You are on page 1of 22

REVIEWS

Targeting the TGF signalling


pathway in disease
Rosemary J.Akhurst1 and Akiko Hata2
Abstract | Many drugs that target transforming growth factor- (TGF) signalling have
been developed, some of which have reached PhaseIII clinical trials for a number of
disease applications. Preclinical and clinical studies indicate the utility of these agents in
fibrosis and oncology, particularly in augmentation of existing cancer therapies, such as
radiation and chemotherapy, as well as in tumour vaccines. There are also reports of
specialized applications, such as the reduction of vascular symptoms of Marfan syndrome.
Here, we consider why the TGF signalling pathway is a drug target, the potential clinical
applications of TGF inhibition, the issues arising with anti-TGF therapy and how these
might be tackled using personalized approaches to dosing, monitoring of biomarkers
as well as brief and/or localized drug-dosing regimens.

Epithelialmesenchymal The transforming growth factor- (TGF) superfamily invivo and innate genetic variation among individuals59.
transition of cytokines, which consists of TGFs, activins, inhibins, This makes the pathway a particular challenge for drug
(EMT). The transformation of Nodal, bone morphogenetic proteins (BMPs), anti- development. Nevertheless, over the past decade several
a keratin-expressing epithelial
Mllerian hormone (AMH; also known as Mllerian- drugs targeting the TGF signalling pathway have been
cell into one with fibroblastic
properties that express
inhibiting factor) as well as growth and differentiation developed by pharmaceutical companies and biotechnol-
mesenchymal markers. factors (GDFs), is conserved through evolution and ogy firms alike. Drug design strategies have been numer-
found in all multicellular organisms1. The TGFs perse ous and include the development of small-molecule
Extracellular matrix
are involved in many cellular processes, including growth inhibitors (SMIs) and monoclonal antibodies, as well
(ECM). Matrix that supports
connective tissue and is inhibition, cell migration, invasion, epithelialmesenchymal as the inhibition of gene expression; some drugs have
composed of proteoglycans, transition (EMT), extracellular matrix (ECM) remodelling reached PhaseIII clinical trials for a number of disease
hyaluronic acid and fibrillar and immune-suppression2. However, although normally applications, particularly fibrosis and oncology. There is
proteins secreted from the cell dynamically regulated and involved in maintenance of an increasing number of preclinical examples of TGF
and rich in bound growth factors.
tissue homeostasis, TGFs are often chronically over- inhibitors that are capable of reducing cancer progression
Fibrosis expressed in disease states, including cancer, fibrosis and and metastasis, and that augment existing cancer therapies
The excess accumulation of inflammation, and this excessive production of TGF (such as radiation therapy in breast cancer) while simul-
fibroblasts and associated
drives disease progression by modulating cell growth, taneously guarding against radiation-induced fibrosis10.
extracellular matrix.
migration or phenotype. The TGF signalling pathway has Additionally, there are novel reports of targeting TGF
therefore become a popular target for drug development. signalling in less prevalent indications, such as reduction
1
Helen Diller Family Knowledge about cellular activities gleaned from of vascular symptoms of Marfan syndrome (MFS)11,12.
Comprehensive Cancer
Center, University of
studying one disease is often applicable to others. For Although there have been many reviews on the pleio-
California at San Francisco, example, inhibition of TGF-induced EMT a process tropic action of TGF during tumorigenesis, which is
San Francisco, California that contributes to cancer progression is a goal not characterized by tumour-suppressing activity of TGF at
94158, USA. only of oncologists but also of cardiovascular surgeons to an early stage of cancer and tumour-promoting activity
2
Cardiovascular Research
prevent neointimal hyperplasia, and of nephrologists and at later stages1316, few focus specifically on drug targets,
Institute, University of
California at San Francisco, pneumologists in the treatment of fibrosis3. In addition, the drug classes and possible therapeutic applications beyond
San Francisco, California immune-modulatory activities of TGF have implications the oncology arena. The translation of anti-TGF thera-
94158, USA. in many diseases, including cancer, cardiovascular disease, pies has been pursued most intensively for oncology;
Correspondence to R.J.A. asthma, rheumatoid arthritis and multiple sclerosis4. however, this Review also discusses the potential of the
e-mail: RAkhurst@cc.ucsf.edu
doi:10.1038/nrd3810
TGF action is highly context-dependent and influ- TGF signalling pathway as a target for non-neoplastic
Published online enced by cell type, culture conditions, interaction with disease therapies and addresses the associated challenges
24 September 2012 other signalling pathways, developmental or disease stage in the development and application of these strategies.

790 | O CTOBER 2012 | VOLUME 11 www.nature.com/reviews/drugdisc

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

The TGF family the SMAD-dependent canonical pathway (BOX1; FIG.1)


The vertebrate genome contains more than 30 pleio- and the SMAD-independent or non-canonical pathways
tropic ligands that belong to the TGF superfamily, (BOX2; FIG.2).
including TGFs, BMPs, GDFs, activins, inhibins, Nodal In the SMAD-dependent pathway, activation of TGF
and AMH1. TGF has a conserved motif of nine cysteine receptor type I (TRI; also known as TGFBR1 and ALK5)
residues, eight of which form a tight cysteine knot, with leads to phosphorylation of receptor-specific SMAD
the ninth being crucial for homodimerization2. Aberrant (RSMAD) proteins. SMAD2 and SMAD3 are substrates
expression and activity of many of the ligands of the of TRI, whereas typeI receptors for BMPs utilize SMAD1,
TGF superfamily are associated with developmental SMAD5 and SMAD8 (FIG.1). Upon phosphorylation by the
defects and human diseases17. Here we focus on TGFs as receptor, RSMADs together with the common mediator
there are currently several clinical trials underway involv- SMAD4 (coSMAD) translocate to the nucleus, where
ing therapies targeting TGF signalling, whereas other they interact with other transcription factors (cofactors)
members of the TGF superfamily are under-represented to regulate transcriptional responses27 (FIG.1). In addition to
in current trials. the canonical role of SMADs as transcription factors,
Three highly homologous isoforms of TGF exist in a novel role for RSMADs in the post-transcriptional
humans: TGF1, TGF2 and TGF3. They share a recep- regulation of microRNA (miRNA) biogenesis has been
tor complex and signal in similar ways but their expression identified28 (FIG. 1). Therefore, the canonical TGF
levels vary depending on the tissue18, and their func- SMAD pathway modulates gene expression both trans
tions are distinct as demonstrated by the phenotypes of criptionally and post-transcriptionally to propagate the
knockout mice1923. Each TGF ligand is synthesized as a physiological and pathological activities of TGF. In the
precursor, which forms a homodimer that interacts with non-canonical pathway, the activated TGF receptor
its latency-associated peptide (LAP) and a latent TGF- complex transmits a signal through other factors, such
binding protein (LTBP), forming a larger complex called as tumour necrosis factor (TNF) receptor-associated
the large latent complex (LLC). The TGF activation factor 4 (TRAF4), TRAF6, TGFactivated kinase 1
process involves the release of the LLC from the ECM, (TAK1; also known as MAP3K7), p38 mitogen-activated
followed by further proteolysis of LAP to release active protein kinase (p38 MAPK), RHO, phosphoinositide
TGF to its receptors2. Matrix metalloproteinase 2 (MMP2) 3kinase (PI3K), AKT (also known as protein kinaseB),
and MMP9 are known to cleave latent TGF. In addition extracellular signal-regulated kinase (ERK), JUN
to MMPs, thrombospondin 1 (THBS1) is known to acti- Nterminal kinase (JNK) or nuclear factor-B (NFB)
vate latent TGF24. Alternatively, upon mechanical stretch, (BOX2; FIG.2). Thus, cellular responses to TGF signal-
V6 integrin can activate TGF by binding to the RGD ling result from the dynamic combination of canonical
motif present in LAP and inducing the release of mature and non-canonical signalling cascades. In addition to the
TGF from its latent complex25,26. complexity generated by the canonical and non-canonical
TGF signalling pathway, TGF signalling can be influ-
TGF signalling enced by different signalling pathways, including the
Proteolytic cleavage, interaction with integrins or pH PI3KAKT, WNT, Hedgehog (HH), Notch, interferon
changes in the local environment are known to acti- (IFN), TNF and RAS pathways (BOX2; FIG.2). Interactions
vate latent TGF and free active TGF for binding to its with several of these pathways can change the output of
receptors at the cell membrane. TGF superfamily mem- TGF signalling from suppressing growth to inducing
bers signal via heteromeric complexes of two related cellular plasticity 29. Nuclear accumulation and transcrip-
Metastasis
The dissemination of tumour
transmembrane typeI and typeII serine/threonine tional activity of RSMADs can also be negatively regulated
cells and re-establishment of kinase receptors. Five typeII receptors and seven typeI through phosphorylation of multiple Ser-Pro and Thr-Pro
tumours at a secondary site. receptors (also termed activin receptor-like kinases residues (in the linker region connecting the MH1 and
(ALKs)) have been identified. Auxilliary coreceptors MH2 domains) by ERK, MAPKs, calcium/calmodulin-
SMAD
(also known as typeIII receptors) that regulate the access dependent protein kinase II and cyclin-dependent kinases
Signal transduction
component of the canonical of TGF superfamily members to signalling receptors (CDKs)30. The mode and outcome of the crosstalk between
transforming growth factor- also exist. Each subfamily of the TGF superfamily of TGF and other signalling pathways vary considerably but
signalling pathway. ligands binds to typeI and typeII receptors (BOX1). are essential to define the activities of TGF in propagating
BMPs can bind to typeI receptors alone and, in their spatially and temporally specific outputs6,31,32.
microRNA
(miRNA). Small (2023
absence, can weakly bind to typeII receptors, but they
nucleotides long) non-coding show highest affinity when both receptors act together. Biological actions of TGF
RNA involved in post- TGF and activin display high affinity only for typeII TGF is involved in a range of biological processes both
translational regulation of receptors and do not normally interact with isolated during embryogenesis and in adult tissue homeostasis.
gene expression. miRNAs
typeI receptors. Binding to the extracellular domains Although the physiological roles of TGF have been
bind to the partially
complementary sequence of typeI and typeII receptors by the dimeric ligand extensively reviewed elsewhere16,3336, the major functions
in the 3-untranslated region induces close proximity and a productive conformation of TGF that are relevant to the topic of this Review are
(3-UTR) of mRNAs and for the intracellular serine/threonine kinase domains briefly outlinedbelow.
negatively regulate their of the receptors, facilitating the phosphorylation and
expression either through
translational inhibition
subsequent activation of the typeI receptor. The activa- Inhibition of cell proliferation. TGF strongly inhibits the
or promotion of mRNA tion of the typeI receptor leads to the propagation of growth of many cell types, including epithelial, endothe-
degradation. signalling by at least two seemingly independent routes: lial, haematopoietic and immune cells37,38. TGF also has

NATURE REVIEWS | DRUG DISCOVERY VOLUME 11 | O CTOBER 2012 | 791

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

Box 1 | Canonical signal transduction pathway of the TGF superfamily of growth factors
The basic framework of the canonical signal transduction pathways of three subfamilies of the transforming growth
factor- (TGF) superfamily TGFs, activins/inhibins/Nodal and bone morphogenetic proteins (BMPs) is highly
conserved. The ligand binds to a specific set of typeI and typeII receptors, which are both serine/threonine kinases,
followed by signal transduction by SMAD proteins31,205. Although each subfamily transmits the signal through a specific
signalling pathway, the interaction among the TGF, activins/inhibins/Nodal and BMP subfamilies is well recognized during
development and in postnatal homeostasis of various organs (BOX3). Upon ligand binding and resultant heterotetrameric
receptor complex formation, the constitutively active typeII receptor phosphorylates the typeI receptor, which in turn
propagates a signal by phosphorylating the receptor-specific SMADs (RSMADs)31,205. Unlike typeI and typeII receptors,
typeIII receptors do not possess kinase activity and are not required for signal transduction; however, they bind to specific
ligands and modulate the signalling pathway either positively or negatively31,205. Phosphorylation of RSMADs at two serine
residues within the extreme carboxyl terminus by typeI receptor kinase activity promotes association with the common
mediator SMAD (coSMAD), SMAD4, resulting in nuclear accumulation and sequence-specific binding to DNA in concert
with other DNA-binding transcription factors that bind distinct sequences adjacent to the SMAD-binding element (SBE)27,
and together these complexes modulate transcription. The inhibitory SMADs (ISMADs), SMAD6 and SMAD7, antagonize
RSMAD activation by competing with RSMADs for typeI receptor interaction and/or by recruiting specific ubiquitin
ligases or phosphatases to the activated receptor complex, thereby targeting it for proteasomal degradation or
dephosphorylation, respectively. SMAD7 inhibits signalling from all branches of the TGF superfamily, whereas SMAD6 is a
specific inhibitor of the BMP signalling pathway. The table indicates the basic molecules in the signal transduction pathway,
including three types of receptors and SMADs, for three subfamilies of the TGF superfamily of ligands: TGFs, activins/
inhibins/Nodal and BMPs.
Molecular TGF pathway* Activin/inhibin/Nodal BMP pathway*
category pathway*
Ligands TGF1, TGF2, TGF3 Activin A, activin B, inhibin A, BMP2, BMP4, BMP5, BMP6, BMP7,
inhibin B, Nodal BMP8A, BMP8B, BMP9, BMP10
TypeI TRI (ALK5), ALK1 ALK4 (ACVR1B or ACTRIB), ALK1 (ACVRL1, SKR3), ALK2 (ACVR1,
receptors (ACVRL1 or SKR3) ALK7 (ACVR1C or ACTRIC) ACTRI), ALK3 (BMPR1A), ALK6 (BMPR1B)
TypeII TRII ACTRIIA, ACTRIIB BMPR2, ACTRIIA, ACTRIIB
receptors
TypeIII TRIII (betaglycan), CRIPTO1 (TDGF1), RGMA, RGMB (DRAGON), RGMC
receptors endoglin, CRIPTO3 CRIPTO3 (TDGF1P3), TRIII (HJV or HFE2), endoglin
(TDGF1P3) (betaglycan)
RSMADs SMAD2, SMAD3 SMAD2, SMAD3 SMAD1, SMAD5, SMAD8
CoSMAD SMAD4 SMAD4 SMAD4
ISMADs SMAD7 SMAD7 SMAD6, SMAD7
*Alternative protein names are listed in brackets. ACTR, activin receptor; ALK, activin receptor-like kinase; BMP, bone morphogenetic
protein; BMPR, BMP receptor; RGM, repulsive guidance molecule; TR, TGF receptor; TDGF, teratocarcinoma-derived growth factor.

pro-apoptotic and differentiation-inducing actions on vascular disorders, such as atherosclerosis, hypertension


epithelial cells; together, these actions result in tumour and pulmonary hypertension, often find signatures of
suppression in the context of cancer 34. TGF in epithelial both upregulation and downregulation of TGF signal-
cells activates transcription of cyclin-dependent kinase ling, as well as complex interactions between this path-
inhibitor1A (CDKN1A) and CDKN2A (which encode way and other ligands of the TGF family, such as BMPs
p21CIP1 and p15INK4B, respectively) to mediate cell cycle (BOX3). This has been confirmed by invitro studies,
arrest at the G1 phase39. Conversely, TGF represses the demonstrating the contradictory effects of TGF in the
transcription of MYC, which encodes a potent transcrip- regulation of vascular cells36,41. Furthermore, the TGF
tional activator of genes that is required for cell prolifer pathway often exhibits contrasting effects in different
ation and growth, and inhibitor of DNA binding (ID) vascular cell types, such as endothelial versus vascular
family genes, which encode transcription factors that pro- smooth muscle cells36. The promiscuous and cell type-
mote cell differentiation and determination40. In oncology, specific action of the TGF pathway on vascular cells
many tumours attenuate TGF growth-inhibitory effects makes the application of targeted TGF signalling thera-
but respond to this ligand in a pro-tumorigenic manner. pies for cardiovascular disease a particular challenge.
Thus, depending on the tumour type and the stage of
tumour progression, TGF may provide potent tumour- Induction of epithelialmesenchymal transition and the
suppressive or tumour-promoting functions directly on myofibroblast phenotype. TGF can induce an EMT
the tumour cell, presumably by mediating differential of both epithelial and endothelial cells. This has con-
gene expression programmes (FIG.3). sequences for disease progression in both cancer and
Unlike the role of TGF signalling during tumorigen- fibrosis3. EMT enhances cellular migration and invasive
esis, the contribution of TGF to vascular disease is more properties, as cell migration requires loss of cellcell
complex and confusing. Studies on clinical samples from contacts and acquisition of fibroblastic characteristics.

792 | O CTOBER 2012 | VOLUME 11 www.nature.com/reviews/drugdisc

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

TGF cell EMT contributes to cancer progression as cells con-


Latent sequently become more migratory and invasive, and they
TGF
can ultimately transition to a myofibroblastic phenotype3.
The myofibroblast further modulates the basic biology of
P P the tumour by increasing ECM elaboration and eliciting
TRI Cytoplasm a tissue contraction process, which results in increased
interstitial fluid pressure (IFP). This has consequences for
P P the efficiency of drug delivery to the tumour 42, as drugs
TRII cannot penetrate tissue under positive IFP. EMT can also
polarize carcinoma cells towards stem cell-like proper-
ties, such as increased tumour-initiating capacity and
tumour cell drug resistance43. Blocking the TGF pathway
can thus have a threefold benefit: the reduction of tumour
SMAD4
invasion and metastasis; the suppression of cancer stem
SMAD7 SMAD2 or SMAD3 P
cell-like properties; and the restoration of negative IFP to
enhance chemotherapeutic drug delivery 44.
In fibrotic conditions, excessive TGF production
induced in the diseased state contributes to EMT elabo-
P miRNA
ration, which can further exacerbate fibrosis, as seen in
SMAD2 or SMAD3 P regulation pulmonary 45,46, cardiac47 and renal48,49 fibrosis, and in
arterial restenosis following surgical trauma50. TGF can
SKI and SNO also promote a proliferative and/or migratory phenotype
on smooth muscle cells that can aggravate some vascu-
SMAD2/SMAD3

lar diseases, including neointimal formation following


vascular surgery 5153.
complex

P
Drosha

SMAD7
SMAD2 or SMAD3 P Extracellular matrix regulation. The ECM is a complex
mG AAAAA structure that surrounds mammalian cells. It is the major
Transcriptional regulation SMAD-mediated
Nucleus
component of connective tissue and is composed of mul-
miRNA processing
tiple proteins, such as collagen, elastin, fibrillin, fibro
Figure 1 | Schematic overview of the canonical, SMAD-dependent TGF nectin, lamin and proteoglycans. Fibrosis is characterized
signalling pathway. The transforming growth factor- (TGF) ligands are synthesized by the accumulation of fibroblasts, which secrete exces-
Nature Reviews | Drug Discovery
as a large latent TGF complex consisting of mature dimeric TGF associated with its sive amounts of ECM. As TGF is widely documented to
latency-associated peptide (LAP) and a latent TGF-binding protein (LTBP) (not shown). increase collagen synthesis and deposition by fibroblasts,
Upon activation, TGF dimers induce heteromeric complex formation between specific
TGF has become a central therapeutic target for different
typeII and typeI receptors (such as TGF receptor type II (TRII) and TRI, respectively).
TypeII receptors then transphosphorylate the typeI receptors, which propagate the types of fibrosis. TGF activity and the synthesis of ECM
signal into the cell by phosphorylating TGF receptor-specific SMADs (RSMADs: proteins are mutually regulated. Several genes encoding
SMAD2 and SMAD3). They form heteromeric complexes with the common mediator ECM proteins that are known to be important in driving
SMAD (coSMAD: SMAD4) and translocate to the nucleus. Once in the nucleus, the fibrosis are directly regulated by TGFSMAD signalling
RSMADcoSMAD complex preferentially associates with the genomic SMAD-binding pathways. There is a reciprocal regulation of TGF by the
element (SBE) in a sequence-specific manner. However, high-affinity binding of the ECM: latent TGF bound to ECM components, such as
RSMADcoSMAD complex with the SBE generally occurs in concert with other fibronectin and fibrillin, is inactive until physiological or
DNA-binding transcription factors that bind to distinct sequences adjacent to the SBE27. pathological processes initiate its release. This is seen in
The nuclear proteins SKI and SNO (also known as SKIL) antagonize the transcriptional MFS, in which the mutation of a fibrillin-encoding gene
regulation by SMADs. An inhibitory SMAD (ISMAD), SMAD7, inhibits the TGF pathway
results in reduced fibrillin levels and a consequent increase
through multiple mechanisms, including by mediating the degradation of the typeI
receptor, inhibiting phosphorylation of RSMADs by the typeI receptor kinase or in levels of unbound TGF; this, in turn, leads to the acti-
inhibiting the formation of the RSMADcoSMAD complex. In addition to regulating vation of TGF signalling, which is possibly responsible
transcription, RSMADs can modulate microRNA (miRNA) biogenesis by facilitating the for the aetiology of many Marfanoid features11,12.
processing of primary miRNA into precursor miRNA in the nucleus. The coSMAD is
not required for the regulation of miRNA biosynthesis by RSMADs. mG and AAAAA Immune-suppression and inflammation. The lethal
represent 5 capping and 3 polyadenylation of mRNAs, respectively. postnatal inflammatory phenotype of Tgfb1knockout
mice19,20,54 demonstrates the important immune-suppres-
sor function of this ligand. The widespread expression
Ecadherin is commonly downregulated in many can- profile of TGF receptors on all immune cell types sug-
cers, and its overexpression can suppress invasion by gests that they have broad activities, including responses
tumour cells. The TGFSMAD pathway mediates the in cytotoxic CD8+ effector Tcells, CD4+ effector T helper
Myofibroblast expression of high mobility group AT-hook 2 (HMGA2), 1 (TH1) and TH2 cells, suppressive regulatory T (TReg) cells,
A contractile fibroblast that which is important for the induction of SNAIL (also natural killer (NK) cells, monocytes, macrophages, neu-
expresses smooth muscle actin
and myosin, and contributes to
known as SNAI1) and SLUG (also known as SNAI2): two trophils and eosinophils (FIG.4). Cell type-specific mouse
disease progression in cancer zinc-finger transcription factors that are known to repress gene knockout studies with Tgfbr2 demonstrate both
and fibrosis. the Ecadherin gene33. In breast and skin cancer, tumour direct and indirect actions of TGF on effector Tcells4.

NATURE REVIEWS | DRUG DISCOVERY VOLUME 11 | O CTOBER 2012 | 793

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

surrounding this topic. First, different laboratories cannot


Box 2 | Non-canonical TGF signalling and crosstalk with other pathways
agree on the specific functions of various TH1, TH2 and
In addition to activating SMAD proteins, transforming growth factor- (TGF) TH17 cell types in disease progression. TH17 cells were
signalling can regulate the activity of a number of signalling molecules, such as TNF implicated as major agonists in inflammatory diseases,
receptor-associated factor 4 (TRAF4), TRAF6, TGF-activated kinase 1 (TAK1), p38 including inflammatory cancer, asthma and autoimmune
mitogen-activated protein kinases (MAPKs), extracellular signal-regulated kinase
disorders55. However, recent studies suggest that the active
(ERK), JUN Nterminal kinase (JNK), RHO GTPases, phosphoinositide 3kinase
(PI3K)AKT and nuclear factorB (NFB), to transmit a signal6. In addition, these
player in disease progression is in fact a TH17derived
non-canonical signals can crosstalk with the SMAD pathways and mutually modulate TH1 cell, or a TH1TH17 cell64. Second, the role of TGF
each other. Both canonical and non-canonical TGF signalling can also be influenced in regulating the balance between TH1 and TH17 differen-
by other signalling pathways, such as the RAS, WNT, Hedgehog, Notch, tumour tiation is in dispute. Despite the widespread acceptance of
necrosis factor (TNF) and interferon pathways6. The exact nature of the crosstalk with a role for TGF in TH17 differentiation6163, more recent
other pathways and biological outcomes is complex and highly context-dependent6. studies have suggested that TGF is totally dispensable
However, some of the crosstalk has been found to modulate the function and stability for the generation of these cells65,66.
of SMAD proteins through post-translational modifications, and to define cell type- With respect to cells of the innate immune system,
and context-specific outcomes by inducing other factors that modulate TGF activity. TGF directly suppresses NKcell-mediated production of
IFN (which is required for the tumour killing activity
of NK cells) through transcriptional effects of SMAD3
TGF has potent growth-suppressing activity on most on the IFN promoter 67. It also polarizes macrophages68
precursor cells of the immune system, particularly T and and neutrophils69 from a typeI, productive phenotype
Bcells of the adaptive arm. TGF is a potent suppressor of (that evolved to attack and devour foreign agents such
Tcell proliferation55 and an inducer of Bcell apoptosis56. as cancer cells) towards a typeII phenotype (that has
Additionally, the ligand can alter the course of immune reduced effector function but produces large quanti-
cell differentiation. Suppressive TReg cells that are driven ties of inflammatory molecules, such as IL6, IL11
by the expression of the transcription factor forkhead and TGF). These molecules can exacerbate the local
box protein P3 (FOXP3) are crucial for maintenance diseased state, resulting in solid tumour progression or
of peripheral immune tolerance as well as regulation of inflammation associated with fibrosis or atherosclerosis4.
tumour immunity and infection. In CD4+ Tcells, Foxp3 In summary, the regulation of the immune system
expression is positively but indirectly regulated by TGF1 by TGF is highly complex and context-dependent.
through enhanced binding of the SMAD2induced It delicately regulates the tolerogenic versus immuno-
transcription factor E2A to the Foxp3 gene promoter, genic arms of the immune system to balance adequate
and by relief from GATA3mediated transcriptional host defence while limiting collateral inflammatory tissue
inhibition of the Foxp3 promoter by competition with damage. The molecular details of this regulation have
TGFinduced Id3 (REF.57). TGF suppresses inflamma- been recently reviewed in depth4,55,64.
tory TH1 and TH2cell differentiation while stimulating
suppressor TReg cells. Overall, TGF-mediated suppres- Targeting TGF signalling
sion of effector CD8+ cytolytic cells and TH cells, together Virtually every component of the TGF pathway has
with TGF dependence for suppressive TReg cell differ- been targeted for drug development (FIG.5) through
entiation, results in the hyper-inflammatory phenotype numerous design strategies (FIG.6). Several have been
seen in Tgfb1/ mice. developed through preclinical to clinical trials (TABLE1)
During tumour progression, excess TGF suppresses and many more have been tested only in preclinical sys-
immune surveillance by attenuating the antitumour func- tems (TABLE2). The drugs that have progressed furthest
tions of CD8+ Tcells, CD4+ Tcells and dendritic cells. in clinical development include anti-ligand antisense
CD4+ Tcell-specific ablation of TGF signalling in trans- oligonucleotides (ASOs) from Antisense Pharma7074,
genic mice expressing dominant negative TRII (DNRII; ligand-competitive peptides from Digna Biotech7578,
also known as CD4TRII and CD4TGFBR2) led to antibodies that target ligands, receptors or associated
the development of autoimmunity 58 and enhanced the proteins spearheaded by Genzyme7981, and SMIs against
differentiation of CD8+ cytotoxic T lymphocytes (CTLs). TGF receptor kinases developed by many companies,
When challenged with tumour cells, these transgenic with Eli Lilly having an active clinical programme in
mice raised a greater tumour-specific CTL response PhaseII development82. The various approaches currently
than wild-type littermates58. Tumour-derived TGF also being investigated are discussed in more detailbelow.
blocks the differentiation of antigen-presenting dendritic
cells59 and modifies chemokine receptor expression to Antisense oligonucleotides and antisense RNA. Antisense
blunt dendritic cell chemotaxis60, further suppressing Pharma uses the strategy of targeting mRNA translation
immune surveillance. using ASOs to downregulate ligand synthesis70,83. Its focus
In addition to having a predominant immune-sup- has been on targeting TGF2, which is produced in exces-
Antisense oligonucleotides pressive function, TGF counterintuitively may have a sive quantities by glioblastoma and pancreatic carcinoma
(ASOs). Short chemically pro-inflammatory role through its effects on TH17 cells cells. Trabedersen (AP12009), a synthetic 18mer phos-
modified oligonucleotides and cells of the innate immune system. TGF, together phorothioate ASO, binds specifically to human TGF2
complementary to a specific
mRNA that can be used to
with interleukin6 (IL6), was reported to be an essential mRNA, and this drug has progressed to a PhaseIII clinical
cause specific knockdown of player in driving pro-inflammatory TH17 lineage differ- trial for oncology applications (BOX4). One of the chal-
targeted gene expression. entiation6163. However, there is considerable controversy lenges of this drug is delivering it directly to the tumour

794 | O CTOBER 2012 | VOLUME 11 www.nature.com/reviews/drugdisc

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

Crosstalk with
other pathways
TGF RAS
Hedgehog
TRII Notch
WNT
IFN
TNF
P P PI3KAKT
TRII TRII TRI
Cytoplasm
P P P P
p38
TRI
ERK
P
R-SMAD R-SMAD

SMAD4

SMAD4
RHO

RAS PI3K TRAF4/TRAF6

ROCK TAK1 NF-B

RAF AKT

JNK p38

R-SMAD

complex
MEK

Drosha
R-SMAD
SMAD4

JUN mG AAAAA
SMAD-mediated
ERK miRNA processing
Nucleus
Figure 2 | Schematic representation of non-canonical TGF signalling and crosstalk with other signalling
pathways. In the non-canonical pathways, the activated transforming growth factor- (TGF)
Nature receptor
Reviews complex
| Drug Discovery
transmits a signal through other factors, such as TNF receptor associated factor 4 (TRAF4) or TRAF6, TGF-activated
kinase 1 (TAK1), p38 mitogen-activated protein kinase (p38MAPK), RHO, phosphoinositide 3kinase (PI3K)AKT,
extracellular signal-regulated kinase (ERK), JUN Nterminal kinase (JNK) or nuclear factor-B (NFB). TGF signalling
can be influenced by pathways other than the canonical and non-canonical TGF signalling pathways, such as the
WNT, Hedgehog, Notch, interferon (IFN), tumour necrosis factor (TNF) and RAS pathways. The crosstalk between
TGF and other pathways defines the activities of TGF to propagate spatial- and temporal-specific signals. miRNA,
microRNA; ROCK, RHO-associated protein kinase; R-SMAD, receptor-specific SMAD; TR, TGF receptor. mG and
AAAAA represent 5 capping and 3 polyadenylation of mRNAs, respectively.

to avoid the off-target toxicity associated with systemic Monoclonal antibodies. The advantages of monoclonal
delivery of first-generation ASOs. In the case of glio antibodies are their specificity and extracellular mecha-
blastoma, this was achieved using intrathecal catheter nism of action an advantage when trying to mop up
delivery directly into the tumour 74. More recently, the excess extracellular ligand. This is tempered by the less
company has started developing intravenous delivery convenient intravenous mode of delivery. However, pro-
approaches for pancreatic cancer, which appear to be longed pharmacokinetic stability permits infrequent drug
effective in mouse models73 and were recently shown to administration. Cambridge Antibody Technologies and
be safe in humans84. Genzyme developed humanized (or murinized for pre-
An antiTGFB2 antisense strategy has also been clinical studies) monoclonal antibodies specific to indi-
used to generate augmented tumour vaccines. Belagen vidual ligands, such as lerdelimumab (CAT152)88,89 and
pumatucelL (Lucanix; NovaRx) is such a drug, in metelimumab (CAT192)90, or with pan-ligand specificity,
which an ~900nucleotide TGFB2 antisense construct such as fresolimumab (GC1008)9193. These antibod-
is transfected into allogeneic non-small-cell lung cancer ies have proceeded through various stages of preclinical
(NSCLC) cells, which are then used as a tumour vaccine. and clinical development. Of these three humanized
Here, drug delivery is not an issue as the drug is in fact antibodies, fresolimumab has progressed furthest in the
genetically engineered NSCLC tumour cell lines. This clinic for both neoplastic and non-neoplastic applica-
tumour vaccine has superior activity compared to con tions. This drug was found to be well tolerated and safe
ventional tumour vaccination approaches85,86. A significant at 15mg perml in PhaseI trials for metastatic melanoma
dose-related survival difference was seen in patients who (MetM) plus renal cell carcinoma93 and at 1mg perml for
received 2.5107 cells per injection, allowing progression the fibrotic disorder focal segmental glomerulosclerosis92.
to a PhaseIII clinical trial87. Lerdelimumab88,89 and metelimumab90, despite passing

NATURE REVIEWS | DRUG DISCOVERY VOLUME 11 | O CTOBER 2012 | 795

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

Tumour-suppressing TGF activity Tumour-progressing TGF activity

TGF
Invasive carcinoma
Tumour-
Normal epithelium or carcinoma in situ initiating cell
EMT Metastasis

Oncogenes
Growth
inhibition Immunosuppression

T cell
TGF
Macrophage
Polarization of IL-11 Osteolysis
recruitment
TANs (N1N2) and PTHRP
TAMs (M1M2) TGF
Bone
Autoinduction
TGF Autoinduction
VEGF
VEGF Vessels
IL-6 CTGF
Stromal modication Angiogenesis
Figure 3 | Biphasic activities of the TGF signalling pathway during tumorigenesis: from the tumour suppressor
to the tumour promoter. Transforming growth factor- (TGF) has biphasic actions during Nature Reviews | Drug
tumorigenesis, Discovery
suppressing
tumorigenesis at early stages but promoting tumour progression later on, which is the underlying paradigm for the
action of TGF during disease progression in general and thus complicates the development of therapies targeting TGF
signalling. The light grey arrows indicate a positive feedforward loop resulting in higher levels of TGF, which is a feature
of non-neoplastic as well as neoplastic diseases. The current goal in cancer therapy is to downmodulate excessive
levels of TGF ligands and to target the tumour-progressing versus the tumour-suppressing arm of TGF action;
the latter goal will almost certainly require more-specific second-generation drugs. CTGF, connective tissue growth
factor; EMT, epithelialmesenchymal transition; IL, interleukin; PTHRP, parathyroid hormone-related protein;
TAMs, tumour-associated macrophages; TANs, tumour-associated neutrophils; VEGF, vascular endothelial growth factor.

safety tests, failed to show efficacy in fibrotic models Biotech, using peptide mimetics of TRIII (also known as
of corneal scarring and systemic sclerosis, respectively, betaglycan and TGFBR3), completed a PhaseIIa clinical
and were therefore discontinued90. Despite a promising trial for scleroderma and skin fibrosis, showing safety and
PhaseI oncology trial of fresolimumab, after Genzyme efficacy when topically applied to skin (TABLE1; BOX4).
was acquired by Sanofi the company made the decision This company has plans to extend to PhaseIIb/III trials
to focus on fibrotic applications of this drug. in 2013 (J. Dotor, personal communication). A peptide
Eli Lilly entered the monoclonal antibody arena with a antagonist of TGF activation, LSKL (Leu-Ser-Lys-Leu),
pan-TGF ligand-specific blocking antibody, LY2382770, binds to a conserved sequence in the LAP region of the
which has progressed to PhaseII trials for kidney fibrosis latent complex and has demonstrated efficacy in reduc-
(TABLE1). Since merging with ImClone, Eli Lilly has also ing TGF signalling invitro97. This antagonist is based on
developed a TRII-blocking antibody, IMCTR1 (REF.94), thrombospondin and specifically blocks TGF activation.
which has just entered clinical trials for breast and colon The issue of peptide drug delivery is not a problem for
cancer (ClinicalTrials.gov identifier: NCT01646203). In topical application; however, to progress to systemic deliv-
addition, Biogen Idec and Stromedix have developed an ery, Digna Biotech has partnered with Flamel Technologies
anti-integrin 6 antibody that prevents the activation to investigate proprietary peptide delivery systems.
of TGF and has been used efficaciously in preclinical
studies of fibrosis and cancer 95; it is in a PhaseII trial Small-molecule inhibitors. There are a plethora of SMIs
for fibrosis (ClinicalTrials.gov identifier: NCT01371305). that specifically target the typeI receptor of TGF to
inhibit the phosphorylation of SMAD2 and SMAD3
Ligand traps and peptides. Genzyme developed a ligand while keeping at least some non-canonical responses,
trap by fusing Fc to the extracellular domain of TRII, such as TAK1 activation, intact. These drugs are gener-
but this construct never reached clinical trials96. However, ally ATP mimetics that bind competitively within the
an alternative ligand trap approach, pursued by Digna hydrophobic ATP binding pocket of the receptor kinase.

796 | O CTOBER 2012 | VOLUME 11 www.nature.com/reviews/drugdisc

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

applied to many other systemic diseases to attenuate the


Box 3 | BMPTGFactivin crosstalk in endothelial cells
activity of the TGF pathway, with the caveat that gene
Within the transforming growth factor- (TGF) superfamily, the crosstalk between therapy is still far from being widely accepted as a thera-
three subfamilies activins/inhibins/Nodal, TGFs and bone morphogenetic proteins peutic approach106.
(BMPs) is well established during development and postnatal homeostasis of various As an approach to stimulate immune destruction
organs206,207. In vertebrates, the BMPSMAD1/SMAD5/SMAD8 and activinNodal
of cancer cells by tumour-infiltrating Tcells, human
TGFSMAD2/SMAD3 signalling pathways execute antagonistic actions in different
developmental contexts by inducing the expression of antagonistic factors, such as
tumour antigen-specific CTLs have been engineered to
inhibitory SMADs (ISMADs: SMAD6 and SMAD7). Some studies have shown that the express DNRII using a clinical grade retrovirus vector.
common mediator SMAD (coSMAD), SMAD4, is rate-limiting; therefore, when one TGFresistant CTLs were found to have a functional
of the two pathways is activated, it can negatively influence the other pathway by advantage over unmodified CTLs in clearing TGF
sequestering SMAD4. In endothelial cells, TGF can signal not only via canonical TGF secreting EpsteinBarr virus (EBV)-positive lymphoma
receptor type I (TRI)SMAD2/SMAD3 but also via activin receptor-like kinase 1 invitro and invivo107, and this approach to therapy has
(ALK1)SMAD1/SMAD5/SMAD8 (REF.36). In contrast to TGFTRI signalling- progressed to a PhaseI clinical trial for EBV-positive
mediated activation of SMAD2 or SMAD3, which leads to endothelium quiescence, lymphoma. A further modification of the CTLs, by engi-
TGFALK1 signalling induces SMAD1/SMAD5/SMAD8 activation and has been neering in an HER2 (also known as ERBB2) chimeric
shown to stimulate endothelial cell migration, proliferation and tube formation, thus
receptor as well as a DNRII, allows the CTLs to target
promoting angiogenesis208. BMP9 was shown to induce SMAD2/SMAD3 and SMAD1/
SMAD5/SMAD8 phosphorylation via signalling mediated by BMP receptor 2 (BMPR2),
HER2-positive tumour cells108111. This approach is in a
activin receptor type II (ACTRII) and ALK1 or ALK2. Cross-activation of TGF-specific PhaseI clinical trial for advanced HER2-positive lung
and BMP-specific receptor-specific SMADs (RSMADs) by a single ligand is believed malignancy, labelled the HERCREEM trial (ClinicalTrials.
to provide a mechanism for the ligand to fine-tune endothelial cell behaviour and gov identifier: NCT00889954).
function36. In summary, the crosstalk among signalling pathways mediated by different Finally, Renova has developed a recombinant TGF3
TGF family ligands exists in every tissue. However, the mechanism and the biological ligand as an anti-scarring agent on the basis of the hypo
outcome of this crosstalk are highly species-, tissue- and context-dependent. thesis that this ligand has activity that is independent of
and antagonistic to TGF1 (REF.112). The drug, adminis-
tered by injection around a surgical wound site, progressed
The chemistry of these compounds has been extensively to a PhaseIII clinical trial, but unfortunately it did not
reviewed98,99 and some molecular structures are shown reach its primary or secondary efficacy endpoints.
in FIG.6. The obvious advantages of these molecules over
most others are their economical production, stability Pre-existing drugs that inhibit TGF. Pre-existing drugs
and ease of oral administration, set against a possible dis- that have been extensively used for other applications
advantage of cross-inhibition of other kinases. The short may act, in part, by inhibiting TGF. Examples are
half-life of these drugs provides the possibility of rapid losartan and candesartan, which are angiotensin typeII
drug withdrawal should adverse events arise. Many suc- receptor inhibitors that were originally developed for the
cessful preclinical studies for metastatic cancer have been treatment of hypertension. Both appear to reduce TGF
undertaken with these SMIs, as reviewed previously100,101. signalling, although the precise molecular mechanisms of
However, the only company to continue pursuit of a TRI- this action are still unclear12,113115. Pirfenidone acts in part
targeted SMI into clinical trials for oncology is Eli Lilly by reducing the fibrotic effects of TGF116 via unknown
with LY2157299 (REF.82) (ClinicalTrials.gov identifier: targets. It is the first approved drug in Europe for idio
NCT01373164). pathic pulmonary fibrosis (IPF), and is in a PhaseIII trial
in the United States117,118. On the other side of the coin,
Other approaches. A novel approach to the suppression some common drugs, including aspirin, elevate circulating
of ligand production has been the preclinical develop- TGF levels, which in certain cases such as arterio
ment of pyrrole-imidazole polyamides that bind with sclerosis correlates with disease suppression119.
sequence specificity to the TGFB1 gene promoter to
attenuate gene expression50,102,103. These large ~17kDa Therapeutic uses of TGF signalling inhibition
polymeric molecules (FIGS5,6) bind within the minor Cancer. TGF has a biphasic action during tumorigenesis,
groove of DNA to prevent transcription factor binding. suppressing tumorigenesis at early stages but promoting
Challenges associated with these drugs include the tumour progression later on (FIG.3). This is a paradigm for
specificity of promoter binding, along with drug delivery the action of TGF during disease progression in general,
issues owing to their large molecular size and the high including that of fibrosis, inflammation and cardiovas-
local concentration required for activity. However, pre- cular disease, and it is rooted in the fact that the normal
clinical studies suggest that these molecules might be function of this ligand is in the regulation of homeostasis.
used in drug-eluting stents for the purpose of reducing During disease progression, TGF signalling can go into
restenosis after coronary or carotid artery surgery 50. override and, once unharnessed, results in more damage
An alternative approach to suppress TGF signalling than good. The main goal in cancer therapy is therefore to
is gene transfer of antagonizing signalling molecules, downmodulate excessive levels of TGF ligands.
such as the inhibitory SMAD7. This approach has been A major challenge in developing TGF inhibitors for
applied in model systems to treat or prevent various cancer therapy has been the fact that these compounds
pathological conditions, including colonic and hepatic are not cytotoxic or cytostatic to most tumour cells
fibrosis, vascular remodelling and diabetic kidney dis- invitro. They were developed to target properties of the
ease104,105. Such an application has the potential to be tumour that are required for cancer progression, including

NATURE REVIEWS | DRUG DISCOVERY VOLUME 11 | O CTOBER 2012 | 797

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

Dendritic cell:
Maturation
Antigen presentation
Chemotaxis

Mast cell: CD8+ T cell:


Chemotaxis Proliferation
Eector function

Macrophage: CD4+ T cell:


Chemotaxis Proliferation
Eector function Eector function
Antigen presentation TGF TH1 and TH2 cells
Inammatory cytokine secretion TReg cells
M1M2 TH17 cells?

Neutrophil:
Eector function B cell:
Inammatory cytokine secretion IgA class switching
N1N2 Activation
Apoptosis

Natural killer cell:


Cytotoxicity
Chemotaxis

Figure 4 | TGF effects on immune cells. Transforming growth factor- (TGF) has effects on most immune cell types.
The figure depicts the activity of TGF on immune cell subsets that is relevant to human diseases. M1M2 and N1N2
indicate polarization of macrophages and neutrophils, respectively, from type I to type II.Nature Reviews | Drug A;
IgA, immunoglobulin Discovery
TH, T
helper; TReg, regulatory T.

migration, invasion and metastasis, as well as effects on higher dose levels of GC1008 and/or longer exposure to
the tumour microenvironment (FIGS3,4). Standard cyto- the drug, and the lesions resolved on drug withdrawal91,93.
toxic screens used by the pharmaceutical industry to To put this toxicity into context, non-melanoma skin
identify anticancer drugs were therefore not relevant, and cancers, such as SCC and keratoacanthoma, develop
therapeutic utility could only be determined by invivo in approximately 1530% of patients with MetM who
efficacy in animal models and ultimately in theclinic. are treated with BRAF inhibitors such as vemurafenib
Two major concerns in TGF drug development have and dabrafenib129, and therapy with sorafenib and TNF
been the inadvertent inhibition of the tumour-suppressing antagonists produced similar findings130,131. Recent data
arm of TGF signalling in cancer 120122 and the develop- from studies with vemurafenib for MetM therapy suggest
ment of adverse side effects unrelated to cancer, such that these lesions arise from pre-existent mutant RAS-
as widespread inflammation, autoimmunity or cardio- containing cells within sun-damaged skin132. Intriguingly,
vascular defects that have been revealed by mouse gene one study of keratoacanthoma that appeared in sorafenib-
knockout studies1921,123. Preclinical studies suggested that treated patients showed somatic TGFBR1 missense muta-
attenuation of TGF-mediated growth inhibition would tions133, one of which was also identified as a causative
not be a major issue96,124,125. However, clinical trials to date82 germline mutation for MSSE128.
have not revealed the cardiac valvulopathy 126 or hyper- Cancer stem cells, or tumour-initiating cells (TICs), are
ostosis and chondrocyte hypertrophy and hyperplasia127 defined by their capacity to self-renew and to initiate and
observed in rat preclinical toxicology studies. Moreover, persistently propagate the entire tumour. Targeting the
there has been no widespread evidence of inflammatory cancer stem cell for destruction or irreversible quiescence
complications in clinical trials reported to date54,82. These is therefore the Holy Grail of oncology, especially as these
reassuring safety findings are supported by evidence cells are exceedingly resistant to both chemotherapy and
from patients with the rare disease multiple self-healing radiotherapy, and are responsible for tumour metastasis
squamous epithelioma (MSSE), who have germline-null and recurrence after therapy 134. Several groups have now
mutations in the gene encoding TRI but develop only reported the phenomenon that TGFinduced EMT can
self-limiting and mostly non-malignant skin lesions128. drive tumour cells towards a more stem cell-like pheno
Intriguingly, in a PhaseI clinical trial of GC1008 for the type characterized by increased expression of stem cell
Tumour-initiating cells
(TICs). The putative cancer treatment of MetM, patients developed skin lesions, ker- markers and enhanced tumour-initiating activity invitro
stem cells that have the ability atoacanthoma or squamous cell carcinoma (SCC) that and invivo43,135. In breast cancer 135, the TGF and WNT
to maintain tumour growth, were similar to the skin abnormalities reported in MSSE, signalling pathways were shown to be the most com-
differentiate into all cell types with the appearance of keratoacanthoma and SCC seem- monly activated signalling pathways in cancer stem cells
of a heterogenous tumour,
and to re-establish secondary
ingly influenced by the extent of exposure to GC1008. that had been fractionated from the bulk tumour on the
tumours with exceedingly These lesions, which appeared on sun-damaged skin, were basis of expression of stem cell markers such as CD44hi
high efficiency. manifested in approximately 25% of patients who received and CD24low. In preclinical studies, TGF inhibitors

798 | O CTOBER 2012 | VOLUME 11 www.nature.com/reviews/drugdisc

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

Table 1 | Summary of clinical trials for TGF inhibitory drugs


Drug; Type Targets Disease Stage Clinical trial Summary of results Refs
company applications identifiers
Trabedersen Antisense oligo TGF2 Glioblastoma PhaseI/IIb NCT00431561 Safe 70,73,74
(AP12009); ligand
Antisense Pancreatic PhaseI NCT00844064 Pancreatic cancer trials 84
Pharma cancer, MetM, continue
colon cancer
Glioblastoma PhaseIII NCT00761280 Glioblastoma trials stopped -
in March 2012 owing to
advances in standard of care
and neurosurgery (BOX4)
Belagen Antisense TGF2 NSCLC PhaseIII NCT00676507 Well tolerated in 75 patients; 8587
pumatucel-L gene-modified survival advantage justifies
(Lucanix); allogeneic further PhaseIII evaluation
NovaRx tumour cell
vaccine
Disitertide Peptide Peptide Skin fibrosis in PhaseII NCT00574613, Preclinical efficacy in 7578
(P144); based on systemic sclerosis NCT00781053 peritoneal fibrosis associated
Digna Biotech TRIII that with peritoneal dialysis, renal
blocks and cardiac fibrosis, corneal
ligand haze and retinal AMD; safety
binding to and efficacy in PhaseIIa
receptors clinical trial for scleroderma/
skin fibrosis
Lerdelimumab Humanized TGF2 Reduction of PhaseIII - Safe; ineffective in reducing 88,89
(CAT152); antibody ligand scarring after (complete) scarring in PhaseIII trial
Cambridge glaucoma surgery
Antibody
Technology
Metelimumab Humanized TGF1 Diffuse systemic PhaseI/II NCT00043706 Ineffective when systemically 90
(CAT192); Antibody ligand sclerosis administered in doses up to
Cambridge 10mg perkg
Antibody
Technology
Fresolimumab Humanized TGF1, Focal segmental PhaseI NCT00464321 Completed and safe; 92
(GC1008); antibody TGF2 and glomerulosclerosis plans to progress
Cambridge TGF3
Antibody ligands Systemic sclerosis PhaseI NCT01284322 Still recruiting -
Technology/ Study ongoing -
Genzyme/
Sanofi Completed, no results -
Myelofibrosis PhaseI NCT01291784 See BOX4 93
IPF PhaseI NCT00125385 See BOX4 93
Renal cell PhaseI NCT00356460 See BOX4 93
carcinoma
Malignant PhaseI NCT00356460 See BOX4 81
melanoma
Metastatic breast PhaseI NCT01401062 Active and recruiting patients -
cancer (with
radiotherapy)
Relapsed PhaseII NCT01112293 Ongoing but not recruiting -
malignant pleural, participants
mesothelioma
LY2382770; Humanized TGF1 Diabetic kidney PhaseII NCT01113801 Safety and efficacy in -
Eli Lilly Antibody disease (fibrosis) protecting kidney function in
patients with diabetic kidney
disease; still recruiting
STX100; Antibody V6 Fibrosis PhaseII NCT01371305 Significant antifibrotic 168
Stromedix integrin activity in preclinical
models of lung, kidney
and liver disease

NATURE REVIEWS | DRUG DISCOVERY VOLUME 11 | O CTOBER 2012 | 799

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

Table 1 (cont.) | Summary of clinical trials for TGF inhibitory drugs


Drug; company Type Targets Disease Stage Clinical trial Summary of results Refs
applications identifiers
LY2157299; Small molecule TRI Advanced-stage PhaseII NCT10038320 See BOX4 82
Eli Lilly kinase melanoma
Recurrent PhaseII NCT01582269 Recruiting; LY2157299 alone or 213
glioblastoma with lomustine therapy versus
lomustine alone in recurrent
glioblastoma
Glioblastoma PhaseII NCT01220271 Recruiting; LY2157299 212
with temozolomide-based
radiochemotherapy in newly
diagnosed malignant glioma
Hepatocellular PhaseII NCT01246986 Recruiting -
carcinoma
Advanced PhaseII NCT01373164 Recruiting; comparison of -
pancreatic gemcitabine with gemcitabine
carcinoma plus LY2157299
Dominant Recombinant TRII Adoptive PhaseI NCT00889954 No update on clinical trials -
negative Tcells transfer of Tcells
TGFBR2-modified expressing HER2
CTLs and NTGFBR2
for lung cancer
(HERCREEM)
Dominant Recombinant TRII TGF-resistant PhaseI NCT00368082 Preclinical efficacy in tumour 107
negative Tcells (a clinical LMP2Aspecific killing of TGF-secreting
TGFBR2-modified grade retrovirus CTLs for EBV-positive lymphoma;
CTLs vector encoding EBV-positive no update on clinical trials
dominant lymphoma
negative TRII)
Avotermin Recombinant TGF3 Scarring PhaseII NCT004322111, The Juvista PhaseII trial had 214
(Juvista); protein NCT00656227 not reached its primary or
Renova secondary efficacy end points
as of February 2011
Pirfenidone; Small IPF, PhaseIII Multiple trials First drug approved for IPF 162
InterMune molecule, not glomerulosclerosis in Europe
TGF-specific and diabetic
kidney disease,
pathological skin
scarring
Losartan; Small AT1 Marfan syndrome PhaseI/II NCT00723801, Reduction of aortic aneurysm 12,182
Merck and Co. molecule, not (MFS) NCT00763893, in mouse model of MFS;
TGF-specific NCT00782327 clinical trials in progress to
reduce aortic root dilation
and cardiac muscle stiffness
in patients with MFS
Tranilast; Small Unknown Corneal primary PhaseIII NCT01003613 Tranilast reduces myofibroblast 185,
Kissei molecule, not pterygium proliferation of corneal 215,216
Pharmaceuticals TGF-specific myofibroblasts invitro and may
be a novel adjuvant therapy
for corneal keloid
IMCTR1; Humanized TRII Mammary and Phase I NCT01646203 Preclinical efficacy against 94
ImClone Systems/ antibody colon cancer primary tumour growth and
Eli Lilly metastasis through multiple
effects on tumour, stroma
and immune cells
AMD, age-related macular degeneration; AT1, angiotensin II type1 receptor; CTL, cytotoxic T lymphocyte; EBV, EpsteinBarr virus; IPF, idiopathic pulmonary
fibrosis; LMP2A, an EBV-specific antigen; MetM, metastatic melanoma; NSCLC, non-small-cell lung cancer; oligo, oligonucleotide; TR, TGF receptor; TGF,
transforming growth factor-; TGFBR2, gene encoding TRII.

have been shown to deplete the stem cell compart- glioblastoma-initiating cells (GICs, which express the
ment in various cancers including breast cancer 135, stem cell markers CD44, ID1, ID3, SOX2 and SOX4)
glioblastoma136138 and chronic myeloid leukaemia139 responded to LY2109761 by downregulating the expres-
which leads to increased lifespan in several mouse sion of stem cell genes. Moreover, patient-derived glio-
models of metastatic cancer. Anido etal.137 showed that blastoma neurospheres transplanted orthotopically into

800 | O CTOBER 2012 | VOLUME 11 www.nature.com/reviews/drugdisc

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

Table 2 | Summary of TGF inhibitory drugs in preclinical development


Drugs; Type Targets Disease applications Stage Summary of results Refs
company
AP11014; Antisense TGF1 ligand Prostate cancer, Preclinical AP11014 significantly reduced TGF1 217
Antisense oligo NSCLC, colorectal secretion by 43100% in different NSCLC,
Pharma cancer colon and prostate cancer cell lines
P17; Peptide Peptide derived Liver and pulmonary Preclinical Preclinical efficacy in peritoneal 76
Digna Biotech from Phage fibrosis, metastatic lung fibrosis associated with peritoneal
Display Technology cancer, angiogenesis, dialysis, lung fibrosis, corneal haze
that targets melanoma, and retinal AMD
TGF1 binding immunosuppression,
to receptor wet AMD
LSKL (academic Peptide Thrombospondin - Preclinical Preclinical efficacy in reducing renal 97
only) injury and proteinuria in a murine model
of diabetic nephropathy
1D11; Mouse Mouse TGF1, Breast cancer Preclinical Safe and efficacious in tumour metastasis 79,80,
R&D Systems antibody TGF2 and in mice 218
TGF3 ligands
SR2F Ligand trap TGF1, TGF3 Breast cancer Preclinical Very safe after lifetime exposure in mice; 125
(academic only) not progressing to clinical trial
Soluble TR2Fc; Ligand trap TGF1, TGF3 Breast cancer Preclinical Safe and efficacious in suppressing 96
Genzyme metastasis in preclinical model of breast
carcinoma; not progressing to clinical trial
LY580276, Small TRI kinase Cancer Preclinical LY2109761 is safe in long-term dosing of 80,156,
LY550410, molecule tumour-bearing mice, and efficacious in 219222
LY364947, reducing metastasis and TICs in mouse
LY2109761*; cancer models
Eli Lilly
SB505124, Small TRI kinase - Preclinical Extensively used invitro; 223225
SB431542; molecule pharmacokinetically unstable invivo
GlaxoSmithKline
SD208, SD093; Small TRI kinase Cancer Preclinical Efficacious in suppressing tumour 146,
Scios molecule metastasis in rodent models; 226,227
programme discontinued after
merger with Johnson & Johnson
Ki26894; Kirin Small TRI kinase Breast cancer Preclinical Not progressing to clinical trial 148,150
Pharmaceuticals molecule
SM16; Small TRI kinase Mesothelioma Preclinical Not progressing to clinical trial 10,155,
Biogen Idec molecule 228,229
GW788388; Small TRI kinase Fibrosis Preclinical Not progressing to clinical trial 230232
GlaxoSmithKline molecule
GB1201, GB1203 Pyrrole- TGFB1 gene Cutaneous and Preclinical Preclinical efficacy in inhibition of 50,
(academic) imidazole promoter corneal scarring, TGFB1 gene expression, which reduced 102,103
polyamide arterial restenosis, corneal scarring and carotid artery
kidney fibrosis restenosis
*Contrary to earlier reports4,16, LY573636 is not a TGFspecific inhibitor. NSCLC, non-small-cell lung cancer; oligo, oligonucleotide; TR, TGF receptor;
TGF, transforming growth factor-; TICs, tumour-initiating cells.

non-obese diabetic/severe combined immunodeficient It has been argued that TGF inhibitors might, how-
mice (NOD/SCID mice, which do not have T cells or ever, release isolated and disseminated tumour (stem)
B cells) responded to LY2109761 by decreasing in size cells from dormancy by initiating proliferation and/or
and reducing their expression of stem cell markers137. disrupting the stem cell niche. A couple of recent studies
The same research team is currently undertaking a may give credence to this notion, as systemic TGF inhibi-
PhaseI/II clinical trial for glioblastoma using the closely tion resulted in increased numbers of circulating tumours
related drug LY2157299 (REF.137). Importantly, they as well as micro- and macro-metastases in mouse models
showed a reduction of CD44 and ID1 RNA levels after of head and neck SCC and mammary cancer invivo140,141.
2months of LY2157299 treatment in tumour biopsy It might therefore be wise to use TGF inhibitors in com-
material from one patient with glioblastoma for whom bination with cytotoxic drugs to coax tumour cells out
a salvage surgical resection was performed both before of their quiescent niche while simultaneously targeting
and after 2months on the trial137. The ability to reduce those that respond proliferatively to TGF inhibition
the number of stem cells in an aggressive tumour such using chemotherapy. This strategy may be highly ben-
as glioblastoma is a majorcoup. eficial for flushing out dormant disseminated tumour

NATURE REVIEWS | DRUG DISCOVERY VOLUME 11 | O CTOBER 2012 | 801

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

Drugs with unknown targets STX-100


or indirect targeting: LSKL
Losartan (AT1 blocker)
Pirfenidone (unknown target)
Tranilast (unknown target)
Active TGF
Latent
TGF Lerdelimumab (CAT-152)
Metelimumab (CAT-192)
Fresolimumab (GC-1008)
TGF 1D11
SR2F (soluble TRII-Fc) LY2382770
P144, P17
IMC-TR1, IMC-MT1
Extracellular milieu

P P SMAD2 or SMAD3 Cytoplasm


AP11014 TRII SMAD2 or SMAD3 P
Trabedersen
Lucanix P P
TRI
SMAD4
LY580276 SD-208
TGF mRNA LY2109761 Ki26894
LY2157299 GW788388
SB-505124 SM16

Pyrrole-imidazole TRX-FOXH1B
polyamide TRX-LEF
Co-TFs

SMAD2 or SMAD3 TF
TGFB1 gene Target gene expression
Nucleus

Figure 5 | Schematic representation of therapeutic approaches for blocking TGF signalling. Transforming growth
factor- (TGF) signalling can be inhibited by: sequestering ligands using soluble receptor ectodomain constructs
(ligand traps) derived from TGF receptor type II (TRII) or TRIII; using TGF-neutralizing antibodies; or with TRII or
Nature Reviews | Drug Discovery
TRI kinase inhibitors. Furthermore, translation of TGF mRNA can be blocked by targeting TGF mRNA with antisense
oligonucleotides, thus preventing the production of the ligand. Different small-molecule kinase inhibitors against TRI
have been developed to block its kinase activity. Peptide inhibitors against specific TGF ligands are also used. Other
approaches block the transformation of TGF from the latent to the active form. Three molecules are shown that either
affect TGF signalling indirectly (losartan) or that have an as-yet-unidentified target (tranilast and pirfenidone). All of
these approaches decrease the initiation of intracellular receptor signalling pathways, such as phosphorylation of
downstream receptor-specific SMADs (RSMADs), and thereby blunt the transcriptional regulation of target genes.
AT1, angiotensin II type1 receptor; co-TFs, co-transcription factors; FOXH1B, forkhead box protein H1B; LEF, lymphoid
enhancer-binding factor; LSKL, Leu-Ser-Lys-Leu peptide; TRX, thioredoxin.

cells, as alluded to by Carlos Arteaga many years ago142. treatment with TGF inhibitors or their derivatives.As
A further cautionary note is warranted, however, on the TGF inhibitors are not directly cytotoxic, the use of these
basis of two reports indicating that TGF may decrease inhibitors in combination with cytotoxic chemothera-
the cancer-initiating cell population of diffuse type gastric peutics may be particularly efficacious. The activation of
carcinoma143 and breast carcinoma144 despite having little TGF signalling in response to chemotherapeutics may
or no effect on cellular proliferation. Finally, TGF inhibi- drive the generation of cancer stem cells (via EMT), result-
tors might act on the stem cell niche by recruiting bone ing in their chemoresistance134. This event may be targeted
marrow mesenchymal stem cell-derived myofibroblasts with TGF inhibitors, as demonstrated by the synergistic
that home in on the primary tumour, contribute to the activity of doxorubicin and TGF inhibitor combination
tumour microenvironment as cancer-associated fibro- therapy on breast cancer growth and metastasis147. Studies
blasts and consequently promote tumour progression145. in multiple myeloma also suggest that TGF inhibitors
Clearly there are tissue- and cell type-specific effects of could potentiate the cytotoxic effects of melphalan and
TGF inhibition that can influence the action of TGF dexamethasone148. In vitro, the exposure of multiple mye-
on the cancer stem cell and its niche146. Understanding loma cells to differentiated versus immature MC3T3E1
the differential molecular mechanisms that elicit these proosteoblastic cells potentiated chemotherapy-induced
variable responses will be critical to a judicious choice of multiple myeloma cell death. As TGF inhibition acts

802 | O CTOBER 2012 | VOLUME 11 www.nature.com/reviews/drugdisc

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

Box 4 | Oncology trials to date


A two-part clinical trial of GC1008 for the treatment of advanced metastatic melanoma (MetM) and renal cell carcinoma
(22 patients) found the drug to be safe and well tolerated with no dose-limiting toxicities (DTLs). Five patients achieved at
least stable disease as assessed by RECIST (response evaluation criteria in solid tumours) criteria, and therefore received
extended treatment. One patient achieved a partial response with a greater than 75% reduction in the target lesion.
The only adverse effect was keratoacanthoma-like lesions in sun-damaged skin of two of the patients with MetM.
However, these resolved on cessation of drug treatment and were not malignant93. Despite these promising results,
the pursuit of GC1008 for oncology was terminated after Genzyme was acquired by Sanofi in late 2011.
Antisense Pharma has had success with trabedersen (AP12009) in glioblastoma, pancreatic cancer and colon cancer.
Preclinical and clinical studies70,71,73 indicate that neutralization of transforming growth factor2 (TGF2)-mediated
immunosuppression, leading to activation of tumour-infiltrating natural killer cells, is the major mode of action.
Intra-tumoural administration of trabedersen to glioblastoma led to shrinkage of the targeted tumour as well as
tumours elsewhere in the brain. Three PhaseI/II studies of trabedersen for recurrent or refractory high-grade glioma
(glioblastoma) and anaplastic astrocytoma showed survival benefit compared with conventional chemotherapy209.
A randomized, controlled PhaseIIb study evaluating the efficacy and safety of two doses (10 and 80mM) of trabedersen
in comparison with standard therapy concluded that patients with glioblastoma on trabedersen had a threefold
enhancement in cognitive function 2 and 3years after therapy compared to standard chemotherapy74. However,
questions have been raised about this most recent study210,211. Wick and Weller211 conceded that although trabedersen
was clinically safe and that TGF inhibitors, in general, show promise for cancer therapy, the conclusions drawn by
Bogdahn etal.74 were premature. Because of other advances in both neurosurgical procedures and first-line standard of
care for patients with glioblastoma212, the SAPPHIRE PhaseIII trial of trabedersen was recently halted owing to patient
recruitment issues (ClinicalTrials.gov identifier: NCT00761280). Nevertheless, the drug has undergone a PhaseI/II trial
for patients with advanced pancreatic cancer, MetM or metastatic colorectal carcinoma, and showed excellent safety
and encouraging survival results (ClinicalTrials.gov identifier: NCT00844064)84.
Eli Lillys clinical small-molecule inhibitor LY2157299 was found to be safe and well tolerated in a PhaseI glioblastoma
trial82. Of 28 patients treated in a dose escalation study (14days on/14days off treatment), at least three patients showed
antitumour effects with durable responses beyond 1year. As a result, the Eli Lilly anti-TGF signalling programme
for oncology continues to be pursued with an ongoing PhaseII trial of LY2157299, with or without gemcitabine, for
hepatocellular carcinoma, glioblastoma and advanced pancreatic cancer, and with lomustine in patients with treatment-
refractory malignant glioma213, plus a new PhaseI trial of IMCTM1, an anti-TGF receptor type II (TRII) antibody.
NovaRxs belagenpumatucelL (Lucanix) has completed an open-label clinical trial of 75 patients with non-small-cell
lung cancer (NSCLC) with a median follow-up of 14.5months (44months for patients with stable disease). One-year,
two-year and five-year survivals were 55%, 35% and 20%, respectively. Individuals who demonstrated an increase in
both cellular and humoral immune reactivity had a significant survival advantage over individuals who showed an
increase in only one measure of immunity (32.5months versus 11.6months; p=0.015). On the basis of these findings,
an international, randomized PhaseIII trial to evaluate the efficacy of belagenpumatucelL in a maintenance setting
has been initiated for patients with stage III/IV NSCLC who have stable disease following frontline chemotherapy87.

within the bone microenvironment to elicit osteoblastic TGF responsiveness. Transduction of CTLs with a virus
differentiation148,149, this combinatorial approach holds encoding a DNRII154 has reached PhaseI clinical trials,
great promise for the treatment of multiple myeloma and recent preclinical data indicate that combining CTL
and other bone metastatic cancers. Likewise, in a mouse therapy with TRI-targeting SMIs may also significantly
model of serous gastric cell carcinoma, Ki26894 had an improve Tcell survival and antitumour Tcell cytotoxic-
additive effect with a fluorouracil analogue in reducing ity 155. Augmenting adoptive Tcell therapy with SMIs may
tumour growth150. Finally, another mechanism whereby be a particularly attractive application of TRI SMIs as
TGF inhibition can augment conventional therapies is in patients need not be exposed to genetically engineered
enhancing drug delivery to the tumour. There are reports Tcells. Moreover, patients might only require short-term
that TGF inhibition can reduce interstitial tumour exposure to the drug for efficacy in this application, thus
pressure44, which enhances the delivery of SMIs, and avoiding the side effects of long-term SMI drug exposure,
regulates vascular leakiness, which enhances the deliv- such as inflammation19, cardiovascular complications126,
ery of nanoparticle-encapsulated drugs, particularly in bone and/or cartilage problems127, subphyseal hyperosto-
highly fibrotic and drug-refractile tumour types such as sis as well as chondrocyte hypertrophy and/or hyperplasia,
pancreatic cancer 151. and reducing the risk of developing SMI drug resistance156.
Adoptive Tcell therapy involves the harvesting and Another clinical application with great promise is
exvivo expansion of autologous tumour-specific CTLs augmenting radiotherapy by inhibiting the TGF path-
followed by their reintroduction into the patient to stimu- way 10,81,157. Radiation not only physically activates latent
late tumour killing 152. Used most extensively in the treat- TGF invitro but also induces the biological release of
ment of MetM and lung cancer, this therapy often fails this growth factor as part of a stress response158. Several
owing to the apoptosis of regrafted CTLs. Preclinical groups have reported the positive role of TGF in sup-
studies suggest that failure may be due to the direct effects porting the DNA damage repair pathway, particularly
of TGF on CTLs153, and strategies to prevent such failure through activation of p53 and phosphorylation of ataxia
include the use of genetically modified CTLs with reduced telangiectasia mutated (ATM) after radiation therapy159.

NATURE REVIEWS | DRUG DISCOVERY VOLUME 11 | O CTOBER 2012 | 803

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

Cl
O N
O N N O
N O N N N
O N
NH2 NH2 NH2
N N F NH
N
H
N HN
N
O
SB-431542 (GlaxoSmithKline) LY2157299 (Eli Lilly) SD208 (Scios) N O SM16 (Biogen Idec)

O O OH O

H N
N
O
O

Tranilast Pirfenidone

H
O N
H
N
N H
H N
O
* N H H
O H N N
H O N
N * N H
N
O H N
H H O O N
N
N O * H H H
O N N N
H H N
N
N
N O O * H
H H O O
N O *
N N
H H H
N O
N
H
Pyrrole-imidazole polyamide N
N
H

Figure 6 | Structures of representative small-molecule inhibitors of TGF signalling. Depicted are the molecular
structures of a selection of small-molecule inhibitors identified to target the transforming growth
Nature factor-
Reviews | Drug (TGF)
Discovery
signalling pathway. SB-431542, LY2157299, SD208 and SM16 are all ATP mimetics that inhibit TGF receptor type I
(TRI; also known as TGFBR1) kinase activity. Pyrrole-imidazole polyamide blocks transcription of the TGFB1 gene.
Pirfenidone and tranilast have unknown molecular mechanisms of action. Dashed lines denote putative hydrogen
bonding with bases in DNA; asterisks indicate positions where hydrogen bonds form with nucleotide residues of DNA
within the TGFB1 gene promoter.

Barcellos-Hoff s group demonstrated that LY2109761 to leukaemia. However, refractory cytopaenias are the
and ID11 both attenuate radiation-induced activation of major cause of morbidity and mortality in sufferers.
p53 and ATM in breast cancer cells invitro and invivo, It was recently shown that reduced expression of SMAD7,
thus preventing DNA repair and accentuating the cyto- an inhibitor of TRI, was a common and significant event
toxic effect of radiation81. Even short-term dosing with observed in CD34+ myeloid progenitor cells in the bone
TRI inhibitors might provide a considerable therapeutic marrow of patients with MDS160. Indeed, low levels of
advantage in potentiating radiotherapy, with the added myeloid SMAD7 expression were seen in most patients
benefit that the local activation of pro-tumorigenic stroma with MDS, regardless of the risk for progression to leu-
and tissue fibrosis a major complication of radiation kaemia. Downregulation of SMAD7 expression sensi-
therapy 10 may also be suppressed by these drugs. tized myeloid precursors to TGF such that even very low
In partnership with Genzyme, this group is currently levels of the ligand elicited an increase in TGF respon-
undertaking a PhaseI trial of fresolimumab in combi- siveness, as defined by P-SMAD2 levels and enhanced
nation with radiotherapy for metastatic breast cancer. immune-suppressive effect, thus providing another
Eli Lilly is also undertaking a PhaseI/IIa trial to test the opportunity to utilize TGF inhibitors for therapeutic
safety and efficacy of LY2157299 in combination with utility in human disease.
temozolomide-based radiochemotherapy in patients Treatment of primary CD34 + haematopoietic
with newly diagnosed malignant glioma157. stem cells with LY2157299 suppressed the activation
of TRI by its ligand. Moreover, in a liver-specific
Myelodysplastic syndrome. Myelodysplastic syndrome TGF1overexpressing transgenic mouse model of
(MDS) is characterized by abnormal myeloid and/or MDS that exhibits severe anaemia, LY2157299 decreased
erythroid differentiation of bone marrow cells that results P-SMAD2 levels in the bone marrow and significantly
in various anaemias and cytopaenias. In one-third of increased the haematocrit of these mice. Importantly,
MDS cases, a high-risk group of patients can progress in ten out of ten primary bone marrow cultures from

804 | O CTOBER 2012 | VOLUME 11 www.nature.com/reviews/drugdisc

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

patients with MDS, administration of LY2157299 signifi- Renal fibrosis has long been thought to be driven by
cantly increased erythroid (burst-forming unit (BFUE)) excess TGF, which results in renal scarring and, ulti-
and myeloid (colony-forming unit (CFU); granulocytic mately, kidney failure170. With the increasing incidence of
monocytic) colony numbers invitro, harbouring great diabetes and associated kidney damage in affluent coun-
promise for the treatment of patients with MDS160. tries, this is a clinical application of growing importance.
Mice overexpressing an active form of TGF1 (from
Fibrosis. IPF is a progressive, chronic and irreversible the liver) develop progressive liver and renal fibrosis170.
lung disease occurring in older adults, and has an Interestingly, although mice overexpressing active TGF1
unknown cause161. The main histological features of IPF develop progressive renal injury, latent TGF1 also has a
are heterogeneous parenchyma, with areas of fibrosis protective role in renal fibrosis through negative effects
and honeycombing alternating with areas of less-affected on inflammation49. TGF1 mediates progressive renal
or normal parenchyma. IPF is characterized by a pro- fibrosis by stimulating the synthesis of ECM production
gressive reduction in lung function, with an estimated while inhibiting its degradation49. TGF1 also mediates
20% survival prospect after 5years, making it more renal fibrosis by inducing the transformation of tubular
lethal than many cancers. The progressive fibrotic reac- epithelial cells into myofibroblasts through EMT in a
tion in IPF is associated with an epithelium-dependent similar way to the process seen in IPF171. Blockade of
fibroblast activation, in which TGF plays a major part16. TGF1 with neutralizing TGF antibodies prevents
TGF1, which is secreted by alveolar epithelial cells in or ameliorates renal fibrosis invivo and invitro, dem-
patients with IPF, drives the process by promoting the onstrating the functional role of TGF1 in EMT and
migration, proliferation and differentiation of resident renal fibrosis. A number of therapeutic interventions
mesenchymal cells. V6 integrin, which binds and that block the action of TGF have resulted in various
activates latent TGF1 and TGF3, is highly induced degrees of improvement in kidney structure and function
following lung injury or fibrosis162. TGF activity 163 then in preclinical studies; such interventions include TGF
promotes activation and differentiation of fibroblasts ASOs, a neutralizing anti-TGF antibody, a soluble TGF
into myofibroblasts, which are specialized contractile receptor, blockade of TGF activation by decorin172, an
cells that cause aberrant ECM deposition, leading to the SMI of TGF receptors, delivery of the inhibitor protein
destruction of lung architecture, scarring 162 and reduced SMAD7 (REF.173) and a THBS1blocking peptide that
lung function. TGF also promotes pulmonary EMT that interferes with TGF activation97. A PhaseI/II trial with
additionally contributes to the expansion of fibroblasts GC1008, a pan-TGF-neutralizing antibody, exhibited
and myofibroblasts164. encouraging efficacy in patients with focal segmental
Pirfenidone, a novel compound that inhibits TGF glomerulosclerosis92, and Eli Lilly is undertaking trials
activity invitro, decreased the rate of decline in vital of its own antiTGF1 monoclonal antibody, LY2382770,
lung capacity and marginally increased progression- in diabetic kidney disease.
free survival in patients with IPF. Pooled data from Cardiac fibrosis is a pathological feature that is com-
two concurrent PhaseIII clinical trials in IPF indicated mon to a number of forms of heart disease, including
improvement in pulmonary function in the pirfenidone- myocardial infarction, ischaemic, dilated and hyper-
treated group165. Currently, there are no US Food and trophic cardiomyopathies and congestive heart failure36.
Drug Administration (FDA)-approved drugs for IPF, The cellular basis of cardiac fibrosis is the aberrant accu-
and pirfenidone is the first such drug to be approved for mulation of collagens and other ECM proteins, which
IPF in Europe. Other approaches to develop TGFbased impair ventricular function and predispose to cardiac
therapies for IPF include gene transfer of a soluble arrhythmias. Because TGF has pleiotropic effects in
TRII construct (as a ligand decoy), which attenuated the cardiovascular system and as cardiac fibrosis is a
injury and fibrosis in bleomycin-induced IPF in mice166. multifactorial disease, the development of an effective
P144 (disitertide; Digna Biotech), a synthetic peptide therapy will require a detailed understanding of the
that attenuates TGF activity and is derived from the role of the TGF signalling pathway in this pathogen-
extracellular domain of betaglycan, was also shown to esis. TGF, a potent stimulator of collagen production
reduce carbon tetrachloride-induced liver injury in by cardiac fibroblasts, is induced in response to cardio-
mice78. P17, another Digna Biotech anti-TGF pep- vascular injury. The TGFSMAD pathway activates
tide, has been shown to be efficacious in attenuation of the transcription of several key fibrotic genes, such as
injury and fibrosis in bleomycin-induced IPF in mice167. those encoding connective tissue growth factor (CTGF),
Although P144 has been used clinically for skin fibrosis, fibronectin, collagens and plasminogen activator inhibi-
drug delivery is an issue for the clinical development of tor 1 (PAI1)36. TGF reduces collagenase production and
P144 for IPF. However, Digna Biotech has recently part- stimulates the expression of tissue inhibitor of metallo
nered with Flamel Technologies to investigate the use of proteinases (TIMPs), resulting in an overall inhibition of
a proprietary drug delivery platform for this application ECM degradation and leading to excessive ECM accu-
(see the press release: Flamel Technologies and Digna mulation. P144 has been investigated in a preclinical
Biotech Announce Multiple Product Development model of cardiac fibrosis77, and losartan can reverse fibro-
Agreement). Other anti-TGF therapies in clinical trials sis in a mouse model of hypertrophic cardiomyopathy 174;
for IPF include the pan TGF-neutralizing antibody however, no drug targeting the TGF pathway has yet
GC1008 (Genzyme) and the V6 integrin-blocking reached clinical trials for this application. A recent
antibody STX100 (Stromedix)168,169. study demonstrated that miR21, which is regulated by

NATURE REVIEWS | DRUG DISCOVERY VOLUME 11 | O CTOBER 2012 | 805

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

SMADs upon TGF activation, is consistently induced the general integrity and elasticity of tissues but also
by cardiac stress. As miR21 plays a part in tumorigen- probably more crucially local TGF signalling 11.
esis by promoting cell proliferation, increased expression Normal fibrillin 1containing microfibrils interact with
of miR21 might contribute to the progression of fibrotic the large latent TGF complex (LLC) to control the
lesions175. ASOs against miR21 might therefore become release of mature, active TGF181. Mutated fibrillin 1 fails
a novel therapeutic approach for treating cardiac fibrosis. to sequester latent TGF, leading to the promiscuous
activation of TGF11. Thus, MFS highlights the critical
Scleroderma. Scleroderma (progressive systemic sclerosis) role of microfibrils in regulating local concentrations
is a systemic autoimmune disorder characterized by skin of TGF and in maintaining the homeostasis, morpho
sclerosis, calcinosis and changes in microvasculature. genesis and function of various organs181. Dilation of the
Increased expression of TRI and TRII in sclerodermal aortic root, which leads to aortic rupture and sudden
fibroblasts suggests that increased production of typeI death, is a major clinical issue for patients with MFS
collagen by autocrine TGF signalling leads to aberrant and patients of the LoeysDietz spectrum who carry
ECM deposition and scarring 36. Therapeutic approaches mutations in TGFBR1, TGFBR2 and SMAD3. A mouse
to scleroderma have included inhibition of TGF activity model of MFS carrying a heterozygous Fbn1 mutation
in sclerotic tissue. Unfortunately, CAT192, a TGF1- developed an aortic aneurism similar to that of patients
neutralizing antibody, did not show evidence of efficacy with MFS. Administration of TGF antagonists, includ-
in a study on the treatment of patients with early-stage ing a TGFneutralizing antibody or the angiotensin II
systemic scleroderma90; however, GC1008 is now in a type1 receptor (AT1) blocker losartan, successfully res-
PhaseI clinical trial for patients with diffuse systemic cued both cardiovascular and non-cardiovascular mani-
sclerosis. Furthermore, topical application of Digna festations of MFS12,182. As losartan is already in widespread
Biotechs P144 peptide inhibitor of TGF1 has shown clinical use for hypertension and has shown no adverse
some efficacy in reducing skin fibrosis in a PhaseII effects, this drug is currently being tested in PhaseI/
clinical trial for systemic sclerosis (see the press release: II clinical trials for MFS (ClinicalTrials.gov identi-
Flamel Technologies and Digna Biotech Announce fiers: NCT00429364, NCT00593710, NCT00683124,
Multiple Product Development Agreement), with the NCT00723801, NCT00763893, NCT00782327 and
caveat that clinical end points for quantifying skin fibrosis NCT01145612) and may plausibly reduce this life-
have not yet been standardized176. threatening manifestation ofMFS.

Restenosis following coronary artery bypass and angio Postoperative scarring in ocular conditions. Trabecul
plasty. The development of fibromuscular intimal ectomy is a surgical procedure designed to reduce intra
hyperplasia following angioplasty and coronary artery ocular fluid pressure in patients with glaucoma. However,
bypass surgery is a major clinical problem and can lead postoperative scarring and fibrotic blockage of the filter-
to coronary artery graft failure. The success of coronary ing bleb that drains excess ocular fluid are serious com-
artery reconstructive procedures is limited by the high plications of this procedure. In preclinical experiments,
incidence of restenosis secondary to intimal hyperpla- administration of neutralizing antibodies against human
sia. TGF1 is a major player in the early development TGF2 (CAT152) exhibited promising inhibition of scar-
of intimal hyperplasia in arteries and peripheral vein ring after glaucoma surgery in rabbits without having any
grafts. The exact mechanism of action of TGF signal- adverse effects. Initial clinical trials with CAT152 amelio-
ling in intimal hyperplasia and subsequent graft failure rated scarring in patients who received trabeculectomy for
is unclear, but it is speculated that TGF1 contributes intractable glaucoma88,183; however, PhaseIII clinical trials
at multiple steps, including EMT, promotion of fibro- were unable to validate such beneficial effects184. Tranilast,
blast, endothelial and vascular smooth muscle cell pro- another incidental TGF inhibitor, has also been used
liferation, increased collagen synthesis and deposition, successfully to reduce reoccurrence of corneal fibrosis,
and induction of fibrosis177. Soluble forms of the small, or primary pterygium, following corneal surgery 185. It is
leucine-rich proteoglycans decorin and fibromodulin, possible that topical application of more-specific TGF
which possess TGFantagonist activity, exhibit potent inhibitors might also be used in treating corneal haze and
intimal hyperplasia-suppressing effects in cultured conjunctival scarring (TABLE1).
human saphenous vein, offering the potential for thera-
peutic benefit after coronary artery bypass surgery 178180. Challenges and considerations for TGF blockade
The novel pyrrole-imidazole polyamide drug class, tar- Individualized responses to TGF blockade. Many dis-
geted to suppress TGFB1 gene transcription, showed effi- eases being tackled with TGF inhibitors, including fibro-
cacy in reducing neointimal hyperplasia and stimulating sis, inflammation, autoimmunity and cancer, are complex
reendothelialization of carotid arteries in a preclinical in nature and show strong genetic predisposition owing
model of arterial injury 50. to innate genetic variation between individuals. It is well
established that there is considerable phenotypic diver-
Marfan syndrome. MFS is a connective tissue disorder sity in the range of responses to reduced TGF signalling
that affects the musculoskeletal, ocular and cardio- invivo, which are dictated by differential inheritance of
vascular systems. It is caused by mutations in the gene germline genetic variants. This is illustrated by the large
encoding an ECM protein, fibrillin 1 (FBN1)181. Growing spectrum of clinical severity and disease manifesta-
evidence suggests that FBN1 mutations perturb not only tions in individuals with mutations in TGF signalling

806 | O CTOBER 2012 | VOLUME 11 www.nature.com/reviews/drugdisc

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

pathway genes8,41,186. Moreover, in mouse models of tumour progression. These include an increased ratio of
cancer, asthma and vascular development, outcomes of liver-enriched inhibitory protein (LIP) to liver-enriched
reduced TGF1 levels are strongly influenced by inter- activating protein (LAP), which are isoforms of CCAAT/
acting genetic modifier loci5,9,186,187. It is therefore most enhancer-binding protein- (C/EBP), a central tran-
rational, economical and safe to preselect patient popu- scription factor that binds within the SMAD transcrip-
lations before initiating anti-TGF drug treatment on tion factor complex to elicit TGF-mediated cytostatic
the basis of surrogate markers of TGF involvement in responses196. Upregulation of SIX1 in breast cancer has
the disease process (such as increased TGF ligand and also been shown to be pivotal in the growth-suppressive
PSMAD levels, and specific disease characteristics) and to tumour-progressive switch197, as has downregulation of
contra-indications of possible adverse side effects (such DAB2 (REF.198). Any of these tumour markers, possibly
as susceptibility to inflammation or certain vascular in combination, may be used in the future to predict
conditions). Peripheral blood may provide non-invasive tumour responses to TGF inhibition.
markers that might be useful in this respect, including Finally, the effects of interactions with other anticancer
the ability to quantitatively screen patient responses to drugs will need to be considered, as some drugs may res-
TGF inhibition on the basis of measuring P-SMAD2 urrect the growth-inhibitory arm of the TGF signalling
levels in peripheral blood mononuclear cells (PMNCs)188. pathway, which would then counter-indicate their combi-
Similarly, potential adverse inflammatory effects could natorial use with TGF inhibitors199. Indeed, the ability to
be predicted by examining specific immunological specifically target the pro-tumorigenic versus the tumour-
responses of PMNCs189192 to TGF inhibition exvivo. suppressive effects of TGF on the tumour cell perse will
require the development of next-generation drugs focusing
Patient selection for TGF inhibitors in oncology. The on these downstream pathways. In the meantime, much
simplest biomarker for patient selection for TGF research still remains to be undertaken to make inroads
inhibitors in oncology is probably high circulating levels into the area of informed patient selection for oncology
of TGF, as one major goal of this therapy is to reduce, applications of TGF inhibitors.
but not totally ablate, TGF signalling 193. As indicated
above, non-invasive biomarkers for predicting patient Drug resistance. In oncology, the development of tumour
responsiveness and efficacy of TGF inhibitors have drug resistance is inevitable134,200,201. It has been docu-
been developed on the basis of measuring P-SMAD2 mented for standard chemotherapy, pathway-targeted
levels in PMNCs. Indeed, PMNCs can be treated with therapies and is even common with anti-angiogenesis
drugs exvivo to determine, and thus predict, individ- inhibitors200,201. Acquired biochemical resistance of
ual patient responses to SMIs188,193. As TGF inhibitors tumour cells to LY2109761 has been observed in a
certainly act to reduce tumour metastasis, the assay of preclinical model of SCC and may have adverse conse-
circulating tumour cell number may also be a useful quences in driving a more stem cell-like phenotype156,
indicator for therapeutic response. Moreover, as many although this remains to be tested. Carefully restricting
of the pro-tumorigenic effects of TGF are mediated TGF inhibitors to short-term or intermittent usage
by immune system modulation, it might be possible should avoid these complications82. Combinatorial and/
to monitor blood cell or plasma protein profiles for or sequential treatment with complementary drugs will
patient selection and for surrogate monitoring of patient also be important. It is clear that oncologists will need an
responses. arsenal of different anticancer drugs to tackle cancer, in
In cancer (as well as non-malignant diseases), the out- much the same way that antibiotics have been developed
come of reduced TGF signalling may be highly depend- to combat infectious diseases.
ent on the innate genetic background of the individual,
especially when considering tumour microenvironment Conclusion and future directions
effects, such as immune surveillance. Elucidating specifi- In conclusion, TGF signalling inhibitors are generally
cally which genetic variants influence signalling output safe and may be efficacious in several clinical applications,
will not only be useful for dissecting the intricacies of especially in desperate cases such as end-stage cancer or
this signalling pathway invivo, but may also provide IPF. The development of these drugs may offer further
predictive markers for the outcome (desired or unde- therapeutic opportunities. Counterintuitively, there have
sired) of TGF signal inhibition. However, for cancer been reports suggesting that inhibition of the TGF sig-
therapeutics, patient selection may be more complex, as nalling pathway may be beneficial in autoimmune dis
the response of both the tumour cell and host (tumour orders, such as multiple sclerosis, through downregulation
microenvironment and normal patient tissue) to TGF of the TH17 pathway 202,203. Recent studies have also sug-
blockade needs to be considered. Tumour biopsy and gested that TGFSMAD3 signalling regulates glucose
genetic analysis (for example, loss of TGFBR2, SMAD2 tolerance and energy homeostasis, and that blockade of
or SMAD4)194,195 might predict whether the tumour the pathway may be used for regulation of diabetes and
retains growth sensitivity to TGF. Molecular and his- obesity 204. The outlook for anti-TGF signalling therapy
tological analyses may also contribute to the prediction for numerous diseases appears bright. At least four com-
of tumour responses to TGF inhibition. The activa- panies are well on their way in clinical drug development,
tion of alternative intracellular signalling pathways and and further scientific and mechanistic studies are war-
transcription factor profiles has been associated with ranted in order to optimize patient selection and drug-
the switch from tumour suppression by TGF to dosing regimens for each disease application.

NATURE REVIEWS | DRUG DISCOVERY VOLUME 11 | O CTOBER 2012 | 807

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

1. Schmierer, B. & Hill, C.S. TGFSMAD signal 28. Davis-Dusenbery, B.N. & Hata, A. Smad-mediated 52. Fu, K. etal. SM16, an orally active TGF typeI
transduction: molecular specificity and functional miRNA processing: a critical role for a conserved receptor inhibitor prevents myofibroblast induction
flexibility. Nature Rev. Mol. Cell Biol. 8, 970982 RNA sequence. RNA Biol. 8, 7176 (2011). and vascular fibrosis in the rat carotid injury model.
(2007). 29. Oft, M. etal. TGF1 and HaRas collaborate in Arterioscler. Thromb. Vasc. Biol. 28, 665671
2. Derynck, R. & Miyazono, K. (eds) The TGF Family modulating the phenotypic plasticity and invasiveness (2008).
(Cold Spring Harbor Press, 2008). of epithelial tumor cells. Genes Dev. 10, 24622477 53. Nishimura, G. etal. EF1 mediates TGF signaling in
3. Derynck, R. & Akhurst, R.J. Differentiation plasticity (1996). vascular smooth muscle cell differentiation. Dev. Cell
regulated by TGF family proteins in development 30. Massague, J. & Gomis, R.R. The logic of TGF 11, 93104 (2006).
and disease. Nature Cell Biol. 9, 10001004 (2007). signaling. FEBS Lett. 580, 28112820 (2006). 54. Diebold, R. J. etal. Early-onset multifocal
4. Flavell, R.A., Sanjabi, S., Wrzesinski, S.H. & 31. Massague, J. How cells read TGF signals. inflammation in the transforming growth factor
Licona-Limon, P. The polarization of immune cells Nature Rev. Mol. Cell Biol. 1, 169178 (2000). 1null mouse is lymphocyte mediated. Proc. Natl
in the tumour environment by TGF. Nature Rev. 32. Mehra, A. & Wrana, J.L. TGF and the Smad signal Acad. Sci. USA 92, 1221512219 (1995).
Immunol. 10, 554567 (2010). transduction pathway. Biochem. Cell Biol. 80, 55. Rubtsov, Y.P. & Rudensky, A.Y. TGF signalling in
5. Mao, J. H. etal. Genetic variants of Tgfb1 act as 605622 (2002). control of Tcell-mediated self-reactivity. Nature Rev.
context-dependent modifiers of mouse skin tumor 33. Padua, D. & Massague, J. Roles of TGF in metastasis. Immunol. 7, 443453 (2007).
susceptibility. Proc. Natl Acad. Sci. USA 103, Cell Res. 19, 89102 (2009). 56. Ramesh, S., Wildey, G.M. & Howe, P.H. Transforming
81258130 (2006). 34. Heldin, C.H., Landstrom, M. & Moustakas, A. growth factor (TGF)-induced apoptosis: the rise and
6. Guo, X. & Wang, X.F. Signaling cross-talk between Mechanism of TGF signaling to growth arrest, fall of Bim. Cell Cycle 8, 1117 (2009).
TGF/BMP and other pathways. Cell Res. 19, 7188 apoptosis, and epithelialmesenchymal transition. 57. Maruyama, T. etal. Control of the differentiation of
(2009). Curr. Opin. Cell Biol. 21, 166176 (2009). regulatory T cells and TH17 cells by the DNA-binding
7. Conidi, A. etal. Few Smad proteins and many Smad- 35. Dooley, S. & ten Dijke, P. TGF in progression inhibitor Id3. Nature Immunol. 12, 8695 (2011).
interacting proteins yield multiple functions and action of liver disease. Cell Tissue Res. 347, 245256 58. Gorelik, L. & Flavell, R.A. Immune-mediated
modes in TGF/BMP signaling in vivo. Cytokine Growth (2012). eradication of tumors through the blockade of
Factor Rev. 22, 287300 (2011). 36. Pardali, E. & Ten Dijke, P. TGF signaling and transforming growth factor signaling in T cells.
8. Akhurst, R.J. TGF signaling in health and disease. cardiovascular diseases. Int. J.Biol. Sci. 8, 195213 Nature Med. 7, 11181122 (2001).
Nature Genet. 36, 790792 (2004). (2012). This study shows that Tcell-specific blockade of
9. Freimuth, J. et al. Epistatic interactions between 37. Derynck, R., Akhurst, R.J. & Balmain, A. TGF TGF signalling using a dominant negative typeII
Tgfb1 and genetic loci, Tgfbm2 and Tgfbm3, signaling in tumor suppression and cancer receptor potentiates immune eradication of
determine susceptibility to an asthmatic stimulus. progression. Nature Genet. 29, 117129 (2001). tumours in mice challenged with live tumour cells.
Proc. Natl Acad. Sci. USA (in the press). 38. Coffey, R.J.Jr. et al. Growth modulation of mouse 59. Yamaguchi, Y., Tsumura, H., Miwa, M. & Inaba, K.
10. Anscher, M. S. etal. Small molecular inhibitor of keratinocytes by transforming growth factors. Contrasting effects of TGF1 and TNF on the
transforming growth factor protects against Cancer Res. 48, 15961602 (1988). development of dendritic cells from progenitors in
development of radiation-induced lung injury. 39. Gomis, R. R. etal. A FoxO-Smad synexpression group mouse bone marrow. Stem Cells 15, 144153
Int. J.Radiat. Oncol. Biol. Phys. 71, 829837 in human keratinocytes. Proc. Natl Acad. Sci. USA (1997).
(2008). 103, 1274712752 (2006). 60. Sato, K. etal. TGF1 reciprocally controls chemotaxis
11. Neptune, E. R. etal. Dysregulation of TGF activation 40. Siegel, P.M., Shu, W. & Massague, J. Mad of human peripheral blood monocyte-derived dendritic
contributes to pathogenesis in Marfan syndrome. upregulation and Id2 repression accompany cells via chemokine receptors. J.Immunol. 164,
Nature Genet. 33, 407411 (2003). transforming growth factor (TGF)-mediated epithelial 22852295 (2000).
12. Habashi, J. P. etal. Losartan, an AT1 antagonist, cell growth suppression. J.Biol. Chem. 278, 61. Bettelli, E. etal. Reciprocal developmental pathways
prevents aortic aneurysm in a mouse model of 3544435450 (2003). for the generation of pathogenic effector TH17 and
Marfan syndrome. Science 312, 117121 (2006). 41. Akhurst, R.J. The paradoxical TGF- vasculopathies. regulatory T cells. Nature 441, 235238 (2006).
13. Roberts, A.B. & Wakefield, L.M. The two faces of Nature Genet. 44, 838839 (2012). 62. Mangan, P. R. etal. Transforming growth factor
transforming growth factor in carcinogenesis. 42. Heldin, C.H., Rubin, K., Pietras, K. & Ostman, A. induces development of the TH17 lineage. Nature
Proc. Natl Acad. Sci. USA 100, 86218623 (2003). High interstitial fluid pressure an obstacle in 441, 231234 (2006).
14. Massague, J. TGF in cancer. Cell 134, 215230 cancer therapy. Nature Rev. Cancer 4, 806813 63. Gutcher, I. etal. Autocrine transforming growth
(2008). (2004). factor1 promotes in vivo TH17 cell differentiation.
15. Levy, L. & Hill, C.S. Alterations in components of 43. Mani, S. A. etal. The epithelial-mesenchymal Immunity 34, 396408 (2011).
the TGF superfamily signaling pathways in human transition generates cells with properties of stem cells. 64. Annunziato, F. & Romagnani, S. Mouse T helper 17
cancer. Cytokine Growth Factor Rev. 17, 4158 Cell 133, 704715 (2008). phenotype: not so different than in man after all.
(2006). 44. Salnikov, A. V. etal. Inhibition of TGF modulates Cytokine 56, 112115 (2011).
16. Hawinkels, L.J. & Ten Dijke, P. Exploring anti-TGF macrophages and vessel maturation in parallel to a 65. Das, J. etal. Transforming growth factor is
therapies in cancer and fibrosis. Growth Factors 29, lowering of interstitial fluid pressure in experimental dispensable for the molecular orchestration of TH17
140152 (2011). carcinoma. Lab. Invest. 85, 512521 (2005). cell differentiation. J.Exp. Med. 206, 24072416
17. Harradine, K.A. & Akhurst, R.J. Mutations of TGF 45. Willis, B. C. etal. Induction of epithelial-mesenchymal (2009).
signaling molecules in human disease. Ann. Med. 38, transition in alveolar epithelial cells by transforming 66. Ghoreschi, K. etal. Generation of pathogenic TH17
403414 (2006). growth factor1: potential role in idiopathic cells in the absence of TGF signalling. Nature 467,
18. Millan, F.A., Denhez, F., Kondaiah, P. & Akhurst, R.J. pulmonary fibrosis. Am. J.Pathol. 166, 13211332 967971 (2010).
Embryonic gene expression patterns of TGF 1, (2005). 67. Laouar, Y., Sutterwala, F.S., Gorelik, L. & Flavell, R.A.
2 and 3 suggest different developmental functions 46. Kim, K. K. etal. Alveolar epithelial cell mesenchymal Transforming growth factor controls T helper type 1
in vivo. Development 111, 131143 (1991). transition develops in vivo during pulmonary cell development through regulation of natural killer
19. Kulkarni, A. B. etal. Transforming growth factor 1 fibrosis and is regulated by the extracellular matrix. cell interferon-. Nature Immunol. 6, 600607 (2005).
null mutation in mice causes excessive inflammatory Proc. Natl Acad. Sci. USA 103, 1318013185 (2006). 68. Mantovani, A., Sozzani, S., Locati, M., Allavena, P. &
response and early death. Proc. Natl Acad. Sci. USA 47. Goumans, M.J., van Zonneveld, A.J. & ten Dijke, P. Sica, A. Macrophage polarization: tumor-associated
90, 770774 (1993). Transforming growth factor induced endothelial- macrophages as a paradigm for polarized M2
20. Shull, M. M. etal. Targeted disruption of the mouse tomesenchymal transition: a switch to cardiac mononuclear phagocytes. Trends Immunol. 23,
transforming growth factor1 gene results in multifocal fibrosis? Trends Cardiovasc. Med. 18, 293298 549555 (2002).
inflammatory disease. Nature 359, 693699 (1992). (2008). 69. Fridlender, Z. G. etal. Polarization of tumor-associated
21. Dickson, M. C. etal. Defective haematopoiesis and 48. Sato, M., Muragaki, Y., Saika, S., Roberts, A.B. & neutrophil phenotype by TGF: N1 versus N2
vasculogenesis in transforming growth factor1 knock Ooshima, A. Targeted disruption of TGF1/Smad3 TAN. Cancer Cell 16, 183194 (2009).
out mice. Development 121, 18451854 (1995). signaling protects against renal tubulointerstitial This study shows that TGF blockade increases
22. Sanford, L. P. etal. TGF2 knockout mice have fibrosis induced by unilateral ureteral obstruction. neutrophil-attracting chemokines, and depolarizes
multiple developmental defects that are non- J.Clin. Invest. 112, 14861494 (2003). the invading tumour-associated neutrophils
overlapping with other TGF knockout phenotypes. 49. Lan, H.Y. Diverse roles of TGF/Smads in renal to become more cytotoxic to tumour cells and to
Development 124, 26592670 (1997). fibrosis and inflammation. Int. J.Biol. Sci. 7, express higher levels of pro-inflammatory cytokines.
23. Proetzel, G. etal. Transforming growth factor3 is 10561067 (2011). 70. Schlingensiepen, R. etal. Intracerebral and
required for secondary palate fusion. Nature Genet. 50. Yao, E. H. etal. A pyrrole-imidazole polyamide intrathecal infusion of the TGF2specific antisense
11, 409414 (1995). targeting transforming growth factor1 inhibits phosphorothioate oligonucleotide AP 12009 in
24. Schultz-Cherry, S., Ribeiro, S., Gentry, L. & restenosis and preserves endothelialization in the rabbits and primates: toxicology and safety.
Murphy-Ullrich, J.E. Thrombospondin binds and injured artery. Cardiovasc. Res. 81, 797804 Oligonucleotides 15, 94104 (2005).
activates the small and large forms of latent (2009). 71. Schlingensiepen, K. H. etal. Targeted tumor therapy
transforming growth factor in a chemically defined This study evaluates the potential use of a with the TGF2 antisense compound AP 12009.
system. J.Biol. Chem. 269, 2677526782 (1994). pyrrole-imidazole polyamide that targets TGFB1 Cytokine Growth Factor Rev. 17, 129139 (2006).
25. Munger, J. S. etal. The integrin v6 binds and expression for preventing restenosis after the 72. Jachimczak, P. etal. The effect of transforming growth
activates latent TGF 1: a mechanism for regulating use of drug-eluting stents. factor-2specific phosphorothioate-anti-sense
pulmonary inflammation and fibrosis. Cell 96, 51. Tsai, S. etal. TGF through Smad3 signaling oligodeoxynucleotides in reversing cellular
319328 (1999). stimulates vascular smooth muscle cell proliferation immunosuppression in malignant glioma.
26. Shi, M. etal. Latent TGF structure and activation. and neointimal formation. Am. J.Physiol. Heart J.Neurosurg. 78, 944951 (1993).
Nature 474, 343349 (2011). Circ. Physiol. 297, H540H549 (2009). 73. Schlingensiepen, K. H. etal. Transforming growth
27. Shi, Y. & Massague, J. Mechanisms of TGF signaling This study demonstrates the role of the TGF factor2 gene silencing with trabedersen (AP 12009)
from cell membrane to the nucleus. Cell 113, SMAD3 pathway in the development of intimal in pancreatic cancer. Cancer Sci. 102, 11931200
685700 (2003). hyperplasia in response to endothelial injury. (2011).

808 | O CTOBER 2012 | VOLUME 11 www.nature.com/reviews/drugdisc

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

74. Bogdahn, U. etal. Targeted therapy for high-grade This report evaluates CAT192, a recombinant 114. Podowski, M. etal. Angiotensin receptor blockade
glioma with the TGF2 inhibitor trabedersen: results human antibody that neutralizes TGF1, in the attenuates cigarette smoke-induced lung injury and
of a randomized and controlled phase IIb study. treatment of early-stage diffuse cutaneous rescues lung architecture in mice. J.Clin. Invest. 122,
Neuro Oncol. 13, 132142 (2011). systemic sclerosis and concludes that there 229240 (2012).
This is the full report of a randomized, open-label, is no evidence of efficacy. 115. Lanz, T. V. etal. Angiotensin II sustains brain
active-controlled, dose-finding PhaseIIb study to 91. Lonning, S., Mannick, J. & McPherson, J.M. inflammation in mice via TGF. J.Clin. Invest. 120,
evaluate the efficacy and safety of trabedersen Antibody targeting of TGF in cancer patients. 27822794 (2010).
(AP12009) administered intra-tumourally by Curr. Pharm. Biotechnol. 12, 21762189 (2011). 116. Zhou, H., Latham, C.W., Zander, D.S., Margolin, S.B.
convection-enhanced delivery in patients with 92. Trachtman, H. etal. A phase1, single-dose study of & Visner, G.A. Pirfenidone inhibits obliterative airway
recurrent or refractory high-grade glioma. fresolimumab, an anti-TGF antibody, in treatment- disease in mouse tracheal allografts. J.Heart Lung
75. Santiago, B. etal. Topical application of a peptide resistant primary focal segmental glomerulosclerosis. Transplant. 24, 15771585 (2005).
inhibitor of transforming growth factor1 ameliorates Kidney Int. 79, 12361243 (2011). 117. Taniguchi, H. etal. The clinical significance of 5%
bleomycin-induced skin fibrosis. J.Invest. Dermatol. 93. Morris, J. C. etal. Phase I/II study of GC1008: a change in vital capacity in patients with idiopathic
125, 450455 (2005). human anti-transforming growth factorbeta (TGF) pulmonary fibrosis: extended analysis of the
This is the first evidence that the peptide inhibitor monoclonal antibody (MAb) in patients with pirfenidone trial. Respir. Res. 12, 93 (2011).
P144 may be efficacious in the treatment of skin advanced malignant melanoma (MM) or renal cell 118. Noble, P. W. etal. Pirfenidone in patients with
fibrosis after topical application. carcinoma (RCC). J.Clin. Oncol. 26, Abstract 9028 idiopathic pulmonary fibrosis (CAPACITY):
76. Llopiz, D. etal. Peptide inhibitors of transforming (2008). two randomised trials. Lancet 377, 17601769 (2011).
growth factor enhance the efficacy of antitumor 94. Zhong, Z. etal. Anti-transforming growth factor 119. Grainger, D. J. etal. The serum concentration of active
immunotherapy. Int. J.Cancer 125, 26142623 receptor II antibody has therapeutic efficacy against transforming growth factor is severely depressed in
(2009). primary tumor growth and metastasis through advanced atherosclerosis. Nature Med. 1, 7479
77. Hermida, N. etal. A synthetic peptide from multieffects on cancer, stroma, and immune cells. (1995).
transforming growth factor1 type III receptor Clin. Cancer Res. 16, 11911205 (2010). 120. Cui, W. etal. TGF1 inhibits the formation of benign
prevents myocardial fibrosis in spontaneously 95. Van Aarsen, L. A. etal. Antibody-mediated blockade skin tumors, but enhances progression to invasive
hypertensive rats. Cardiovascular Res. 81, 601609 of integrin v6 inhibits tumor progression in vivo by spindle carcinomas in transgenic mice. Cell 86,
(2009). a transforming growth factor-regulated mechanism. 531542 (1996).
This study shows that P144 inhibits the Cancer Res. 68, 561570 (2008). 121. Bierie, B. & Moses, H.L. Gain or loss of TGF signaling
TGF1dependent signalling pathway and 96. Muraoka, R. S. etal. Blockade of TGF inhibits in mammary carcinoma cells can promote metastasis.
prevents myocardial fibrosis. mammary tumor cell viability, migration, and Cell Cycle 8, 33193327 (2009).
78. Ezquerro, I. J. etal. A synthetic peptide from metastases. J.Clin. Invest. 109, 15511559 122. Bierie, B. etal. Abrogation of TGF signaling
transforming growth factor type III receptor inhibits (2002). enhances chemokine production and correlates with
liver fibrogenesis in rats with carbon tetrachloride 97. Lu, A., Miao, M., Schoeb, T.R., Agarwal, A. & prognosis in human breast cancer. J.Clin. Invest. 119,
liver injury. Cytokine 22, 1220 (2003). Murphy-Ullrich, J.E. Blockade of TSP1dependent 15711582 (2009).
79. Nam, J. S. etal. An anti-transforming growth factor TGF activity reduces renal injury and proteinuria in a 123. Larsson, J. etal. Abnormal angiogenesis but intact
antibody suppresses metastasis via cooperative murine model of diabetic nephropathy. Am. J.Pathol. hematopoietic potential in TGF type I receptor-
effects on multiple cell compartments. Cancer Res. 178, 25732586 (2011). deficient mice. EMBO J. 20, 16631673 (2001).
68, 38353843 (2008). 98. Yingling, J.M., Blanchard, K.L. & Sawyer, J.S. 124. Akhurst, R.J. TGF antagonists: why suppress a
80. Ganapathy, V. etal. Targeting the transforming growth Development of TGF signalling inhibitors for cancer tumor suppressor? J.Clin. Invest. 109, 15331536
factor pathway inhibits human basal-like breast therapy. Nature Rev. Drug Discov. 3, 10111022 (2002).
cancer metastasis. Mol. Cancer 9, 122 (2010). (2004). 125. Yang, Y. A. etal. Lifetime exposure to a soluble TGF
81. Bouquet, F. etal. TGF1 inhibition increases the 99. Bonafoux, D. & Lee, W.C. Strategies for TGF antagonist protects mice against metastasis without
radiosensitivity of breast cancer cells in vitro and modulation: a review of recent patents. Expert Opin. adverse side effects. J.Clin. Invest. 109, 16071615
promotes tumor control by radiation in vivo. Ther. Pat. 19, 17591769 (2009). (2002).
Clin. Cancer Res. 17, 67546765 (2011). 100. Saunier, E.F. & Akhurst, R.J. TGF inhibition for This is the first demonstration that long-term
82. Rodon Ahnert, J. etal. First human dose (FHD) study cancer therapy. Curr. Cancer Drug Targets 6, exposure to a TGF antagonist in mice does
of the oral transforming growth factor- receptor I 519532 (2006). not manifest any major adverse effects.
kinase inhibitor LY2157299 in patients with 101. Akhurst, R.J. Large- and small-molecule inhibitors of 126. Anderton, M. J. etal. Induction of heart valve lesions
treatment-refractory malignant glioma. J.Clin. Oncol. transforming growth factor-beta signaling. Curr. Opin. by small-molecule ALK5 inhibitors. Toxicol. Pathol. 39,
29, Abstract 3011 (2011). Investig. Drugs 7, 513521 (2006). 916924 (2011).
83. Schlingensiepen, K.H., Fischer-Blass, B., Schmaus, S. 102. Washio, H. etal. Transcriptional inhibition of 127. Frazier, K. etal. Inhibition of ALK5 signaling induces
& Ludwig, S. Antisense therapeutics for tumor hypertrophic scars by a gene silencer, pyrrole- physeal dysplasia in rats. Toxicol. Pathol. 35,
treatment: the TGF2 inhibitor AP 12009 in clinical imidazole polyamide, targeting the TGF1 promoter. 284295 (2007).
development against malignant tumors. Recent J.Invest. Dermatol. 131, 19871995 (2011). 128. Goudie, D. R. etal. Multiple self-healing squamous
Results Cancer Res. 177, 137150 (2008). 103. Chen, M. etal. Pretranscriptional regulation of Tgf1 epithelioma is caused by a disease-specific spectrum
84. Oettle, H. etal. Final results of a phaseI/II study in by PI polyamide prevents scarring and accelerates of mutations in TGFBR1. Nature Genet. 43, 365369
patients with pancreatic cancer, malignant melanoma, wound healing of the cornea after exposure to alkali. (2011).
and colorectal carcinoma with trabedersen. Mol. Ther. 18, 519527 (2010). 129. Chapman, P. B. etal. Improved survival with
J.Clin. Oncol. 30, Abstract 4034 (2012). 104. Chen, H. Y. etal. The protective role of Smad7 in vemurafenib in melanoma with BRAF V600E
85. Nemunaitis, J. etal. Phase II study of diabetic kidney disease: mechanism and therapeutic mutation. N.Engl. J.Med. 364, 25072516 (2011).
belagenpumatucelL, a transforming growth factor -2 potential. Diabetes 60, 590601 (2011). 130. Arnault, J. P. etal. Keratoacanthomas and squamous
antisense gene-modified allogeneic tumor cell vaccine 105. Kapoor, S. Smad7 gene transfer therapy: therapeutic cell carcinomas in patients receiving sorafenib.
in non-small-cell lung cancer. J.Clin. Oncol. 24, applications beyond colonic fibrosis. Eur. J.Clin. J.Clin. Oncol. 27, e59e61 (2009).
47214730 (2006). Invest. 38, 876877; author reply 878880 (2008). 131. Esser, A.C., Abril, A., Fayne, S. & Doyle, J.A.
This paper reports the first PhaseII trial of the 106. Sheridan, C. Gene therapy finds its niche. Nature Acute development of multiple keratoacanthomas
non-viral, gene-based, allogeneic tumour cell Biotech. 29, 121128 (2011). and squamous cell carcinomas after treatment with
vaccine of antisense TGF2transfected 107. Foster, A. E. etal. Antitumor activity of EBV-specific infliximab. J.Am. Acad. Dermatol. 50, S75S77
NSCLC cells. This vaccine shows great promise Tlymphocytes transduced with a dominant negative (2004).
in increasing survival rates for NSCLC. TGF receptor. J.Immunother. 31, 500505 (2008). 132. Su, F. etal. RAS mutations in cutaneous squamous-
86. Nemunaitis, J. etal. Phase II trial of 108. Rainusso, N. etal. Immunotherapy targeting HER2 cell carcinomas in patients treated with BRAF
BelagenpumatucelL, a TGF2 antisense gene with genetically modified T cells eliminates tumor- inhibitors. N.Engl. J.Med. 366, 207215 (2012).
modified allogeneic tumor vaccine in advanced initiating cells in osteosarcoma. Cancer Gene Ther. 133. Arnault, J. P. etal. Skin tumors induced by sorafenib;
non small cell lung cancer (NSCLC) patients. 19, 212217 (2012). paradoxic RAS-RAF pathway activation and oncogenic
Cancer Gene Ther. 16, 620624 (2009). 109. Nakazawa, Y. etal. PiggyBac-mediated cancer mutations of HRAS, TP53, and TGFBR1. Clin. Cancer
87. Bazhenova, L., Carrier, E., Shawler, D. & Fakhrai, H. immunotherapy using EBV-specific cytotoxic Tcells Res. 18, 263272 (2012).
Long-term survival in a phaseII study of expressing HER2specific chimeric antigen receptor. 134. Singh, A. & Settleman, J. EMT, cancer stem cells and
belagenpumatucelL (antisense TGF vaccine) in Mol. Ther. 19, 21332143 (2011). drug resistance: an emerging axis of evil in the war on
non small-cell lung cancer (NSCLC). Cancer Res. 110. Ahmed, N. etal. HER2specific T cells target primary cancer. Oncogene 29, 47414751 (2010).
72, Abstract 5367 (2012). glioblastoma stem cells and induce regression of 135. Shipitsin, M. etal. Molecular definition of breast
88. Mead, A.L., Wong, T.T., Cordeiro, M.F., Anderson, I.K. autologous experimental tumors. Clin. Cancer Res. tumor heterogeneity. Cancer Cell 11, 259273
& Khaw, P.T. Evaluation of anti-TGF2 antibody as a 16, 474485 (2010). (2007).
new postoperative anti-scarring agent in glaucoma 111. Ahmed, N. etal. Regression of experimental 136. Ikushima, H. etal. Autocrine TGF signaling maintains
surgery. Invest. Ophthalmol. Vis. Sci. 44, 33943401 medulloblastoma following transfer of HER2specific tumorigenicity of glioma-initiating cells through
(2003). Tcells. Cancer Res. 67, 59575964 (2007). Sry-related HMG-box factors. Cell Stem Cell 5,
89. CAT152 0201 Trabeculectomy Study Group. CAT152 112. Laverty, H. G. etal. Effects of avotermin (transforming 504514 (2009).
trabeculectomy study. Ophthalmology 114, 1950 growth factor 3) in a clinically relevant pig model 137. Anido, J. etal. TGF receptor inhibitors target the
(2007). of long, full-thickness incisional wounds. J.Cutan. CD44high/Id1high glioma-initiating cell population in
90. Denton, C. P. etal. Recombinant human anti- Med. Surg. 14, 223232 (2010). human glioblastoma. Cancer Cell 18, 655668
transforming growth factor 1 antibody therapy 113. Cohn, R. D. etal. Angiotensin II type 1 receptor (2010).
in systemic sclerosis: a multicenter, randomized, blockade attenuates TGFinduced failure of This is a demonstration of the targeting of
placebo-controlled phaseI/II trial of CAT192. muscle regeneration in multiple myopathic states. tumour-initiating cells by TGF blockade in
Arthritis Rheum. 56, 323333 (2007). Nature Med. 13, 204210 (2007). a mouse model of glioblastoma.

NATURE REVIEWS | DRUG DISCOVERY VOLUME 11 | O CTOBER 2012 | 809

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

138. Penuelas, S. etal. TGF increases glioma-initiating 158. Barcellos-Hoff, M.H. & Akhurst, R.J. Transforming 181. Ramirez, F. & Dietz, H.C. Marfan syndrome:
cell self-renewal through the induction of LIF in human growth factor in breast cancer: too much, too late. from molecular pathogenesis to clinical treatment.
glioblastoma. Cancer Cell 15, 315327 (2009). Breast Cancer Res. 11, 202 (2009). Curr. Opin. Genet. Dev. 17, 252258 (2007).
139. Naka, K. etal. TGFFOXO signalling maintains 159. Kirshner, J. etal. Inhibition of transforming growth 182. Holm, T. M. etal. Noncanonical TGF signaling
leukaemia-initiating cells in chronic myeloid factor1 signaling attenuates ataxia telangiectasia contributes to aortic aneurysm progression in Marfan
leukaemia. Nature 463, 676680 (2010). mutated activity in response to genotoxic stress. syndrome mice. Science 332, 358361 (2011).
This is a demonstration of the targeting of Cancer Res. 66, 1086110869 (2006). This report demonstrates a significant contribution
leukaemia-initiating cells by TGF blockade in 160. Zhou, L. etal. Reduced SMAD7 leads to of the non-canonical TGF signalling pathway to
a mouse model of chronic myeloid leukaemia. overactivation of TGF- signaling in MDS that can be aortic aneurysm progression in MFS.
140. Bragado, P. etal. Microenvironmental signals dictate reversed by a specific inhibitor of TGF- receptor I 183. Siriwardena, D. etal. Human antitransforming growth
disseminated tumor cells (DTCs) fate through kinase. Cancer Res. 71, 955963 (2011). factor 2 monoclonal antibody a new modulator
regulation of TGFII and p38. Cancer Res. 72 161. Cerri, S., Spagnolo, P., Luppi, F. & Richeldi, L. of wound healing in trabeculectomy: a randomized
(Suppl.1), Abstract 5234 (2012). Management of idiopathic pulmonary fibrosis. placebo controlled clinical study. Ophthalmology 109,
141. Biswas, T., Yang, J., Zhao, L. & Sun, L.Z. Clin. Chest Med. 33, 8594 (2012). 427431 (2002).
Development of murine models of breast cancer 162. Sheppard, D. Transforming growth factor : a central 184. Khaw, P. etal. A phaseIII study of subconjunctival
metastasis for the evaluation of the efficacy of TGF modulator of pulmonary and airway inflammation and human anti-transforming growth factor 2 monoclonal
inhibitors as therapeutic agents. Cancer Res. 72 fibrosis. Proc. Am. Thorac. Soc. 3, 413417 (2006). antibody (CAT152) to prevent scarring after first-time
(Suppl. 1), Abstract 1366 (2012). 163. Katsumoto, T.R., Violette, S.M. & Sheppard, D. trabeculectomy. Ophthalmology 114, 18221830
142. Ohmori, T., Yang, J.L., Price, J.O. & Arteaga, C.L. Blocking TGF via inhibition of the v6 integrin: (2007).
Blockade of tumor cell transforming growth factors a possible therapy for systemic sclerosis interstitial This study reports the results from a PhaseIII
enhances cell cycle progression and sensitizes human lung disease. Int. J.Rheumatol. 2011, 208219 (2011). study examining the efficacy of CAT152, a
breast carcinoma cells to cytotoxic chemotherapy. This report summarizes the inhibition of monoclonal antibody to TGF2, in preventing the
Exp. Cell Res. 245, 350359 (1998). V6integrin as an attractive therapeutic option progression of fibrosis in patients undergoing
143. Ehata, S. etal. Transforming growth factor decreases for systemic sclerosis and interstitial lung disease first-time trabeculectomy. The study found
the cancer-initiating cell population within diffuse-type mediated by TGF. no difference between CAT152 and placebo in
gastric carcinoma cells. Oncogene 30, 16931705 164. Chapman, H.A. Epithelial-mesenchymal interactions preventing the failure of trabeculectomy.
(2011). in pulmonary fibrosis. Annu. Rev. Physiol. 73, 185. Fukuda, K., Chikama, T., Takahashi, M. & Nishida, T.
144. Tang, B. etal. Transforming growth factor can 413435 (2011). Long-term followup after lamellar keratoplasty in
suppress tumorigenesis through effects on the 165. Azuma, A. Pirfenidone treatment of idiopathic a patient with bilateral idiopathic corneal keloid.
putative cancer stem or early progenitor cell and pulmonary fibrosis. Ther. Adv. Respir. Dis. 6, 107114 Cornea 30, 14911494 (2011).
committed progeny in a breast cancer xenograft (2012). 186. Benzinou, M. etal. Mouse and human strategies
model. Cancer Res. 67, 86438652 (2007). This report summarizes the outcome of a number identify PTPN14 as a modifier of angiogenesis
145. Quante, M. etal. Bone marrow-derived of clinical trials of pirfenidone, which is the first and hereditary haemorrhagic telangiectasia.
myofibroblasts contribute to the mesenchymal stem antifibrotic agent to be approved for the treatment Nature Commun. 3, 616 (2012).
cell niche and promote tumor growth. Cancer Cell 19, of IPF. 187. Bonyadi, M. etal. Mapping of a major genetic
257272 (2011). 166. Yamada, M. etal. Gene transfer of soluble modifier of embryonic lethality in TGF1 knockout
146. Hayashi, T. etal. Transforming growth factor transforming growth factor type II receptor by in vivo mice. Nature Genet. 15, 207211 (1997).
receptor I kinase inhibitor down-regulates cytokine electroporation attenuates lung injury and fibrosis. 188. Farrington, D. L. etal. Development and validation of
secretion and multiple myeloma cell growth in the J.Clin. Pathol. 60, 916920 (2007). a phosphorylated SMAD ex vivo stimulation assay.
bone marrow microenvironment. Clin. Cancer Res. 167. Arribillaga, L. etal. Therapeutic effect of a peptide Biomarkers 12, 313330 (2007).
10, 75407546 (2004). inhibitor of TGF on pulmonary fibrosis. Cytokine 53, 189. Liu, W. etal. CD127 expression inversely correlates
This is the first report of a chemical inhibitor of 327333 (2011). with FoxP3 and suppressive function of human CD4+
TRI showing antitumour effects via activity on the 168. Crunkhorn, S. Deal watch: Biogen acquires Stromedix T reg cells. J.Exp. Med. 203, 17011711 (2006).
tumour microenvironment in a mouse model of to pursue novel fibrosis therapy. Nature Rev. Drug 190. McClymont, S. A. etal. Plasticity of human regulatory
multiple myeloma. Discov. 11, 260 (2012). T cells in healthy subjects and patients with type 1
147. Bandyopadhyay, A. etal. Doxorubicin in combination 169. Allison, M. Stromedix acquisition signals growing diabetes. J.Immunol. 186, 39183926 (2011).
with a small TGF inhibitor: a potential novel therapy interest in fibrosis. Nature Biotech. 30, 375376 191. You, S. etal. Adaptive TGFdependent regulatory
for metastatic breast cancer in mouse models. (2012). Tcells control autoimmune diabetes and are a
PLoS ONE 5, e10365 (2010). 170. Liu, Y. Renal fibrosis: new insights into the privileged target of antiCD3 antibody treatment.
148. Takeuchi, K. etal. TGF inhibition restores terminal pathogenesis and therapeutics. Kidney Int. 69, Proc. Natl Acad. Sci. USA 104, 63356340
osteoblast differentiation to suppress myeloma 213217 (2006). (2007).
growth. PLoS ONE 5, e9870 (2010). 171. Fragiadaki, M. & Mason, R.M. Epithelial-mesenchymal 192. Clay, T.M., Hobeika, A.C., Mosca, P.J., Lyerly, H.K. &
149. Yin, J. J. etal. TGF signaling blockade inhibits PTHrP transition in renal fibrosis evidence for and against. Morse, M.A. Assays for monitoring cellular immune
secretion by breast cancer cells and bone metastases Int. J.Exp. Pathol. 92, 143150 (2011). responses to active immunotherapy of cancer.
development. J.Clin. Invest. 103, 197206 (1999). 172. Alan, C., Kocoglu, H., Altintas, R., Alici, B. & Clin. Cancer Res. 7, 11271135 (2001).
150. Shinto, O. etal. Combination effect of a TGF Resit Ersay, A. Protective effect of decorin on acute 193. Baselga, J. etal. TGF signalling-related markers in
receptor kinase inhibitor with 5FU analog S1 on ischaemia-reperfusion injury in the rat kidney. cancer patients with bone metastasis. Biomarkers 13,
lymph node metastasis of scirrhous gastric cancer in Arch. Med. Sci. 7, 211216 (2011). 217236 (2008).
mice. Cancer Sci. 101, 18461852 (2010). 173. Terada, Y. etal. Gene transfer of Smad7 using 194. Bornstein, S. etal. Smad4 loss in mice causes
151. Kano, M. R. etal. Improvement of cancer-targeting electroporation of adenovirus prevents renal fibrosis spontaneous head and neck cancer with increased
therapy, using nanocarriers for intractable solid in post-obstructed kidney. Kidney Int. 61, S94S98 genomic instability and inflammation. J.Clin. Invest.
tumors by inhibition of TGF signaling. Proc. Natl (2002). 119, 34083419 (2009).
Acad. Sci. USA 104, 34603465 (2007). This report indicates that genetic transfer of 195. Lu, S. L. etal. Loss of transforming growth factor
152. Liu, Y. & Zeng, G. Cancer and innate immune system Smad7 prevents unilateral ureteral obstruction- type II receptor promotes metastatic head-and-neck
interactions: translational potentials for cancer induced renal fibrosis, suggesting that SMAD7 may squamous cell carcinoma. Genes Dev. 20,
immunotherapy. J.Immunother. 35, 299308 be applicable for the treatment of renal fibrosis. 13311342 (2006).
(2012). 174. Lim, D. S. etal. Angiotensin II blockade reverses 196. Gomis, R.R., Alarcon, C., Nadal, C., Van Poznak, C. &
153. Wang, L. etal. Immunotherapy for human renal cell myocardial fibrosis in a transgenic mouse model of Massague, J. C/EBP at the core of the TGF cytostatic
carcinoma by adoptive transfer of autologous human hypertrophic cardiomyopathy. Circulation 103, response and its evasion in metastatic breast cancer
transforming growth factor insensitive CD8+ T cells. 789791 (2001). cells. Cancer Cell 10, 203214 (2006).
Clin. Cancer Res. 16, 164173 (2010). 175. van Rooij, E. & Olson, E.N. Searching for miR-acles in 197. Micalizzi, D. S. etal. The Six1 homeoprotein induces
154. Zhang, Q. etal. Blockade of transforming growth cardiac fibrosis. Circ. Res. 104, 138140 (2009). human mammary carcinoma cells to undergo
factor- signaling in tumor-reactive CD8+ T cells 176. Merkel, P. A. etal. Validity, reliability, and feasibility of epithelial-mesenchymal transition and metastasis
activates the antitumor immune response cycle. durometer measurements of scleroderma skin disease in mice through increasing TGF signaling.
Mol. Cancer Ther. 5, 17331743 (2006). in a multicenter treatment trial. Arthritis Rheum. 59, J.Clin. Invest. 119, 26782690 (2009).
155. Wallace, A. etal. Transforming growth factor 699705 (2008). 198. Hannigan, A. etal. Epigenetic downregulation of
receptor blockade augments the effectiveness of 177. McCaffrey, T.A. TGF signaling in atherosclerosis and human disabled homolog 2 switches TGF from a
adoptive Tcell therapy of established solid cancers. restenosis. Front. Biosci. (Schol. Ed.) 1, 236245 tumor suppressor to a tumor promoter. J.Clin. Invest.
Clin. Cancer Res. 14, 39663974 (2008). (2009). 120, 28422857 (2010).
This is the first demonstration of the use of a 178. Friedl, R. etal. Intimal hyperplasia and expression of 199. Law, B. K. etal. Rapamycin potentiates transforming
chemical TGF receptor inhibitor augmenting transforming growth factor1 in saphenous veins and growth factor induced growth arrest in
adoptive Tcell therapy via its immunomodulatory internal mammary arteries before coronary artery nontransformed, oncogene-transformed, and human
effects. surgery. Ann. Thorac. Surg. 78, 13121318 (2004). cancer cells. Mol. Cell. Biol. 22, 81848198
156. Connolly, E. C. etal. Outgrowth of drug-resistant 179. Ranjzad, P., Salem, H.K. & Kingston, P.A. Adenovirus- (2002).
carcinomas expressing markers of tumor aggression mediated gene transfer of fibromodulin inhibits 200. Engelman, J.A. & Settleman, J. Acquired resistance
after long term TRI/II kinase inhibition with neointimal hyperplasia in an organ culture model of to tyrosine kinase inhibitors during cancer therapy.
LY2109761. Cancer Res. 71, 111 (2011). human saphenous vein graft disease. Gene Ther. 16, Curr. Opin. Genet. Dev. 18, 7379 (2008).
157. Zhang, M. etal. Blockade of TGF signaling by 11541162 (2009). 201. Corcoran, R. B. etal. BRAF gene amplification can
the TGFRI kinase inhibitor LY2109761 enhances 180. Kapur, N. K. etal. Inhibition of transforming growth promote acquired resistance to MEK inhibitors in
radiation response and prolongs survival in factor restores endothelial thromboresistance in vein cancer cells harboring the BRAF V600E mutation.
glioblastoma. Cancer Res. 71, 71557167 (2011). grafts. J.Vasc. Surg. 54, 11171123.e1 (2011). Sci. Signal. 3, ra84 (2010).

810 | O CTOBER 2012 | VOLUME 11 www.nature.com/reviews/drugdisc

2012 Macmillan Publishers Limited. All rights reserved


REVIEWS

202. Acharya, M. etal. v Integrin expression by DCs is 216. Isaji, M. etal. Inhibitory effects of tranilast on 228. Suzuki, E. etal. A novel small-molecule inhibitor of
required for TH17 cell differentiation and development the proliferation and functions of human transforming growth factor type I receptor kinase
of experimental autoimmune encephalomyelitis in pterygium-derived fibroblasts. Cornea 19, 364368 (SM16) inhibits murine mesothelioma tumor
mice. J.Clin. Invest. 120, 44454452 (2010). (2000). growth in vivo and prevents tumor recurrence after
203. Melton, A. C. etal. Expression of v8 integrin on 217. Schlingensiepen, K. H. etal. The TGF1 antisense surgical resection. Cancer Res. 67, 23512359
dendritic cells regulates TH17 cell development and oligonucleotide AP 11014 for the treatment of (2007).
experimental autoimmune encephalomyelitis in mice. non-small cell lung, colorectal and prostate cancer: 229. Kim, S. etal. Systemic blockade of transforming
J.Clin. Invest. 120, 44364444 (2010). preclinical studies. J.Clin. Oncol. 22, Abstract 3132 growth factor signaling augments the efficacy of
204. Yadav, H. etal. Protection from obesity and diabetes (2004). immunogene therapy. Cancer Res. 68, 1024710256
by blockade of TGF/Smad3 signaling. Cell Metab. 218. Nam, J. S. etal. Bone sialoprotein mediates the tumor (2008).
14, 6779 (2011). cell-targeted prometastatic activity of transforming This is the first demonstration of the use of a
205. Massague, J., Seoane, J. & Wotton, D. Smad growth factor in a mouse model of breast cancer. chemical TGF receptor inhibitor, acting via its
transcription factors. Genes Dev. 19, 27832810 Cancer Res. 66, 63276335 (2006). immunomodulatory effects, to enhance the
(2005). 219. Melisi, D. etal. LY2109761, a novel transforming efficacy of an adenovirus expressing IFN-based
206. Keller, B. etal. Interaction of TGF and BMP signaling growth factor receptor typeI and typeII dual immunotherapy.
pathways during chondrogenesis. PLoS ONE 6, inhibitor, as a therapeutic approach to suppressing 230. Gellibert, F. etal. Discovery of 4-{4-[3-(pyridin-2yl)-
e16421 (2011). pancreatic cancer metastasis. Mol. Cancer Ther. 7, 1Hpyrazol-4yl]pyridin-2yl}-N-(tetrahydro2H-
207. Montero, J.A., Lorda-Diez, C.I., Ganan, Y., Macias, D. 829840 (2008). pyran4yl)benzamide (GW788388): a potent,
& Hurle, J.M. Activin/TGF and BMP crosstalk 220. Fransvea, E., Angelotti, U., Antonaci, S. & Giannelli, G. selective, and orally active transforming growth
determines digit chondrogenesis. Dev. Biol. 321, Blocking transforming growth factor upregulates factor type I receptor inhibitor. J.Med. Chem. 49,
343356 (2008). Ecadherin and reduces migration and invasion of 22102221 (2006).
208. Goumans, M. J. etal. Activin receptor-like kinase hepatocellular carcinoma cells. Hepatology 47, 231. Petersen, M. etal. Oral administration of GW788388,
(ALK)1 is an antagonistic mediator of lateral TGF/ 15571566 (2008). an inhibitor of TGF typeI and II receptor kinases,
ALK5 signaling. Mol. Cell 12, 817828 (2003). 221. Mazzocca, A., Fransvea, E., Lavezzari, G., Antonaci, S. decreases renal fibrosis. Kidney Int. 73, 705715
This study demonstrates that TGF regulates & Giannelli, G. Inhibition of transforming growth (2008).
the activation state of the endothelium via two factor receptor I kinase blocks hepatocellular 232. Tan, S.M., Zhang, Y., Connelly, K.A., Gilbert, R.E. &
opposing typeI receptors, ALK1 and TRI carcinoma growth through neo-angiogenesis Kelly, D.J. Targeted inhibition of activin receptor-like
(also known as ALK5). regulation. Hepatology 50, 11401151 (2009). kinase 5 signaling attenuates cardiac dysfunction
209. Hau, P. etal. Inhibition of TGF2 with AP12009 in 222. Zhang, B., Halder, S.K., Zhang, S. & Datta, P.K. following myocardial infarction. Am. J.Physiol. Heart
recurrent malignant gliomas: from preclinical to phase Targeting transforming growth factor signaling in Circ. Physiol. 298, H1415H1425 (2010).
I/II studies. Oligonucleotides 17, 201212 (2007). liver metastasis of colon cancer. Cancer Lett. 277,
210. Chamberlain, M.C. Convection-enhanced delivery 114120 (2009). Acknowledgements
of a transforming growth factor2 inhibitor 223. Inman, G. J. etal. SB431542 is a potent and specific The authors give thanks to J.Dotor, J.McPherson and
trabedersen for recurrent high-grade gliomas: efficacy inhibitor of transforming growth factor superfamily J.Yingling for useful discussions. Work in the authors laborato-
real or imagined?, in reference to Bogdahn etal. typeI activin receptor-like kinase (ALK) receptors ries is funded by grants from the US National Cancer Institute,
(Neuro-Oncology 2011;13:132142). Neuro Oncol. ALK4, ALK5, and ALK7. Mol. Pharmacol. 62, 6574 the US National Heart, Lung and Blood Institute (NHLBI), the
13, 558559; author reply 561552 (2011). (2002). US National Institute of Arthritis and Musculoskeletal and
211. Wick, W. & Weller, M. Trabedersen to target 224. Hjelmeland, M. D. etal. SB431542, a small Skin Diseases, the US National Institute of General Medical
transforming growth factor-: when the journey molecule transforming growth factor-receptor Sciences, the Bouque Foundation and the Helen Diller Family
is not the reward, in reference to Bogdahn etal. antagonist, inhibits human glioma cell line Comprehensive Cancer Center to R.J.A., and from the NHLBI,
(Neuro-Oncology 2011;13:132142). Neuro Oncol. proliferation and motility. Mol. Cancer Ther. 3, the American Heart Association and the LeDucq Foundation to
13, 559560; author reply 561552 (2011). 737745 (2004). A.H. We apologize to all colleagues whose work could not be
212. Roldan Urgoiti, G.B., Singh, A.D. & Easaw, J.C. 225. DaCosta Byfield, S., Major, C., Laping, N.J. & cited owing to space restrictions.
Extended adjuvant temozolomide for treatment Roberts, A.B. SB505124 is a selective inhibitor of
of newly diagnosed glioblastoma multiforme. transforming growth factor type I receptors ALK4, Competing interests statement
J.Neurooncol. 108, 173177 (2012). ALK5, and ALK7. Mol. Pharmacol. 65, 744752 The authors declare no competing financial interests.
213. Azaro, A. etal. The oral transforming growth factor (2004).
(TGF-) receptor I kinase inhibitor LY2157299 plus 226. Uhl, M. etal. SD208, a novel transforming growth
lomustine in patients with treatment-refractory factor receptor I kinase inhibitor, inhibits growth FURTHER INFORMATION
malignant glioma: the first human dose study. and invasiveness and enhances immunogenicity of Rosemary J. Akhursts homepage:
J.Clin. Oncol. 30, Abstract 2042 (2012). murine and human glioma cells invitro and invivo. http://cancer.ucsf.edu/research/akhurst-lab
214. Occleston, N. L. etal. Discovery and development of Cancer Res. 64, 79547961 (2004). Akiko Hatas homepage:
avotermin (recombinant human transforming growth This is the first report of the use of a chemical http://www.cvri.ucsf.edu/Scientist/Hata.shtml
factor 3): a new class of prophylactic therapeutic for inhibitor of TRI showing efficacy in reduced ClinicalTrials.gov website: http://www.clinicaltrials.gov
the improvement of scarring. Wound Repair Regen. tumour growth and invasion as well as increased Flamel Technologies and Digna Biotech Announce Multiple
19 (Suppl. 1), S38S48 (2011). immunogenicity in a mouse model of glioblastoma. Product Development Agreement (27 June 2011 press
215. Platten, M. etal. N[3,4dimethoxycinnamoyl]- 227. Ge, R. etal. Inhibition of growth and metastasis of release): http://www.dignabiotech.com/comun/fichero.
anthranilic acid (tranilast) inhibits transforming mouse mammary carcinoma by selective inhibitor
asp?id=105
growth factor relesase and reduces migration and of transforming growth factor type I receptor
invasiveness of human malignant glioma cells. kinase in vivo. Clin. Cancer Res. 12, 43154330 ALL LINKS ARE ACTIVE IN THE ONLINE PDF
Int. J.Cancer 93, 5361 (2001). (2006).

NATURE REVIEWS | DRUG DISCOVERY VOLUME 11 | O CTOBER 2012 | 811

2012 Macmillan Publishers Limited. All rights reserved

You might also like