WO2009106096A1

WO2009106096A1 - Treatment of resistant tumors with trifunctional antibodies

Info

Publication number: WO2009106096A1
Application number: PCT/EP2008/001551
Authority: WO
Inventors: Carsten Lindemann; Michael Kluge; Petra Schroeder; Kirsten Dettmar; Judith Atz; Ulrike Grimm
Original assignee: Fresenius Biotech GmbH
Current assignee: Neovii Biotech GmbH
Priority date: 2008-02-27
Filing date: 2008-02-27
Publication date: 2009-09-03
Anticipated expiration: 2010-08-27
Also published as: WO2009106321A1

Abstract

The invention relates to the field of cancer treatment using therapeutic antibodies. More specifically, the invention relates to a trifunctional bispecific monoclonal antibody with specificities against HER-2 and a T-cell specific cell surface protein for use in a method of treatment of HER-2 over-expressing tumors that exhibit a resistance against monospecific anti-HER-2 antibodies and/or tyrosine kinase inhibitors.

Description

Treatment of resistant tumors with trifunctional antibodies

Field of the invention

The present invention relates to the field of cancer treatment using therapeutic antibodies. More specifically, the invention relates to a trifunctional bispecific monoclonal antibody with specificities against HER-2 and a T-cell specific cell surface protein for use in a method of treatment of HER-2 over-expressing tumors that exhibit a resistance against monospecific anti-HER-2 antibodies and/or tyrosine kinase inhibitors.

Background of the invention

Her2/neu, also known as ErbB2, belongs to the human epidermal growth factor receptor (EGFR) gene family or HER family. Her2/neu encodes a tyrosine kinase receptor (HER-2), which is over-expressed in approximately 25% of invasive breast cancers. HER-2 over- expression has consistently been found to confer resistance to cytotoxic and endocrine therapy and to account for an aggressive biological behaviour, thereby resulting in shorter disease-free and overall survival in both, patients with early and advanced breast cancer. Upon binding of a growth factor, receptors of the EGFR family dimerize using HER-2 as their preferred binding partner. Heterodimerization induces intrinsic receptor tyrosine kinase mediated autophosphorylation and subsequent activation of downstream signalling components via the MAPK and PI3K pathways, resulting in unresisted growth and enduring survival of the tumour cells. Therapy with the anti-HER-2 humanized monoclonal antibody trastuzumab (Herceptin , Genentech, South San Francisco, CA, US) has become a standard therapy in patients with HER-2 positive tumours, e.g. breast cancer and other types of tumors. There are several modes of action described for Herceptin (Nahta et al. Cancer Letters 2006a; 232: 123- 138). One is the effect of activating the immune system by binding the antibody to the tumor cell surface, referred to as antibody dependent cellular cytotoxicity (ADCC). Another important effect is the blocking of the surface receptor, thereby disturbing signal transduction. It is commonly accepted that the blocking of the dimerization and the signal transduction plays a more significant role for the potency of Herceptin^® than ADCC. For this reason, Herceptin^® also acts apoptotic in the absence of the immune system (see e.g. EP 865448).

However, therapy with Herceptin^® is only suitable for patients showing a high expression rate of HER-2 on the tumor cell surface. HER-2 over-expression is generally due to gene amplification and has been defined by immunohistochemistry as being highest (3+) when receptor levels approach 2 million, or medium intensity (2+) when receptor levels are approximately 500,000, whereas normal levels of HER-2 are reported to be 20,000 per cell. The level of HER-2 gene amplification in human cancer cells can be classified by fluorescence in situ hybridization (FISH). A therapy with Herceptin^® is only believed to be promising in tumors of patients, which tumors can be classified 2+/3+ and FISH+, respectively. It has been thought that the HER-2 expression in the other remaining patient groups is not sufficiently high to achieve a satisfactory effect when using monospecific antibodies such as Herceptin^®. However, it was found that trifunctional bispecific antibodies also produce an anti-tumor effect in 1+ classified tumor cells (EP 1820513). The objective response rates to trastuzumab monotherapy are quite low, ranging from 12% to 34% for a median duration of nine months, i.e. up to 88% of the patients show a de novo resistance against trastuzumab. Of the patients showing an initial response to trastuzumab- based therapy, however, a majority acquires a secondary resistance within one year of treatment initiation. Identification of novel agents that inhibit the growth of trastuzumab-resistant cells is important for improving the survival of metastatic breast cancer patients whose tumors over-express HER-2 (2+/3+/FISH+). It is proposed that resistance against Herceptin^® mainly develops, if either the binding of the monospecific antibody to HER-2 is blocked, or the interruption of the signal transduction is not blocked although the binding occurred, or other tumor growth and survival promoting signal pathways have replaced the HER-2 signal pathway. Binding of the antibody to the surface receptor can be blocked due to a change of the surface epitope, or by masking the epitopes through other proximal surface proteins, e.g. the membrane-associated glycoprotein mucin-4 (MUC4). It is thought that several reasons may explain the absence of an effect on signal transduction despite of the binding of the antibody. Among others, the intracellular part may be permanently activated, independently from a stimulation of the extracellular part. For these reasons, Herceptin^® resistant cells may also be resistant against other monospecific antibodies, e.g. pertuzumab, which bind another epitope than Herceptin^® thereby inhibiting dimerization of HER-2 and blocking signal transduction. To tackle this kind of resistance, tyrosine kinase inhibitors have been developed which target the tyrosine binding site in the intracellular part of HER-2. An example of such an intracellular tyrosine kinase inhibitor is lapatinib (Tykerb^®) (GlaxoSmithKline, Research Triangle Park, NC, US). However, there may be patient groups which exhibit a general resistance against HER-2 tyrosine kinase inhibitors, administered alone or in combination with Herceptin^®. hi these patients, signal transduction initiates growth, even though the function of the transmembrane receptor, e.g. HER-2, is blocked intra- as well as extracellularly. Resistance against tyrosine kinase inhibitors may originate from the high expression of molecular transporter proteins, such as P-glycoprotein (Pgp) and Breast Cancer Resistance Protein (BCRP) (Polli et al., Drug Metab. Dispos. 2008). Thus, there still remains the need to provide a treatment for patients having tumors being resistant against monospecific anti- HER-2 antibodies and/or tyrosine kinase inhibitors.

It is therefore an object of the invention to provide a trifunctional, bispecific antibody that is effective against tumor cells which are resistant against monospecific anti-HER-2 antibodies, in particular Herceptin^®, and/or tyrosine kinase inhibitors.

Summary of the invention

It has now surprisingly been found that a trifunctional, bispecific antibody can be used in a method of treatment of tumors exhibiting a de novo or secondary resistance against a monospecific anti-HER-2 antibody, trastuzumab (Herceptin^®), as concluded from the Examples disclosed herein.

Accordingly, the present invention relates to a trifunctional, bispecific antibody for use in the treatment of a HER-2 over-expressing tumor, wherein the tumor is or becomes resistant against one or more monospecific anti-HER-2 antibody and/or one or more tyrosine kinase inhibitor, as defined in the claims. Furthermore, a pharmaceutical composition comprising one or more trifunctional, bispecific antibody and one or more anti-HER-2 antibody is provided as it is defined in the claims. Finally, a kit comprising one or more trifunctional, bispecific antibody and one or more anti-HER-2 antibody as defined in the claims is also provided. The various aspects of the invention as defined in the independent claims and the preferred embodiments contained in the dependent claims are herewith incorporated by reference.

Detailed description of the preferred embodiments

According to a first aspect, the present invention relates to a trifunctional, bispecific antibody having the following properties: (a) binding to a T-cell specific cell surface protein, (b) binding to the tumor-associated antigen HER-2 on a tumor cell, and (c) binding to Fc-gamma-receptor type I and/or type III positive cells, for use in the treatment of a HER-2 over-expressing tumor, wherein the tumor is or becomes resistant against one or more monospecific anti-HER-2 antibody and/or one or more tyrosine kinase inhibitor. A trifunctional, bispecific antibody for use in the present invention may be prepared in accordance with the procedures described in EP 763128, EP 826696 and EP 1820513. Preferably, the trifunctional, bispecific antibody binds to the T-cell specific cell surface protein CD3. In a particular advantageous embodiment, the trifunctional, bispecific antibody is an anti-HER-2 x anti-CD3 antibody binding to Fc-gamma-receptors type I and/or III. In particular, the Fc portion comprises the isotype combination rat- IgG2b/mouse-IgG2a. A particularly preferred antibody is ertumaxomab. Ertumaxomab is an intact bispecific antibody targeting HER-2 and CD3 with selective binding of activatory Fcγ type I/III receptors.

The trifunctional, bispecific antibody described above is used in the treatment of a HER-2 over-expressing tumor, wherein the tumor is or becomes resistant against one or more monospecific anti-HER-2 antibody and/or one or more tyrosine kinase inhibitor. As used herein, the term "is or becomes resistant" means that the tumor does not respond to the respective agent, i.e. shows a de novo resistance, or, respectively, the initial responders demonstrate disease progression within a certain period after initiation of the treatment, i.e. develop a secondary resistance (see also Bartsch et al., 2007). The treatment of the HER-2 over-expressing tumor preferably comprises administration of one or more monospecific anti-HER-2 antibodies and/or one or more tyrosine kinase inhibitors. The one or more monospecific anti-HER-2 antibody may be selected from, e.g. trastuzumab (Herceptin^®) and/or pertuzumab (Omnitarg^®, 2C4), or any other monospecific anti-HER-2 antibody suitable for the described treatment. The antibodies trastuzumab and pertuzumab are both commercially available (Genentech, US). The term "tyrosine kinase inhibitor" refers to a molecule, in particular a small molecule, which can pass the cell membrane and target the intracellular domain of HER-2 and/or the intracellular domain of any binding partner of HER-2 (e.g. Nahta et al., 2007). Preferred examples of such tyrosine kinase inhibitors are lapatinib (Tykerb)/GW572016, GlaxoSmithKline, NC, US), gefitinib (Iressa^®, ZD1839), imatinib (Gleevec^®, STI-571), erlotinib (Tarceva^®), lanafamib (Sarasar^®), sorafinib and/or sunitimib (see also Bartsch et al., 2007). It is believed that, based on the present disclosure, one of average skill in the art can define a protocol for use of the one or more monospecific anti-HER-2 antibody and/or one or more tyrosine kinase inhibitor.

HER-2 is reported to be usually present on a cell at a level of 20,000 receptors per cell. Due to HER-2 gene amplification or chromosome 17 polysomy in human cancer cells HER-2 will typically be over-expressed in tumor cells from a number of primary as well as secondary tumors. The term "HER-2 over-expressing tumor", as used herein, refers to a tumor expressing HER-2 at a level of about 50,000 to about 10,000,000 receptors/tumor cell, preferably at least about 75,000, 100,000, 125,000, 150,000, 200,000, 300,000, 400,000, 500,000, 1,000,000, 2,000,000 receptors/tumor cell to 10,000,000 receptors/tumor cell. The level of expression of HER-2 on the tumor cells can be determined in accordance with standard procedures known in the art. Preferably, the expression level of HER-2 is quantified by flow cytometry, as described in the experimental section below (see, e.g. Example 1). In a preferred embodiment, the status of HER-2 over-expression is evaluated by histochemical analysis using the HercepTest (DAKO, CA, US). The HercepTest is approved by the US Food and Drug Administration (FDA) for determining the suitability for trastuzumab treatment. The test provides an evaluation system for HER-2 comprising four steps, 0, 1+, 2+, 3+, referring to an approximate number of expressed target antigens on the surface of a tumor cell. In accordance with this system, cells expressing less than 20,000 HER-2 molecules on the target cell are classified as negative; cells with an expression of more than about 20,000 and up to about 100,000-110,000 molecules are classified as 1+; cells with an expression of up to 500,000 molecules as 2+, and cells with an expression of between about 2,000,000 and about 10,000,000 molecules are classified as 3+. hi accordance with a preferred embodiment, the HER-2 over-expressing tumor is classified by a value in the HercepTest of 2+ and/or 3+. Alternatively, or in addition to the above procedures for determining the level of expression of HER-2 receptors per tumor cell described above, the status of HER- 2 expression can be determined by fluorescence in-situ hybridization (FISH). The FISH assay was initially approved by the FDA for assessing prognosis and predicting response to standard chemotherapy and has now also been approved for determining the eligibility for trastuzumab treatment. The assay is commercially available (PathVysion test; Vysis, IL, US). It is particularly preferred that the HER-2 over-expressing tumor is a FISH positive (FISH+) tumor. The treatment described herein may be particular advantageous in case of tumors classified as FISH+ and 2+ and/or FISH+ and 3+. The HER-2 over-expressing tumor to be treated may be a breast tumor, ovarian tumor, prostate tumor, colon tumor, pancreas tumor, stomach tumor, esophagus tumor, endometrium tumor, skin tumor, oropharynx tumor, larynx tumor, cervix tumor, bladder tumor, preferably a carcinoma, more preferably an adenocarcinoma and/or a squamous cell carcinoma.

In another preferred embodiment, the treatment of HER-2 over-expressing tumors additionally comprises the administration of one or more monospecific antibody against an antigen other than HER-2, preferably a member of the HER family such as epidermal growth factor receptor (EGFR), HER-3, and/or HER-4. Particularly preferred are one or more anti-EGFR antibodies, in particular the monospecific anti-EGFR antibody cetuximab. Cetuximab is a chimeric monoclonal antibody targeting the extracellular domain of EGFR.

In another aspect, the invention relates to a pharmaceutical composition comprising (i) one or more trifunctional, bispecific antibody; and (ii) one or more anti-HER-2 antibody; optionally further comprising (iii) one or more monospecific antibody against an antigen other than HER-2, preferably one or more anti-EGFR antibodies, and/or (iv) one or more tyrosine kinase inhibitor. The components (i) to (iv) may take any form as defined above and any combination thereof.

According to the still further aspect of the invention, there is provided a kit comprising (i) one or more trifunctional, bispecific antibody, and (ii) one or more anti-HER-2 antibody; optionally further comprising (iii) one or more monospecific antibody against an antigen other than HER-2, preferably one or more anti-EGFR antibodies; and/or (iv) one or more tyrosine kinase inhibitor; wherein component (i) to (iv) are the above-described embodiments taken either alone or in combination with other embodiments described herein. In a still further aspect, it is provided a bispecific antibody or a fragment thereof, having the following properties: (a) binding to a tumor-associated antigen HER-2 on a tumor cell, and (b) binding to immunocompetent cells, preferably Fc-gamma-receptor type I and/or type III positive cells, in particular binding to an epitope on the Fc-gamma-receptor or to CD3, for use in the treatment of a HER-2 over-expressing tumor, wherein the tumor is or becomes resistant against one or more tyrosine kinase inhibitor. Preferably, the antibody fragment is a single chain antibody (scFv).

Description of the Figures

Figure 1. Schematic antibody structure of ertumaxomab.

A hybrid-hybridoma derived intact bispecific antibody with specificities against HER-2 and CD3 combining the two subclasses mouse IgG2a and rat IgG2b, which are evolutionary related and highly homologous Ig-subclasses.

Figure 2. Analysis of proliferation of JIMT-I and BT-474 cells incubated with trastuzumab; cell proliferation analyzed at cell density of 2.5xlO⁴ cells/mL.

Figure 3. Analysis of proliferation of JIMT-I and BT-474 cells incubated with trastuzumab; cell proliferation analyzed at cell density of 5x10⁴ cells/mL.

Figure 4. Analysis of ertumaxomab mediated cellular cytotoxicity towards JIMT-I cells.

(1^st set of experiments). Data show mean % residual tumor cells (plus standard deviation).

Figure 5. Analysis of ertumaxomab mediated cellular cytotoxicity towards SK-BR-3 cells. (1^st set of experiments). Data show mean % residual tumor cells (plus standard deviation).

Figure 6. Analysis of ertumaxomab mediated cellular cytotoxicity towards SK-OV-3 cells. (1^st set of experiments). Data show mean % residual tumor cells (plus standard deviation). Figure 7. Analysis of ertumaxomab mediated cellular cytotoxicity towards JIMT-I cells. (2^nd set of experiments). Data show mean % residual tumor cells (plus standard deviation).

Figure 8. Analysis of ertumaxomab mediated cellular cytotoxicity towards BT-474 cells. (1^st set of experiments). Data show mean % residual tumor cells (plus standard deviation).

The present invention is further described by the following examples which should not be construed as limiting the scope of the invention.

Example 1: Characterization of the tumor target cells

The functionality and cytotoxic activity of the test item ertumaxomab (anti-HER- 2 x anti-CD3) was investigated by an in vitro test system using the trastuzumab resistant HER-2 positive human tumor cell line JIMT-I (breast). As control test systems for the activity of trastuzumab the HER-2 positive human tumor cell lines SK-O V-3 (ovary), BT- 474 (breast) and SK-BR-3 (breast) were used. The features of the cell lines used in this study are summarized in Table 1.

The expression level of HER-2 in these cell lines was quantified by flow cytometry using a murine antibody directed against HER-2 (clone 9G6.10, Alexis) and fluorescence calibration beads (Quifikit, DAKO). 5xl0⁵-10⁶ tumor cells were incubated with 20 μg/mL anti-HER-2 antibody followed by staining with saturating concentrations of FITC- conjugated anti-murine IgG antibodies. All antibody incubations were conducted for 30 min at 2— 8°C. In parallel beads which have known numbers of binding sites for secondary anti-mouse IgG antibodies were stained with fluorescein isothiocyanate conjugated anti- mouse antibodies (Dianova). This allows to correlate the fluorescence intensity signals with HER-2 binding sites per cell.

The expression levels of the tumor target cells used in these examples are shown in Table 2. The number of binding sites per cell was related to the clinical HER-2 score according to Ross et al. MoI Cell Proteomics 2004; 3(4): 379-398. All cell lines are known to have an amplification of the Her-2/neu gene but show different levels of expression. Whereas JUVIT-1 cells are classified for the HER-2 status 2+ BT-474, SK-BR-3 and SK-OV-3 cells are classified as 3+ according to Ross et al., supra.

Table 2: Expression level of tumor target cells

To verify the resistance of JIMT-I cells to trastuzumab, JIMT-I tumor cells were incubated with trastuzumab and cell proliferation was determined by thymidine incorporation into the cellular DNA. As a control, BT-474 tumor cells that are sensitive to trastuzumab were used.

In detail, tumor cells were incubated for 9 days following the instructions of the manufacturer in medium supplemented with 10 and 100 μg/mL trastuzumab. As a control, cells were incubated without trastuzumab. Cells were then seeded into 96-well plates at 2.5xlO⁴ cells or 5xlO⁴ cells/well and pulsed with thymidine for 18 h at 37°C, 5%CO₂ and 95 relative humidity (rH). Each concentration was tested in 32 well of a 96-well plate. The whole plates were frozen and stored at -20°C until further use. For determination of the thymidine incorporation, cells were thawed for 1 h at 37°C. The medium was decanted and cells were detached with trypsin for 20 min at 37°C. Using a cell harvester, cell suspensions were collected onto a filter membrane and counted using a beta-counter. Results are determined as counts per minute (cpm). For further analysis proliferation was calculated as % of control (without trastuzumab):

% proliferation

cpm control

Mean values and standard deviations are derived from 32 wells per antibody concentration. Proliferation of the control cell line BT-474 was inhibited by trastuzumab at a concentration of 10 μg/mL and 100 μg/mL, whereas JIMT-I cells were not inhibited by trastuzumab even at 100 μg/mL (Table 3, Table 4, Figure 2, Figure 3).

Table 3: Analysis of proliferation of JIMT-I and BT-474 cells incubated with trastuzumab at 5xlO⁴/mL.

Table 4: Analysis of proliferation of JIMT-I and BT-474 cells incubated with trastuzumab at 2.5xlO⁴/mL.

In this example, the cytotoxic in vitro activity of ertumaxomab towards the HER-2 over- expressing human breast cancer cell line JIMT-I (FISH+/2+) was investigated. The JIMT- 1 cell line was isolated from a trastuzumab resistant patient with breast cancer and represents a model for trastuzumab (Herceptin^®) resistance in vitro and in vivo (Tanner et al., MoI. Cancer Ther. 2004; 3(12): 1585-1592). JIMT-I cells are characterized by an amplification of the Her-2/neu oncogene without any mutations in the coding sequence, a low shedding of HER-2 and a lack of growth inhibition by transtuzumab in vitro. Although JIMT- 1 cells express HER-2 only at a level of 100,000 molecules per cell (Mocanu et al., Cancer Lett. 2005; 227(2): 201-212) the expression levels of ER, PR, HER-I, HER-3 and HER-4 are similar to that of the trastuzumab-sensitive cell line SK-BR-3 (Mocanu et al, supra; Szollosi et al., Cancer Res. 1995; 55: 5400-5407). Trastuzumab resistance of JIMT- 1 cells is supposed to be based on reduced availability and a lack of activation of HER-2 (Nagy et al., Cancer Res. 2005; 65(2): 473-482). They express MUC4, a membrane- associated mucin that contributes to the masking of membrane proteins. It was concluded that masking of HER-2 in JIMT-I may lead to diminished trastuzumab binding. This study shows low binding of trastuzumab to JIMT-I cells when compared to SK-BR-3 cells that also over-express HER-2 but have no MUC4 expression.

In this example, the trastuzumab resistant phenotype of the JIMT-I cell line has been verified in vitro. Cell proliferation of JIMT-I cells was not decreased after incubation in the presence of 100 μg/mL trastuzumab for 9 days. In contrast, the trastuzumab sensitive cell line BT-474 showed a strong reduction of proliferation after incubation even with 10 μg/mL for 9 days.

Example 2: Binding of ertumaxomab and trastuzumab to tumor target cells

To confirm binding of ertumaxomab and trastuzumab to tumor target cells, 5x10⁵- 1x10⁶ JIMT-I cells were incubated with 4 μg/mL ertumaxomab or trastuzumab for 30 min at 2°-

8°C. Binding of antibodies was detected by using FITC-conjugated secondary antibodies directed against rat IgG (for analysis of ertumaxomab binding) or antibodies directed against human IgG (for detection of trastuzumab binding) in saturating concentrations.

This second incubation was conducted for 30 min at 2-8°C. After a washing step, mean fluorescence intensity (MFI) was measured by flow cytometry. Cells were gated to exclude cell debris and dead cells. As a control, binding analysis was performed with SK-OV-3

(ovary), BT-474 and SK-BR-3 (breast) tumor cells.

Binding of ertumaxomab and trastuzumab to JIMT-I cells was demonstrated (Table 5, Table 6), but weak compared to the binding to SK-OV-3, BT-474 and SK-BR-3 cells.

Table 5: Binding of ertumaxomab to tumor cells.

Table 6: Binding of trastuzumab to tumor cells

Example 3: Analysis of ertumaxomab mediated killing of JIMT-I cells

The test item ertumaxomab is supposed to induce a specific cell-mediated elimination of tumor cells in the presence of peripheral blood mononuclear cells (PBMC). To assess the cytotoxic in vitro activity, the test item was added in varying concentrations to co-cultures of tumor target cells and mononuclear cells at an effector to target ratio of 10:1.

In detail, SK-BR-3, SK-OV-3, BT-474 and JIMT-I tumor cells were seeded with 10⁴ cells per well in 96-well plates and incubated at 37°C and 5% CO2 for 24 h to assure adherence of the cells. The tumor cells were incubated for 3 days with medium containing mononuclear cells in the presence of varying concentrations of ertumaxomab or trastuzumab. Each concentration was tested in 8 wells of a 96-well plate. After 3 days incubation at 37°C and 5% CO2 the residual surviving tumor cells were quantified using the XTT-method:

Mitochondrial dehydrogenases of viable cells cleave the tetrazolium ring, yielding purple formazan. The absorbance of the resulting purple solution was measured spectrophotometrically at a wavelength of 450-500 nm. This method determines the enzymatic activity of viable cells which is directly correlated to the number of viable cells per sample. Data acquisition was performed with the software Magellan (Tecan). For further analysis the activity of residual tumor cells was calculated as percentage of residual tumor cells using the following formula: [mean absorbance sample - mean absorbance medium control]x 100

% residual tumor cells =-

[mean absorbance control without antibody - mean absorbance medium control] Mean values / SD are derived from 8 wells per antibody concentration.

In a first set of experiments ertumaxomab and trastuzumab were used at 0.33 ng/mL to 125 ng/mL (Table 7, Table 8, Table 9). In this setting the HER-2 over-expressing (3+) tumor cell lines SK-BR-3 (breast) and SK-OV-3 (ovary) were used as positive control for trastuzumab mediated cellular cytotoxicity.

Table 7: Analysis of ertumaxomab and trastuzumab mediated cellular cytotoxicity towards JIMT-I tumor cells.

Table 8: Analysis of ertumaxomab and trastuzumab mediated cellular cytotoxicity towards SK-BR-3 tumor cells.

(ng/mL)

% residual SK-BR-

91 .5 90.2 85 .7 82 .2 73 .9 71.9 67.2 62 .4 52 .6 49 .3 3 cells mean standard deviation 1. 1 0.4 0. 6 1. 2 1. 6 2.0 1.4 1. 6 1. 3 1. 3

Table 9: Analysis of ertumaxomab and trastuzumab mediated cellular cytotoxicity towards SK-O V-3 tumor cells.

In a second set of experiments ertumaxomab and trastuzumab were used at different, overlapping concentrations (ertumaxomab: 0.069 to 35.2 ng/mL; trastuzumab: 11.7 to 230.9 μg/mL). For this experiment BT-474 tumor cells were used as a positive control for trastuzumab mediated cellular cytotoxicity (Table 10, Table 11).

Table 10: Analysis of ertumaxomab and trastuzumab mediated cellular cytotoxicity towards JIMT-I tumor cells, n.d. = not determined.

Table 11 : Analysis of ertumaxomab and trastuzumab mediated cellular cytotoxicity towards BT-474 tumor cells, n.d. = not determined.

Ertumaxomab mediated a concentration-dependent decrease of SK-BR-3, SK-O V-3 and JIMT-I tumor cells. Ertumaxomab and trastuzumab are both efficient in killing BT-474 (Fig. 8), SKBR-3 (Figure 5) and SK-OV-3 tumor cells (Figure 6) in vitro. The efficiency of ertumaxomab was higher than that of trastuzumab for these cell lines. Trastuzumab showed no dose-dependent cytotoxicity towards JIMT-I tumor cells (Figure 4, Figure 7). Ertumaxomab is able to mediate a cellular cytotoxicity towards JIMT-I cells in vitro. This is in accordance with the finding that in vivo JIMT-I cells are prone to cellular cytotoxicity (Barok et al., MoI. Cancer Ther. 2007; 6(7): 2065-2072). Ertumaxomab and trastuzumab are both efficient in killing BT-474, SK-OV-3 and SK-BR-3 cells in vitro. The efficiency of ertumaxomab was higher than that of trastuzumab for these cell lines. Trastuzumab showed no cytotoxicity against JIMT-I cells. The observed different cytotoxic activities may be explained by the different mode of action.

In summary, this example shows that ertumaxomab efficiently kills the human breast cancer cell line JIMT-I in the presence of PBMC in a dose-dependent manner. No cytotoxic effect on JIMT-I cells could be observed with trastuzumab under the same experimental conditions. These in vitro data are believed to be predictable for use of the invention in a patient, because the efficient killing activity of ertumaxomab was shown for an accepted model of trastuzumab resistance.

Without being bound to a particular scientific theory, based on the above findings, it is contemplated that there is a different mode of action that is responsible for the different functionality of trastuzumab vs. ertumaxumab. Even though ertumaxomab binds to a different epitope of HER-2 than trastuzumab, it is believed that epitope binding is not a major contributor to resistance, as cross-resistance of Herceptin-resistant tumor cells to alternative monospecific antibodies (pertuzumab) does occur (Nahta et al., 2006b). This functionality may be independent from the level of binding to the tumor cells (cf. Example 2). The cytotoxicity of trastuzumab is believed to be mainly based on the interruption of signal transduction, whereas that of ertumaxumab results from an enhanced ADCC (cf. e.g. Table 7).

References

EP 0763128 B 1 Method for producing heterologous bispecific antibodies

EP 0826696 Bl Use of bi- and trispecific antibodies for inducine tumor immunity

EP 0865448 Apoptosis induced by monoclonal antibody anti-HER-2

EP 1820513 Al Destruction of tumor cells expressing low to medium levels of tumor associated target antigens by trifunctional bispecific antibodies Barok, M., Isola, J., Palyi-Krekk, Z., Nagy, P., Juhasz, I., Vereb, G., Kauraniemi, P., Kapanen, A., Tanner, M., Vereb, G., Szδllδsi, J. Trastuzumab causes antibody- dependent cellular cytotoxicity-mediated growth inhibition of submacroscopic JIMT-I breast cancer xenografts despite intrinsic drug resistance. MoI. Cancer Ther. 2007; 6(7): 2065-2072.

Bartsch, R., Wenzel, C, Zielinski, C. C, Steger, G. G. HER-2-positive breast cancer: hope beyond trastuzumab. BioDrugs. 2007; 21(2): 69-77.

Kallioniemi, O. P., Kallioniemi, A., Kurisu, W., Thor, A., Chen, L. C, Smith, H. S., Waldman, F. M., Pinkel, D., Gray J. W. ERBB2 amplification in breast cancer analyzed by fluorescence in situ hybridization. Proc Natl Acad Sci U S A. 1992; 89(12): 5321-5325.

Mocanu, M. M., Fazekas, Z., Petras, M., Nagy, P., Sebestyen, Z., Isola, J., Timar, J., Park, J. W., Vereb, G., Szollosi, J. Associations ofErbB2. betal-integrin and lipid rafts on Herceptin (Trastuzumab) resistant and sensitive tumor cell lines. Cancer Lett. 2005; 227(2): 201-212.

Nagy, P., Friedlander, E., Tanner, M., Kapanen, A. I., Carraway, K. L., Isola, J., Jovin, T. M. Decreased accessibility and lack of activation ofErbB2 in JIMT-I, a herceptinresistant, MUC4-expressing breast cancer cell line. Cancer Res. 2005; 65(2): 473-482.

Polli, J. W., Humphreys, J. E., Harmon, K. A., Castellino, S., O'Mara, M. J., Olson, K. L., St John-Williams, L. A., Koch, K. M., Serabj it-Singh, C. J. The Role of Efflux and Uptake Transporters in Lapatinib (Tykerb, GW 572016) Disposition and Drug Inter actions. Drug Metab. Dispos. 2008 Jan 23; Epub.

Nahta, R., Esteva, F. J. Herceptin: mechanisms of action and resistance. Cancer Letters 2006a ; 232: 123-138. Nahta, R., Yu, D., Hung, M.-C, Hortobagyi, G.N. and Esteva, F.J., Mechanisms of Disease: understanding resistance to HER2 -targeted therapy in human breast cancer, Nature Clinical Practice 2006b; 3(5), 269-280.

Nahta, R., Yuan, L. X., Du, Y., Esteva, F. J. Lapatinib induces apoptosis in trastuzumab- resistant breast cancer cells: effects on insulin-like growth factor I signaling. MoI Cancer Ther. 2007;6(2):667-674.

Rhodes, A., Borthwick, D., Sykes, R., Al-Sam, S., Paradiso, A. The use of cell line standards to reduce HER-2/neu assay variation in multiple European cancer centers and the potential of automated image analysis to provide for more accurate cut points for predicting clinical response to trastuzumab. Am J Clin Pathol. 2004; 122(1): 51-60.

Ross, J. S., Fletcher, J. A., Bloom, K. J., Linette, G. P., Stec, J., Symmans, W. F., Pusztai, L., Hortobagyi, G. N. Targeted therapy in breast cancer. MoI. Cell. Proteomics 2004; 3(4): 379-398.

Szollosi, J., Balazs, M., Feuerstein, B. G., Benz, C. C, Waldman, F. M. ERBB-2 (HER2/neu) gene copy number, pl85HER-2 overexpression, and intratumor heterogeneity in human breast cancer. Cancer Res. 1995; 55: 5400-5407.

Tanner, M., Kapanen, A. I., Junttila, T., Raheem, O., Grenman, S., EIo, J., Elenius, K., Isola, J. Characterization of a novel cell line established from a patient with Herceptinresistant breast cancer. MoI. Cancer Ther. 2004; 3(12): 1585-1592

Claims

1. A trifunctional, bispecific antibody having the following properties: (a) binding to a T-cell specific cell surface protein,

(b) binding to the tumor-associated antigen HER-2 on a tumor cell, and

(c) binding to Fc-gamma-receptor type I and/or type III positive cells, for use in the treatment of a HER-2 over-expressing tumor, wherein the tumor is or becomes resistant against one or more monospecific anti- HER-2 antibody and/or one or more tyrosine kinase inhibitor.

2. The use of claim 1, wherein the said treatment comprises administration of one or more monospecific anti-HER-2 antibody and/or one or more tyrosine kinase inhibitor.

3. The use of claim 1 or 2, wherein the one or more monospecific anti-HER-2 antibody is trastuzumab (Herceptin^®) and/or pertuzumab (Omnitarg^®, 2C4).

4. The use of any of claims 1 to 3, wherein the one or more tyrosine kinase inhibitor (TKI) is lapatinib (Tykerb), gefitinib (Iressa^®, ZDl 839), imatinib (Gleevec^®, STI-

571), erlotinib (Tarceva^®), lanafamib (Sarasar^®), sorafinib and/or sunitimib.

5. The use of any of the preceding claims, wherein said HER-2 over-expressing tumor expresses HER-2 at a level of about 50,000, preferably 75,000, more preferably about 100,000 to about 10,000,000 receptors/tumor cell; preferably of greater than about 150,000 to about 10,000,000 receptors/tumor cell.

6. The use of claim 5, wherein the HER-2 over-expressing tumor is classified by a value in the HercepTest of 2+ or 3+.

7. The use of any of the preceding claims, wherein the HER-2 over-expressing tumor is a FISH positive (FISH+) tumor.

8. The use of any of the preceding claims, wherein the tumor is classified as (i) 2+ and FISH+ or (ii) 3+ and FISH+.

9. The use of any of the preceding claims, wherein said HER-2 over-expressing tumor is selected from the group consisting of breast tumor, ovarian tumor, prostate tumor, colon tumor, pancreas tumor, stomach tumor, esophagus tumor, endometrium tumor, skin tumor, oropharynx tumor, larynx tumor, cervix tumor, bladder tumor, preferably carcinoma, more preferably adenocarcinoma or squamous cell carcinoma.

10. The use of any of the preceding claims, wherein the T-cell specific cell surface protein is CD3.

11. The use of claim 10, wherein said trifunctional bispecific antibody is an anti-HER-2 x anti-CD3 antibody binding to Fc-gamma-receptors type I and/or type III.

12. The use of claim 11, wherein the Fc portion comprises the isotype combination rat- IgG2b/mouse-IgG2a.

13. The use of any of the preceding claims, wherein said treatment further comprises administration of one or more monospecific antibody against an antigen other than HER-2, preferably one or more anti-EGFR antibodies, more preferably the monospecific anti-EGFR antibody cetuximab.

14. A pharmaceutical composition comprising (i) one or more trifunctional, bispecific antibody; and (ii) one or more anti-HER-2 antibody; optionally further comprising (iii) one or more monospecific antibody against an antigen other than HER-2, preferably one or more anti-EGFR antibodies, and/or (iv) one or more tyrosine kinase inhibitor; wherein component (i) to (iv) are as further defined in any of claims

1-13.

15. A kit comprising (i) one or more trifunctional, bispecific antibody, and (ii) one or more anti-HER-2 antibody; optionally further comprising (iii) one or more monospecific antibody against an antigen other than HER-2, preferably one or more anti-EGFR antibodies; and/or (iv) one or more tyrosine kinase inhibitor; wherein component (i) to (iv) are as further defined in any of claims 1-13.

16. A bispecific antibody or a fragment thereof, preferably wherein the antibody fragment is a single chain antibody (scFv), having the following properties:

(a) binding to the tumor-associated antigen HER-2 on a tumor cell, and

(b) binding to immunocompetent cells, for use in the treatment of a HER-2 over-expressing tumor, wherein the tumor is or becomes resistant against one or more tyrosine kinase inhibitor.