Immunoreactivities of polyclonal and monoclonal anti-T and anti-Tn antibodies with human carcinoma cells, grown in vitro and in a xenograft model.

Human polyclonal, monospecific anti-T and -Tn antibodies were found to be reactive in ELISA tests with human ovarian (IGROV-1, OVCAR-3 and SKOV-3), breast (SKBr-3 and T47D)- and oral (KB)-carcinoma cell lines, but less so or non-reactive with normal epithelia and fibroblasts. The direct binding radioimmunoassay, using 125I-labeled human antibodies, to the IGROV-1 cancer cells was inhibited by homologous unlabeled antibodies of the same concentration, but not by the respective immunodominant haptenic monosaccharides (Gal for T and GalNAc for Tn). Rodent ascitic monoclonal anti-T (Ca3114 and Ca3741) and anti-Tn (Ca3250, Ca3268 and Ca3638) antibodies were also reactive with the ovarian- and breast-cancer cells, as measured by FACS and ELISA tests, but to a lower extent than the polyclonal human antibodies. Both the monoclonal anti-T (Ca3741) and anti-Tn (Ca3250 and Ca3638) antibody-binding reactivities were significantly inhibited by the haptenic free monosaccharides. Addition of the above MAbs to IGROV-1 ovarian-cancer or T47D breast-cancer cells cultured in vitro resulted in significant cytological change and inhibition of the viability of the tumor cells, but not of normal epithelial breast cells. This effect on viability was shown to be complement-independent, yet it was profoundly influenced by the concentration of the serum added to the assay medium. In vivo biodistribution of the anti-T (Ca3114) and anti-Tn (Ca3638) MAbs administered i.p. to athymic IGROV-1 tumor-bearing CD1 female nude mice revealed higher 125I-labeled antibody accumulation in the tumor xenografts and in their lung tissues, as compared with other organs of the same mice tested. The above results thus suggest the feasibility of utilizing these antibodies in immunotherapy and drug targeting.