Tumor-associated antigens in gynecologic cancer

If the study of tumor immunology is to have a profound impact on clinical medicine, certain hypotheses must be proven to be valid. First and foremost, it must be demonstrated that malignant tissue possesses antigenic substances (probably protein moieties) that are unique to that particular malignant process. In addition, these antigenic substances must be very similar in histologically similar tumors. Second, the host defense mechanisms must be capable of reacting to these tumor-associated antigens. The reaction is, of course, necessary in order to develop both diagnostic and therapeutic routes of application. The reaction of the immunologic system to these tumor-associated antigens could be monitored as an early serodiagnostic tool for subclinical cancer, and the cytotoxic reaction holds great promise as an immunotherapeutic tool. The essence of tumor immunologic research can thus be stated in the form of the following questions: 1. Do histologically similar cancers from identical primary sites share common tumor-associated antigens? 2. Does the immunologic system react to these antigens? 3. Can this reaction be assayed on one hand for serodiagnosis and augmented on the other for immunotherapy? Specific antigens have been found in animal tumors and have been divided into two classes: the viral induced tumors, which share common antigens when caused by the same viral agent, and carcinogen-induced tumors, which appear to have unique antigenic determinants for each tumor. In recent years a great many human tumors have been found to have tumor-associated antigens; these include colonic carcinoma, neuroblastoma, melanoma, soft tissue and osteogenic sarcoma, bladder carcinoma and Burkitt's lymphoma. This report includes evidence for the existence of such antigens in adenocarcinoma of the ovary and squamous cell carcinoma of the cervix. The laboratory evidence that has been presented would suggest that there are both a cell-mediated response and humoral response to the antigenic determinants of these two gynecologic cancers. It would appear that the mediated (lymphocyte) effect is considerably more cytotoxic and definitive than the humoral factors measured. In addition, the allogenic experiments would suggest strongly that indeed (at least with regard to these two cancers) histologically similar cancers from the same organ share common antigenic determinants. The identification and isolation of these tumor-associated antigens appears complex. The complexity is increased when one studies patients afflicted with these cancers for plasma carcinoembryonic antigens. This antigen, which was thought to be specific for adenocarcinoma of the colon, is found in the blood of a significant number of patients with adenocarcinoma of the ovary and squamous cell carcinoma of the cervix.