Cytotoxicity mediation of cells evidencing surface expression of mcsp

This application is a divisional of U.S. patent application Ser. No. 10/949,846, filed Sep. 24, 2004, which is a continuation-in-part of U.S. patent application Ser. No. 10/810,744, filed Mar. 26, 2004, now U.S. Pat. No. 7,393,531, issued Jul. 1, 2008, which is a continuation-in-part of U.S. patent application Ser. No. 10/762,129, filed Jan. 20, 2004, now U.S. Pat. No. 7,361,342, issued Apr. 22, 2008, which is a continuation-in-part of U.S. patent application Ser. No. 10/743,451, filed Dec. 19, 2003, now abandoned, which is a continuation of U.S. patent application Ser. No. 10/348,231, filed Jan. 21, 2003, now U.S. Pat. No. 7,009,040, issued Mar. 7, 2006, the contents of each of which are herein incorporated by reference.

This invention relates to the diagnosis and treatment of cancerous diseases, particularly to the mediation of cytotoxicity of tumor cells; and most particularly to the use of cancerous disease modifying antibodies (CDMAB), optionally in combination with one or more chemotherapeutic agents, as a means for initiating the cytotoxic response. The invention further relates to binding assays, which utilize the CDMAB of the instant invention.

Melanoma-associated chondroitin sulfate proteoglycan (MCSP) was identified independently by several investigators who developed monoclonal antibodies to human metastatic melanoma cell lines. Several antibodies were found to react with a specific antigen associated with the melanoma cell surface. The independent development of these antibodies led to the multiplicity of names for the target antigen, all of which were subsequently determined to be MCSP. MCSP has therefore also been referred to as high molecular weight melanoma associated antigen (HMW-MAA), human melanoma proteoglycan (HMP), melanoma-associated proteoglycan antigen (MPG) and melanoma chondroitin sulfate proteoglycan (mel-CSPG), and has been identified as the antigen of various specific antibodies, some of which have been set out below. MCSP was also found to be over 80 percent homologous with the rat proteoglycan NG2 and is hence also referred to by that name.

MCSP is a glycoprotein-proteoglycan complex consisting of an N-linked glycoprotein of 250 kDa and a proteoglycan component >450 kDa. The core glycoprotein is present on the surface of melanoma cells, either as a free glycoprotein or modified by the addition of chondroitin sulfate. The molecular cloning of MCSP led to the identification of several structural features. There are 3 extracellular domains containing a total of 10 cysteines (5 potential disulfide bridges), 15 possible N-linked glycosylation sites, and 11 potential chondroitin sulfate attachment sites. The transmembrane segment has a single cysteine, however the functional significance of that residue has not been established. The cytoplasmic domain has 3 threonine residues that may serve as sites for phosphorylation by protein kinase C, although it has not yet been shown that MCSP is phosphorylated.

It has been shown that MCSP is expressed in the majority of melanoma cancers, and it was originally thought that it had a very limited distribution on normal cells and other tumor types. One early study that led to this conclusion used immunohistochemistry (IHC) on normal and tumor tissues fixed with formaldehyde or methanol in order to determine the distribution of MCSP using anti-MCSP antibody B5. In this study, antibody B5 was found to react with 17 out of 22 melanoma tumors tested, 2 out of 2 astrocytomas tested, and none of the 23 carcinomas tested. Out of 22 normal tissues tested, B5 was found to bind only skin keratinocytes, lung alveolar epithelium and capillary endothelium.

Another study examined the tissue distribution of MCSP as defined by anti-MCSP antibody 9.2.27 using frozen tissue sections. Again, reactivity was found in all melanoma tissues and cell lines tested, but there was no reactivity in any of the 6 various carcinoma tumors tested. Out of the 7 fetal tissues tested, reactivity was only observed in the skin and faintly in the aorta while in adult tissues, reactivity was only seen in 3 out of 13 tissues tested.

A subsequent study examined the distribution of MCSP using the anti-MCSP antibodies B5, 9.2.27, 225.28S and A0122, all of which recognize distinct epitopes of MCSP. This study was performed on frozen tissues. It was found that all of the anti-MCSP antibodies had similar staining patterns, reacting with normal and malignant tumors of neural, mesenchymal and epithelial origin, that were previously thought to be MCSP negative. Specifically, the antibody B5 reacted with various epithelial, connective, neural and muscular tissues in the 24 organs that were tested, and reacted with 28 out of 34 various tumors tested. The authors explained that the differences between their findings and previous reports were due to the use of improved and more consistent IHC techniques, noting that choice of fixative was important, presumably leading to the conclusion that an important characteristic of the MCSP antigen is its sensitivity to the processing steps involved in IHC.

A further study was carried out in order to localize MCSP at the ultrastructural level. Immunolocalization studies using electron microscopy demonstrated that MCSP was localized almost exclusively to microspikes, a microdomain of the melanoma cell surface that may play a role in cell-cell contact and cell-substratum adhesion.

The molecular cloning of MCSP in 1996 enabled northern blot analysis of MCSP expression in tumor cell lines and normal human tissues using MCSP cDNA probes. Out of 8 various tumor cell lines tested, expression of MCSP was observed only in the melanoma cell line. MCSP expression was not seen in any of the 16 normal adult and 4 normal fetal tissues tested. The discrepancies found in different studies of tissue localization of MCSP indicate that further study may be required to elucidate the actual expression patterns of this antigen or to account for the differences that have been reported.

Since proteoglycans have been known to mediate cell-cell and cell-extracellular matrix (ECM) interactions, the role of MCSP in these processes has been investigated. MCSP has been shown to stimulate α₄β₁-integrin mediated adhesion and spreading of melanoma cells, and it has also been proposed that signaling through the MCSP core protein induces recruitment and tyrosine phosphorylation of p130^cas which may regulate cell adhesion and motility, contributing to tumor invasion and metastasis. The combination of these results indicated that MCSP may function to enhance adhesion of melanoma cells by both activating integrins and stimulating pathways that lead to cytoskeletal rearrangement.

MCSP has also been found to associate with membrane-type 3 matrix metalloproteinase (MT3-MMP), likely through the chondroitin sulfate component of MCSP. It has been suggested that MT3-MMP expression in melanomas in vivo could promote the degradation of ECM proteins in the vicinity of the growing tumor, providing space in which the tumor can expand. Therefore, the association between MT3-MMP and MCSP may be an activation step to promote melanoma invasion.

Several in vitro assays using anti-MCSP antibodies have been carried out to examine the role of MCSP in processes linked to tumor invasion and metastasis. The role of MCSP in anchorage-independent growth was assessed using the antibody 9.2.27. Melanoma cells cultured in soft agar containing 9.2.27 showed a 67-74 percent specific decrease in their colony formation. These findings suggested that MCSP might be involved in cell-cell interaction, and contribute to anchorage-independent growth. The same authors also examined the effects of blocking MCSP with 9.2.27 in assays measuring the adhesion of M14 melanoma cells on basement membranes of bovine aorta endothelial (BAE) cells. The effect of 9.2.27 treatment was compared to treatment with a control monoclonal antibody W6/32 (directed against all class I histocompatibility antigens). M14 control cells and M14 cells pretreated with antibody were plated on basement membranes of BAE cells. A significant inhibition of 27 percent in cell adhesion was observed in 9.2.27 treated cells, whereas no significant effect was observed in W6/32 treated cells. A more striking effect of cell pretreatment with 9.2.27 was the inhibition of cell spreading which was verified at the ultrastructural level using scanning electron microscopy.

Many of the antibodies that were developed against melanoma cells and determined to specifically recognize MCSP have been tested in both in vitro and in vivo assays to determine their anti-cancer effects.

Monoclonal antibody 9.2.27 recognizes the core glycoprotein component of MCSP and was one of the first antibodies investigated for tumor suppressing properties. Bumol et al. investigated 9.2.27 and a diphtheria toxin A (DTA) conjugate of 9.2.27 for immunotherapy of melanoma tumors grown in nude mice. In vitro cytotoxicity assays were first carried out by measuring the effects of both 9.2.27 and 9.2.27-DTA conjugate on protein synthesis in M21 human melanoma cells as indicated by protein incorporation of [³⁵S]methionine. The 9.2.27-DTA conjugate significantly inhibited protein synthesis in M21 melanoma cells, though a greater effect was seen with unconjugated DTA. There was only minimal effect achieved by 9.2.27 alone. Both the 9.2.27 and 9.2.27-DTA conjugate were investigated for anti-tumor effects in human melanoma tumor-bearing nude mice. M21 tumor mince was implanted subcutaneously and allowed to establish growth for 3 days, then mice were treated at day 3 and at 3 day intervals thereafter with either 9.2.27 or 9.2.27-DTA conjugate. Tumor volumes of treated mice were compared to those of untreated control mice. At day 18 (the last day for which data was reported), 9.2.27 treated mice showed a 64 percent inhibition of tumor growth while 9.2.27-DTA conjugate treated mice showed a 52 percent inhibition of tumor growth, compared to untreated controls. In this initial study the authors concluded that 9.2.27 and 9.2.27-DTA conjugate were approximately equivalent in their effect on suppression of growth of M21 melanoma tumors in nude mice. While this initial study reports in vivo suppression of tumor growth by treatment with 9.2.27, several subsequent studies, including those by the same authors, have demonstrated that naked 9.2.27 did not exhibit any anti-tumor effects in vivo. Collectively, as outlined below, the experiments carried out to investigate the utility of using 9.2.27 to treat human tumors have demonstrated that, although cancer cells were targeted by 9.2.27, no anti-cancer activity resulted from treatment with the naked antibody.

A phase I clinical trial was carried out which was designed to give large doses of 9.2.27 in anticipation of later therapeutic studies with 9.2.27 immunoconjugates. Eight patients with malignant melanoma whose tumors reacted with 9.2.27 by flow cytometry and/or immunoperoxidase staining, received single doses of antibody intravenously, twice weekly on a dose escalating scale of 1, 10, 50, 100 and 200 mg. Although none of the patients experienced significant toxicity and 9.2.27 localized to the metastatic melanoma nodules, no clinical responses were observed.

In a later study, 9.2.27 was conjugated to the chemotherapeutic drug doxorubicin (DXR), and the conjugate was investigated for growth inhibition of melanoma in vitro and in vivo. Growth inhibition of M21 cells treated with the DXR-9.2.27 conjugate was measured using a [³H]thymidine incorporation assay. The conjugate showed specific dose-dependent growth inhibition of the M21 target cells and no effect on an MCSP negative control cell line. No in vitro assays were carried out examining effects of 9.2.27 alone. To investigate the DXR-9.2.27 conjugate in vivo, M21 cells were injected subcutaneously and allowed to establish a tumor for 8-10 days. Injections were given intravenously at day 10 and at 3 day intervals thereafter for 30 days. Significant suppression of tumor growth was seen only in mice treated with the DXR-9.2.27 conjugate. Both DXR treatment alone and 9.2.27 treatment alone failed to suppress tumor growth; a mixture of 9.2.27 and DXR showed similar negative effects.

Another study was carried out investigating the effects of a 9.2.27 conjugate. Schrappe et al. conjugated the chemotherapeutic agent 4-desacetylvinblastine-3-carboxyhydrazide (DAVLBHY) to 9.2.27 and tested its effect on human gliomas. Nude mice were injected with U87MG (a human glioma cell line) cells subcutaneously and the animals were treated on days 2, 5, 7, and 9. Tumor volume was most effectively reduced by the 9.2.27-DAVLBHY conjugate. Control groups, which were treated with either PBS or 9.2.27 alone, developed fast growing tumors and there was no reduction in tumor volume in 9.2.27 treated mice compared to mice treated with PBS.

Antibody 225.28S was made against the human M21 melanoma cell line, and was initially described as reacting with a high molecular weight melanoma associated antigen. This molecule was subsequently shown to be the same molecule as MCSP. An early study tested the cytolytic ability of 225.28S, an IgG_2a, on a human melanoma cell line and compared it to another anti-MCSP antibody, clone 653.40S that was an IgG₁. 225.28S and 653.40S were determined to recognize the same, or spatially close, antigenic determinants on MCSP. It was found that neither antibody could lyse melanoma cells in conjunction with complement in in vitro assays. Both antibodies could mediate lysis of target melanoma cells in an antibody-dependent cell-mediated (ADCC) cytotoxicity assay, with 225.28S exhibiting a higher lytic activity than 653.40S. However, lysis of melanoma cells was only obtained with a significantly higher effector/target cell ratio than had been reported by others using anti-melanoma antigen antibodies. The authors concluded that the lack of cytolytic activity of these antibodies in conjunction with human complement and the high effector/target cell ratio required for lysis to occur in ADCC suggested that the injection of monoclonal antibodies into melanoma patients was not likely to cause the destruction of tumor cells. The authors suggested that the immunotherapeutic use of these antibodies should be limited to utilizing them as carriers of radioisotope, chemotherapeutic or toxic agents.