Published OnlineFirst May 12, 2009; DOI: 10.1158/0008-5472.

CAN-08-2861

Research Article

The Trifunctional Antibody Ertumaxomab Destroys Tumor

Cells That Express Low Levels of Human Epidermal
Growth Factor Receptor 2
1 1 1 2 1,2
Michael Jager, Alexandra Schoberth, Peter Ruf, Jurgen Hess, and Horst Lindhofer
1
TRION Research GmbH, Martinsried, Germany and 2TRION Pharma GmbH, Munich, Germany

Abstract The humanized monoclonal anti-HER2/neu antibody trastuzu-

mab inhibits growth of tumor cell lines that strongly express the
Human epidermal growth factor receptor 2 (HER2/neu) is an
HER2/neu antigen (6). Several clinical studies have shown the
important target for the treatment of the breast cancers in
greatest benefit from trastuzumab treatment (7, 8) among women
which it is overexpressed. However, no approved anti-HER2/
with metastatic breast cancers (scored 3+ or 2+ by immunohis-
neu therapy is available for the majority of breast cancer
tochemisty), and with HER2 gene amplifications (confirmed by
patients, who express HER2/neu at low levels (with scores of
fluorescent in situ hybridization; FISH). Based on these results,
1+ or 2+/fluorescence in situ hybridizationnegative). The
assessment of the HER2/neu status is absolutely required for all
trifunctional antibody ertumaxomab targets HER2/neu, CD3,
breast cancer patients who may be considered for trastuzumab
and activating Fc; receptors. In presence of ertumaxomab,
therapy. However, trastuzumab therapy cannot be offered to the
tri-cell complexes consisting of tumor cells, T cells, and
majority of breast cancer patients who have low levels of HER2/neu
accessory cells form to cause tumor cell lysis. In a phase I trial
expression (scored 1+ and 2+) and negative FISH results.
with metastatic breast cancer patients, ertumaxomab could be
Ertumaxomab is a new member of a family of trifunctional
applied safely and resulted in radiographically confirmed
bispecific antibodies (anti-HER2/neu  anti-CD3). Kiewe and
clinical responses. In this study, we compare ertumaxomab-
colleagues (9) presented the first promising clinical data on the
and trastuzumab-mediated killing of cancer cell lines that
safety and efficacy of ertumaxomab in the treatment of metastatic
express HER2/neu at low and high levels. Under optimal
breast cancer patients with different HER2/neu expression levels.
conditions for trastuzumab-mediated destruction of HER2/
Such as BiUII and catumaxomab, which target EpCAM instead of
neu-overexpressing cells, only ertumaxomab was able to
HER2/neu, ertumaxomab evokes a concerted interaction of
mediate the elimination of tumor cell lines that express
different immune cell types directed against the tumor (10, 11).
HER2/neu at low levels (1+). Ertumaxomab-mediated activity
Such as BiUII, ertumaxomab might simultaneously recruit and
was accompanied by a Th1-based cytokine release, a unique
activate FcgRI and FcgRIII-positive accessory cells (i.e., monocytes,
mode of action of trifunctional antibodies. Competitive
macrophages, natural killer cells, and dendritic cells) through its
binding studies with trastuzumab and 520C9 mapped the
unique isotype combination (mouse IgG2a and rat IgG2b), leading
binding site of ertumaxomab to the extracellular regions II
to the phagocytosis of the tumor cells (12). The importance of the
and III of the HER2/neu ectodomain. This site is distinct from
mouse/rat hybrid Fc region in the process of immunization has
the binding site of trastuzumab, so that HER2/neu-expressing
been shown in immunocompetent mouse tumor models, using the
tumor cells can be eliminated by ertumaxomab in the
trifunctional surrogate antibody BiLu (anti-human EpCAM x anti-
presence of high amounts of trastuzumab. The ability of
murine CD3; refs. 13, 14). As evidenced by Riesenberg and
ertumaxomab to induce cytotoxicity against various tumor
colleagues (15), perforin-mediated cytotoxicity may contribute to
cell lines, including those with low HER2/neu antigen density,
the antitumor response. Taken together, these data suggest that
may provide a novel therapeutic option for breast cancer
ertumaxomab may induce the formation of a tri-cell-complex
patients who are not eligible for trastuzumab treatment.
consisting of HER2/neu+ tumor cells, CD3+ T cells, and FcgR+
[Cancer Res 2009;69(10):42706]
accessory cells, leading to efficient elimination of tumor cells.
In this study, we report on the high efficacy of the trifunctional
antibody ertumaxomab to specifically eradicate different HER2/
Introduction neu-positive tumor cell lines, accompanied by the activation of a
The proto-oncogene HER2 codes for the human epidermal Th1-type cytokine pattern. In contrast to the monospecific
growth factor receptor 2 (HER2/neu), which is overexpressed in humanized antibody trastuzumab, ertumaxomab destroys tumor
20% to 30% of breast cancer patients (13). HER2/neu over- cell lines with high and also with low HER2/neu expression. In
expression is usually based on gene amplification. It is correlated addition, we are able to show that trastuzumab and ertumaxomab
with a poor prognosis, reducing progression-free outcomes and recognize different epitopes on Her2/neu. The possible clinical
overall survival (4, 5). HER2/neu is an important target for implications of these findings are discussed.
antibody-mediated therapy in breast cancer patients.
Materials and Methods
Antibodies, effector cells, and target cell lines. We used freshly
Requests for reprints: Horst Lindhofer, TRION Pharma GmbH, Frankfurter Ring harvested peripheral blood mononuclear cells (PBMC) from healthy donors
193a, 80807 Munich, Germany. Phone: 49-89-324266100; Fax: 49-89-324266199; E-mail:
horst.lindhofer@trionpharma.de.
as effector cells. The cells were purified by density centrifugation through
I2009 American Association for Cancer Research. Ficoll Histopaque (PAN Biotech) at 897  g, 15 min, 20jC. They were
doi:10.1158/0008-5472.CAN-08-2861 washed twice with PBS without Mg2+ or Ca2+ (PAN Biotech), and

Cancer Res 2009; 69: (10). May 15, 2009 4270 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on April 10, 2016. 2009 American Association for Cancer
Research.
 Published OnlineFirst May 12, 2009; DOI: 10.1158/0008-5472.CAN-08-2861

Ertumaxomab Lyses Tumor Cells with Low HER2/neu Levels

Figure 1. HER2/neu-expressing profiles of used cell lines: A, SK-BR-3, (B ) HCT-8, (C ) BT-20, and (D ) SK-LU-1. FACS, fluorescence activated cell sorter
histogram, FL1-H = 2502A + secondary detection antibody rat anti-mouse IgG H+L FITC, mouse IgG2a isotype control Me361 (TRION Research) + secondary
detection antibody rat anti-mouse IgG H+L FITC; MFI, mean fluorescence intensity; SABC, specific antigen binding capacity as determined by DAKO QIFIKIT.

centrifuged at 458  g, for 10 min, at 20jC. The supernatant was removed,
and the pellet was resuspended in 20 mL RPMI 1640, supplemented with
2 mmol/L glutamine, 1 mmol/L sodium pyruvate, 1 mmol/L nonessential
amino acids, and 10% FCS (PAN Biotech). Cell number and viability
were determined using a Neubauer counting chamber after trypan blue
staining (Sigma-Aldrich). The breast cancer cell lines SK-BR-3 (ATCC
HTB-30) and BT-20 (ATCC HTB-19), the human ileocaecal adenocarcinoma
cell line HCT-8 (ATCC CCL-244), and the lung cancer cell line SK-LU-1
(ATCC HTB-57) served as target cells in cytotoxicity assays with PBMC
effector cells, followed by 2,3-bis[2-methoxy-4-nitro-5-sulfophenyl]-
2H-tetrazolium-5-carboxanilide inner salt (XTT) cell proliferation assess-
ment (16). The trifunctional antibody ertumaxomab with anti-HER2/neu 
anti-CD3 antigen specificities was manufactured by TRION Pharma. The
HER2/neu-specific monoclonal mouse antibodies 2502A and 520C9 (ATCC
HB-8696; ref. 17) were produced by TRION Research. 2502A constitutes the
HER2/neu binding arm of ertumaxomab. The humanized IgG1 antibody
trastuzumab (Roche) also recognizes the HER2/neu protein.
Quantitative determination of the cell surface antigen HER2/neu.
HER2/neu antigen expression on the surface of tumor target cell lines was
quantified using DAKO QIFIKIT (DAKO Cytomation), according to the
manufacturers protocol. HER2/neu-specific 2502A was used as the primary
antibody for HER2/neu detection and quantification. HER2/neu antigen
quantity is indicated as specific antibody-binding capacity units after
subtraction of isotype control (mouse IgG2a) background.
FISH analysis. Target cell lines (e.g., SK-BR-3 or HCT-8) were spun down
on cytospin slides, and pretreated with pepsin (Sigma-Aldrich) for 5 min
at 37jC. Subsequently, cells were fixed with formaldehyde (Sigma-Aldrich)
and dehydrated in alcohol (70 100%). Probe hybridization was then
performed with SPEC HER2/CEN 17 (Zytovision), overnight at 37jC. The
SPEC HER2/CEN 17 Dual Color Probe is a mixture of an orange
fluorochrome directlabeled CEN 17 probe, specific for the a satellite
centromeric region of chromosome 17 (D17Z1), and a green fluorochrome
directlabeled SPEC HER2 probe, specific for the HER2 gene at 17q12. After
counterstaining with 4,6-diamidino-2-phenylindole (Zytovision), slides were
evaluated at 100 oil magnification, applying a computerized image
analysis (MDS; Applied Imaging).
Fluorescence-activated cell sorter binding competition analysis.
Highly positive HER2/neu SK-BR-3 cells were preincubated with varying
concentrations of either 520C9 (10,000100 ng/mL) or trastuzumab (10,000
100 ng/mL) for 10 min, followed by addition of a constant concentration of
ertumaxomab (1,000 ng/mL), and further incubation for 45 min. After a
washing step, ertumaxomab cell binding was detected by fluorescence-
activated cell sorting (FACS) with an anti-rat IgG (H+L) FITC-labeled
secondary detection antibody (Dianova).
Cytotoxicity assay. Effector cells (2  105; PBMC) and target cells
(SK-BR-3, HCT-8, BT-20, or SK-LU-1) at varying E/T ratios were coincubated
in flat-bottomed 96-well plates (Greiner) with either trastuzumab or
ertumaxomab and combinations of different antibodies concentrations as
indicated. PBMCs coincubated with tumor cells alone (allogeneic setting)
were used as controls. After 4 d, PBMCs were discarded by washing, and
proliferation of tumor cells was analyzed using the XTT Cell Proliferation
kit II, as described by the manufacturer (Roche). Absorbance was measured
in a Versamax microplate reader (Molecular Devices), and raw data were
analyzed in Excel XP (Microsoft). Percent cytotoxicity was calculated as
follows: [(absorbance allogeneic reaction absorbance sample)/(absor-
bance allogeneic reaction absorbance PBMC)]  100. All tests were
performed in duplicate, and experiments were repeated with PBMC from
three different healthy donors. It should be noted that the value of

www.aacrjournals.org 4271 Cancer Res 2009; 69: (10). May 15, 2009

Downloaded from cancerres.aacrjournals.org on April 10, 2016. 2009 American Association for Cancer
Research.
 Published OnlineFirst May 12, 2009; DOI: 10.1158/0008-5472.CAN-08-2861

Cancer Research

allogeneic cytotoxicity in the absence of respective antibodies in each
sample was always below 13%. In addition, by using these specific assay
conditions all antibody entities such as trastuzumab or ertumaxomab
completely failed to inhibit growth of respective tumor target cells (HCT-8)
or showed only minimal cytotoxic activity (SK-BR-3 cells) in the absence of
PBMCs (data not shown). The HER2/neu specificity of ertumaxomab was
analyzed by preincubation of the PBMC effector and HCT-8 target cells with
excess amounts of the anti-HER2/neu blocking antibody 2502A (200, 500, or
2,000 ng/mL) for 15 min, followed by addition of different concentrations of
ertumaxomab (100 0.001 ng/mL).
Measurement of cytokines. Ertumaxomab- or trastuzumab-induced
cytokine release was determined by culturing PBMCs from healthy donors
with target cells (HCT-8) for 24 h in 96 flat-bottomed well plates.
Ertumaxomab (100-0.001 ng/mL) or trastuzumab (5,000-0.001 ng/mL) were
added at different concentrations. After 24 h, the supernatants were
collected and frozen at 20jC. Cytokines were measured with the human
Th1/Th2 cytometric bead array kit (BD Biosciences) comprising interleukin
(IL)-2, IL-6, IFN-g, and tumor necrosis factor-a (TNF-a; data not shown for
TNF-a). Data acquisition and analysis were performed using a FACS-Calibur
with the cytometric bead array software (BD Biosciences).
Analysis of cytokine-induced cytotoxicity. Cytokine-induced killing of
the target cell line HCT-8 was analyzed by transferring the supernatants of
cytotoxicity assay samples to 3  104 freshly plated tumor cells (HCT-8).
Tumor cell proliferation was analyzed using the XTT Cell Proliferation kit II
(Roche). Extinction was measured in a Versamax microplate reader
(Molecular Devices). Raw data were analyzed in Excel XP (Microsoft). All
experiments were performed in duplicates.
superior ability of ertumaxomab to lyse cells that express high
levels of HER2/neu under unfavorable E/T ratios.
Cytotoxicity of ertumaxomab and trastuzumab against
tumor cells with a low HER2/neu expression profile. The
efficacious killing of SK-BR-3 tumor cells by ertumaxomab, even at
low effector cell numbers, led to the hypothesis that ertumaxomab
might also be able to eliminate tumor cells that express HER2/neu
at low levels (1+). Based on this hypothesis, a cytotoxicity assay was
established to analyze the antitumor efficacy of ertumaxomab,
compared with trastuzumab, on cell lines that express low levels of
HER2/neu: HCT-8, BT-20, and SK-LU-1. Each of the three cell lines,
derived from different carcinomas (colon, breast, and lung), was
completely killed in the presence of ertumaxomab in a dose range
of z5 ng/mL. In contrast, trastuzumab entirely failed to exert any
cytotoxic effects on HCT-8, BT-20, and SK-LU-1 cancer cells, even at
high concentrations up to 5,000 ng/mL, and with E/T ratios of
20:1 (Fig. 3A, C, and D)the conditions that are optimal for
trastuzumab induced SK-BR-3 cell lysis (Fig. 2A). Interestingly, the
cytotoxic antitumor potential of ertumaxomab remained compa-
rable in these experiments, even at an unfavorable E/T ratio of 7:1
(Fig. 3B).
Cytokines induced by ertumaxomab in the presence of HCT-
8 target cells and PBMC. As both CD3+ T cells and accessory
immune cells are redirected by ertumaxomab to target cells, the
interaction of these immune cell types can be assessed by detecting
the relevant cytokines secreted into the supernatant of the

Results
Quantification of HER2/neu expression on target cells. To
evaluate the influence of the HER2/neu expression level on
ertumaxomab- and trastuzumab-mediated cytotoxicity, the
amount of HER2/neu antigen on the surface of different human
cancer cell lines (SK-BR-3/breast, BT20/breast, HCT-8/colon, and
SK-LU-1/lung) was determined. As expected, SK-BR-3 cells that are
scored 3+ (17, 18) stained intensively with anti-HER2/neu
monoclonal antibody (mAb) 2502A. Consistent with the intensive
staining, the DAKO QIFI test displayed a high specific antigen
binding capacity (Fig. 1A). The HCT-8 cells showed a significantly
weaker HER2/neu surface density, as did the other cell lines tested
(BT-20 and SK-LU-1). Subsequently, the FACS analyses were
completed by FISH. Thereby, SK-BR-3 cells revealed amplification
of the HER2/neu gene locus, whereas HCT-8, BT-20, and SK-LU-1
cells did not (Fig. 1A, B, C, and D). In summary, the target cell lines
HCT-8, BT-20, and SK-LU-1 showed low levels of HER2/neu
expression (1+), whereas SK-BR-3 cells scored 3+ for HER2/neu,
in agreement with the DAKO HercepTest classification.
Cytotoxicity of ertumaxomab and trastuzumab against
tumor cells with a high HER2/neu expression profile.
Trastuzumab is well-known to promote tumor cell death in
HER2/neu-overexpressing cell lines such as SK-BR-3 (17, 18).
We compared the ability of ertumaxomab to inhibit SK-BR-3
growth with trastuzumab at effector (PBMC) to target ratios (E/T)
of 20:1 and 5:1. Different antibody concentrations of trastuzumab
(5,0000.001 ng/mL) and ertumaxomab (100 0.001 ng/mL) were
added to allogeneic cell samples. As shown in Fig. 2A, SK-BR-3 cells
were completely lysed by both ertumaxomab and trastuzumab at
an E/T ratio of 20:1. However, at suboptimal E/T conditions (5:1), Figure 2. Cytotoxicity against SK-BR-3 cells (HER2 high, HER2 ampl+)
trastuzumab-induced killing of SK-BR-3 cells declined to f40%, at E/T ratios of (A) 20:1 and (B) 5:1 mediated by trastuzumab (5,0000.001
even at high concentrations (5,0001,000 ng/mL; Fig. 2B). In ng/mL) or ertumaxomab (1000.001 ng/mL). Cytotoxicity experiments were
performed thrice and samples measured in duplicates with 2  105 PBMC
contrast, maximal target cell killing remained at ertumaxomab of 3 different healthy donors; sample: PBMC + SK-BR-3 + antibody; allogeneic
concentrations as low as 25 ng/mL. These results show the reaction: PBMC + SK-BR-3 cells. Points, mean; bars, SD.

Cancer Res 2009; 69: (10). May 15, 2009 4272 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on April 10, 2016. 2009 American Association for Cancer
Research.
 Published OnlineFirst May 12, 2009; DOI: 10.1158/0008-5472.CAN-08-2861

Ertumaxomab Lyses Tumor Cells with Low HER2/neu Levels

Figure 3. Cytotoxicity against (A ) HCT-8 cells (HER2 low, HER2 ampl ; E/T, 20:1); (B) HCT-8 cells (E/T, 7:1); (C ) BT-20 cells (HER2 low, HER2 ampl ; E/T, 20:1)
and (D ) SK-LU-1 cells (HER2 low, HER2 ampl ; E/T, 20:1) mediated by trastuzumab (5,0000.001 ng/mL) or ertumaxomab (1000.001 ng/mL). Cytotoxicity
experiments were performed thrice and samples measured in duplicates with 2  105 PBMC of 3 different healthy donors; sample: PBMC + HCT-8 + antibody;
allogeneic reaction: PBMC + HCT-8 cells. Points, mean; bars, SD.

respective cell growth media. Doses of ertumaxomab ranging from
1 to 100 ng/mL were able to stimulate high levels of the pro-
inflammatory cytokines [i.e., IL-6; IFN-g; TNF-a (data not shown)]-
after 24 hours of incubation time (Fig. 4A and B). Of note, in the
presence of accessory cells, T cells, and tumor cells, ertumaxomab
also induced the production of IL-2 (Fig. 4C). A strong Th1
response is suggested by the high IFN-g and IL-2 levels that were
present in samples treated with ertumaxomab. Furthermore, IL-6
secretion was increased, indicating a proinflammatory response with
the contributions of accessory cells (19). In contrast, trastuzumab
merely stimulated IL-6 secretion, and at significantly lower levels
(Fig. 4B). Interestingly, supernatant transfer experiments showed
that all the cytokines induced by ertumaxomab did not have any
effect on the growth of HCT-8 cells (Fig. 4AC). These results strongly
support the view that ertumaxomab-induced killing of HCT-8 cells is
specific and not mediated by the cytokines per se and depends on
physical contact of the redirected effector cells and HCT-8 cells.
Our results show that the release of proinflammatory cytokines
and IL-2, which is stimulated by ertumaxomab, reflects the
engagement of both Fcg receptor+ accessory immune cells and
CD3+ T-cells. The less pronounced IL-6 secretion induced by
trastuzumab is probably evoked by binding to FcgR on accessory
cells alone.
Dependence of ertumaxomab-mediated lysis on HER2/neu
binding. To address the question of whether the provoked
cytotoxic effects of ertumaxomab on HCT-8 target cells depend
on the expression of the HER2/neu antigen, we initiated blocking
experiments with an HER2/neu antibody. We preincubated the
cytotoxicity samples with excess amounts of anti-HER2/neu
antibody 2502A. In three independent experiments with effector
cells from different donors, we observed a significant dose-
dependent reduction of tumor cell killing at 2502A blocking
concentrations between 200 and 2,000 ng/mL (Fig. 4D). These
results show that direct binding of ertumaxomab to HER2/neu is
mandatory for target cell destruction.
Inhibition of ertumaxomab binding to HER2/neu by
antibody 520C9. To define the anti-HER2/neu binding region of
ertumaxomab, competitive binding analysis of trastuzumab with
mAb 520C9 were performed on SK-BR-3 cells. MAb 520C9 was able
to inhibit ertumaxomab binding but trastuzumab was not. This
result suggests that 520C9 and ertumaxomab recognize similar
epitopes (Fig. 5A). Indeed, the HER2/neu binding site of the mAb
520C9 was previously mapped to the extracellular domain of
HER2/neu (amino acid positions 243370), covering parts of
subdomains II and III (20). This 520C9 site is distinct from the
binding site of trastuzumab, which is located in region IV (21). We
conclude that trastuzumab and ertumaxomab have different
binding epitopes on HER2/neu, which do not interfere with each
other. Having shown that ertumaxomab and trastuzumab recog-
nize different HER2/neu epitopes (Fig. 5A), we further asked
whether ertumaxomab was able to induce the killing of target
cells that express low levels of HER2/neu (e.g., HCT-8) in the

www.aacrjournals.org 4273 Cancer Res 2009; 69: (10). May 15, 2009

Downloaded from cancerres.aacrjournals.org on April 10, 2016. 2009 American Association for Cancer
Research.
 Published OnlineFirst May 12, 2009; DOI: 10.1158/0008-5472.CAN-08-2861

Cancer Research

presence of trastuzumab. Both antibodies, ertumaxomab and
trastuzumab, were used in combination in competitive cytotoxicity
assays (Fig. 5B). Effector and HCT-8 target cells were preincubated
with a constant concentration of trastuzumab (5,000 ng/mL),
followed by the addition of ertumaxomab (50 ng/mL). In this
setting, trastuzumab alone (5,000 ng/mL) failed again to lyse
HCT-8 target cells. Ertumaxomab-mediated maximal cell lysis was
present at concentration as low as 50 ng/mL. No difference was
observed in the killing efficacy of ertumaxomab between samples
with (100-fold excess) trastuzumab or without. The observation
that excess amounts of trastuzumab do not hamper the lysis of
target cells by ertumaxomab confirms that the two antibodies
recognize different HER2/neu epitopes (Fig. 5A).
Discussion
It is well-established that tumor therapy with the anti-HER2/neu
antibody trastuzumab achieves significant survival benefits in
patients with HER2/neu overexpressing metastatic breast cancer,
whether trastuzumab is used with or without chemotherapy as a
first-line therapy until disease progression (7, 8). Moreover, a
retrospective analysis of breast cancer patients enrolled during
clinical trials, with tumors scored 2+ or 3+ by immunohistochem-
istry, showed the most beneficial treatment effects in patients who
carried FISH-positive tumors. This retrospective analysis indicates
that breast cancer HER2/neu overexpression most frequently
correlates with gene amplification (22). Unfortunately, overexpres-
sion of the HER2/neu antigen is only detectable in 20% to 30% of
breast cancer patients (1, 2), leaving the majority with fewer
therapeutic options.
This study describes the cytotoxic capacity of the trifunctional
antibody ertumaxomab for tumor cell lines that express HER2/neu
at high and low levels, in the presence of immune effector cells
(PBMC) in vitro. Only slight differences could be assessed in the
killing efficiency of SK-BR-3 cells between trastuzumab and
ertumaxomab, at a relatively high E/T ratio of 20:1. However, at
more unfavorable E/T ratios (7:1 or lower), probably resembling the
situation at the tumor site, only ertumaxomab was able to
efficiently lyse SK-BR-3 cells (Fig. 2). This observed superiority of
ertumaxomab over trastuzumab is based on the particular mode of
action that all members of the trifunctional antibody family (e.g.,
ertumaxomab, catumaxomab, Bi20/FBTA05, BiLu, and BiUII) have
in common. These therapeutic antibodies induce a simultaneous
recruitment and activation of T cells and accessory cells, leading to
the destruction of targeted tumor cells by different killer
mechanisms (9, 12, 13, 15, 23).

Figure 4. A-C, influence of ertumaxomab-induced cytokines IFN-g, IL-6, and IL-2 on freshly plated HCT-8 cells (HER2 low, HER2 ampl ). Cytokine levels of
supernatants of ertumaxomab or trastuzumab cytotoxicity experiments were analyzed after 24 h. Supernatants containing ertumaxomab- or trastuzumab-induced
cytokines were transferred to freshly plated HCT-8 cells (1  104). HCT-8 growth (o) was measured after a 3-d incubation period with the XTT proliferation assay.
As a control was a sample of HCT-8 cells incubated with RPMI1640 medium, which was set as a standard of 100% HCT-8 growth; points, mean. Ab, antibody.
D, HER2/neu specificity of ertumaxomab-induced cytotoxicity against HCT-8 cells: Cytotoxicity experiments were done in duplicates with 2  105 PBMC of 3 different
healthy donors and an E/T ratio of 6:1. PBMC and target cells were preincubated with 200 ng/mL (E), 500 ng/mL (o), or 2,000 ng/mL (X) of the anti-HER2/neu
antibody 2502A followed by addition of ertumaxomab (1000.001 ng/mL). Ertumaxomab-induced cytotoxicity was compared with samples without preincubation of
2502A (n). Points, mean; bars, SD.

Cancer Res 2009; 69: (10). May 15, 2009 4274 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on April 10, 2016. 2009 American Association for Cancer
Research.
 Published OnlineFirst May 12, 2009; DOI: 10.1158/0008-5472.CAN-08-2861

Ertumaxomab Lyses Tumor Cells with Low HER2/neu Levels

Figure 5. A, FACS competition binding: inhibition of ertumaxomab (ertu; 1 Ag/mL) binding to SK-BR-3 cells in the presence of varying ratios of 520C9 or trastuzumab
(trastu) compared with ertumaxomab (10:1, 1:1, 1:10); all experiments were performed thrice; as control, the binding capacity of ertumaxomab, 520C9, and trastuzumab
to SK-BR-3 cells was analyzed; points, mean; bars, SD. B, cytotoxicity of ertumaxomab (50 ng/mL) induced against HCT-8 cells (HER2 low, HER2 ampl ) after
preincubation with a 100-fold higher concentration of trastuzumab (5,000 ng/mL). As controls were samples with trastuzumab (5,000 ng/mL) and ertumaxomab (50 ng/mL)
alone. Cytotoxicity experiments were done in duplicate with 2  105 PBMC of 3 different healthy donors and a HCT-8 cell concentration of 1  104 (E/T, 20:1); sample:
PBMC + HCT-8 + antibody; allogeneic reaction: PBMC + HCT-8 cells. Points, mean; bars, SD.

The unique mode of action of ertumaxomab led us to the
hypothesis that tumor cells with low HER2/neu expression profiles,
categorized as immunohistochemisty 1+ and FISH negative, might
be potential targets for ertumaxomab therapy. To address this
question, we chose three tumor cell lines originating from different
tissues (breast, colon, and lung), each of which has a detectable low
HER2/neu-expressing profile, identified by quantitative FACS
measurements and FISH analysis. Cytotoxicity assays revealed that
ertumaxomab has a strong killing activity against all three cell lines
at concentrations above 5 ng/mL, and at E/T ratios of 20:1 or even
lower (Fig. 3). Trastuzumab, which acts against tumor cells mainly
by means of antibody-dependent cellular cytotoxicity (17, 24),
without contributions from T cells, completely failed to inhibit
growth of these three cell lines, even at high concentrations.
High quantities of IFN-g, TNF-a(data not shown), and IL-6 were
detected in the supernatants of mixed HCT-8/PBMC samples to
which ertumaxomab had been added. IL-2 secretion was stimulat-
ed as well as IFN-g, a strong indication of a Th1-type T-cell
activation. The induction of these cytokines confirms findings
described previously for other trifunctional antibodies (10, 25), and
supports the view that their mode of action is mainly independent
of the targeted antigen. The cytokines IFN-g, TNF-a, and IL-6 were
also released 3 or 6 hours after ertumaxomab infusion in a clinical
trial with metastatic breast cancer patients, confirming the
relevance of our in vitro results for the clinic (9).
Trastuzumab induced significant amounts of only IL-6, which is
most likely attributable to accessory cells stimulated by Fcg
receptor engagement (Fig. 4B; ref. 26). These results further
emphasize the substantial differences in the mechanisms of action
between ertumaxomab and trastuzumab.
Because the transfer of ertumaxomab-treated mixed cell culture
supernatants had no growth inhibition effect on freshly plated
HCT-8 cells, the killing of this cell line was not induced by the
detected mediator substances, such as TNF-a or IFN-g. Although
sufficient at low levels, the necessity of HER2/neu antigen
expression for ertumaxomab-provoked tumor cell destruction

www.aacrjournals.org 4275 Cancer Res 2009; 69: (10). May 15, 2009

Downloaded from cancerres.aacrjournals.org on April 10, 2016. 2009 American Association for Cancer
Research.
 Published OnlineFirst May 12, 2009; DOI: 10.1158/0008-5472.CAN-08-2861

Cancer Research

was finally shown by HER2/neu blocking experiments. Preincuba-
tion of HCT-8 target cells with the parental bivalent anti-HER2/neu
antibody 2502A inhibited ertumaxomab-induced killing up to 60%.
This inhibition shows that physical contact of ertumaxomab with
the tumor cell is required for its efficacy (Fig. 4D). This finding has
been confirmed in vivo for the trifunctional surrogate antibody
BiLu, which has no effect on the growth of target antigen negative
tumor cells, whereas it mediates full protection against antigen-
positive tumor cells in mice (13).
An important result, with possible consequences for clinical use,
was that ertumaxomab and trastuzumab recognize two different
HER2/neu epitopes. The HER2/neu binding site for trastuzumab is
located in the cysteine-rich extracellular subdomain IV (amino
acids 529627; ref. 21). Ertumaxomab binding is not competitive
with trastuzumab. It is competitive with mAb 520C9 (Fig. 5A),
whose HER2/neu binding site was previously mapped to the
extracellular HER2/neu subdomains II and III (amino acid positions
243370; ref. 20). The HER2/neu binding epitope of ertumaxomab
must be distinct from that of trastuzumab. In consequence, patients
that have already received trastuzumab and are refractory to
trastuzumab treatment may benefit from ertumaxomab adminis-
tration because no interference in binding or killing efficacy was
observed in coincubation experiments (Fig. 5B). Therefore, a
subsequent or even simultaneous application of both antibodies
could be reasonable. In conclusion, ertumaxomab may give new
treatment opportunities for breast cancer patients with HER2/neu
expression, independent of the expression profile that is currently
under investigation in phase II clinical studies.
Disclosure of Potential Conflicts of Interest
H. Lindhofer: commercial research support, TRION Research GmbH; ownership
interest and patents, TRION Pharma GmbH. The other authors disclosed no potential
conflicts of interest.
Acknowledgments
Received 7/25/08; revised 1/28/09; accepted 3/25/09.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
We thank Melanie Honcia, Annette Arbter, Melanie Goelden, and Kathrin Gasow
for their expert technical assistance.

References
1. Owens MA, Horten BC, Da Silva MM. HER2 amplifi-
cation ratios by fluorescence in situ hybridization and
correlation with immunohistochemistry in a cohort of
6556 breast cancer tissues. Clin Breast Cancer 2004;5:
639.
2. Ross JS, Fletcher JA, Bloom KJ, et al. Targeted therapy
in breast cancer: the HER-2/neu gene and protein. Mol
Cell Proteomics 2004;3:37998.
3. van de Vijver MJ, Mooi WJ, Peterse JL, Nusse R.
Amplification and over-expression of the neu oncogene
in human breast carcinomas. Eur J Surg Oncol 1988;14:
1114.
4. Slamon DJ, Clark GM, Wong SG, Levin WJ, Ullrich A,
McGuire WL. Human breast cancer: correlation of
relapse and survival with amplification of the HER-2/
neu oncogene. Science 1987;235:17782.
5. Tandon AK, Clark GM, Chamness GC, Ullrich A,
McGuire WL. HER-2/neu oncogene protein and
prognosis in breast cancer. J Clin Oncol 1989;7:
11208.
6. Hudziak RM, Lewis GD, Winget M, Fendly BM,
Shepard HM, Ullrich A. p185HER2 monoclonal antibody
has antiproliferative effects in vitro and sensitizes
human breast tumor cells to tumor necrosis factor.
Mol Cell Biol 1989;9:116572.
7. Slamon DJ, Leyland-Jones B, Shak S, et al. Use of
chemotherapy plus a monoclonal antibody against
HER2 for metastatic breast cancer that overexpresses
HER2. N Engl J Med 2001;344:78392.
8. Cobleigh MA, Vogel CL, Tripathy D, et al. Multina-
tional study of the efficacy and safety of humanized
anti-HER2 monoclonal antibody in women who have
HER2-overexpressing metastatic breast cancer that has
progressed after chemotherapy for metastatic disease.
J Clin Oncol 1999;17:263948.
9. Kiewe P, Hasmuller S, Kahlert S, et al. Phase I trial of
the trifunctional anti-HER2 x anti-CD3 antibody ertu-
maxomab in metastatic breast cancer. Clin Cancer Res
2006;12:308591.
10. Zeidler R, Reisbach G, Wollenberg B, et al.
Simultaneous activation of T cells and accessory
cells by a new class of intact bispecific antibody
results in efficient tumor cell killing. J Immunol 1999;
163:124652.
11. Burges A, Wimberger P, Kumper C, et al. Effective
relief of malignant ascites in patients with advanced
ovarian cancer by a trifunctional anti-EpCAM x anti-
CD3 antibody: a phase I/II study. Clin Cancer Res 2007;
13:3899905.
12. Zeidler R, Mysliwietz J, Csanady M, et al. The
Fc-region of a new class of intact bispecific antibody
mediates activation of accessory cells and NK cells and
induces direct phagocytosis of tumour cells. Br J Cancer
2000;83:2616.
13. Ruf P, Lindhofer H. Induction of a long-lasting
antitumor immunity by a trifunctional bispecific
antibody. Blood 2001;98:252634.
14. Morecki S, Lindhofer H, Yacovlev E, Gelfand Y, Ruf P,
Slavin S. Induction of long-lasting antitumor immunity
by concomitant cell therapy with allogeneic lympho-
cytes and trifunctional bispecific antibody. Exp Hematol
2008;36:9971003.
15. Riesenberg R, Buchner A, Pohla H, Lindhofer H. Lysis
of prostate carcinoma cells by trifunctional bispecific
antibodies (a EpCAM x a CD3). J Histochem Cytochem
2001;49:9117.
16. Roehm NW, Rodgers GH, Hatfield SM, Glasebrook
AL. An improved colorimetric assay for cell proliferation
and viability utilizing the tetrazolium salt XTT.
J Immunol Methods 1991;142:25765.
17. Lewis GD, Figari I, Fendly B, et al. Differential
responses of human tumor cell lines to anti-p185HER2
monoclonal antibodies. Cancer Immunol Immunother
1993;37:25563.
18. Sarup JC, Johnson RM, King KL, et al. Characteriza-
tion of an anti-p185HER2 monoclonal antibody that
stimulates receptor function and inhibits tumor cell
growth. Growth Regul 1991;1:7282.
19. Jones SA. Directing transition from innate to
acquired immunity: defining a role for IL-6. J Immunol
2005;175:34638.
20. Boyer CM, Pusztai L, Wiener JR, et al. Relative
cytotoxic activity of immunotoxins reactive with
different epitopes on the extracellular domain of the
c-erbB-2 (HER-2/neu) gene product p185. Int J Cancer
1999;82:52531.
21. Cho HS, Mason K, Ramyar KX, et al. Structure of the
extracellular region of HER2 alone and in complex with
the Herceptin Fab. Nature 2003;421:75660.
22. Dowsett M, Bartlett J, Ellis IO, et al. Correlation
between immunohistochemistry (HercepTest) and fluo-
rescence in situ hybridization (FISH) for HER-2 in 426
breast carcinomas from 37 centres. J Pathol 2003;199:
41823.
23. Stanglmaier M, Faltin M, Ruf P, Bodenhausen A,
Schroder P, Lindhofer H. Bi20 (FBTA05), a novel
trifunctional bispecific antibody (anti-CD20 x anti-
CD3), mediates efficient killing of B-cell lymphoma cells
even with very low CD20 expression levels. Int J Cancer
2008;123:11819.
24. Cooley S, Burns LJ, Repka T, Miller JS. Natural killer
cell cytotoxicity of breast cancer targets is enhanced by
two distinct mechanisms of antibody-dependent cellular
cytotoxicity against LFA-3 and HER2/neu. Exp Hematol
1999;27:153341.
25. Ruf P, Jager M, Ellwart J, Wosch S, Kusterer E,
Lindhofer H. Two new trifunctional antibodies for the
therapy of human malignant melanoma. Int J Cancer
2004;108:72532.
26. Clynes RA, Towers TL, Presta LG, Ravetch JV.
Inhibitory Fc receptors modulate in vivo cytoxicity
against tumor targets. Nat Med 2000;6:4436.

Cancer Res 2009; 69: (10). May 15, 2009 4276 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on April 10, 2016. 2009 American Association for Cancer
Research.
 Published OnlineFirst May 12, 2009; DOI: 10.1158/0008-5472.CAN-08-2861

The Trifunctional Antibody Ertumaxomab Destroys Tumor

Cells That Express Low Levels of Human Epidermal Growth
Factor Receptor 2
Michael Jger, Alexandra Schoberth, Peter Ruf, et al.

Cancer Res 2009;69:4270-4276. Published OnlineFirst May 12, 2009.

Updated version Access the most recent version of this article at:
doi:10.1158/0008-5472.CAN-08-2861

Cited articles This article cites 26 articles, 11 of which you can access for free at:
http://cancerres.aacrjournals.org/content/69/10/4270.full.html#ref-list-1

Citing articles This article has been cited by 3 HighWire-hosted articles. Access the articles at:
http://cancerres.aacrjournals.org/content/69/10/4270.full.html#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Subscriptions Department at pubs@aacr.org.

Permissions To request permission to re-use all or part of this article, contact the AACR Publications
Department at permissions@aacr.org.

Downloaded from cancerres.aacrjournals.org on April 10, 2016. 2009 American Association for Cancer
Research.