Tumor Antigens and Tumor-Host Relationships

The view that immunological reactions play a role in modifying or even controlling malignant disease has now gained wide acceptance. This has developed from well­ substantiated investigations showing that many, although by no means all, experi­ mental animal tumors induced by chemical carcinogens or oncogenic viruses, as well as those of unknown (spontaneous) etiology, elicit immune rejection responses against the tumor (I, 2). Consequently, hosts preimmunized against these tumors (e.g. by implantation of radiation-attenuated cells) will reject a subsequent challenge with the same tumor. Comparable evidence for host defenses in human malignant disease is not avail­ able, apart from limited stu4ies indicating that autografted tumor cells do not readily take (3). There are, however, a small number of impressive examples from clinical investigations suggesting, but not ·proving, that in certain circumstances immunological reactions may control malignant disease. These include reports of spontaneous remissions in melanoma, neuroblastoma, and renal cell carcinoma (4), as well as remissions occurring in Burkitt's lymphoma patients receiving minimal chemotherapy (5). Conversely, there is substantial evidence of increased incidences of malignant tumors in patients receiving continuous immunosuppressive therapy follC'wing organ transplantation (6). Finally there is an increasing number of studies showing that procedures designed to heighten patients' general immunocompetence or to specifically boost tumor-immune responses may have therapeutic potential. Broadly speaking, however, the major body of evidence supporting the concept that human tumors express neoantigens similar to those detected in experimental animal tumors has been derived from in vitro assays of cell-mediated and humoral antibody responses to tumor cells (7). In addition there are a number of reports showing that cancer patients exhibit delayed hypersensitivity responses to tumor extracts (8).