial Sloan Kettering Cancer Center and Weill Cornell Medical College
New York, NY
Suzanne Louise Topalian, MD
The Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins
Baltimore, MD
VELCHETI AND SCHALPER
Basic Overview of Current Immunotherapy Approaches in

Cancer
Vamsidhar Velcheti, MD, and Kurt Schalper, MD, PhD
OVERVIEW
Recent success of immunotherapy strategies such as immune checkpoint blockade in several malignancies has established
the role of immunotherapy in the treatment of cancer. Cancers use multiple mechanisms to co-opt the host-tumor immune
interactions, leading to immune evasion. Our understanding of the host-tumor interactions has evolved over the past few
years and led to various promising new therapeutic strategies. This article will focus on the basic principles of immunotherapy, novel pathways/agents, and combinatorial immunotherapies.
ver the past decade, the role of the immune system in

controlling tumorigenesis and tumor progression has
been well established. Although the role of the adaptive
(e.g., lymphocyte-mediated) immune responses has been
extensively addressed, the function of the innate immune
responses is less well understood. Accumulating evidence
shows a correlation between tumor-infiltrating lymphocytes
(TILs) in cancer tissue and favorable prognosis in various
malignancies. In particular, the presence of CD8+ cytotoxic
T cells and the ratio of CD8+ effector T cells/CD4+/forkhead
box P3+ regulatory T cells (Tregs) seems to correlate with
improved prognosis and long-term survival in many solid
tumors.1-9 T-cell antigen receptors (TCR) on T lymphocytes
engage with antigenic peptides presented on the cell surface
in the context of the major histocompatibility complex
(MHC). The TCR is a disulfide-linked membrane-anchored
heterodimeric protein complex composed of highly variable
alpha and beta chains and the invariable CD3 chain molecules.10 To respond to the myriad of possible foreign antigens, the adaptive immune system maintains a highly
diverse repertoire of TCR configurations through a process
known as somatic V(D)J recombination. Under normal circumstances, T cells with a diverse TCR repertoire circulate in
the body patrolling for evidence of foreign peptides presented in the surface of cells because of infection or cancer.10 When a T cell encounters a tumor antigen, this results
in activation, clonal proliferation/expansion, and a cytolytic
response. The innate and adaptive immune systems interact
to mediate the anticancer immune surveillance and immune editing.11 Dysfunctional tumor immune interactions
leading to immune evasion are key events in tumorigenesis
and metastasis. 12 Tumors achieve this through diverse

mechanisms leading to impaired antigen recognition or
by creating a highly immunosuppressive tumor microenvironment. Reduced tumor epitope recognition can occur
through epigenetic and post-transcriptional silencing or by
alterations in the antigen-presenting/peptide-processing
machinery. The presence of an immune-suppressive tumor microenvironment can be a consequence of diverse
factors (alone or in combination), including the enrichment
in regulatory cells (e.g., Tregs, immune inhibitory B cells, and
myeloid-derived suppressor cells [MDSCs]), upregulation of
co-inhibitory lymphocyte signals (e.g., membrane checkpoint ligands/receptors), elevated tolerogenic enzymes
(e.g., indoleamine 2,3-dioxygenase-1, arginase-1), reduced
immunoglobulin (Ig)-mediated opsonization, and the presence of a metabolically unfavorable milieu for immune cells
(Fig. 1).13
Novel anticancer immunostimulatory therapies harnessing pre-existing (ineffective) immune responses by targeting
the immune checkpoint pathways have shown remarkable clinical activity across several tumor types. However, a
majority of patients do not benefit from these agents, and
this is likely because of the complexity/multiplicity of
mechanisms in play, the heterogeneity in the immune
contexture across tumors, and varying (and perhaps dynamic) tumor immunogenicity. Thus, drugs targeting various
mechanisms of immune tolerance and combinations thereof
are being actively investigated. Most immunotherapeutic
agents currently in development could broadly be categorized into: (1) drugs targeting the tumor immune evasion via
blockade of negative regulatory signals (e.g., co-inhibitory
From the Department of Hematology and Oncology, Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH; Departments of Pathology and Medicine (Medical Oncology), Yale School
of Medicine, New Haven, CT.
Corresponding author: Vamsidhar Velcheti, MD, Cleveland Clinic, 9500 Euclid Ave., R35, Cleveland, OH 44143; email: velchev@ccf.org.
298
CURRENT IMMUNOTHERAPY APPROACHES IN CANCER
checkpoints and tolerogenic enzymes) and (2) agents that

directly stimulate immunogenic pathways (e.g., agonists
of costimulatory receptors). Additional immunostimulatory
strategies include enhancers of antigen presentation (e.g.,
vaccines), the use of exogenous recombinant cytokines,
oncolytic viruses, and cell therapies using native or modified
antigen-competent immune cells. In this study, we will
summarize the current state of these therapies and discuss
their most salient biologic features and clinical perspectives.
TARGETING IMMUNE TOLERANCE VIA

CO-INHIBITORY CHECKPOINTS
After TCR engagement with an antigen (e.g., signal 1), the
activation and mounting of a T-cell response is context
dependent and finely modulated by diverse coregulatory
signals (e.g., signal 2) ultimately deciding the response fate.
This process comprises a complex mechanism of simultaneous and overlapping co-inhibitory and costimulatory
pathways resulting in elimination of self-reactive T cells and
amplification of the antigen-specific immune responses
(Fig. 1). Several drugs targeting these pathways are currently
in clinical development, and some have received regulatory
approval by the U.S. Food and Drug Administration (FDA) for
use in various solid tumors (Table 1).
Co-inhibitory Checkpoint Pathways

CTLA-4. CTLA-4 is expressed mainly on T cells, with some
expression in other immune cells including B lymphocytes
and fibroblasts.14,15 In the priming phase of antigen presentation and following TCR-peptide complex engagement,
surface CTLA-4 acts as a negative regulatory receptor of
T cells.15 In naive T lymphocytes, CTLA-4 is expressed at a
very low level and is rapidly upregulated upon TCR engagement.15,16 The level of CTLA-4 is variable across T-cell
subtypes and is regulated by the calcium/calcineurininduced transcription factor NFATc1 (nuclear factor of activated T cells 1).17,18 Both CD4 helper and CD8 effector
T cells express CTLA-4 when activated; however, the level of
KEY POINTS
Novel anticancer immunotherapies have shown

remarkable clinical activity across several tumor types.
Tumors evade the immune system through multiple
regulatory pathways and cells.
Immune evasive pathways appear to be nonredundant
and hence the possibility of enhancing endogenous
immune responses by therapeutic blockade of these
pathways.
Majority of solid tumors do not mount an adequate
endogenous immune response, hence response from
immune checkpoint inhibitors is modest.
Understanding the mechanisms of epitope loss or
suppression of an endogenous immune response could
uncover novel therapeutic options.
expression is prominently higher in CD4 T cells. Transport

and cell-surface targeting (exocytosis and endocytosis) of
CTLA-4 is tightly regulated by complex mechanisms not
completely understood, some of them independent of T-cell
activation.19 CTLA-4 is a homolog of CD28, which is a key
costimulatory receptor on T cells, and hence it competes for
the binding of the same ligands, CD80 and CD86. However,
CTLA-4 has a higher affinity than CD28 for both ligands,
resulting in interference with the immune synapse and T-cell
inactivation.20 Therapeutic antiCTLA-4 monoclonal antibodies have shown remarkable clinical activity in advanced melanoma, and the precise mechanism of action
is not completely understood. In addition to the expected
mechanism of disrupting the CD28 activation on T cells,
antiCTLA-4 antibodies promote depletion of Tregs in the
tumor microenvironment. This effect is seem to be mediated by innate cells in the tumor expressing high levels of
FcgRIV.21
PD-1/PD-L1 axis. PD-1 is a co-inhibitory receptor expressed
on several immune cells. PD-1 is highly expressed on activated T cells, B lymphocytes, natural killer (NK) cells, and
MDSCs.22,23 In the effector phase of the immune response,
PD-1 expression is induced by TCR-antigen engagement and
by common g-chain cytokines like interleukin (IL)-2, IL-7, IL15, and IL-21. PD-1 has two known ligands, PD-L1 and PD-L2,
both of which are expressed at relatively low levels in
healthy tissues. Peripheral PD-L1 and PD-L2 expression is
believed to mediate the natural immune tolerance to avoid
tissue autoimmune responses and damage after a sustained
inflammatory response.24,25 PD-L2 has higher affinity for the
PD-1 receptor than PD-L1; when expressed in equal levels, it
outcompetes PD-L1 for PD-1 engagement.25,26 PD-L2 is
expressed in low levels compared with PD-L1; however, in
the presence of Th2 signals, PD-L2 expression is upregulated.25 The precise role of PD-L2 in cancer immunity, tolerance, and PD-1 receptor blockade is yet unclear. PD-1
engagement with its ligands leads to effector T-cell exhaustion and apoptosis.27-29 In addition, PD-1 engagement
with PD-L1 on tumors may render the tumor cells resistant to
lysis by cytotoxic T-lymphocytes (CTLs) and Fas-induced
apoptosis.30,31 PD-L1 expression/upregulation has been
documented in various tumors, including melanoma, non
small cell lung cancer (NSCLC), breast cancer, and squamous cell head and neck cancer.1,8,9,32-34 Notably, diverse studies show that tumor PD-L1 expression is
associated with increased benefit to PD-1 axis blockers,
particularly in melanoma and nonsquamous NSCLC. The
activation of key oncogenic pathways, including phosphoinositide 3-kinase (PI3K) and mitogen-activated protein kinase (MAPK), can increase PD-L1 expression in model
systems.35-37 Activating mutations in tyrosine kinases such
as EGFR and BRAF and loss of phosphatase and tensin homolog is also associated with PD-L1 expression via increased
MAPK signaling.36-38 In addition, PD-L1 is potently induced
by Th1 signals, and the majority of tumors upregulate PD-L1
in response to a proinflammatory milieu or the presence
299
FIGURE 1. Summary of Tumor Immune-Evasion Mechanisms Acting Through Antigenic Silencing and ImmuneSuppressive Microenvironment
Reprinted with permission, Cleveland Clinic Center for Medical Art & Photography 2016 All Rights Reserved.
of antitumor immune pressure. This is mediated by proinflammatory cytokines including interferon gamma (IFN-g)
and IL-4 through STAT1 and IFN regulatory factor-1.
T-cell Ig and ITIM domain. T-cell immunoglobulin and ITIM
domain (TIGIT) is a T-cell co-inhibitory receptor of the immunoglobulin (Ig) superfamily member. TIGIT is expressed
by subsets of CD4+ T cells (memory and regulatory), CD8+
T cells, and NK cells. The ligands for TIGIT are Ig-like
transmembrane cell adhesion molecules called nectins,
CD155 (poliovirus receptor [PVR]), CD112 (PVRL2), and
(lower-affinity) CD113 (PVRL3, NECTIN-3).39 The PVR also
binds to CD226, which is a costimulatory receptor on T cells
playing an important role in antiviral and antitumoral
300
responses.40,41 Upon activation by its cognate ligands, TIGIT

suppresses NK cell killing and CD4+ T-cell activation and
induces tolerance by increasing dendritic cell production of
anti-inflammatory cytokines such as IL-10.42 TIGIT is highly
expressed in tumor-infiltrating CD8+ T cells and is often
coexpressed with other co-inhibitory receptors, particularly
PD-1.43 In preclinical models, combined blockade of PD-L1
and TIGIT resulted in restoring antitumor immunity better
than blockade of PD-L1 alone.43
Lymphocyte activation gene 3. Lymphocyte activation gene
3 (LAG-3) is a T-cell co-inhibitory transmembrane receptor
belonging to the Ig superfamily. LAG-3 is expressed by activated T cells (CD4+ and CD8+) and NK cells.44,45 This receptor
TABLE 1. Drugs Targeting Pathways in Clinical Development or Approved by the U.S. Food and Drug Administration
Co-inhibitory
Checkpoint
Receptor Distribution
Ligand
CTLA-4
Activated CD8 T-cells,

regulatory T cells
CD80, CD86
PD-1
TIM-3
Mechanism of Immune
Evasion
Drugs Targeting
Tumor Type
Cytoplasmic tail interacts with Ipilimumab,

protein kinase C-M (PKC-M)
tremelimumab
and recruitment of SHP-2;
ectodomain competition
with counter receptor
(CD28)
Melanoma,
lung cancer
Activated CD8 T cells,

PD-L1, PD-L2
regulatory T cells, DCs,
NK cells, Macrophages
Recruitment of SH2-domain AntiPD-1s: nivolumab,

containing protein tyrosine
pembrolizumab,
phosphatases to the ITSM
pidilizumab; antiPDcytoplasmic region of
L1s: avelumab,
PD-1 then inhibits
atezolizumab,
downstream signals of
durvalumab
the T-cell receptor
Multiple solid
tumors
T-cells, B-cells, DCs, NK

cells, macrophages
SHP-2 phosphatases are

MBG453
recruited in an ITIMdependent manner and Lck,
CD45, CD148 at the
immune synapse to
suppress TCR signaling
Multiple solid
tumors
VISTA (also known CD11b myeloid cells, DCs, VISTAimmunoglobulin

as PD-1 homolog)
T cells
fusion protein
Cytoplasmic tail domain has JNJ-61610588

two protein kinase C
binding sites and also
proline residues that could
function as binding sites;
VISTA could potentially
function as both a receptor
and a ligand and its
signaling inhibits IL-2
production and activation
of T cells
Multiple solid
tumors
LAG-3
T cells, B cells, NK cells
MHC-II molecule
Binds to MHC-II molecule on LAG525, BMS-986016

APC and inhibiting the TCR
signaling; ectodomain
competition with counter
receptor (CD4)
Solid tumors,
CLL, NHL,
MM
TIGIT
T cells, NK cells
Nectins, CD155 (PVR),

CD112 (PVRL2), and
(lower-affinity) CD113
(PVRL3)
TIGIT binds with CD155 and N/A

CD112 with high and low
affinity, respectively;
recruitment of SHP
phosphatases that suppress
cytokine production by T
cells and by production of
IL-10 by DCs; ectodomain
competition with the
stimulatory counter
receptor (CD266)
N/A
Galectin-9
phosphatidylserine
Abbreviations: DC, dendritic cell; NK, natural killer; TIM-3, t-cell immunoglobulin and mucin domain-3; VISTA, v-domain immunoglobulin-containing suppressor of T-cell activation; IL,
interleukin; LAG-3, lymphocyte activation gene 3; TCR, T-cell receptor; CLL, chronic lymphocytic leukemia; NHL, non-Hodgkin lymphoma; MM, multiple myeloma; TIGIT, t-cell
immunoglobulin and ITIM domain; PVR, poliovirus receptor; N/A, not applicable.
binds with the MHC class II (MHC-II) molecule and also interacts with TCR/CD3 complexes on T cells, inhibiting the
calcium response to CD3 stimulation. Blockade of LAG-3
promotes T-cell activation, proliferation, and effector function.46 Regulatory T-cell activity is highly dependent on LAG-3
expression, and, thus, LAG-3 plays a crucial role in lymphocyte
homoeostasis and function.47 In addition, certain tumor types
express MHC-II, and interaction of these tumor cells with
LAG-3 activates antiapoptotic lymphocyte signals.48
V-domain Ig-containing suppressor of T-cell activation.
V-domain Ig-containing suppressor of T-cell activation (VISTA)
is another transmembrane Ig superfamily member.49 It is
expressed on T cells and CD11b+ antigen-presenting cell

(APC)/myeloid cells, but not in B lymphocytes.50,51 The
interactions and binding partners of VISTA on APCs are not
clearly understood. VISTA on APCs and soluble VISTA-Ig
fusion protein are considered as inhibitory ligands.49,52
VISTA is not known to be expressed on tumor cells;
however, it is found in MDSCs, tumor-associated macrophages, and dendritic cells. Blocking VISTA in murine
models enhances the CD4+ and CD8+ T cell proliferation and
proinflammatory cytokine production.
T-cell Ig and mucin domain-3. The T-cell Ig and mucin
domain-3 (TIM) family of co-inhibitory receptors are
301
transmembrane proteins containing a single IgV domain

followed by a variable-length mucin domain and a
cytoplasmic tail with a tyrosine-based signaling motif.53
There are three members of the TIM family (TIM-1, -3,
and -4) identified in humans, but the role of TIM-1 and -4
remain unclear. TIM-3 is expressed on IFN-gsecreting T
helper 1 CD4+ and CD8+ T cells and on T helper 17 cells.54
TIM-3 expression is upregulated by antigenic stimulation
and in response to proinflammatory cytokines. Galectin-9 is
the proposed ligand for TIM-3, and interaction with TIM-3
induces tolerance and T-cell exhaustion.55-57 Coexpression
of both TIM-3 and PD-1 can be seen in antigen-specific T cells
and promotes T-cell exhaustion that may not readily be
reversed by PD-1 inhibitors alone. In murine models,
combined treatment with antiTim-3 and antiPD-L1
resulted in considerable antitumor immune responses and
reduction in tumor size compared with either treatment
alone.58
ENHANCING ENDOGENOUS ANTITUMOR

IMMUNE RESPONSES
Immune surveillance plays a major role in tumor elimination.
Tumors overcome the prolonged dormancy/latency phase
through a process of immune editing, resulting in selective
pressure eradicating or diminishing the most immunogenic
clones. By the time of clinical manifestation, many tumors
are proficient in immune evasion and hence do not mount a
strong antitumor immune response.13 The presence of a preexisting endogenous antitumor immune response is indispensable for maintaining tumor dormancy and a key
factor in inducing immune reinvigoration using the immune
checkpoint blockers CTLA-4 and PD-1/PD-L1. This could
explain why these drugs are effective only in a fraction of
patients with advanced malignancies, and a majority of the
patients do not benefit to these agents. Several genetic
and epigenetic mechanisms are involved in antigen loss/
silencing, rendering tumors less visible to the immune
system.13 These changes impact the degree of effector T-cell
infiltration and favor an imbalance toward accumulation of
Tregs. Understanding the mechanisms of epitope loss or
suppression of an endogenous immune response could
uncover novel therapeutic options. Reciprocal to the
blockade of co-inhibitory receptors, diverse strategies are
used to directly stimulate the antitumor rejection, including
vaccines, agonistic antibodies for costimulatory receptors,
immunostimulatory cytokines, oncolytic viruses, and cell
therapies (Fig. 1). Although many of these strategies are
being clinically developed, there is often limited information
about fundamental biologic aspects of such interventions.
Therapeutic Cancer Vaccine

Antigen presentation is the first step in generating an immune response. Cancer vaccines generate and augment the
adaptive antitumor response largely by increased tumor
antigen presentation. Cancer vaccines are broadly of two
types: either active whole-cell vaccines or specific peptide
302
antigen preparations. There are two types of tumor antigens

that are considered targets for immunotherapy: tumorspecific antigens (TSAs) and tumor-associated antigens
(TAAs). TSAs are highly tumor specific and expressed only in
tumor cells, whereas TAAs are more widely expressed in
both tumor and nontumor cells. Use of TSAs can avoid offtarget autoimmune adverse events, but tumor responses
are limited by the presence of the target antigen. TAAs are
expressed in higher levels in tumors relative to nontumor
cells, but they generate weaker immune response and
higher off-target autoimmune events. APCs, like dendritic
cells or macrophages, display these vaccine antigens to
activate B and T cells, resulting in increased antitumor responses.59 In general, cancer vaccines are administered with
an adjuvant nonspecific immune stimulant to increase immune reactivity. Cancer cells present TAAs and TSAs via
HLA molecules on the cell surface. Identification and reintroduction of these peptides may amplify antitumor cytotoxic responses. Targets for TAA or TSA peptide vaccination
are identified by cloning complementary DNAs that encode
TAAs and TSAs that are recognized by CD8 T cells. Diverse
RNA expression platforms are used to identify genes that are
differentially expressed between tumor cells and normal
counterparts. Several TAAs/TSAs are potential candidates
for vaccines and are in clinical development in various tumor
types. Most of the vaccine peptide targets are restricted to
specific HLA haplotypes, and hence efficacy may be limited
selected patients. Also, the elicited antitumor responses are
restricted to the peptide target and hence may not be
sufficient for a clinically meaningful tumor response. Solid
tumors like melanoma and lung cancer have high rates of
somatic mutations and could result in multiple mutant
neoantigens. Personalized vaccine approaches targeting
specific mutant neoepitopes detected in a particular tumor
and matched with the patient HLA type are under investigation. Using a whole tumor cell vaccine may allow
dendritic cells to process and present multiple antigens and
elicit a stronger polyclonal cytotoxic T-cell response. Heat
shock proteins such as HSP-3 gp96, localized in the endoplasmic reticulum, are thought to serve as a chaperon for
peptides on their way to MHC-I and -II molecules. Allogeneic
tumor cells transfected with complementary DNA for gp96Ig have been used as vaccines.60 These cells continually
secrete heat shock protein gp96 along with its chaperoned
antigens, thereby augmenting antigen presentation and
efficiency of tumor-rejecting CD8 T cells.61 Several therapeutic cancer vaccines are currently in clinical trials alone or
in combination with other immunotherapeutic approaches.
Sipuleucel-T is the only therapeutic cancer vaccine that is
FDA approved. It is a fusion protein consisting of recombinant prostate acid phosphatase, which needs to be incubated
with the patients isolated APCs ex vivo. However, the
company producing this vaccine filed bankruptcy, likely because of the difficulties and cost of continuously developing
clinical-grade preparations. Granulocyte macrophage colonystimulating factor (GM-CSF) is used as an immune stimulant
to activate the dendritic cells ex vivo.62
Costimulatory Checkpoint Pathways on Cancer

Glucocorticoid-induced
tumor
necrosis
factor
receptor. Glucocorticoid-induced tumor necrosis factor receptor (GITR) is a member of the tumor necrosis factor receptor
superfamily. It is expressed on the surface of multiple types of
immune cells, including Tregs, effector T cells, B lymphocytes, NK
cells, and activated dendritic cells.63 GITR activation increases
the function of T cells, NK cells, APCs, and B lymphocytes.
Signaling through GITR enhances CD4+ and CD8+ T cell proliferation in conjunction with TCR stimulation and reduces Tregmediated suppression. GITR signaling may thereby enhance host
immune responses against tumor and aid in tumor rejection.64
GITR agonistic antibodies in combination with PD-1 blockade
were demonstrated to be synergistic in murine models.65
OX40. OX40 is a member of the TNFRSF. OX40 is expressed by
both CD4+ and CD8+ T cells during the antigen-priming phase,66
in response to TCR/CD3 cross-linking and in the presence of
proinflammatory cytokines in the tumor microenvironment.67
OX40 is not expressed in resting or nonactivated T cells.68 A
substantial proportion of TILs, both CD4 and CD8, express
OX40, possibly following antigen engagement with TCR.69,70
Binding of OX40 receptor to its ligand OX40L induces proliferation and activation of the effector T-cell response.71 Agonistic antibodies to OX40 promote effector T-cell response
and induce tumor regression experimental models.71
4-1BB. Similar to the other costimulatory receptors, 4-1BB
(CD137) is also a member of the TNFRSF. 4-1BB is expressed on
diverse immune cell types and transiently upregulated upon
activation in CD8 and CD4 T cells.72 In addition, 4-1BB has been
detected in dendritic cells and NK T cells.73,74 Recent results
suggest that 4-1BB agonists could reprogram Tregs into cytotoxic
CD4 T cells with antitumor activity.75 4-1BB agonists have been
demonstrated to induce tumor regression in several murine
models, and these effects were more pronounced when combined with other immunotherapy strategies such has negative
checkpoint antagonists, oncolytic viruses, and T-cell therapy.76,77
CD40. CD40 is a type I transmembrane glycoprotein belonging
to the TNFRSF. CD40 is expressed on APCs such has dendritic
cells, activated monocytes, and macrophages. In addition,
CD40 is present on endothelial and vascular smooth muscle
cells.78 CD40 binds to the CD40 ligand (CD40L, CD154), which is
primarily expressed on activated CD4 T cells. This interaction
of CD40/CD40L increases expression of other costimulatory
molecules and MHC peptides and stimulates production of
proinflammatory cytokines. The CD40/CD40L pathway has also
been proposed as a critical step in the initiation of an adaptive
immune response.79 CD40 agonists demonstrate antitumor
immune responses in murine models, and clinical activity in
humans has been noted in early clinical trials.80-82
Cytokines
Cytokines are a group of relatively small proteins (approximately 520 kDa) that play a critical role in cell signaling,
allowing immune cells to communicate and respond in a

synchronous/organized manner. They allow the communication between physically distant cells/tissues and modulate
the intensity and duration of the immune response against
target antigens. Cytokines are produced by nearly all immune
and most nonimmune stromal cells, including lymphocytes,
myeloid cells, APCs, fibroblasts, and endothelial cells. In addition, diverse cytokines can be produced by tumor cells.83
Proinflammatory cytokines can promote effector T-cell proliferation and activation. Therapeutic manipulation of the
cytokine environment is a promising approach to cancer
immunotherapy. In addition, the modulation of cytokines can
directly affect tumor cells, leading to apoptosis and inhibition
of proliferation.84,85 Among clinically used cytokines, highdose IL-2 has been approved for treatment of metastatic renal
cell carcinoma and melanoma. Several additional cytokines
are currently in clinical development.
1. IFNs: IFNs are cytokines expressed by nearly all normal
cells and have a critical role in modulating cellular
interactions with the stroma. There are three categories of IFNs (e.g., types IIII). Type I IFNs include IFN-a
and IFN-b and are the most used in anticancer
immunotherapy. Type I IFNs induce the expression of
MHC-I on tumor cells and promote maturation of
dendritic cells.86,87 Type I IFNs also activate effector
T cells, NK cells, and macrophages, induce direct tumor
cell apoptosis, and alter the tumor neovasculature.85
IFNs can also stimulate the secretion of IL-4, resulting
in activation of B cells. IFN-a was extensively investigated in clinical trials in various settings, and
pegylated recombinant IFN-a-2b is FDA approved for
adjuvant treatment in melanoma. Clinical activity of
IFN as a single agent is modest; however, it may be a
useful component in combinatorial strategies with
other immunotherapies.
2. IL-2: IL-2 and other members of the IL-2 family are
considered as T-cell growth factors. IL-2 acts via the
IL-2 receptor, a trimeric complex composed of a (CD25),
b (CD122), and g (CD132) chains.88,89 The b and g
chain are involved in intracellular signaling, whereas
the ligand-specific a chain is only involved in cytokine
binding.88 The affinity of the IL-2 interaction with
T cells depends on the chains present in the cell
surface. The a chain is inducible and is found only in
T cells. The b and g chains are expressed on T cells,
B cells, and NK cells. However, B cells and NK cells only
have an intermediate affinity IL-2 receptor. Stimulation with IL-2 induces proliferation and enhanced
cytotoxicity of NK cells and promote differentiation of
B cells.90 IL-2 preferentially activates the high-affinity
IL-2R, driving the expansion of high-affinity IL-2R
expressing cells including immunosuppressive CD4+
Tregs, which limits anticancer activity of recombinant
human IL-2 (rhIL-2; aldesleukin). The high-affinity
IL-2R is also expressed on vascular and pulmonary
303
endothelial cells, which could contribute to rhIL-2

mediated toxicity via capillary leak syndrome.91 IL-2
induces tumor regression and T- and NK-cell proliferation in patients with melanoma and renal cell
carcinoma. Moreover, objective responses are seen in
nearly 20% of patients, but they are associated with
notable grade 3 to 4 toxicities, mainly relating to
capillary leak syndrome.92 Although the use of IL-2 was
FDA approved for use in renal cell carcinoma and
melanoma nearly 2 decades ago, its use is limited by its
safety profile. A novel fusion protein comprising of
a circularly permuted IL-2 and IL-2Ra (ALKS 4230)
designed to selectively activate the intermediateaffinity IL-2R is currently in clinical development.
3. IL-15: IL-15 has structural similarity with IL-2, and both
bind the shared IL-2Rbg receptor and activate intracellular signaling by JAK/STAT, MAPK, and PI3K
pathways in T and NK cells. Unlike IL-2, IL-15 does not
engage the high-affinity IL-2Ra constitutively upregulated on Tregs.93 Therefore, IL-15 does not induce the
expansion of these immunosuppressive cells. The other
key difference of IL-15 as compared with IL-2 is that
IL-15 promotes early T-cell stimulation and activation
without causing activation-induced cell death (AICD) of
T cells.94 Studies using IL-15 bound to the soluble IL15Ra demonstrate a stronger biologic and antitumor
activity compared with unbound IL-15.95,96 Novel fusion
peptides of IL-15/IL-15Ra are in clinical development as
monotherapy and in combination with other immunotherapies, including immune checkpoint inhibitors.
4. IL-21: IL-21 has high homology with IL-2 and shares the
gc receptor and a cytokine-specific a-receptor. IL-21 is
primarily produced by activated CD4 T cells and induces T-cell proliferation and enhancement of CD8+
T cell and NK cellmediated cytotoxicity without
promoting AICD. 97,98 IL-21 agonists enhance the
in vivo CD8 T-cell proliferation and induce durable
tumor responses in murine models.99 In early clinical
trials, IL-21 agonists appear to be well tolerated and
have signs of activity.100
5. IL-7: IL-7 is a cytokine of the IL-2 family of T-cell growth
factors and signals through the gc receptor subunit
promoting survival and proliferation signals in effector
T cells.101 IL-7 is a homeostatic cytokine, induces
memory T cells, and enhances the T-cell repertoire
diversity. The receptor IL-7R is expressed on immature
B-cell progenitors, and IL-7 appears to be critical in
B-cell development. Unlike IL-2, IL-7 selectively expands
CD8 T cells over Tregs.102 IL-7 agonists are currently in
clinical development and appear to be well tolerated
and promote antitumor immune responses.103
Oncolytic virus therapy. Oncolytic viruses (OVs) are a relatively new class of cancer therapeutics with a distinct
mechanism of action. OV consist of either native or
304
engineered viruses capable of preferentially replicating in

cancer cells. OV infection induces direct oncolytic activity
and also augments the endogenous antitumor immune
response. They can also induce immunogenic cell death by
release of danger-associated molecular pattern signals such
as calreticulin, high-mobility group protein B1, and TAAs.104
OVs present potent danger signals to dendritic cells and
efficiently cross present TAAs to induce adaptive antitumor
immune response.105 The administration route of OVs
seems to be a key factor in determining the initial host
immune response. Intravenous and intra-arterial delivery
results in rapid elimination of the OV by circulating antibodies of the humoral defense. Intratumoral injection can
also result in limited OV replications because of intracellular
and microenvironmental antiviral defense responses in infected tumor cells. OV replication and tumoricidal activity
can be improved with pharmacologic agents, such as histone
deacetylase inhibitors, immune-modulating drugs, or by
altering the expression of genes that block antiviral defense
mechanisms.106,107 Several types of viruses are currently
being investigated as potential OVs, including herpesvirus,
poxvirus, picornavirus, adenovirus, paramyxovirus, parvovirus, reovirus, Newcastle disease virus, and rhabdovirus.108
To date, the only FDA-approved OV is an oncolytic herpes
simplex virus 1expressing GM-CSF (talimogene laherparepvec, or T-Vec). T-Vec is approved for use as an intratumor
injection in metastatic melanoma. Several other OVs are
currently in clinical development.
Adoptive T-cell therapy. TCR interaction with immunogenic
tumor cell-surface MHC-peptide complexes triggers a cascade
of tumor-specific immune responses resulting in tumor cell
recognition/elimination. The amount, affinity, and functionality of the T-cell clones expanded to attack the tumor are
proportional to the effectiveness in tumor elimination. The
rationale for adoptive T-cell therapy (ACT) is to artificially
enrich the amount of T cells that are able to recognize TSAs
and kill tumor cells. To this end, T cells with tumor antigen
specificity are enriched or genetically manipulated to develop
predetermined antigenic specificity and potency ex vivo. The
earliest attempts of ACT harvested TILs from surgical specimens. After isolation and selection, the TILs were expanded
in vitro using T-cell growth factors such as recombinant
IL-2.109 These autologous T-cell preparations are administered back to the patient after chemotherapy-induced
lymphodepletion aimed to eliminate Tregs and competing
mechanisms of the immune system.110 The adoptively
transferred T cells maintain specificity to tumor antigens and
recognize the antigens presented on the tumor cell surface
by the MHC-I complex.111 In early phase trials in melanoma,
ACT with TILs achieved objective responses of 50%70%,
including a few complete responses.112,113 However, TILs are
difficult to collect/expand, and the usual low frequency of
TAA-specific cells limits their use in many patients. Circulating
tumor-specific T cells (CTLs) can be collected, enriched, and
expanded ex vivo and used for ACTs.114,115 However, CTLs
with TCR specificity for TAAs are mostly of low affinity due to
close resemblance to self-antigens and natural negative

selection of high-affinity clones. Using novel molecularengineering approaches, transgenic TCRs with high-antigen
specificity for TAAs and optimized affinity can be produced.
This approach can augment the ability of transferred T cells to
induce clinical responses while maintaining high specificity for
tumor antigens.116,117 However, TAA recognition by T cells is
MHC restricted. This MHC restriction is a major limitation
because solid tumors evade immune recognition by downregulating MHC expression and TAA processing and presentation.13 In addition, allelic and haplotypic diversity of HLA
peptides among individuals is a challenge for systematic offthe-shelf production of transgenic TCR T cells for clinical use.
For effective response of transgenic T cells, it is required to
individualize production of these TCRs or create a vast library
of cells that can recognize TAAs within the MHC context of
multiple HLA types.
To overcome such limitations imposed by MHC restriction,
chimeric antigen receptors (CARs) were designed.118 CARs
are recombinant proteins comprising of the TCR exodomain,
which recognizes antigen and the cytoplasmic domain of the
TCR zeta (z) chain that mediates T-cell signaling. In addition,
CARs could be designed to include costimulatory signals and
T-cell activation moieties. These CARs can be transfected
into immune effector cells using a variety of technologies
enabling rapid ex vivo production of TAA-specific T cells.118
Unlike conventional TCRs, CARs can recognize antigens that
are unprocessed and independent of the MHC complex and
proteasomal processing. In addition, CARs can also recognize
carbohydrate, ganglioside, proteoglycan, and heavily glycosylated protein.118 Early results, particularly among patients with CD19-positive hematologic malignancies, are
encouraging, with responses over 50% in heavily pretreated
patients.119,120 However, clinical studies in solid tumors have
been largely disappointing, except one trial including 19
patients with neuroblastoma (three patients with complete
response, one patient with partial response, and one with
stable disease).120,121 Reasons for the lack of efficacy in solid
tumors include the relative absence of unique TAAs, inefficient
trafficking of CAR T cells in the tumor microenvironment, and,
most importantly, the highly immunosuppressive milieu within
the tumor bed.120 CAR gene constructs including costimulatory
molecules can enhance the antitumor activity.122,123 Newer
approaches use CAR constructs capable of generating CAR
T cells redirected for universal cytokine killing (TRUCKs). These
TRUCK T cells can release proinflammatory cytokines at the
time of target interaction, augmenting T-cell activation and
infiltration by innate immune cells to eliminate antigennegative cancer cells.124 Another recent approach is using
bispecific T-cell engager (BiTE) antibody constructs linking

onto a single polypeptide chain (e.g., the minimal binding
domains of monoclonal antibodies) for tumor-associated
surface antigens and the T-cell receptorassociated molecule
CD3. Recently, blinatumomab, a BiTE antibody targeting CD19,
was approved by the FDA for Philadelphia chromosomenegative, relapsed, or refractory B-cell precursor acute lymphoblastic leukemia. Several BiTe antibodies are currently
under clinical investigation.125 Despite their promising clinical
activity, these treatments have on- and off-target toxicities in
nontumor tissues that express the targeted antigen. It is
therefore critical to use highly TSAs for such approaches and to
optimize the affinity and specificity of the CAR and the lymphodepleting regimens used prior to ACT. Another major
complication with ACT is cytokine release syndrome (CRS). This
is a consequence of simultaneous and uncontrolled release of
several proinflammatory cytokines, including IL-6, TNF-a, and
IFN-g, secondary to unbalanced T-cell activation. Clinically, CRS
manifests with high fever, hypotension, and potential multiorgan failure, but is often self-limited or rapidly reversed
with anticytokine-directed therapies like tocilizumab (IL-6R
antagonist).126
CONCLUSION
Recent success of novel anticancer immunotherapies has led
to a new era of cancer treatment. Immunotherapies can
elicit clinically meaningful and durable responses in multiple
tumor types. Unfortunately, the response rates to immunotherapy are still modest, and, clearly, more efforts are
required to improve outcomes with these treatments.
Development of predictive biomarkers or personalized/
precision immunotherapy could help overcome some of
these limitations.
A key factor underlying the limited response seen in
diverse trials is the complexity in the host-immune tumor interactions and the existence of multiple redundant
mechanisms of tumor-mediated immune suppression. Also,
clear understanding of the fundamentals of tumor antigen
production/maintenance, antigenic evolution, and tumor
immune heterogeneity are essential. Increased efforts in
basic and translational research should also shed light on the
structural and functional regulation of most coregulatory
immune pathways and their specific role in diverse human
malignancies. Understanding the dynamic interactions between tumor cells and immune system will allow us to
personalize immunotherapies and design optimal combinatorial approaches to improve outcomes for patients with
advanced malignancies.
References
1.
Schalper KA, Brown J, Carvajal-Hausdorf D, et al. Objective measurement and clinical significance of TILs in non-small cell lung cancer.
2.
Bremnes RM, Busund LT, Kilvr TL, et al. The role of tumor
infiltrating lymphocytes in development, progression, and prognosis
of non-small cell lung cancer. J Thorac Oncol. Epub 2016 Feb 1.
305
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
306
Garca-Martnez E, Gil GL, Benito AC, et al. Tumor-infiltrating immune

cell profiles and their change after neoadjuvant chemotherapy predict
response and prognosis of breast cancer. Breast Cancer Res. 2014;16:
488.
Kroeger DR, Milne K, Nelson BH. Tumor infiltrating plasma cells are
associated with tertiary lymphoid structures, cytolytic T cell responses, and superior prognosis in ovarian cancer. Clin Cancer Res.
Epub 2016 Jan 13.
Lee HE, Chae SW, Lee YJ, et al. Prognostic implications of type and
density of tumour-infiltrating lymphocytes in gastric cancer. Br J
Cancer. 2008;99:1704-1711.
Li J, Liao Y, Ding T, et al. Tumor-infiltrating macrophages express
interleukin-25 and predict a favorable prognosis in patients with
gastric cancer after radical resection. Oncotarget. Epub 2016 Jan 31.
Polcher M, Braun M, Friedrichs N, et al. Foxp3(+) cell infiltration and
granzyme B(+)/Foxp3(+) cell ratio are associated with outcome in
neoadjuvant chemotherapy-treated ovarian carcinoma. Cancer
Immunol Immunother. 2010;59:909-919.
Schalper KA, Velcheti V, Carvajal D, et al. In situ tumor PD-L1 mRNA
expression is associated with increased TILs and better outcome in
breast carcinomas. Clin Cancer Res. 2014;20:2773-2782.
Velcheti V, Schalper KA, Carvajal DE, et al. Programmed death ligand-1
expression in non-small cell lung cancer. Lab Invest. 2014;94:107-116.
Thaxton JE, Li Z. To affinity and beyond: harnessing the T cell receptor
for cancer immunotherapy. Hum Vaccin Immunother. 2014;10:
3313-3321.
Kim R, Emi M, Tanabe K. Cancer immunoediting from immune surveillance to immune escape. Immunology. 2007;121:1-14.
Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation.
Cell. 2011;144:646-674.
Schreiber RD, Old LJ, Smyth MJ. Cancer immunoediting: integrating
immunitys roles in cancer suppression and promotion. Science. 2011;
331:1565-1570.
Quandt D, Hoff H, Rudolph M, et al. A new role of CTLA-4 on B cells in
thymus-dependent immune responses in vivo. J Immunol. 2007;179:
7316-7324.
Allison JP, Krummel MF. The yin and yang of T cell costimulation.
Science. 1995;270:932-933.
Perkins D, Wang Z, Donovan C, et al. Regulation of CTLA-4 expression
during T cell activation. J Immunol. 1996;156:4154-4159.
Gibson HM, Hedgcock CJ, Aufiero BM, et al. Induction of the CTLA-4
gene in human lymphocytes is dependent on NFAT binding the
proximal promoter. J Immunol. 2007;179:3831-3840.
Chan DV, Gibson HM, Aufiero BM, et al. Differential CTLA-4 expression
in human CD4+ versus CD8+ T cells is associated with increased NFAT1
and inhibition of CD4+ proliferation. Genes Immun. 2014;15:25-32.
Mead KI, Zheng Y, Manzotti CN, et al. Exocytosis of CTLA-4 is dependent on phospholipase D and ADP ribosylation factor-1 and
stimulated during activation of regulatory T cells. J Immunol. 2005;
174:4803-4811.
Greene JL, Leytze GM, Emswiler J, et al. Covalent dimerization of
CD28/CTLA-4 and oligomerization of CD80/CD86 regulate T cell
costimulatory interactions. J Biol Chem. 1996;271:26762-26771.
Simpson TR, Li F, Montalvo-Ortiz W, et al. Fc-dependent depletion
of tumor-infiltrating regulatory T cells co-defines the efficacy of
anti-CTLA-4 therapy against melanoma. J Exp Med. 2013;210:
1695-1710.
Good-Jacobson KL, Szumilas CG, Chen L, et al. PD-1 regulates germinal
center B cell survival and the formation and affinity of long-lived
plasma cells. Nat Immunol. 2010;11:535-542.
Keir ME, Butte MJ, Freeman GJ, et al. PD-1 and its ligands in tolerance
and immunity. Annu Rev Immunol. 2008;26:677-704.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
Liang SC, Latchman YE, Buhlmann JE, et al. Regulation of PD-1, PD-L1,
and PD-L2 expression during normal and autoimmune responses. Eur J
Immunol. 2003;33:2706-2716.
Ghiotto M, Gauthier L, Serriari N, et al. PD-L1 and PD-L2 differ in their
molecular mechanisms of interaction with PD-1. Int Immunol. 2010;
22:651-660.
Dai S, Jia R, Zhang X, et al. The PD-1/PD-Ls pathway and autoimmune
diseases. Cell Immunol. 2014;290:72-79.
Dong H, Zhu G, Tamada K, et al. B7-H1, a third member of the B7
family, co-stimulates T-cell proliferation and interleukin-10 secretion.
Nat Med. 1999;5:1365-1369.
Kinter AL, Godbout EJ, McNally JP, et al. The common gamma-chain
cytokines IL-2, IL-7, IL-15, and IL-21 induce the expression of programmed death-1 and its ligands. J Immunol. 2008;181:6738-6746.
Latchman Y, Wood CR, Chernova T, et al. PD-L2 is a second ligand for
PD-1 and inhibits T cell activation. Nat Immunol. 2001;2:261-268.
Azuma T, Yao S, Zhu G, et al. B7-H1 is a ubiquitous antiapoptotic
receptor on cancer cells. Blood. 2008;111:3635-3643.
Hirano F, Kaneko K, Tamura H, et al. Blockade of B7-H1 and PD-1 by
monoclonal antibodies potentiates cancer therapeutic immunity.
Cancer Res. 2005;65:1089-1096.
McLaughlin J, Han G, Schalper KA, et al. Quantitative assessment of
the heterogeneity of PD-L1 expression in non-small-cell lung cancer.
JAMA Oncol. 2016;2:46-54.
Schalper KA, Carvajal-Hausdorf D, McLaughlin J, et al. Clinical significance of PD-L1 protein expression on tumor-associated macrophages in lung cancer. J Immunother Cancer. 2015;3(Suppl 2):415.
Vassilakopoulou M, Avgeris M, Velcheti V, et al. Evaluation of PD-L1
expression and associated tumor-infiltrating lymphocytes in laryngeal
squamous cell carcinoma. Clin Cancer Res. 2016;22:704-713.
Lastwika KJ, Wilson W III, Li QK, et al. Control of PD-L1 expression by
oncogenic activation of the AKT-mTOR pathway in non-small cell lung
Song M, Chen D, Lu B, et al. PTEN loss increases PD-L1 protein expression and affects the correlation between PD-L1 expression and
clinical parameters in colorectal cancer. PLoS One. 2013;8:e65821.
Jiang X, Zhou J, Giobbie-Hurder A, et al. The activation of MAPK in
melanoma cells resistant to BRAF inhibition promotes PD-L1 expression that is reversible by MEK and PI3K inhibition. Clin Cancer Res.
2013;19:598-609.
Chen N, Fang W, Zhan J, et al. Upregulation of PD-L1 by EGFR activation mediates the immune escape in EGFR-Driven NSCLC: implication for optional immune targeted therapy for NSCLC patients with
EGFR mutation. J Thorac Oncol. 2015;10:910-923.
Yu X, Harden K, Gonzalez LC, et al. The surface protein TIGIT suppresses T cell activation by promoting the generation of mature
immunoregulatory dendritic cells. Nat Immunol. 2009;10:48-57.
Gilfillan S, Chan CJ, Cella M, et al. DNAM-1 promotes activation of
cytotoxic lymphocytes by nonprofessional antigen-presenting cells
and tumors. J Exp Med. 2008;205:2965-2973.
Pauken KE, Wherry EJ. TIGIT and CD226: tipping the balance between
costimulatory and coinhibitory molecules to augment the cancer
immunotherapy toolkit. Cancer Cell. 2014;26:785-787.
Chan CJ, Martinet L, Gilfillan S, et al. The receptors CD96 and CD226
oppose each other in the regulation of natural killer cell functions. Nat
Immunol. 2014;15:431-438.
Johnston RJ, Comps-Agrar L, Hackney J, et al. The immunoreceptor
TIGIT regulates antitumor and antiviral CD8(+) T cell effector function.
Cancer Cell. 2014;26:923-937.
Workman CJ, Dugger KJ, Vignali DA. Cutting edge: molecular analysis
of the negative regulatory function of lymphocyte activation gene-3.
J Immunol. 2002;169:5392-5395.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
Workman CJ, Vignali DA. Negative regulation of T cell homeostasis

by lymphocyte activation gene-3 (CD223). J Immunol. 2005;174:
688-695.
Hannier S, Triebel F. The MHC class II ligand lymphocyte activation
gene-3 is co-distributed with CD8 and CD3-TCR molecules after their
engagement by mAb or peptide-MHC class I complexes. Int Immunol.
1999;11:1745-1752.
Huang CT, Workman CJ, Flies D, et al. Role of LAG-3 in regulatory
T cells. Immunity. 2004;21:503-513.
Hemon P, Jean-Louis F, Ramgolam K, et al. MHC class II engagement by
its ligand LAG-3 (CD223) contributes to melanoma resistance to
apoptosis. J Immunol. 2011;186:5173-5183.
Wang L, Rubinstein R, Lines JL, et al. VISTA, a novel mouse Ig superfamily ligand that negatively regulates T cell responses. J Exp Med.
2011;208:577-592.
Le Mercier I, Chen W, Lines JL, et al. VISTA regulates the development
of protective antitumor immunity. Cancer Res. 2014;74:1933-1944.
Flies DB, Han X, Higuchi T, et al. Coinhibitory receptor PD-1H preferentially suppresses CD4 T cell-mediated immunity. J Clin Invest.
2014;124:1966-1975.
Lines JL, Pantazi E, Mak J, et al. VISTA is an immune checkpoint
molecule for human T cells. Cancer Res. 2014;74:1924-1932.
Meyers JH, Sabatos CA, Chakravarti S, et al. The TIM gene family
regulates autoimmune and allergic diseases. Trends Mol Med. 2005;
11:362-369.
Hastings WD, Anderson DE, Kassam N, et al. TIM-3 is expressed on
activated human CD4+ T cells and regulates Th1 and Th17 cytokines.
Eur J Immunol. 2009;39:2492-2501.
Zhu C, Anderson AC, Schubart A, et al. The Tim-3 ligand galectin-9
negatively regulates T helper type 1 immunity. Nat Immunol. 2005;6:
1245-1252.
Sharma S, Sundararajan A, Suryawanshi A, et al. T cell immunoglobulin
and mucin protein-3 (Tim-3)/Galectin-9 interaction regulates influenza A virus-specific humoral and CD8 T-cell responses. Proc Natl
Acad Sci USA. 2011;108:19001-19006.
Sakuishi K, Jayaraman P, Behar SM, et al. Emerging Tim-3 functions in
antimicrobial and tumor immunity. Trends Immunol. 2011;32:
345-349.
Sakuishi K, Apetoh L, Sullivan JM, et al. Targeting Tim-3 and PD-1
pathways to reverse T cell exhaustion and restore anti-tumor immunity. J Exp Med. 2010;207:2187-2194.
Bitton RJ. Cancer vaccines: a critical review on clinical impact. Curr
Opin Mol Ther. 2004;6:17-26.
Yamazaki K, Nguyen T, Podack ER. Cutting edge: tumor secreted heat
shock-fusion protein elicits CD8 cells for rejection. J Immunol. 1999;
163:5178-5182.
Strbo N, Garcia-Soto A, Schreiber TH, et al. Secreted heat shock
protein gp96-Ig: next-generation vaccines for cancer and infectious
diseases. Immunol Res. 2013;57:311-325.
Kantoff PW, Higano CS, Shore ND, et al; IMPACT Study Investigators.
Sipuleucel-T immunotherapy for castration-resistant prostate cancer.
N Engl J Med. 2010;363:411-422.
Ronchetti S, Zollo O, Bruscoli S, et al. GITR, a member of the TNF
receptor superfamily, is costimulatory to mouse T lymphocyte subpopulations. Eur J Immunol. 2004;34:613-622.
Shimizu J, Yamazaki S, Takahashi T, et al. Stimulation of CD25(+)CD4(+)
regulatory T cells through GITR breaks immunological self-tolerance.
Nat Immunol. 2002;3:135-142.
Lu L, Xu X, Zhang B, et al. Combined PD-1 blockade and GITR
triggering induce a potent antitumor immunity in murine cancer
models and synergizes with chemotherapeutic drugs. J Transl Med.
2014;12:36.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
Bansal-Pakala P, Halteman BS, Cheng MH, et al. Costimulation of CD8

T cell responses by OX40. J Immunol. 2004;172:4821-4825.
Cannons JL, Lau P, Ghumman B, et al. 4-1BB ligand induces cell division, sustains survival, and enhances effector function of CD4 and
CD8 T cells with similar efficacy. J Immunol. 2001;167:1313-1324.
Croft M. Control of immunity by the TNFR-related molecule OX40
(CD134). Annu Rev Immunol. 2010;28:57-78.
Petty JK, He K, Corless CL, et al. Survival in human colorectal cancer
correlates with expression of the T-cell costimulatory molecule OX-40
(CD134). Am J Surg. 2002;183:512-518.
Serghides L, Bukczynski J, Wen T, et al. Evaluation of OX40 ligand as a
costimulator of human antiviral memory CD8 T cell responses:
comparison with B7.1 and 4-1BBL. J Immunol. 2005;175:6368-6377.
Redmond WL, Ruby CE, Weinberg AD. The role of OX40-mediated costimulation in T-cell activation and survival. Crit Rev Immunol. 2009;
29:187-201.
Dawicki W, Watts TH. Expression and function of 4-1BB during CD4
versus CD8 T cell responses in vivo. Eur J Immunol. 2004;34:743-751.
Melero I, Johnston JV, Shufford WW, et al. NK1.1 cells express 4-1BB
(CDw137) costimulatory molecule and are required for tumor immunity elicited by anti-4-1BB monoclonal antibodies. Cell Immunol.
1998;190:167-172.
Lai C, August S, Behar R, et al. Characteristics of immunosuppressive
regulatory T cells in cutaneous squamous cell carcinomas and role in
metastasis. Lancet. 2015;385:S59.
Akhmetzyanova I, Zelinskyy G, Littwitz-Salomon E, et al. CD137 agonist
therapy can reprogram regulatory T cells into cytotoxic CD4+ T cells
with antitumor activity. J Immunol. 2016;196:484-492.
Bartkowiak T, Curran MA. 4-1BB agonists: multi-potent potentiators
of tumor immunity. Front Oncol. 2015;5:117.
Chen S, Lee LF, Fisher TS, et al. Combination of 4-1BB agonist and PD-1
antagonist promotes antitumor effector/memory CD8 T cells in a
poorly immunogenic tumor model. Cancer Immunol Res. 2015;3:
149-160.
van Kooten C, Banchereau J. CD40-CD40 ligand. J Leukoc Biol. 2000;67:
2-17.
Vonderheide RH, Glennie MJ. Agonistic CD40 antibodies and cancer
therapy. Clin Cancer Res. 2013;19:1035-1043.
Beatty GL, Chiorean EG, Fishman MP, et al. CD40 agonists alter tumor
stroma and show efficacy against pancreatic carcinoma in mice and
humans. Science. 2011;331:1612-1616.
Beatty GL, Torigian DA, Chiorean EG, et al. A phase I study of an agonist
CD40 monoclonal antibody (CP-870,893) in combination with gemcitabine in patients with advanced pancreatic ductal adenocarcinoma.
Clin Cancer Res. 2013;19:6286-6295.
Vonderheide RH, Flaherty KT, Khalil M, et al. Clinical activity and
immune modulation in cancer patients treated with CP-870,893, a
novel CD40 agonist monoclonal antibody. J Clin Oncol. 2007;25:
876-883.
Dranoff G. Cytokines in cancer pathogenesis and cancer therapy. Nat
Rev Cancer. 2004;4:11-22.
Lee S, Margolin K. Cytokines in cancer immunotherapy. Cancers
(Basel). 2011;3:3856-3893.
Wagner TC, Velichko S, Chesney SK, et al. Interferon receptor expression regulates the antiproliferative effects of interferons on
cancer cells and solid tumors. Int J Cancer. 2004;111:32-42.
Trepiakas R, Pedersen AE, Met O, et al. Addition of interferon-alpha
to a standard maturation cocktail induces CD38 up-regulation and
increases dendritic cell function. Vaccine. 2009;27:2213-2219.
Basham TY, Bourgeade MF, Creasey AA, et al. Interferon increases HLA
synthesis in melanoma cells: interferon-resistant and -sensitive cell
lines. Proc Natl Acad Sci USA. 1982;79:3265-3269.
307
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
105.
106.
107.
108.
308
Krieg C, Letourneau S, Pantaleo G, et al. Improved IL-2 immunotherapy

by selective stimulation of IL-2 receptors on lymphocytes and endothelial cells. Proc Natl Acad Sci USA. 2010;107:11906-11911.
Waldmann TA. The interleukin-2 receptor on normal and malignant lymphocytes. Adv Exp Med Biol. 1987;213:129-137.
Begley CG, Burton JD, Tsudo M, et al. Human B lymphocytes express the
p75 component of the interleukin 2 receptor. Leuk Res. 1990;14:263-271.
McNally A, Hill GR, Sparwasser T, et al. CD4+CD25+ regulatory T cells
control CD8+ T-cell effector differentiation by modulating IL-2 homeostasis. Proc Natl Acad Sci USA. 2011;108:7529-7534.
Fisher RI, Rosenberg SA, Fyfe G. Long-term survival update for highdose recombinant interleukin-2 in patients with renal cell carcinoma.
Cancer J Sci Am. 2000;6:S55-S57.
Liao W, Lin JX, Leonard WJ. Interleukin-2 at the crossroads of effector
responses, tolerance, and immunotherapy. Immunity. 2013;38:13-25.
Marks-Konczalik J, Dubois S, Losi JM, et al. IL-2-induced activationinduced cell death is inhibited in IL-15 transgenic mice. Proc Natl Acad
Sci USA. 2000;97:11445-11450.
Stoklasek TA, Schluns KS, Lefranois L. Combined IL-15/IL-15Ralpha
immunotherapy maximizes IL-15 activity in vivo. J Immunol. 2006;177:
6072-6080.
Epardaud M, Elpek KG, Rubinstein MP, et al. Interleukin-15/
interleukin-15R alpha complexes promote destruction of established tumors by reviving tumor-resident CD8+ T cells. Cancer Res.
2008;68:2972-2983.
Habib T, Nelson A, Kaushansky K. IL-21: a novel IL-2-family lymphokine
that modulates B, T, and natural killer cell responses. J Allergy Clin
Immunol. 2003;112:1033-1045.
Brandt K, Bulfone-Paus S, Foster DC, et al. Interleukin-21 inhibits
dendritic cell activation and maturation. Blood. 2003;102:4090-4098.
He H, Wisner P, Yang G, et al. Combined IL-21 and low-dose IL-2
therapy induces anti-tumor immunity and long-term curative effects
in a murine melanoma tumor model. J Transl Med. 2006;4:24.
Petrella TM, Tozer R, Belanger K, et al. Interleukin-21 has activity in
patients with metastatic melanoma: a phase II study. J Clin Oncol.
2012;30:3396-3401.
Schluns KS, Kieper WC, Jameson SC, et al. Interleukin-7 mediates the
homeostasis of nave and memory CD8 T cells in vivo. Nat Immunol.
2000;1:426-432.
Rosenberg SA, Sportes
` C, Ahmadzadeh M, et al. IL-7 administration to
humans leads to expansion of CD8+ and CD4+ cells but a relative
decrease of CD4+ T-regulatory cells. J Immunother. 2006;29:313-319.
Sportes
` C, Babb RR, Krumlauf MC, et al. Phase I study of recombinant
human interleukin-7 administration in subjects with refractory malignancy. Clin Cancer Res. 2010;16:727-735.
Bartlett DL, Liu Z, Sathaiah M, et al. Oncolytic viruses as therapeutic
cancer vaccines. Mol Cancer. 2013;12:103.
Chiocca EA, Rabkin SD. Oncolytic viruses and their application to
cancer immunotherapy. Cancer Immunol Res. 2014;2:295-300.
Alvarez-Breckenridge C, Kaur B, Chiocca EA. Pharmacologic and
chemical adjuvants in tumor virotherapy. Chem Rev. 2009;109:
3125-3140.
Fulci G, Breymann L, Gianni D, et al. Cyclophosphamide enhances
glioma virotherapy by inhibiting innate immune responses. Proc Natl
Acad Sci USA. 2006;103:12873-12878.
Kaufman HL, Kohlhapp FJ, Zloza A. Oncolytic viruses: a new class of
immunotherapy drugs. Nat Rev Drug Discov. 2015;14:642-662.
109. Rosenberg SA, Packard BS, Aebersold PM, et al. Use of tumorinfiltrating lymphocytes and interleukin-2 in the immunotherapy of
patients with metastatic melanoma. A preliminary report. N Engl J
Med. 1988;319:1676-1680.
110. Muranski P, Boni A, Wrzesinski C, et al. Increased intensity lymphodepletion and adoptive immunotherapyhow far can we go? Nat
Clin Pract Oncol. 2006;3:668-681.
111. Dudley ME, Wunderlich JR, Shelton TE, Even J, Rosenberg SA. Generation of tumor-infiltrating lymphocyte cultures for use in adoptive
transfer therapy for melanoma patients. J Immunother. 2003;26:
332-342.
112. Rosenberg SA, Yang JC, Sherry RM, et al. Durable complete responses
in heavily pretreated patients with metastatic melanoma using T-cell
113. Dudley ME, Yang JC, Sherry R, et al. Adoptive cell therapy for patients
with metastatic melanoma: evaluation of intensive myeloablative
chemoradiation preparative regimens. J Clin Oncol. 2008;26:
5233-5239.
114. Bollard CM, Gottschalk S, Leen AM, et al. Complete responses of
relapsed lymphoma following genetic modification of tumor-antigen
presenting cells and T-lymphocyte transfer. Blood. 2007;110:
2838-2845.
115. Straathof KC, Bollard CM, Popat U, et al. Treatment of nasopharyngeal
carcinoma with Epstein-Barr virusspecific T lymphocytes. Blood.
2005;105:1898-1904.
116. Morgan RA, Dudley ME, Wunderlich JR, et al. Cancer regression in
patients after transfer of genetically engineered lymphocytes. Science. 2006;314:126-129.
117. Robbins PF, Morgan RA, Feldman SA, et al. Tumor regression in patients with metastatic synovial cell sarcoma and melanoma using
genetically engineered lymphocytes reactive with NY-ESO-1. J Clin
Oncol. 2011;29:917-924.
118. Dotti G, Gottschalk S, Savoldo B, et al. Design and development of
therapies using chimeric antigen receptor-expressing T cells. Immunol
Rev. 2014;257:107-126.
119. Grupp SA, Kalos M, Barrett D, et al. Chimeric antigen receptormodified T cells for acute lymphoid leukemia. N Engl J Med. 2013;
368:1509-1518.
120. Dai H, Wang Y, Lu X, et al. Chimeric antigen receptors modified T-cells
for cancer therapy. J Natl Cancer Inst. 2016;108:djv439.
121. Louis CU, Savoldo B, Dotti G, et al. Antitumor activity and long-term
fate of chimeric antigen receptor-positive T cells in patients with
neuroblastoma. Blood. 2011;118:6050-6056.
122. Zhao Z, Condomines M, van der Stegen SJ, et al. Structural design of
engineered costimulation determines tumor rejection kinetics and
persistence of CAR T cells. Cancer Cell. 2015;28:415-428.
123. Curran KJ, Seinstra BA, Nikhamin Y, et al. Enhancing antitumor efficacy
of chimeric antigen receptor T cells through constitutive CD40L expression. Mol Ther. 2015;23:769-778.
124. Chmielewski M, Abken H. TRUCKs: the fourth generation of CARs.
Expert Opin Biol Ther. 2015;15:1145-1154.
125. Klinger M, Benjamin J, Kischel R, et al. Harnessing T cells to fight cancer
with BiTE antibody constructs - past developments and future directions. Immunol Rev. 2016;270:193-208.
126. Davila ML, Sauter C, Brentjens R. CD19-targeted T cells for hematologic malignancies: clinical experience to date. Cancer J. 2015;21:
470-474.
Author Index
Abramson, Vandana
Accordino, Melissa K.
Ajani, Jaffer A.
Al-Sukhun, Sana
Aldape, Kenneth D.
Alfano, Catherine M.
Anders, Carey K.
Antonarakis, Emmanuel S.
Antoniou, Antonis C.
Atlas, Jennifer
Attard, Gerhardt
Azim, Hatem A.
Back, Anthony L.
Basch, Ethan M.
Bass, Adam J.
Beltran, Himisha
Bernard, Brandon
Blanchard, Charmaine
Bleyer, Archie
Bradbury, Angela R.
Burg, Mary Ann
Cahill, Daniel P.
Chagpar, Anees
Chang, Susan M.
Cheville, Andrea L.
Chi, Dennis
Chingos, Diana T.
Chung, Arlene E.
Cook, Elise
Cowan, Renee
Davidson, Nancy E.
DeMatteo, Ronald P.
Dent, Rebecca
Dhakal, Binod
Dietz, Andreas
Domchek, Susan M.
Duska, Linda
Ellisen, Leif W.
Ersek, Jennifer L.
Figlin, Robert A.
Fisch, Michael J.
Friedman Ross, Lainie
Ganti, Apar K.
Garcia-Aguilar, Julio
Gershman, Boris
Glynne-Jones, Rob
Goffin, John R.
Gourley, Charlie
Halpern, Michael T.
Hari, Parameswaran
Hayes, Daniel F.
Hlubocky, Fay J.
34
200
104
58
75
241
34
131
44
83
131
23
271
67
104
131
119
58
185
251
231
75
18
75
241
153
67
200
185
153
23
281
34
210
176
44
164
14
83
113
200
251
176
92
119
92
223
143
231
210
292
271
.....................................................................................................................................................
............................................................................................................................................
.....................................................................................................................................................................
...................................................................................................................................................................
.......................................................................................................................................................
..............................................................................................................................................
..................................................................................................................................................................
...........................................................................................................................
..................................................................................................................................................
..........................................................................................................................................................................
.............................................................................................................................................................
.....................................................................................................................................................................
.............................................................................................................................................................
...................................................................................................................................................................
........................................................................................................................................................................
..............................................................................................................................................................
.........................................................................................................................................................
..............................................................................................................................................
.......................................................................................................................................................................
..................................................................................................................................................
................................................................................................................................................................
.....................................................................................................................................................................
.....................................................................................................................................................................
.................................................................................................................................................................
......................................................................................................................................................
................................................................................................................................................................................
...............................................................................................................................................................
.............................................................................................................................................................
..................................................................................................................................................................................
......................................................................................................................................................................
........................................................................................................................................................
..............................................................................................................................................
.........................................................................................................................................................................
.......................................................................................................................................................................
.....................................................................................................................................................................
.....................................................................................................................................................
............................................................................................................................................................................
........................................................................................................................................................................
................................................................................................................................................................
...............................................................................................................................................................
..............................................................................................................................................................
...........................................................................................................................................
.......................................................................................................................................................................
......................................................................................................................................................
..............................................................................................................................................................
...........................................................................................................................................................
....................................................................................................................................................................
...............................................................................................................................................................
...................................................................................................................................................
.................................................................................................................................................
...............................................................................................................................................................
...............................................................................................................................................................
Huang, Paul
Hutson, Thomas E.
Ison, Gwynn
Karnes, R. Jeffrey
Kehoe, Sean
Keller, Jesse
Kesserwan, Chimene
Kim, Edward S.
Kluetz, Paul G.
Kosty, Michael P.
Krishnan, Amrita
Kurian, Allison W.
Leary, Alexandra
Lee, Alex
Lee, Jeeyun
Legnini, Mark W.
Mackay, Helen J.
McCabe, Mary S.
Mehta, Minesh P.
Mitchell, Sandra A.
Morris, Michael J.
Mustian, Karen
Nankivell, Matthew
Nichols, Kim E.
Ostroff, Jamie S.
Perou, Charles
Pickard, Todd
Pollack, Seth M.
Powell, Melanie
Raghavan, Derek
Rini, Brian I.
Ruddy, Kathryn J.
Ruff, Paul
Sabatini, Mary E.
Schalper, Kurt
Schrag, Deborah
Shahrokni, Armin
Shanafelt, Tait D.
Shulman, Lawrence N.
Sledge, George W.
Sweeney, Christopher J.
Tai, Eric
Tan, Tira
Thoreson, Gregory R.
Unger, Joseph M.
Vapiwala, Neha
Velcheti, Vamsidhar
Viale, Pamela
Vij, Ravi
Walker, Joan L.
Warren, Graham W.
Yom, Sue S.
..............................................................................................................................................................................
......................................................................................................................................................
281
113
83
119
153
210
251
83
67
3
210
44
153
281
104
9
143
231
75
67
131
241
153
251
223
18
3
281
164
9
113
23
58
14
298
92
164
271
58
18
119
185
34
113
185
119
298
3
210
143
223
176
................................................................................................................................................................................
...........................................................................................................................................................
.............................................................................................................................................................................
..............................................................................................................................................................................
...............................................................................................................................................
.......................................................................................................................................................................
.........................................................................................................................................................................
....................................................................................................................................................................
.............................................................................................................................................................
.............................................................................................................................................................
.............................................................................................................................................................
.........................................................................................................................................................................................
...............................................................................................................................................................................
....................................................................................................................................................................
............................................................................................................................................................
............................................................................................................................................................
.............................................................................................................................................................
.........................................................................................................................................................
.........................................................................................................................................................
..................................................................................................................................................................
...................................................................................................................................................
....................................................................................................................................................................
..............................................................................................................................................................
.........................................................................................................................................................................
.................................................................................................................................................................................
...............................................................................................................................................................
................................................................................................................................................................
......................................................................................................................................................................
.............................................................................................................................................................................
..............................................................................................................................................................
..........................................................................................................................................................................................
................................................................................................................................................................
.......................................................................................................................................................................
..................................................................................................................................................................
...........................................................................................................................................................
...........................................................................................................................................................
.............................................................................................................................................
..........................................................................................................................................................
...................................................................................................................................
.............................................................................................................................................................................................
..............................................................................................................................................................................................
.............................................................................................................................................
..........................................................................................................................................................
.................................................................................................................................................................
..................................................................................................................................................
..................................................................................................................................................................................
............................................................................................................................................................................................
..................................................................................................................................................................
................................................................................................................................................
..............................................................................................................................................................................
asco.org/edbook | ASCO EDUCATIONAL BOOK
309
Amgen
Merck & Co., Inc.
2016 ASCO EDUCATIONAL BOOK
This publication is supported by an

educational donation provided by:
AMERICAN SOCIETY OF CLINICAL ONCOLOGY
2016 EDUCATIONAL BOOK

Support for this program is funded through
Collective Wisdom: The Future of Patient-Centered Care and Research

A PEER-REVIEWED, INDEXED PUBLICATION
52nd Annual Meeting | June 37, 2016 | Chicago, Illinois | Volume 36
Vol. 36

2016 EdBook

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2016 EdBook

Uploaded by

Copyright:

Available Formats

Amgen

Merck & Co., Inc.

2016 ASCO EDUCATIONAL BOOK

This publication is supported by an

AMERICAN SOCIETY OF CLINICAL ONCOLOGY

2016 EDUCATIONAL BOOK

Collective Wisdom: The Future of Patient-Centered Care and Research

American Society of Clinical Oncology Educational Book

2016 American Society of Clinical Oncology, Alexandria, VA

20152016 Cancer Education Committee

2016 Educational Book Expert Panel

2016 Annual Meeting Supporters

Letter From the Editor

Value in Oncology: Balance Between Quality and Cost

Womens Health Issues for BRCA Mutation Carriers

Collective Wisdom: Lobular Carcinoma of the Breast

Less Is More: The Evolving Surgical Approach to Breast Cancer

Strategies for Sustainable Cancer Care

asco.org/edbook | 2016 ASCO EDUCATIONAL BOOK

Less Is More: A Multidisciplinary Conversation on Treatment Options

Targeted Therapies in Hormone ReceptorPositive Breast Cancer

Treatment of Premenopausal Women With Endocrine-Sensitive Breast Cancer

Triple-Negative Breast Cancer: Is Change on the Horizon?

Updates and Controversies in HER2-Positive Breast Cancer

Emerging Therapeutic Options for HER2-Positive Breast Cancer

CANCER PREVENTION, HEREDITARY GENETICS, AND EPIDEMIOLOGY

Evolving Recommendations on Prostate Cancer Screening

Refining Breast Cancer Risk Assessment: Additional Genes, Additional Screening

CARE DELIVERY AND PRACTICE MANAGEMENT

2016 ASCO EDUCATIONAL BOOK | asco.org/edbook

Integrating Patient-Reported Outcomes Into Real-Life Medical Decisions

Optimizing Team-Based Oncology Care: Building the Village

Oncology Advanced Practitioners Bring Advanced Community Oncology Care

Quality and Value: Measuring and Utilizing Both in Your Practice

The Oncology Care Model: The Man Behind the Curtain

CENTRAL NERVOUS SYSTEM TUMORS

Targeted Therapy in Brain Metastases: Ready for Primetime?

The Future of Immunotherapy in Glioblastoma

Treatment of Low-Grade Gliomas in the Era of Genomic Medicine

DEVELOPMENTAL THERAPEUTICS AND TRANSLATIONAL RESEARCH

asco.org/edbook | 2016 ASCO EDUCATIONAL BOOK

Biosimilars: Here and Now

Clinical Trial Eligibility Criteria: Tradition Versus Reality

Immunotherapy: Beyond Checkpoint Inhibitors

Overcoming Therapeutic Resistance by Targeting Cancer Inflammation

GASTROINTESTINAL (COLORECTAL) CANCER

Questions on Treatment of Locally Advanced Rectal Cancer

GASTROINTESTINAL (NONCOLORECTAL) CANCER

Designing Clinical Trials to Achieve Breakthrough Results in Upper Gastrointestinal Malignancies

2016 ASCO EDUCATIONAL BOOK | asco.org/edbook

Localized Pancreatic Cancer: Multidisciplinary Management With Incorporation of ASCO Guidelines

GENITOURINARY (NONPROSTATE) CANCER

Renal Cell Carcinoma: Systemic Treatment, Evolving TKIs, and Immuno-Oncology

GENITOURINARY (PROSTATE) CANCER

Oligometastatic Disease in Prostate Cancer: Treating the Patient or the Scan?

Divide and Conquer: Epithelial Gynecologic Cancers Beyond BRCA

Intraperitoneal Chemotherapy for Ovarian Cancer: Trials and Tribulations

Neoadjuvant Chemotherapy: Location, Location, Location

Practical Challenges in Endometrial Cancer

Symptom Management for the Patient with Gynecologic Cancer

HEAD AND NECK CANCER

HEALTH SERVICES RESEARCH AND QUALITY OF CARE

Removing Barriers to Clinical Trial Participation