Cytotoxicity mediation of cells evidencing surface expression of trop-2

This application is a divisional of U.S. patent application Ser. No. 11/709,898, filed on Feb. 22, 2007, which claims benefit of the filing date of U.S. Provisional Patent Application No. 60/776,466, filed on Feb. 24, 2006, the contents of which are herein incorporated by reference.

This invention relates to the isolation and production of cancerous disease modifying antibodies (CDMAB) and to the use of these CDMAB in therapeutic and diagnostic processes, optionally in combination with one or more chemotherapeutic agents. The invention further relates to binding assays which utilize the CDMAB of the instant invention.

TROP-2 is a cell surface glycoprotein expressed on most carcinomas, as well as some normal human tissues. It was initially defined as a molecule recognized by two murine monoclonal antibodies raised to a human choriocarcinoma cell line BeWo that recognized an antigen on human trophoblast cells (Faulk 1978). The same molecule was independently discovered by other investigators which led to multiple names describing the same antigen. Hence, TROP-2 was also referred to as GA733-1 and epithelial glycoprotein-1 (EGP-1) (Basu 1995, Formaro 1995).

The TROP-2 gene is an intronless gene that was thought to have been formed through the retroposition of a homologous gene GA 733-2 (also known as epithelial glycoprotein-2, EpCAM and Trop-1) via an RNA intermediate. The TROP-2 gene has been mapped to chromosome 1p32 (Calabrese 2001). The protein component of TROP-2 has a molecular mass of approximately 35 kilodaltons. Its mass may be increased by 11-13 kilodaltons with heterogeneous N-linked glycosylation of its extracellular domain. There are many cysteine residues in the extracellular domain which could form disulfide bridge sites. TROP-2 is a substrate for protein kinase C, a Ca²⁺ dependent protein kinase and the intracellular serine 303 residue has been shown to be phosphorylated (Basu 1995). It has also been shown that crossing-linking of TROP-2 with anti-TROP-2 antibodies transduced a calcium signal as shown by a rise in cytoplasmic Ca²⁺ (Ripani 1998). These data support signal transduction as a physiological function of TROP-2, although to date no physiological ligand has been identified. Recently an association between TROP-2 expression and cancer has been shown as TROP-2 was identified as a member of a group of genes reported to be the most highly overexpressed in ovarian serous papillary carcinoma compared to normal ovarian epithelium in a large-scale gene expression analysis using cDNA microarray technology (Santin 2004).

The expression profile of TROP-2 has been elucidated through immunohistochemistry (IHC) and flow cytometery studies using many different TROP-2 antibodies. Anti-TROP-2 antibodies 162-25.3 and 162-46.2 were produced through immunization of mice with the human choriocarcinoma cell line BeWo, and were investigated for their reactivity to a series of tumor and lymphoid cell lines and peripheral blood mononuclear cells. In this study both antibodies appeared to be trophoblast specific, staining 3 of the 4 choriocarcinoma cell lines tested, while none of the other lymphoid or tumor cell lines (representing fibrosarcoma, cervical sarcoma, colon carcinoma, melanoma, neuroblastoma, erythroleukemia) were stained in an indirect immunofluorescence FACS assay. In addition, none of the normal peripheral blood cells were stained. The antibodies were tested for staining of formalin-fixed paraffin-embedded placenta tissue sections and frozen normal sections of liver, kidney, spleen, thymus and lymph node tissues. The placenta tissue sections were stained with both antibodies, while there was no staining of the other normal tissues (Lipinski 1981). These two antibodies have strictly been reported for use in in vitro diagnostic studies.

Anti-TROP-2 antibody MOv16 was generated through the immunization of mice with a crude membrane preparation of poorly differentiated ovarian carcinoma OvCa4343/83. MOv16 was tested for reactivity to a series of frozen tissue sections of benign and malignant ovarian tumors. MOv16 reacted with 31 of 54 malignant ovarian tumors and 2 of 16 benign ovarian tumors. Of the 5 mucinous ovarian tumors that were tested, MOv16 was completely unreactive. MOv16 was also tested for reactivity to frozen sections of non-ovarian malignant tumors where it was found to bind 117 of 189 breast carcinoma sections and 12 of 18 lung carcinoma sections. MOv16 was completely unreactive on 16 non-epithelial tumors that were tested (including liposarcomas, chondrosarcomas, endotheliomas, histiocytomas and dysgerminomas). When tested on frozen normal tissue sections, MOv-16 was reactive with breast, pancreas, kidney and prostate sections. MOv16 reactivity was reported to be negative on lung, spleen, skin, ovary, thyroid, parotid gland, stomach, larynx, uterus and colon sections, though the number of tissue sections that were used was not reported. The authors noted that frozen tissue sections were used because MOv16 was unreactive to paraffin embedded tissues (Miotti 1987). This antibody has also only been reported for use in in vitro diagnostic studies.

Anti-TROP-2 antibody Rs7-3G11 (RS7) was generated through the immunization of mice with a crude membrane preparation derived from a surgically removed human primary squamous cell carcinoma of the lung. IHC was used to examine the staining of RS7 on frozen sections of human tumor and normal tissues. RS7 bound to 33 of the 40 sections representing tumors of the breast, colon, kidney, lung, prostate and squamous cell cancer. Of the normal tissues RS7 bound to 16 of 20 sections of breast, colon, kidney, liver, lung and prostate tissues while none of the five sections of spleen tissue were stained. In this study the authors noted that it appeared that antigen density in tumors was higher than in normal epithelial tissues (Stein 1990).

Additional studies of the tissue specificity of RS7 were carried out on both tumor and normal tissues. RS7 was tested on a panel of frozen tumor sections and bound to 65 of the 77 sections representing tumors of the lung, stomach, kidney, bladder, colon, breast, ovary, uterus and prostate. There was no binding to the 5 lymphomas tested. RS7 was tested on a panel of 85 frozen human normal tissue sections composed of a total of 24 tissue types. 39 sections of 13 normal tissues (lung, bronchus, trachea, esophagus, colon, liver, pancreas, kidney, bladder, skin, thyroid, breast and prostate) were stained by RS7. The authors of this study noted that in the tissues in which positive staining was observed, the reactivity was generally restricted to epithelial cells, primarily in ducts or glands. It was also noted that this study was limited to frozen sections since it was observed that RS7 was not reactive on formalin-fixed paraffin-embedded sections (Stein 1993).

Polyclonal anti-TROP-2 antibodies were prepared by immunizing mice with a synthetic peptide corresponding to amino acid positions between 169 and 182 of the cytoplasmic domain of human TROP-2. The polyclonal antibodies were tested on a tissue array slide that contained formalin-fixed human esophageal hyperplasia and carcinoma tissues. Ten of the 55 carcinoma specimens displayed heavy staining with the polyclonal antibodies, while the mild hyperplasia tissue stained very weakly, indicating expression levels may be related to malignant transformation (Nakashima 2004).

Overall, IHC reactivity patterns obtained with different anti-TROP-2 antibodies were consistent. Expression in cancer was seen primarily in carcinomas, and most carcinomas were reactive. In normal tissues, expression appeared to be limited to cells of epithelial origin, and there was some evidence that staining of carcinomas was stronger than staining of corresponding normal epithelial tissues.

In addition to being used in IHC studies, antibody RS7 was tested in in vivo models with initial experiments consisting of tumor targeting studies in nude mouse xenograft models. Radiolabeled RS7 injected i.v. was shown to accumulate specifically in the tumor of mice bearing either Calu-3 (lung adenocarcinoma) or GW-39 (colon carcinoma) tumors (Stein 1990). Further studies were done to investigate the biodistribution of radiolabeled RS7 in a xenograft system and to study the therapeutic potential of RS7 as an immunoconjugate. In this study the therapeutic efficacy of ¹³¹I-labeled RS7 F(ab′)₂ was investigated in nude mice bearing Calu-3 human lung adenocarcinoma xenografts. Three weeks following inoculation of the mice with Calu-3 cells, when the tumors had reached a size of approximately 0.3-0.9 grams, groups of 6-7 mice were treated with a single dose i.v. of either 10.0 mCi ¹³¹I-RS7-F(ab′)₂ or 1.5 mCi 131I-RS7-F(ab′)₂ and compared to a similar group of untreated control mice. The single dose of 1.0 mCi ¹³¹I-RS7-F(ab′)₂ resulted in tumor growth suppression for approximately 5 weeks, while the single dose of 1.5 mCi ¹³¹I-RS7-F(ab′)₂ resulted in tumor regression, and the mean tumor size did not exceed the pre-therapy size until the eighth week after radioantibody injection. Mice receiving the 1.5 mCi ¹³¹I-RS7-F(ab′)₂ dose experienced a mean body weight loss of 18.7 percent, indicating there was toxicity associated with the treatment. In this study, effects of treatment with naked RS7 or the F(ab′)₂ fragment of RS7 were not tested (Stein 1994a). Another study was done to test the efficacy of ¹³¹I-RS7 in a MDA-MB-468 breast cancer xenograft model. Groups of ten mice bearing MDA-MB-468 tumors of approximately 0.1 cm³ were treated with a single dose i.v. of either 250 microcuries ¹³¹I-RS7 or 250 microcuries ¹³¹I-Ag8 (an isotype matched control antibody). Groups of six mice were treated with a single dose i.v. of 30 micrograms of either unlabeled RS7 or Ag8. Complete regression of the tumors (except for one animal that had a transient reappearance of tumor) was seen in the animals treated with ¹³¹I-RS7, which lasted for the duration of the 11 week observation period. Tumor regression was also seen in ¹³¹I-Ag8 treated mice, though was only observed between 2 weeks and 5 weeks with tumors either persisting or continuing to grow for the remainder of the study. Tumor growth of mice that received unlabeled RS7 or Ag8 was not inhibited and there did not appear to be any differences in the mean tumor volume of RS7 treated mice compared to the Ag8 treated mice. Two additional groups of 10 mice bearing larger MDA-MB-468 tumors of approximately 0.2-0.3 cm³ were treated with a slightly higher single dose of either 275 microcuries ¹³¹I-Rs7 or 275 microcuries ¹³¹Ag8 and compared to a similar group of untreated mice. Tumor volume was measured weekly for 15 weeks. Although in this case there was a significant difference in tumor growth between the ¹³¹I-RS7 treated mice compared to the untreated mice, there was no significant difference in the tumor growth of the ¹³¹I-RS7 compared to the ¹³¹I-Ag8 treated mice, indicating a portion of the efficacy may have been due to non-specific effects of the radiation. Unlabeled antibodies were not tested in mice containing 0.2-0.3 cm³ tumors (Shih 1995).

There have been numerous additional studies examining the efficacy of RS7 as an immunoconjugate with an attempt to select the optimal radiolabel for radioimmunotherapy (Stein 2001a, Stein 2001b, Stein 2003). A humanized version of RS7 has also been generated, however it has only been tested in preclinical xenograft models as a radioconjugate (Govindan 2004). These studies show similar positive effects as the previously described studies with RS7, however in one study, even when radiolabeled RS7 was delivered at a previously determined maximum tolerable dose, toxicity occurred leading to death in some mice (Stein 2001a). Although effective treatment of xenograft tumors in mice was achieved with radiolabeled RS7 in these studies, naked RS7 was not evaluated.

Immunizing mice with neuramindase pre-treated H3922 human breast carcinoma cells produced the anti-TROP-2 monoclonal antibody BR110 (as disclosed in U.S. Pat. No. 5,850,854, refer to Prior Patents section). By immunohistology, using human frozen tissue specimens, BR110 was shown to react with a wide range of human carcinoma specimens including those of the lung, colon, breast, ovarian, kidney, esophagus, pancreas, skin, lung and tonsil. No human normal tissue sections were tested. In vitro studies demonstrated that BR110 had no ADCC or CDC activity on the human carcinoma cell lines H3396 or H3922. In vitro studies analyzing the cytotoxicity of BR110-immunotoxins was performed on the human cancer cell lines H3619, H2987, MCF-7, H3396 and H2981. The EC₅₀ for the cell lines tested was 0.06, 0.001, 0.05, 0.09 and >5 micrograms/mL respectively. No cytotoxicity data was disclosed for the naked BR110 antibody. No in vivo data was disclosed for the naked or immunoconjugated BR110.

A number of additional antibodies have been generated that target TROP-2, such as MR54, MR6 and MR23 which were generated from immunization of mice with the ovarian cancer cell line Colo 316 (Stein 1994b) and antibody T16 which was generated by immunization of mice with the T24 bladder cancer cell line (Fradet 1984). The use of these antibodies has been limited to biochemical characterization of the TROP-2 antigen and cell line and tissue expression studies. There have been no reports of anti-cancer efficacy of these antibodies, either in vitro or in vivo. RS7 was the only antibody that was tested for therapeutic efficacy in preclinical cancer models, with its use being limited to a carrier of radioisotope. There are no reports of any naked TROP-2 antibodies exhibiting therapeutic efficacy in preclinical cancer models either in vitro or in vivo.

Monoclonal Antibodies as Cancer Therapy: Each individual who presents with cancer is unique and has a cancer that is as different from other cancers as that person\'s identity. Despite this, current therapy treats all patients with the same type of cancer, at the same stage, in the same way. At least 30 percent of these patients will fail the first line therapy, thus leading to further rounds of treatment and the increased probability of treatment failure, metastases, and ultimately, death. A superior approach to treatment would be the customization of therapy for the particular individual. The only current therapy which lends itself to customization is surgery. Chemotherapy and radiation treatment cannot be tailored to the patient, and surgery by itself, in most cases is inadequate for producing cures.

With the advent of monoclonal antibodies, the possibility of developing methods for customized therapy became more realistic since each antibody can be directed to a single epitope. Furthermore, it is possible to produce a combination of antibodies that are directed to the constellation of epitopes that uniquely define a particular individual\'s tumor.

Having recognized that a significant difference between cancerous and normal cells is that cancerous cells contain antigens that are specific to transformed cells, the scientific community has long held that monoclonal antibodies can be designed to specifically target transformed cells by binding specifically to these cancer antigens; thus giving rise to the belief that monoclonal antibodies can serve as “Magic Bullets” to eliminate cancer cells. However, it is now widely recognized that no single monoclonal antibody can serve in all instances of cancer, and that monoclonal antibodies can be deployed, as a class, as targeted cancer treatments. Monoclonal antibodies isolated in accordance with the teachings of the instantly disclosed invention have been shown to modify the cancerous disease process in a manner which is beneficial to the patient, for example by reducing the tumor burden, and will variously be referred to herein as cancerous disease modifying antibodies (CDMAB) or “anti-cancer” antibodies.

At the present time, the cancer patient usually has few options of treatment. The regimented approach to cancer therapy has produced improvements in global survival and morbidity rates. However, to the particular individual, these improved statistics do not necessarily correlate with an improvement in their personal situation.