Cells responsible for tumor surveillance in man: effects of radiotherapy, chemotherapy, and biologic response modifiers.

Currently, the most probable theory of tumor surveillance is neither the existence of any tumor-specific, antigen-dependent, T-cell-mediated cytotoxic effect that could eliminate spontaneous tumors in man and that could be used for some kind of vaccination against tumors, nor the complete absence of any surveillance or defense systems against tumors. What is probable is the cooperation of a number of antigen-independent, relatively weakly cytotoxic or possibly only cytostatic humoral and cellular effects, including nutritional immunity, tumor necrosis factor, certain cytokines, and the cytotoxic effects mediated by macrophages, NK cells, NK-like cells, and certain stimulated T-cells. One question remaining to be solved is why these antigen-independent effects do not attack normal cells. A number of plausible hypotheses are discussed. The hypothetical surveillance system is modulated both by traditional cancer treatment and by attempts at immunomodulation. Radiotherapy reduced the T-helper cell function for almost a decade, but not those of macrophages or NK cells. T-cell changes have no prognostic implication, supporting, perhaps, the suggestion of a major role for macrophages and NK cells. Cyclic adjuvant chemotherapy reduces the peripheral lymphocyte population and several lymphocyte functions but not NK activity. Most of the parameters were normalized some years following treatment, but NK activity remained elevated and Th/Ts cell ratio was still decreased. This might possibly be taken to support the surveillance role of NK cells. Bestatin increases the frequency of lymphocytes forming rosettes with sheep red blood cells (but not their mitogenic responses), enhances NK activity, and augments the phagocytic capacity of granulocytes and monocytes (but not their cytotoxic activity). Improved survival with Bestatin treatment following chemotherapy has been observed in patients with melanoma Stages 1b and II and in patients with acute nonlymphatic leukemia, where BCG also seems active, although possibly only in patient groups with less than 49% complete remissions.