Combination of anti-EP-CAM antibody with a chemotherapeutic agent

[0001] This present invention relates to the combination of antibodies that specifically bind to the EP-CAM antigen with chemotherapeutic agents that affect cell growth by blocking progression of the cell cycle in G₂/M and their use in therapy of cancers which express the antigen.

[0002] The conventional therapeutic approaches to cancer include surgery, radiotherapy and chemotherapy in various combinations; however, response rates have not improved significantly in the last 20 years. The major limitation of chemotherapy and radiotherapy is the non-selective targeting of both normal and tumour cells that results in toxic side effects. In the search for less toxic and more specific treatment alternatives, various types of immunotherapy have been investigated. Among these modalities, strategies based on monoclonal antibodies have been applied to a broad spectrum of malignancies (Riethmüller et al. Curr Opin Immun 1992, 4, 647-655 and Riethmüller et al. Curr Opin Immunol 1993, 5, 732-739). The utility of monoclonal antibodies is based upon their clonal antigen specificity, i.e. molecular recognition of specific epitopes which may comprise an antigen and to bind to these antigens with high affinity. Monoclonal antibodies can bind to antigens expressed uniquely or preferentially on the surface of malignant cells, and hence can be used to specifically target and destroy tumour cells. Antibodies may be constructed as delivery vehicles for drugs or DNA, or as conjugates with radionuclides. Binding of naked antibody to target cells may also activate innate antitumour immune functions such as antibody-dependent cell-mediated cytotoxicity (ADCC) and complement-mediated cytotoxicity (CMC), either of which may result in lysis or phagocytosis of the targeted cell. Both ADCC and CMC are antibody-dose-related immune functions and it is therefore desirable to get as much antibody bound to target cells as possible. One way of achieving this objective is to increase the level of expression of the relevant antigen which may effectively increase antibody functions such as, for example, ADCC of the target cells by virtue of getting more antibody bound to the cells (Fogler et al. Cancer Research 48: 6303-6308 (1988)).

[0003] One antigen of importance in cancer therapy is the Ep-CAM antigen (human pan-carcinoma antigen). This antigen is a transmembrane glycoprotein which has been reported to function as a cell adhesion molecule (Litvinow et al. J. Cell Biology 125: 437446, 1994) and is also known as the 17-1A antigen, 40 kD antigen, EGP40, GA733-2, KSA and ESA but may be known by other names or descriptions in the literature as well. It is expressed on the baso-lateral surface of a majority of simple cuboidal or columnar, pseudo stratified columnar and transitional epithelia and at generally higher levels in tumour cells. Epithelial cells are known to be the most important cell type in the development of human malignancies. Thus more than 90% of all malignant tumours are carcinomas, and therefore of epithelial origin (Acta Anatomica; 156 (3); 217-226 (1996)). Although the Ep-CAM antigen is expressed on most tumour cells of epithelial origin there are examples of cells of epithelal origin that do not express Ep-CAM such as adult epithelial tissues, epidermal adult keratinocytes, gastric parietal cells, thymic cortical epithelium, myoepithelial cells and hepatocytes. The phenotype of a malignant cell plays an important role in the efficacy of monoclonal antibodies. While tumour specific antigens have proven to be elusive, differences in expression of the antigens between normal cells and tumour cells have provided a means to target antibodies to tumours. It would be clinically advantageous to improve on these differences by enhancement of antigen homogeneity and density of expression on tumour cells.

[0004] Interferons are well-known to alter cell phenotypes by increasing expression of tumour antigens as well as many normal antigens, e.g. Class I HLA. For example, human recombinant interferon-α and interferon-γ can increase the expression of human tumour antigens TAG-72 and CEA (Greiner et al. Cancer Res 44: 3208-3214 (1984)). Interferon exposure induced a more homogeneous CEA-positive tumour cell population which shed more CEA from the cells surface, which was confirmed by in vivo studies with human carcinoma xenografts in athymic mice. Treatment with interferon-γ enhanced TAG-72 and CEA expression on ovarian or gastrointestinal tumour cells in patients' malignant ascites (Greiner et al. J Clin Oncol 10: 735-746 (1992)). The effects of interferons on cells are myriad and range from direct cytotoxicity to immunopotentiation to antiproliferative activity. None of the effects of interferons on antigen expression have been directly ascribed to interference with cell cycle progression.

[0005] Briefly, cell cycle progression refers to the sequence of events between one mitotic division and another in a cell. A quiescent resting phase (G₀) is followed by a growth phase (G₁), then by a DNA synthesis phase (S). A second growth phase of cell enlargement (G₂) and DNA replication (M phase) is followed by division of the cell into two progeny cells. Any interference with the cell machinery may inhibit all cycle progression at any phase of the cell cycle. For example, specific chemotherapeutic agents may block progression in either G₂ or M or in both G₂ and M (G₂/M). In other words exposure to certain drugs e.g. chemotherapeutic agents will for example, arrest individual cells in G₂ and/or M until eventually most, or all of the cells in a population become arrested in G₂ and/or M (G₂/M). In HeLa cells, for example, the G₁, S, G₂ and M phase take 8.2, 6.2, 4.6 and 0.6 hours, respectively. The period between mitoses is called interphase. Cells may have different doubling times, depending on their developmental stage or tissue type. The variation in doubling times is usually a function of the time spent in G₁ (A Dictionary of Genetics, 5th edition, R C King and W D Stansfield, Oxford University Press, 1997).

[0006] While a few proteins have been identified as produced solely at certain phases of the cell cycle, and therefore can serve as markers of cell cycle status, most others are produced across the cell cycle but at higher or lower levels at certain points. Variation of antigen density across the cell cycle is typical for the sarcoma antigens p102 and p200 (Song S, Anticancer Research 16(3A): 1171-5 (1996)), the leukaemia/lymphoma-associated antigen JD118 (Czuczman et al. Cancer Immunology, Immunotherapy 36(6): 387-96 (1993)), and the gastric tumour antigen PC1 (Wei et al., J of Oncology 9(3): 179-82 (1987)). A few tumour antigens have been reported to be cell-cycle independent, e.g. liver metastases 3H4 (Wulf et al., J. Cancer Research and Clinical Oncology 122(8): 476-82 (1996)) and small cell lung cancer antigens (Fargion et al., Cancer Research 46: 2633-2638 (1986)).

[0007] Surprisingly, it has been found that pre-treatment with a drug, for example a chemotherapeutic agent known to block cell cycle progression at S and/or G₂/M results in a significant increase in the density of the Ep-CAM antigen population and thus in greater targeting of anti-Ep-CAM antibodies to Ep-CAM expressing tumours. In lytic antibodies this translates into an increased susceptibility to antibody-dependent cytolysis. This perturbation of tumour cell phenotype has a significant impact on the biological effectiveness of therapeutic antibodies, and provides synergistic benefit to standard chemotherapeutic regimens. Furthermore, this increase in Ep-CAM antigen expression appears to be tumour specific. In other words, pre-treatment with chemotherapeutic agents that block the cell cycle at S and/or G₂/M does not seem to affect Ep-CAM antigen expression in non-tumour cells.

[0008] Accordingly, the present invention provides a combination of an Ep-CAM antibody and a chemotherapeutic agent that is capable of arresting Ep-CAM antigen expressing cells in S or G₂/M, preferably in G₂/M.

[0009] Examples of anti-Ep-CAM antibodies are ING1 (Colcher et al., Proc. Natl. Acad. Sci. USA, 78 (5), 3199 to 3203 (1981) and Laio et al, Human Antibody Hybridomas 1(2), 66-76 (1990)); 17-1A e.g. Panorex (Herlyn et al, Proc. Natl. Acad. Sci. USA 76: 1438-1452 (1979) and Herlyn et al, Hybridoma 1985; 5 (suppl. 1) S3 to S10); and NR-LU-10 (Okabe et al, Cancer Research, 44, 5273 to 5278 (1984)).

[0010] Panorex (Adjuqual®) is a 17.1A mouse monoclonal antibody. It is marketed by Glaxo Wellcome in Germany for the post-operative adjuvant therapy of colorectal cancer.

[0011] An example of an anti-Ep-CAM antibody is one with the variable light chain cDNA sequence as set out in FIG. 15 and the heavy chain cDNA sequence as set out in FIG. 16. (known as humanised 323/A3/IgG₁). Two further preferred examples of anti-Ep-CAM antibodies are those with the variable light chain cDNA sequence as set out in FIG. 15 and heavy chain cDNA sequences as set out in FIG. 17 or 18 respectively (known as humanised 323/A3 IgG₄ and IgG₂cys respectively).

[0012] A preferred example of an anti-Ep-CAM antibody is 17.1A, most preferably Panorex.

[0013] Specific anti-Ep-CAM antibodies can be given on their own or in combination with other anti-Ep-CAM antibodies. Examples of such anti-Ep-CAM antibody combinations are an anti-Ep-CAM antibody with the variable light chain cDNA sequence as set out in FIG. 15 and the heavy chain cDNA sequence as set out in FIG. 16 in combination with ING1. Thus throughout the specification reference to an anti-Ep-CAM antibody includes antibody combinations of various anti-Ep-CAM antibodies, preferably non-competing anti-Ep-CAM antibodies targeting different epitopes on the same Ep-CAM antigen.

[0014] Examples of chemotherapeutic agents which are capable of arresting Ep-CAM antigen expressing cells in G₂/M are vinorelbine, cisplatin, mytomycin, paclitaxel, carboplatin, oxaliplatin and CPT-II (camptothecin).

[0015] Vinorelbine tartrate is a semisynthetic vinca alkaloid with the chemical name 3′,4′-didehydro-4′-deoxy-C′-norvincaleukoblastine [R—(R*,R*)-2,3-dihydroxybutanedioate (1:2)(salt)]. Vinorelbine tartrate is used in combination with other chemotherapy agents such as cisplatin or as a single agent in the treatment of various solid tumours particularly non-small cell lung, advanced breast, and hormone refractory prostate cancers. The brand name Navelbine® is used in North America and Europe. Navelbine® is administered intravenously as a single-agent or in combination therapy typically at doses of 20-30 mg/m² on a weekly basis. An oral formulation of vinorelbine is in clinical development.

[0016] Cisplatin has the chemical name cis-diamminedichloroplatinum. Cisplatin is used in the treatment of metastatic testicular tumours as a combination therapy, as single and combination therapy in metastatic ovarian tumours, as well as a single agent in advanced bladder cancer. Cisplatin is manufactured by Bristol-Myers Squibb under the brand names of Platinol® and Platinol-AQ®. Cisplatin is also used in the following types of cancer, typically in combination therapy: non-small cell and small cell lung cancers, head and neck, endometrial, cervical, and non-Hodgkin's lymphoma. Cisplatin is typically administered intravenously in doses ranging from 15-150 mg/m² once every 3 to 4 weeks, or daily for 5 days repeated every 3 or 4 weeks. However, higher and more frequent doses are occasionally administered and the route of administration could be different than intravenous, such as intra-arterial or intraperitoneal.

[0017] Carboplatin has the chemical name platinum, diammine [1,1-cyclobutane-dicarboxylato(2)-0,0′]-(SP-4-2). Carboplatin is usually administered in combination with other cytotoxics such as paclitaxel and etoposide. It is used in the treatment of advanced ovarian cancer, non-small cell lung cancer as well as in many of the same types of cancer as cisplatin is used. The brand name of carboplatin manufactured by Bristol-Myers Squibb is Paraplatin®. Carboplatin is typically administered intravenously at 300-400 mg/m², or to a target area under the drug concentration versus time curve (AUC) of 4-6 mg/ml-min using the patient's estimated glomerular filtration rate (GFR). Higher doses up to around 1600 mg/m² divided over several, usually five, days may also be administered.

[0018] Paclitaxel has the chemical name 5β, 20 epoxy-1,2α,4,7β,10β,13α-hexahydroxytax-11-en-9-one 4,10-diacetate 2-benzoate 13-ester with (2R,3S)—N-benzoyl-3-phenylisoserine. Paclitaxel is manufactured by Bristol-Myers Squibb as Taxol®. It is used to treat a variety of carcinomas including ovarian, breast, non-small cell lung, head and neck. Typical doses include 135-175 mg/m² as either a 3 or 24 hour intravenous infusion given every 3 or 4 weeks. Higher doses up to around 300 mg/m² have also been administered.

[0019] Besides the active ingredient, the drug products provided by manufacturers typically contain a diluent such as sterile water, dextrose 5% in water or 0.9% sodium chloride in water with additional excipients such as Cremophor vehicle added to make for example, paclitaxel soluble.

[0020] More detailed information on treatment regimes, dosages and compositions etc can be obtained from standard reference books such as: Martindale, The Extra Pharmacopoeia, 31st edition, edited by J E F Reynolds, London, Royal Pharmaceutical Society, 1996 and the Physicians Desk reference, 49th Edition, 1995, Medical Economics Data Production Company, Montvale.

[0021] Other chemotherapeutic agents that may cause cells to accumulate in G₂/M include anthracyclines e.g. doxorubicin and aclarubicin; carmustine (BCNU), camptothecin, 9-nitro-camptothecin, cyclophosphamide and its derivatives, docetaxel, etoposide, Razoxane (ICRF-187), alkyllyso-phospholipids e.g. ilmofosine; methotrexate, MST-16, taxanes, vinblastine, vincristine and teniposide (VM-26) (again see Martindale, The Extra Pharmacopoeia, 31st edition, edited by J E F Reynolds, London, Royal Pharmaceutical Society, 1996) and flavonoids e.g. apigenin and genistein (see The Merck Index, 12th edition, Merck Research Laboratories, Merck and Co Inc, 1996). In addition, adozelesin (a class of pyrazole compounds) (Cancer Research 1992, Oct. 15; 52 (2): 5687 to 5692)), Bistratene A (Mutation Research 1996, Mar. 1; 367 (3): 169 to 175), cycloxazoline (Cancer Chemotherapy & Pharmacology 1994; 33(5): 399 to 409), imidazoarcridinone, melephan (Experimental Cell Biology 1986; 54 (3): 138 to 148 and International Journal of Cancer 1995, Jul. 17; 62 (2): 170 to 175), merbarone (Environmental & Molecular Mutagenesis 1997; 29 (1): 16 to 27 and Cancer Research 1995, Apr. 1; 55 (7): 1509 to 1516) and oracin (FEBS Letters 1997, Jan. 2; 400 (1): 127 to 130) are also believed to cause cells to accumulate in G₂/M generally all topo II inhibitors, e.g. to potecan (abpi, 1998-1999), all vinca derivatives and all DNA damaging agents including radiation are also believed to arrest cells in G₂/M.

[0022] Moreover, 5FU has been reported to arrest cells in G₂/M (Oncology Research 1994; 6(7): 303-309) and it is therefore believed that 5FU and compounds similar to 5FU such as UFT, methotrexate, capecitabine and Gemcitabine will increase Ep-Cam expression in some tissues. Similarly, tomudex (Raloxifen) which is known to arrest cells in the S phase is believed to increase Ep-Cam expression.

[0023] The term “chemotherapeutic agent” throughout the specification is therefore not limited to cytotoxic therapy, but also encompasses cytostatic therapy and any other drugs capable of stopping cells in G₂/M. It should be further noted that radiotherapy is capable of arresting cells in G₂/M and that throughout the specification the term chemotherapeutic can therefore be substituted with “radiotherapy”.

[0024] Throughout the specification reference to a chemotherapeutic agent includes combinations of one or more specific chemotherapeutic agents which arrest Ep-CAM expressing tumour cells in G₂/M. Examples of typical combinations are vinorelbine with cisplatin and paclitaxel with carboplatin; oxaliplatin with 5FU; cyclophosphamide with methotrexate and 5FU; cyclophosphamide with doxorubicin and 5FU.

[0025] While it is possible for the chemotherapeutic agent to be administered alone it is preferable to present it as a pharmaceutical composition comprising an active ingredient, as defined above, together with an acceptable carrier therefor. Each carrier must be “acceptable” in the sense of being compatible with the other ingredients of the composition and not injurious to the recipient.

[0026] Preferred combinations of an Ep-Cam antibody and a chemotherapeutic agent(s) that are capable of arresting Ep-CAM antigen expressing cells in S or G₂/M are: Panorex in combination with any of the following chemotherapeutic agents: UFT, Capecitabine, CPT-II, Oxaliplatin, 5FU, 5FU continuous infusion, Paclitaxel, Docetaxel, Cyclophosphamide, Methotrexate, Doxorubicin, Navelbine (iv and oral), Epirubicin, Mitoxantrone, Raloxifen, Cisplatin, Mitomycin, Carboplatinum, Gemcitabine, Etoposide and Topotecan.

[0027] Particularly preferred combinations are Panorex with CPT-II, 5FU (continuous infusion), Oxaliplatin, Capecitibine, UFT and Tomudex (Raloxifen).

[0028] These Panorex combinations are useful in the treatment of cancer, particualrly in the treatment of colorectal cancer, breast cancer, gastric cancer, prostate cancer and non-small-cell lung cancer.

[0029] Specifically, the following combinations are particualrly preferred for colorectal cancer: Panorex in combination with: UFT (optionally with Leucovorin); Capecitabine; Oxaliplatin (optionally with 5FU); CPT-II, 5FU (optionally with Eniluracil or Levamisole or Leucovorin); 5FU protacted continuous infusion; and Mitomycin.

[0030] Preferred combinations for the treatment of breast cancer are: Panorex in combination with Paclitaxel; Docetaxel; Cyclophosphamide (optionally with 5FU and either Methotrexate or Doxorubicin); Navelbine (iv and/or oral); Doxorubicine; Epirubicin; Mitoxantrone; and Raloxifin.

[0031] Preferred combinations for the treatment of gastric cancer are: Panorex in combination with Cisplatin; 5FU; Mitomycin; and Carboplatinum.

[0032] A preferred combination for the treatment of prostatic cancer is: Panorex in combination with Mitoxantrone.

[0033] Preferred combinations for the treatment of non-small-cell lung cancer are: Panorex in combination with: Navelbine; Cisplatin; Carboplatin; Paclitaxel; Docetaxel; Gemcitabine; Topotecan; and Etoposide.

[0034] Information regarding dosing of Panorex and the above chemotherapeutic agents given in combination with Panorex can be found in standard reference books such as ABPI, Compendium of Data Sheets and Summaries of Product Characteristics, Datapharm Publications Limited, 1998-1999.

[0035] The compositions include those suitable for oral, rectal, nasal, topical (including buccal and sublingual), vaginal, parenteral (including subcutaneous, intramuscular, intravenous and intradermal) or transdermal administration. The compositions may conveniently be presented in unit dosage form and may be prepared by any methods well known in the art of pharmacy. Such methods include the step of bringing into association the active ingredient with the carrier which constitutes one or more accessory ingredients. In general, the compositions are prepared by uniformly and intimately bringing into association the active ingredient with liquid carriers or finely divided solid carriers or both, and then if necessary shaping the product.

[0036] Compositions of the chemotherapeutic agent suitable for oral administration may be presented as discrete units such as capsules, cachets or tablets each containing a predetermined amount of the active ingredient; as a powder or granules; as a solution or suspension in an aqueous or non-aqueous liquid; or as an oil-in-water-liquid emulsion or a water-in-oil liquid emulsion. The active ingredient may also be presented as a bolus, electuary or paste.

[0037] A tablet may be made by compression or moulding, optionally with one or more accessory ingredients. Compressed tablets may be prepared by compressing in a suitable machine the active ingredient in a free-flowing form such as a powder or granules, optionally mixed with a binder (e.g. povidone, gelatin, hydroxypropylmethyl cellulose), lubricants, inert diluent, preservative, disintegrant (eg. sodium starch glycollate, cross-linked povidone, cross-linked sodium carboxymethyl cellullose) surface-active or dispersing agent. Moulded tablets may be made by moulding in a suitable machine a mixture of the powdered compound moistened with an inert liquid diluent. The tablets may optionally be coated or scored and may be formulated so as to provide slow or controlled release of the active ingredient therein using, for example, hydroxypropylmethyl cellulose in varying proportions to provide the desired release profile. Tablets may optionally be provided with an enteric coating to provide release in parts of the gut other than the stomach.

[0038] Compositions suitable for oral use as described above may also include buffering agents designed to neutralise stomach acidity. Such buffers may be chosen from a variety of organic or inorganic agents such as weak acids or bases admixed with their conjugated salts.

[0039] Composition suitable for topical administration in the mouth include lozenges comprising the active ingredient in a flavoured basis, usually sucrose and acacia or tragacanth; pastilles comprising the active ingredient in an inert basis such as gelatine and glycerin, or sucrose and acacia and mouthwashes comprising the active ingredient in a suitable carrier.

[0040] Compositions for rectal administration may be presented as a suppository with suitable base comprising for example cocoa butter or a salicylate.

[0041] Compositions suitable for vaginal administration may be presented as pessaries, tampons, creams, gels, pastes, foams or spray formulations containing in addition to the active ingredient such carriers as are known in the art to be appropriate.

[0042] Compositions suitable for parenteral administration include aqueous and non-aqueous isotonic sterile injection solutions which may contain anti-oxidants, buffers, bacteriostats and solutes which render the compositions isotonic with the blood of the intended recipient; and aqueous and non-aqueous sterile suspensions which may include suspending agents and thickening agents, such as liposomes or other microparticulate systems which are designed to target the compounds to blood components or one or more organs. The compositions may be presented in unit-dose or multi-dose sealed containers, for example, ampoules and vials, and may be stored in a freeze-dried(lyophilized) condition requiring only the addition of sterile liquid carrier, for example water for injections, immediately prior to use. Extemporaneous injection solutions and suspensions may be prepared from sterile powders, granules and tablets of the kind previously described.

[0043] Compositions suitable for transdermal administration may be presented as discrete patches adapted to remain in intimate contact with the epidermis of the recipient for a prolonged period of time. Such patches suitably contain the active ingredient as an optionally buffered, aqueous solution of, for example, 0.1 0.2M concentration with respect to said compound. As one particular possibility, the active ingredient may be delivered from the patch by iontophoresis as generally described in Pharmaceutical Research, 3 (6), 318 (1986).

[0044] It should be understood that in addition to the ingredients particularly mentioned above the compositions in question, for example, those suitable for oral administration may include such further agents as sweeteners, thickeners and flavouring agents.

[0045] The dosage range of the chemotherapeutic agent to be co-administered with the antibody may typically be between about 1 to 1000 mg/M² (based on patient body surface area) or about 2-30 mg/kg (based on patient body weight), depending on the chemotherapeutic agent(s) used. Thus, for example, vinorelbine (navelbine) would typically be administered at a dosage of about 20 to 30 mg/m², cisplatin at about 15 to 100 mg/m² carboplatin at about 300 to 600 mg/m² and paclitaxel at about 100 to 300 mg/m², preferably around 135 to 175 mg/m². Another way of expressing dosage is by their AUC value. For example carboplatin would typically be administered at a dose calculated as AUC=4 to 6 mg/ml-min. Generally, the doses of chemotherapeutic agents are lower when given in combination with another chemotherapeutic agent and/or antibody than if given on their own as the single chemotherapeutic agent. The doses of chemotherapeutic agents that will be co-administered with anti Ep-CAM antibody(ies) will likely be the standard doses for the type of carcinoma treated or lower doses. In general the highest tolerated doses of the chemotherapeutic agents are administered either alone or in combination.

[0046] The anti-Ep-CAM antibodies of the present invention preferably have the structure of a natural antibody or a fragment thereof. Antibodies typically comprise two heavy chains linked together by disulphide bonds and two light chains. Each light chain is linked to a respective heavy chain by disulphide bonds. Each heavy chain has at one end a variable domain followed by a number of constant domains. Each light chain has a variable domain at one end and a constant domain at its other end. The light chain variable domain is aligned with the variable domain of the heavy chain. The light chain constant domain is aligned with the first constant domain of the heavy chain. The constant domains in the light and heavy chains are not involved directly in binding the antibody to antigen.

[0047] The variable domains of each pair of light and heavy chains form the antigen binding site. The domains on the light and heavy chains have the same general structure and each domain comprises a framework of four regions, whose sequences are relatively conserved, connected by three complementarity determining regions (CDRs). The four framework regions largely adopt a beta-sheet conformation and the CDRs form loops connecting, and in some cases forming part of the beta-sheet structure. The CDRs are held in close proximity by the framework regions and with the CDRs from the other domain, contribute to the formation of the antigen binding site, which in the case of the present invention is the formation of an anti-Ep-CAM binding site. CDRs and framework regions of antibodies may be determined by reference to Kabat et al (“Sequences of proteins of immunological interest” U.S. Dept. of Health and Human Services, U.S. Government Printing Office, 1987).

[0048] The preparation of an antibody in which the CDRs are derived from a different species than the framework of the antibody's variable domains is disclosed in EP-A-0239400. The CDR's may be derived from a rodent or primate monoclonal antibody. The framework of the variable domains and the constant domains of such altered antibodies are usually derived from a human antibody. Such a humanised antibody should not elicit as great an immune response when administered to a human compared to the immune response mounted by a human against a wholly foreign antibody such as one derived from a rodent.

[0049] The antibody preferably has the structure of a natural antibody or a fragment thereof. Throughout the specification reference to antibody therefore comprises not only a complete antibody but also fragments such as a (Fab′) 2 fragment, a Fab fragment, a light chain dimer or a heavy chain dimer. The antibody may be an IgG such as IgG₁, IgG₂, IgG₃ or IgG₄; or IgM, IgA, IgE or IgD or a modified variant thereof, including those that may be conjugated to other molecules such as radionuclides, enzymes etc. Typically, the constant region is selected according to the functionality required. Normally an IgG1 will demonstrate lytic ability through binding to complement and will mediate ADCC (antibody dependent cell cytotoxicity). An IgG₄ antibody will be preferred if a non-cytotoxic antibody is required. Antibodies according to the present invention also include bispecific antibodies such as, for example, the 17-1A antibody disclosed in Mack et al, The Journal of Immunology, 1997, 158: 3965-3970. Antibodies of the present invention may be murine, chimaeric or humanised with the preferred antibody being humanised antibody.

[0050] There are four general steps to humanise a monoclonal antibody. These are:

• [0051] (1) determining the nucleotide and predicted amino acid sequence of the starting antibody light and heavy variable domains;
• [0052] (2) designing the humanised antibody, i.e. deciding which antibody framework region to use during the humanising process;
• [0053] (3) the actual humanising methodologies/techniques; and
• [0054] (4) the transfection and expression of the humanised antibody.

[0055] More specifically,

Step 1: Determining the Nucleotide and Predicted amino acid sequence of the Antibody Light and Heavy Chain Variable Domains

[0056] To humanise an antibody only the amino acid sequence of the antibody's heavy and light chain variable domains needs to be known. The sequence of the constant domains is irrelevant because these do not contribute to the reshaping strategy. The simplest method of determining an antibody variable domain amino acid sequence is from cloned cDNA encoding the heavy and light variable domain.

[0057] There are two general methods for cloning a given antibody's heavy and light chain variable domain cDNAs: (1) via a conventional cDNA library, or (2) via the polymerase chain reaction (PCR). Both of these methods are widely known. Given the nucleotide sequence of the cDNAs, it is a simple matter to translate this information into the predicted amino acid sequence of the antibody variable domains.

Step 2: Designing the Humanised Antibody

[0058] There are several factors to consider in deciding which human antibody sequence to use during the humanisation. The humanisation of light and heavy chains are considered independently of one another, but the reasoning is basically similar for each.

[0059] This selection process is based on the following rationale: a given antibody's antigen specificity and affinity is primarily determined by the amino acid sequence of the variable region CDRs. Variable domain framework residues have little or no direct contribution. The primary function of the framework regions is to hold the CDRs in their proper spatial orientation to recognise the antigen. Thus the substitution of rodent CDRs into a human variable domain framework is most likely to result in retention of their correct spatial orientation if the human variable domain framework is highly homologous to the rodent variable domain from which they originated. A human variable domain should preferably be chosen therefore that is highly homologous to the rodent variable domain(s).

[0060] A suitable human antibody variable domain sequence can be selected as follows:

• [0061] 1. Using a computer program, search all available protein (and DNA) databases for those human antibody variable domain sequences that are most homologous to the rodent antibody variable domains. The output of a suitable program is a list of sequences most homologous to the rodent antibody, the percent homology to each sequence, and an alignment of each sequence to the rodent sequence. This is done independently for both the heavy and light chain variable domain sequences. The above analyses are more easily accomplished if only human immunoglobulin sequences are included.
• [0062] 2. List the human antibody variable domain sequences and compare for homology. Primarily the comparison is performed on lengths of CDRs, except CDR 3 of the heavy chain which is quite variable. Human heavy chains and Kappa and Lambda light chains are divided into subgroups; Heavy chain 3 subgroups, Kappa chain 4 subgroups, Lambda chain 6 subgroups. The CDR sizes within each subgroup are similar but vary between subgroups. It is usually possible to match a rodent antibody CDR to one of the human subgroups as a first approximation of homology. Antibodies bearing CDRs of similar length are then compared for amino acid sequence homology, especially within the CDRs, but also in the surrounding framework regions. The human variable domain which is most homologous is chosen as the framework for humanisation.
  Step 3: The Actual Humanising Methodologies/Techniques

[0063] An antibody may be humanised by grafting the desired CDRs onto a human framework according to EP-A-0239400. (see also P. T. Jones et al, Nature 321: 522 (1986); L. Reichman et al, Nature 332: 323 (1988); Verhoeyen M. et al, Science 239: 1534 (1988) and J. Ellis et al, The Journal of Immunology, 155: 925-937 (1995)). A DNA sequence encoding the desired reshaped antibody can therefore be made beginning with the human DNA whose CDRs it is wished to reshape. The rodent variable domain amino acid sequence containing the desired CDRs is compared to that of the chosen human antibody variable domain sequence. The residues in the human variable domain are marked that need to be changed to the corresponding residue in the rodent to make the human variable region incorporate the rodent CDRs. There may also be residues that need substituting in, adding to or deleting from the human sequence.

[0064] Oligonucleotides are synthesised that can be used to mutagenise the human variable domain framework to contain the desired residues. Those oligonucleotides can be of any convenient size. One is normally only limited in length by the capabilities of the particular synthesiser one has available. The method of oligonucleotide-directed in vitro mutagenesis is well known.

[0065] Alternatively humanisation may be achieved using the recombinant polymerase chain reaction (PCR) methodology of WO92/07075. Using this methodology, a CDR may be spliced between the framework regions of a human antibody.

[0066] In general, the technique of WO92/07075 can be performed using a template comprising two human framework regions, AB and CD and between them, the CDR which is to be replaced by a donor CDR. Primers A and B are used to amplify the framework region AB, and primers C and D used to amplify the framework region CD. However, the primers B and C each also contain, at their 5′ ends, an additional sequence corresponding to all or at least part of the donor CDR sequence. Primers B and C overlap by a length sufficient to permit annealing of their 5′ ends to each other under conditions which allow a PCR to be performed. Thus, the amplified regions AB and CD may undergo gene splicing by overlap extension to produce the humanised product in a single reaction.

Step 4: The Transfection and Expression of the Reshaped Antibody

[0067] Following the mutagenesis reactions to reshape the antibody, the mutagenised DNAs can be linked to an appropriate DNA encoding a light or heavy chain constant region, cloned into an expression vector, and transfected into host cells, preferably mammalian cells. These steps can be carried out in routine fashion. A reshaped antibody may therefore be prepared by a process comprising

• [0068] (a) preparing a first replicable expression vector including a suitable promoter operably linked to a DNA sequence which encodes at least a variable domain of an Ig heavy or light chain, the variable domain comprising framework regions from a human antibody and the CDRs required for the humanised antibody of the invention.
• [0069] (b) preparing a second replicable expression vector including a suitable promoter operably linked to a DNA sequence which encodes at least the variable domain of a complementary Ig light or heavy chain respectively;
• [0070] (c) transforming a cell line with the first or both prepared vectors; and
• [0071] d) culturing said transformed cell line to produce said altered antibody.

[0072] Preferably the DNA sequence in step (a) encodes both the variable domain and the or each constant domain of the human antibody chain. The humanised antibody can be recovered and purified. The cell line which is transformed to produce the altered antibody may be Chinese Hamster Ovary (CHO) cell line or an immortalised mammalian cell line, which is advantageously of lymphoid origin, such as a myeloma, hybridoma, trioma or quadroma cell line. The cell line may also comprise a normal lymphoid cell, such as a B-cell, which has been immortalised by transformation with a virus, such as the Epstein-Barr virus. Most preferably, the immortalised cell line is a myeloma cell line or a derivative thereof. The expression system of choice is the glutamine synthetase expression system described in WO87/00462 (see also P. E. Stephens et al, Nucleic Acid Res. 17: 7110 (1989) and C. R. Bebbington et al, Bio/Technology 10: 169 (1992)).

[0073] Although the cell line used to produce the humanised antibody is preferably a mammalian cell line, any other suitable cell line, such as a bacterial cell line or a yeast cell line, may alternatively be used. For single antibody chains, it is envisaged that E. coli—derived bacterial strains could be used. The antibody obtained is checked for functionality. If functionality is lost, it is necessary to return to step (2) and alter the framework of the antibody.

[0074] Once expressed, the whole antibodies, their dimers, individual light and heavy chains, or other immunoglobulin forms of the present invention can be purified according to standard procedures of the art, including ammonium sulfate precipitation, affinity columns, column chromatography, gel electrophoresis and the like (see generally Scopes, R, Protein Purification, Springer-Verlag, N.Y. (1982)). Substantially pure immunoglobulins of at least about 90 to 95% homogeneity are preferred and 98 to 99% or more homogeneity most preferred, for pharmaceutical uses. Once purified, partially or to homogeneity as desired, an antibody may then be used therapeutically.

[0075] Antibodies are typically provided as a pharmaceutical composition comprising a pharmaceutically acceptable carrier or diluent and, as active ingredient, an antibody according to the invention. The antibody and pharmaceutical compositions thereof are particularly useful for parenteral administration i.e. subcutaneously, intramuscularly or intravenously.

[0076] The compositions for parenteral administration will commonly comprise a solution of the antibody or a cocktail thereof dissolved in an acceptable carrier, preferably an aqueous carrier. A variety of aqueous carriers can be used, eg. sterile water for injection, 0.9% sodium chloride in water, 5% dextrose in water and Lactated Ringers solution. These solutions are sterile and generally free of particulate matter. These compositions may be sterilised by conventional, well known sterilisation techniques. The compositions may contain pharmaceutically acceptable auxiliary substances as required to approximate physiological conditions such as pH adjusting and buffering agents, toxicity adjustment agents and the like, for example sodium acetate, sodium chloride, potassium chloride, calcium chloride, sodium lactate, etc. The concentration of antibody in these formulations can vary widely, for example from less than about 0.5%, usually at or at least about 1% to as much as 15 or 20% by weight and will be selected primarily based on fluid volumes, viscosities, etc. in accordance with particular mode of administration selected.

[0077] Thus, a typical pharmaceutical composition for intramuscular injection could be made to contain 1 ml sterile buffered water, and 50 mg of antibody. A typical composition for intravenous infusion could be made up to contain 250 ml of sterile Ringers solution and 150 mg of antibody. Actual methods for preparing parenterally administrable compositions will be known or apparent to those skilled in the art, particularly, those trained in the preparation of parenteral products and are described in more detail in, for example, Remmington's Pharmaceutical Science, 15th Ed., Mack Publishing Company, Easton, Pa. (1990).

[0078] The antibodies of this invention can be lyophilised for storage and reconstituted in a suitable carrier prior to use. This technique has been shown to be effective with conventional immunoglobulins. Any suitable lyophilisation and reconstitution techniques can be employed. It will be appreciated by those skilled in the art that lyophilization and reconstitution can lead to varying degrees of antibody activity loss (eg. with conventional immune globulins, IgM antibodies tend to have greater activity loss than IgG antibodies) and that use levels may have to be adjusted to compensate.

[0079] The dosage range of the antibody in accordance with the invention is about 0.5 to 1000 mg/m², preferably about 0.5 to 250 mg/m², more preferably, between 0.5 and 100 mg/m² and 0.5 and 50 mg/m² and most preferably between 5 and 25 mg/m² such as for example, 15 mg/m².

[0080] Similarly, expressed in mg per dose, the dosages of the antibody may be about 1 to 2000 mg per dose, preferably about 1 to 500 mg per dose, more preferably between 1 to 200 mg per dose and between 1 to 100 mg per dose and most preferably between 10 and 50 mg per dose such as, for example 30 mg per dose.

[0081] Single or multiple administrations of the compositions can be carried out with dose levels and pattern being selected by the treating physician. In any event, the pharmaceutical formulations should provide a quantity of the antibody(ies) sufficient to effectively treat the patient.

[0082] Typically, the chemotherapeutic agent and antibody will be presented as separate pharmaceutical compositions for co-administration, but they may also be formulated as a single pharmaceutical formulation. In this way both the antibody and the chemotherapeutic agent are presented to the patient within one and the same composition.

[0083] One or more distinct chemotherapeutic agents and one or more distinct anti-Ep-CAM antibodies may be co-administered in all aspects of the present invention. Thus reference to a chemotherapeutic agent comprises one or more distinct chemotherapeutic agent(s). If there is more than one chemotherapeutic agent, these may be administered either individually each on its own and/or together as a chemotherapeutic agent cocktail. Similarly, reference to antibody comprises one or more distinct anti-Ep-CAM antibody(ies). If there is more than one antibody, these may again be administered either individually each on its own and/or together as a cocktail. Additionally, the chemotherapeutic agent(s) are typically administered separately from the antibody(ies) but they may also be administered together as a chemotherapeutic agent(s)/antibody(ies) cocktail.

[0084] Co-administration of the chemotherapeutic agent/radiotherapy with the antibody comprehends administration substantially simultaneously of both the chemotherapeutic agent/radiotherapy and the antibody. Essentially, the rational behind co-administration is to allow sufficient exposure of Ep-CAM expressing tumour cells to a chemotherapeutic agent/radiotherapy known to block cell cycle progression at G₂/M to achieve the desired increase in Ep-CAM antigen density prior to exposure of the same tumour cells to an anti-Ep-CAM antibody thereby achieving greater targeting of anti-Ep-CAM antibodies to Ep-CAM expressing tumours. Co-administration therefore comprises any mode of administering a chemotherapeutic agent/radiotherapy in conjunction with an anti-Ep-CAM antibody that will achieve this result. Throughout the specification the term “combination of an anti-Ep-CAM antibody with a chemotherapeutic agent” refers to one wherein the chemotherapeutic agent/radiotherapy and the anti-Ep-CAM antibody have been co-administered.

[0085] Preferably the chemotherapeutic agent is administered simultaneously with the antibody or more preferably before the antibody. Thus the chemotherapeutic agent may be administered on the same day as the antibody, either together or within hours of each other but may also be administered up to about two months beforehand, typically, about one or two weeks beforehand and more typically less than a week beforehand, say one to three days beforehand.

[0086] Additionally, co-administration also includes administering more than one dose of antibody within several weeks after one or more doses of chemotherapeutic agent, in other words the chemotherapeutic agent need not be re-administered again with every subsequent administration of the antibody, but may be administered just once or intermittently during the course of antibody treatment. Co-administration also comprises administration of the chemotherapeutic agent up to 3 weeks after the antibody, preferably within a week and more preferably within a few days such as one to five days.

[0087] The antibody may be administered several times daily. Similarly the chemotherapeutic agent may be infused continuously over several hours or even days.

[0088] The present invention also provides a method of treating mammalian patients, preferably humans, afflicted with cancer which comprises co-administering a chemotherapeutic agent which is capable of arresting Ep-CAM antigen expressing cells in G₂/M in combination with an anti-Ep-CAM antibody. Preferably, the chemotherapeutic agent is given simultaneously and more preferably prior to administration of the antibody.

[0089] The cancers which may be treated particularly effectively with this combination therapy are primary or metastatic cancers of any histologic or histogenetic origin that express the Ep-CAM antigen. This includes, for example, prostate cancers, lung cancers, breast cancers, colon cancers, pancreatic cancers and ovarian cancers.

[0090] Dosing schedules for the treatment method of the present invention can be adjusted to account for the patient characteristics, disease state, characteristics of the chemotherapeutic agent and characteristics of the anti-Ep-CAM antibody. The goal of dosing schedules under this invention will be to administer anti-Ep-CAM antibody in a manner that will expose the Ep-CAM expressing tumour cells to the anti-Ep-CAM antibody at a time when antigen expression is likely to be increased due to exposure to chemotherapy which is known to block cell cycle progression at G₂/M. Additionally, as much as possible a dosing schedule convenient for the patient must be maintained.

[0091] Preferred dosing schedules for administration of the anti-Ep-CAM antibody and chemotherapy include: administering the anti-Ep-CAM antibody once every one or two weeks, preferably once every three or four weeks or a combination thereof for as long as necessary. The chemotherapeutic agent is given according to the established regimen for that agent or a regimen which will allow exposure of Ep-CAM expressing tumour cells to be arrested in G₂/M. Preferred dosing schedules vary with the chemotherapy agent and disease state but include, for example, once weekly, once every three or four weeks, or daily for several (e.g. 3-5) days repeated every three or four weeks for as long as necessary. Dosing of the anti-Ep-CAM antibody may take place on the same day or different days as indicated for the chemotherapeutic agent. Adjustment of the dosing schedule or strength of dose to prevent or decrease toxicity or side effects may take place with either the anti-Ep-CAM antibody or the chemotherapy agent.

[0092] For example, the preferred dosing schedule for co-administration of vinorelbine and cisplatin in combination with humanised 323/A3 (IgG₁) is administration of humanised 323/A3 (IgG₁) at a dose of 30 mg/m² once a week for as long as necessary but typically for a period of 3 to 4 weeks, followed by a 30 mg/m² dose every other week thereafter for as long as necessary. Vinorelbine is administered at a dose 25 mg/m² on day 1, 8, 15 and 22. Cisplatin is given only once at a dose of 100 mg/m² on day 1. Thereafter the vinorelbine/cisplatin regime is repeated every 28 days for as long as necessary. Preferably, vinorelbine, cisplatin and humanised 323/A3 (IgG₁) are administered at the same time on day one over a period of about 2 to 3 hours.

[0093] Another example of a preferred dosing schedule is the administration of paclitaxel/carboplatin in combination with humanised 3231A3 (IgG₁), wherein 323/A3 (IgG₁) is administered as for the vinorelbine/cisplatin example above and paclitaxel and carboplatin are given at a dose of 225 mg/m² and AUC=6.0 respectively, on day 1, with a repeat dosage every 28 days thereafter for as long as necessary. Again, paclitaxel, carboplatin and humanised 323/A3 (IgG₁) are preferably administered together on day 1 over a period of about 2 to 3 hours.

[0094] Other preferred dosage schedules which comprise the combination of 323/A3 (IgG₁) with any of navelbine, cisplatin or taxol on their own would comprise similar dosages and administration schedules, using just one anticancer agent instead of two.

[0095] When the preferred anti-Ep-CAM antibody is Panorex, the dosage of antiobdy is between 10 to 500 mg per dose, preferably 100 mg per dose.

[0096] A further aspect of the present invention is a method of increasing antibody binding of anti-Ep-CAM antibodies to Ep-CAM expressing cells by co-administering to a patient a chemotherapeutic agent capable of arresting cells in G₂/M together with said anti-Ep-CAM antibody.

[0097] By co-administering a chemotherapeutic agent according to the present invention together with an Ep-CAM antibody, it is possible to increase antibody binding by about 2 to 10 fold, preferably by more than 4 fold, more preferably by more than 6 fold and most preferably by more than 8 fold.

[0098] FIG. 1.

[0099] Ep-CAM is expressed across the cell cycle, but at higher density and greater homogeneity on cells in S (dotted line) and in G₂/M (dashed line) phases than in G₀/G₁ cells (solid line). This pattern of expression has been documented in a number of other human colon, prostrate, and lung tumour cell lines.

[0100] FIG. 2.

[0101] Cell cycle arrest is a prominent feature of adenocarcinoma cells exposed in vitro to Navelbine (NVB; 30 nM) plus Cisplatin (CDDP; 5 μM), or Taxol (TAX; 80 nM) plus Carboplatin (CPBDA; 100 μM), compared to media alone, 5-Fluorouracil (5FU), interferon-alpha (IFN-alpha; 100 U/ml), or interferon-gamma (IFN-gamma; 100 U/ml). The area of each bar is divided to indicate the percentage of cells in G₀/G₁ and in S+G₂/M phases; the height of each bar indicates the average number of Ep-CAM molecules per cell within the population. Cells in S phase and in G₂/M phase express higher levels of Ep-CAM (FIG. 1), and the agents which blocked cell cycle progression had overall increased Ep-CAM expression

[0102] FIG. 3.

[0103] The expression of Ep-CAM antigen was quantified on a variety of adenocarcinoma cell lines as well as primary cultures of normal human cells. Cultured cells were exposed sequentially to media, or to 30 nM Navelbine followed by 5 μM Cisplatin (NVB+CDDP), or to 80 nM Taxol followed by 100 μM Carboplatin (TAX+CPBDA). The 4 adenocarcinoma cells expressed higher antigen levels subsequent to exposure to cycle-specific drug combinations, whereas the 4 normal cells did not show any increase in antigen expression, which remained undetectable in 2 of the normal cell populations.

[0104] FIG. 3a.

[0105] The binding of Panorex, a related murine monoclonal antibody with specificity for the Ep-CAM antigen, was evaluated after a 15 minute incubation with HT29 adenocarcinoma cells which had been cultured with Navelbine plus Cisplatin or with Taxol as previously described. A significant increase (34%) in antibody binding was seen on the cells treated with Navelbine plus Cisplatin; 82% of these cells were arrested in S or G₂/M cycle phase compared to 21% of the control cells. (A smaller increase (8%) in antibody binding was seen for cells treated with Taxol, but in this expeiment only 57% of the cells were cycle-arrested) as is shown in FIG. 3a.

[0106] FIG. 4.

[0107] The ability of human peripheral blood ADCC effector cells to lyse tumour target cells incubated with humanized 323/A3 (IgG₁) (a humanized monoclonal antibody having specificity for the Ep-CAM antigen and capable of interacting with Fc receptors on human effector cells) in vitro was improved when the target cells had been pre-treated with NAVELBINE (30 nM) plus Cisplatin (5 μM).

[0108] FIG. 5.

[0109] Treatment of human tumour xenograft-bearing mice with a cell-cycle-specific cytotoxic agent promoted improved localization of antibody specific for Ep-CAM to the tumours.

[0110] FIG. 6.

[0111] Humanised 323/A3 (IgG₁) Kappa Light Chain Amino Acid Sequence

[0112] FIG. 7.

[0113] Humanised 323/A3 (IgG₁) Heavy Chain Amino Acid Sequence

[0114] FIG. 8.

[0115] Vector Map of pEE6

[0116] FIG. 9.

[0117] Vector Map of pEE12

[0118] FIG. 10.

[0119] Vector Map of pEE18

[0120] FIG. 11

[0121] Humanised 323/A3 (IgG_4cys) Kappa Light Chain Amino Acid Sequence

[0122] FIG. 12

[0123] Humanised 323/A3 (IgG_4cys) variant Heavy Chain Amino Acid Sequence

[0124] FIG. 13

[0125] Humanised 323/A3 (IgG_2cys) Kappa Light Chain Amino Acid Sequence

[0126] FIG. 14

[0127] Humanised 323/A3 (IgG_2cys) Heavy Chain Amino Acid Sequence

[0128] FIG. 15

[0129] Humanised 323/A3 (IgG₁) light chain cDNA Sequence

[0130] FIG. 16

[0131] Humanised 323/A3 (IgG₁) Heavy chain cDNA Sequence

[0132] FIG. 17

[0133] Humanised 323/A3 (IgG₄) heavy chain cDNA Sequence

[0134] FIG. 18

[0135] Humanised 323/A3 (IgG_2cys) heavy chain cDNA Sequence

[0136] The following examples illustrate the invention.

[0137] Populations of PC-3 prostatic adenocarcinoma cells were evaluated for distribution in G₀/G₁, S, and G₂/M phases of the cell cycle as well as Ep-CAM expression. Cells were gently trypsinized and mechanically detached from the culture flasks and resuspended in calcium-and magnesium-free phosphate-buffered saline containing bovine serum albumin and NaN₃. Exactly 2×10⁵ cells were stained with FITC-323/A3 murine IgG antibody or FITC-murine IgG (control). Cells were fixed with cold paraformaldehyde, then permeabilized for DNA staining with Tween-20. Cellular DNA was stained with propidium iodide and RNase A. Listmode data were acquired on a FACScan flow cytometer (Becton Dickinson Immunocytometry Systems) equipped with a 488 nm laser using Cell Fit software. Cell cycle analysis was done using SOBR modelling (where possible, otherwise manual estimations were employed) on Cell Fit. EP-CAM antigen expression as detected by 323/A3 binding was evaluated separately using histogram analysis in Win List (Verity Software House).

[0138] FIG. 1 shows that Ep-CAM is expressed across the cell cycle, but at higher density and greater homogeneity on cells in S (dotted line) and in G₂/M (dashed line) phases than in G₀/G₁ cells (solid line). This pattern of expression has been documented in a number of other human colon, prostate, and lung tumor cell lines.

[0139] Adenocarcinoma cell lines were exposed to the various drugs or combinations of drugs as indicated in FIG. 2. Subconfluent cells were exposed to Navelbine or Taxol for up to 24 hours, then washed and exposed to Cisplatin or Carboplatin, respectively, overnight. Cells were exposed to 5FU for 24 hours, and for 2-5 days to the interferons. Cells were washed and cultured for another 2-5 days prior to analysis for antigen expression and cell cycle status as described in Example 1. Antigen expression was quantified by comparison of the binding of fluorescein-conjugated 323/A3 to cultured cells with binding to calibrated microbead standards.

[0140] Cell cycle analysis demonstrated that only 6.3% of the media control cells were in S and G₂/M phases combined, compared to 39.4% of NVB+CDDP and 82.6% of TAX+CPBDA cells, both combinations of which caused significant increases in Ep-CAM antigen expression (as demonstrated in FIG. 2). Antigen expression was not significantly increased in cells exposed to 5FU, IFN-α, or IFN-γ, which had only 7.9%, 12%, and 11.5%, respectively, of cells in S+G₂/M phase. Thus, only the drugs which caused accumulation of cells in S or G₂/M phases were able to cause a significant increase in Ep-CAM antigen expression.

[0141] The binding of Panorex, a related murine monoclonal antibody with specificity for the Ep-CAM antigen, was evaluated after a 15 minute incubation with HT29 adenocarcinoma cells which had been cultured with Navelbine plus Cisplatin or with Taxol as previously described. A significant increase (34%) in antibody binding was seen on the cells treated with Navelbine plus Cisplatin; 82% of these cells were arrested in S or G₂/M cycle phase compared to 21% of the control cells. (A smaller increase (8%) in antibody binding was seen for cells treated with Taxol, but in this expeiment only 57% of the cells were cycle-arrested) as is shown in FIG. 3a.

[0142] The expression of Ep-CAM antigen was quantified on a variety of adenocarcinoma cell lines as well as primary cultures of normal human cells. Cultured subconfluent cells were exposed sequentially to media, or to 30 nM Navelbine followed by 5 μM Cisplatin (NVB+CDDP), or to 80 nM Taxol followed by 100 μM Carboplatin (TAX+CPBDA). Cells were washed with media and cultured for another 2-5 days prior to analysis for antigen expression as described in Examples 1 and 2.

[0143] FIG. 3 clearly shows that the 4 adenocarcinoma cells expressed higher antigen levels subsequent to exposure to cycle-specific drug combinations, whereas the 4 normal cells did not show any increase in antigen expression, which remained undetectable in 2 of the normal cell populations.

[0144] Adenocarcinoma cells were exposed to drugs as described in Examples 1 and 2 above, and then harvested and seeded into 96-well plates for use as target cells in a ⁵¹Cr-release cytotoxicity assay. Target cells were cultured overnight with ⁵¹Cr, and then washed. Human peripheral blood mononuclear cells which had been allowed to adhere overnight were added at a 50:1 effector: target ratio, and the ADCC cultures were incubated for 6 hours. Supernatants were collected and counted for radioactivity, and the percentage specific release was calculated. (see FIG. 4).

[0145] FIG. 4 clearly shows that PC-3 prostatic adenocarcinoma cells are better targets for human ADCC activity after exposure to Navelbine/Cisplatin compared to controls which have not been exposed to these chemotherapeutic agents. This effect may be due directly to increased antigen expression and thereby increased antibody binding, decreased modulation of the Ep-CAM antigen, increased fragility of the target cells, or a combination of the above.

[0146] Human colon adenocarcinoma (HT-29) tumours were initiated by subcutaneous implantation into female CD-1 nude mice (Charles River). When the tumours reached 200-300 mg, animals were divided into groups of five. Navelbine was injected intravenously at a dosage of 28 mg/kg on days 1 and 5. A control group was dosed with 5-fluorouracil (5-FU) intraperitoneally at 20 mg/kg on days 1 and 5. On day 6, humanised 323/A3 IgG_4cys-TMT (a humanized monoclonal antibody chelator conjugate with specificity for the Ep-CAM antigen) was labelled with lutetium-177 and injected intravenously via the lateral tail vein. Each mouse received 4.1 μg protein/2.09 μCi lutetium-177/0.2 ml injection. Blood, spleen, liver, lung, kidney, femur and tumour were harvested on days 1, 3 and 5 post-antibody for direct gamma counting (see FIG. 5 for results).

[0147] FIG. 5 shows that pre-treatment with Navelbine increases antibody targeting to Ep-CAM positive tumours whilst pre-treatment with 5-FU does not.

1. Purpose/Summary

[0148] The cDNAs encoding the humanized 323/A3 antibody light and heavy chains (see FIGS. 15 and 16 respectively) were genetically engineered into a single Celltech glutamine synthetase (GS) expression plasmid, pEE18 (see FIG. 10), and used to transfect murine NSO cells.

2. Materials and Methods

2.1 Materials

[0149] NSO cells were obtained from Celltech Biologics plc, Slough, SL1 4EN, Berkshire, UK. The expression plasmids pEE6HCMV and pEE12 (see FIGS. 8 and 9) were obtained from Celltech Biologics plc, Slough.

[0150] 2.2 The pEE6hmcv plasmid (see FIG. 8) encoding full length humanised heavy chain DNA was digested with Bam HI and Bgl II to liberate the 3.2 kb fragment that contained the DNA encoding the heavy chain under the transcriptional control of the major immediate early promoter of the human cytomegalovirus. This fragment was cloned into the Bam HI site of pEE12 (FIG. 9) that contained the DNA encoding the humanised light chain. (See FIG. 6 for humanised 323/A3 (IgG₁) Kappa light chain amino acid sequence and FIG. 7 for the humanised 323/A3(IgG₁) Heavy chain amino acid sequence. See FIG. 10 for schematic representation of the pEE18 plasmid encoding 323/A3 (IgG₁) heavy and light chains.

2.2.2 Transfection and Selection of NSO Cells

2.2.2.1 Tissue Culture

[0151] All single cell culture activities were performed in isolated rooms that contained a single laminar flow hood and single incubator dedicated solely to the use of NSO cells in the production of stable cell lines secreting humanised 323/A3(IgG₁). No other NSO cells lines, human cell lines or virus transformed cell lines were used within this environment.

[0152] A vial of NSO cells was revived and grown in 1:1:1 medium composed of DMEM:RPMI-1640:Sigma PFHM (1:1:1) to a cell density between 0.5 and 1×10⁶ mL. For electroporation, the cells were harvested by centrifugation and washed once with PBS. pEE18 plasmid DNA encoding 323/A3 (IgG₁) was digested with Sal I, heat inactivated at 65° C. for 15 minutes, precipitated with ethanol and air-dried. The dried DNA pellet was resuspended in PBS to a concentration of 0.5 μg/mL and 100 μL aliquoted into a 2 mm electroporation cuvette (BTX). Washed NSO cells were resuspended at 1.2×10⁷/ml and 400 μL added to the cuvette to give a final density of 10⁶ mL in a final volume of 0.5 mL. Electroporation was at 300 V for 1 msec in a BTX 8209 GenePulser followed by incubation on ice for 5-10 minutes. The electroporation mixture was resuspended at 10⁵ cells/mL with 1:1:1 medium and distributed over 96-well plates at 50 μL/well. The following day, wells were fed with 150 μL GS medium (Gln-free IMDM, 1=X GS and nucleoside supplement, 5% DFBS) to begin the GS selection process such that all wells had a final concentration of 3% DFBS.

2.2.2.2 Specific Production Rate (SPR)

[0153] Selected cell lines grown in GS media (3% DFBS) were seeded at a density of 0.2×10⁶ cells/mL in T-25 flasks (Costar) that contained 5 mL of GS media (3% DFBS). Cells were incubated overnight at 37° C. for 24 hours after which an aliquot of each culture supernatant was removed. The supernatants were used in the human IgG ELISA assay to determine the concentration of secreted humanised 323/A3(IgG₁). The SPR value was derived by multiplying the concentration of 323/A3 (IgG₁) antibody in the supernatant times the volume (5.0) and is expressed as μg/10⁶ cells/24 hours.

2.2.2.3 Cryopreservation of Cells

[0154] Selected cell lines were routinely harvested when cell density was greater than 0.2×10⁶ cells/mL. An appropriate volume of cells was removed and subjected to centrifugation at 1,000×g for 5 minutes at 22° C. The cell pellet was gently resuspended to 1-4×10⁶ cells/mL with ice-cold freezing media consisting of 20% (v/v) FBS/10-% (v/v) DMSO/GS Media (sterile filtered). Each 1.0 mL of the cell suspension was aliquoted into a 1.8 ml cryopreservation vial (NUNC) and gradually frozen overnight in a Cryo 1° C. Freezing Container (Nalgene) that had been placed in a −70° C. freezer. The vials were then removed from the container and stored in the vapour phase of a liquid nitrogen freezer.

[0155] Twenty vials of each cell line, including a low humanised 323/A3(IgG₁) producer were frozen down as described above and stored initially in the vapour phase of an MVE Cryogenics XLC440 liquid nitrogen freezer. The cells were subsequently transferred and stored in the vapour phase of an MVE Cryogenics XLC500 liquid nitrogen freezer.

1. Purpose Summary

[0156] The cDNAs encoding the humanized antibody 323/A3(IgG_4cys) (a humanised 323/A3 antibody) antibody light and heavy chains (see FIGS. 15 and 17 were genetically engineered into a single Celltech glutamine synthetase (GS) expression plasmid, pEE18, and used to transfect murine NSO cells.

2. Materials and Methods

2.1 Materials (as for Example 6 above)

[0157] 2.2 Creation of humanised 323/A3 (IgG_4cys pEE18 Expression Plasmid The pEE6HMCV plasmid (see FIG. 8) encoding full length humanized heavy chain DNA was digested with BAM HI and Bgl II to liberate a 3.2 kb fragment that contained the DNA encoding the heavy chain under the transcriptional control of the major immediate early promoter of the human cytomeglovirus. This fragment was cloned into the Bam HI site of pEE12 that contained the DNA encoding the humanized light chain (See FIG. 11 for humanised 323/A3(IgG₄) Kappa Light Chain Amino Acid Sequence and FIG. 12 for the 323/A3IgG_4cys variant Heavy Chain Amino Acid Sequence). See FIG. 10 for schematic representation of the pEE18 plasmid encoding 323/A3 heavy and light chains.

2.2.2 Transfection and Selection of NSO Cells: see Example 6 above.

1. Purpose/Summary

[0158] The cDNAs encoding the humanized 323/A3(IgG_2cys) antibody heavy and light chains were genetically engineered into a single Celltech glutamine synthethase (GS) expression plamid, pEE18, and used to transfect murine NSO cells.

2. Materials and Methods

2.1 Materials as for Examples 6 and 7 above

2.2 Creation of 323/A3 (IgG_2cys) pEE18 Expression for Plasmid

[0159] The pEEE6 hcmv plasmid encoding full length humanized heavy chain DNA was digested with Bam HI and Bgl II to liberate 3.2 kb fragment that contained the DNA encoding the heavy chain under the transcriptional control of the major immediate early promoter of the human cytomegalovirus. This fragment was cloned into the Bam II site of pEE12 that contained the DNA encoding the humanized light chain (See FIG. 13 for 323/A3(IgG_2cys) Kappa Light Chain Amino Acid Sequence and FIG. 14 for the 323/A3(IgG_2cys) Heavy Chain Amino Acid Sequence). See FIG. 10 for schematic representation of the pEE18 plasmid encoding 323/A3 (IgG_2cys) heavy and light chains.

2.2.2 Transfection and Selection of NSO Cells—See Examples 6 and 7 above.