EFFECTOR MECHANISMS IN CANCER IMMUNITY
Immunological studies performed in animals developing experimental tumors following exposure to MCA and other chemical carcinogens clearly showed that the immunity directed against rejection antigens is T-cell–mediated, since it can be transferred among mice within the same strain through the intravenous injection of immune cells, but not through the transfer of immune serum. Tumor-specific cytotoxic T cells (CTL) are generally MHC-restricted CD8+ lymphocytes, because they kill target tumor cells through recognition of an antigenic epitope bound to an appropriate MHC-I molecule expressed on the surface of the tumor cell. Additional data suggest that CD4+ lymphocytes are crucial to the develop-ment of CD8+ CTL.
Active CD8+ CTL can be derived from the peripheral blood of cancer patients by propagating the cells in the presence of interleukin-2 (IL-2) under special conditions of ap-propriate antigenic stimulation, which over time appear to expand the CD8+ component ca-pable of expressing tumor-specific CTL activity. In extensive studies conducted at the Surgery Branch of the National Cancer Institute, it has been shown that tumor-infiltrating lymphocytes (TIL) obtained by expanding lymphocyte populations present in surgically re-sected melanoma lesions often represent an enriched population of CTL capable of recog-nizing and killing autologous tumor cells. Taken together, these observations led to clini-cal studies in which melanoma patients were treated by infusion of IL-2 and their own TIL, which had been expanded ex vivo to large numbers (1011–1013). The results of these stud-ies showed that some patients did benefit from these experimental treatments.
Monoclonal antibodies have been extensively utilized in studies aimed at finding out whether antibody-mediated immunity plays any significant role in anticancer immunity. Although there is very little evidence that antibodies are relevant to controlling tumor growth and spread, recent laboratory and clinical studies have revealed some surprising facts. Often cancer patients have in their serum antibodies that specifically recognize anti-gens expressed or released by their own tumors. These antibodies have led to the identifi-cation of several new tumor-associated antigens, and their possible clinical relevance is currently being evaluated. Moreover, several monoclonal antibodies are being assessed as therapeutics for a number of human cancers, and one of them, as previously discussed, is currently approved for treatment of breast cancers overexpressing the HER-2 gene. There-fore, it is important to realize that, while antibodies in general may not play a major role in natural antitumor immunity, appropriately engineered monoclonal antibodies targeting cer-tain tumor-associated antigens may represent important therapeutic tools against some types of cancer. The results of these clinical studies also suggest that successful approaches to immunotherapy utilizing either monoclonal antibodies or lymphocytes are not necessar-ily aimed at restoring what may be defective in the host immune response of the cancer pa-tient. What these immunotherapeutic interventions seem to achieve is the mobilization of antitumoral immunological processes, mediated either by cells or by proteins, to specifi-cally destroy autologous tumor cells.
Natural killer (NK) cells are a special subset of lymphocytes capable of killing tumor cells by a less specific mechanism than the one utilized by HLA-restricted CTL. NK cells preferentially recognize target cells that are low or nonexpressors of MHC molecules, a phenotype frequently exhibited by tumor cells as a way to avoid recognition by host CTL. Activation of NK cells can be achieved in vitro by propagating and expanding them in the presence of IL-2. These lymphokine-activated killer (LAK) cells have been utilized in clin-ical studies similar to those using the combination of TIL/IL-2; however, their lack of specificity supports the conclusion that HLA-restricted CTL are the most effective killers of autologous tumor cells.
Cytokines have been intensely studied because of their ability to modulate an exten-sive array of immune functions. We will discuss only IL-2 and the granulocyte-macrophage colony-stimulating factor (GM-CSF), since these two immuno-augmenting cytokines are of major importance not only for tumor immunology, but also for immunotherapeutic ap-proaches that have shown some real promise in recent clinical studies.
IL-2 is a well-known activator of T-cell immunity, mostly by stimulating helper T cells to proliferate following their interaction with antigen-presenting cells (APCs); in turn, activated helper T cells (TH1 cells) allow for the subsequent stimulation and proliferation of CTL, which will be ultimately responsible for destroying the target tumor cells. The key role played by IL-2 in the activation and maturation of the cell-mediated immune response explains to a large extent why IL-2 has become an approved drug for immunotherapy of melanoma and renal cell cancer. It is reasonable to expect that the uses of IL-2 may be ex-panded to other forms of cancer and possibly to AIDS immunotherapy.
GM-CSF was identified more recently than IL-2, and only after its discovery was it realized that one key function of this cytokine is the activation of APCs. In a seminal study aimed at comparing the immunogenicity of tumor cells engineered to express a wide vari-ety of cytokines, it was shown that GM-CSF–expressing cells inactivated by irradiation and injected as a cancer vaccine into appropriate animal hosts were able to protect vaccinated animals from challenge with a lethal dose of unmodified tumor cells. Moreover, it was also shown that this kind of vaccination had some efficacy in a therapeutic setting, one in which vaccinated animals already carried a tumor burden. The results of this work helped refocus the attention of tumor immunologists on APCs as the central element of an effective im-mune response. In other words, APCs appear to explain the difference between antigenicity and immunogenicity, insofar as they could confer immunogenicity to antigenic entities that would have otherwise been not only inef-fective, but counterproductive to tumor immunity . These animal studies also prompted a number of human clinical studies utilizing autologous tumor cells, either engi-neered to produce GM-CSF or simply mixed with it, as therapeutic vaccines for various forms of cancer.