The significance of cytotoxicity of lymphocyte populations in coloncarcinoma patients in comparison to normal controls using a new statistical method of evaluation.

Lymphocytotoxicity was studied in patients suffering from colon carcinoma. 125Iododeoxyuridin-labeled established colon carcinoma cells were used as target cells in a cytotoxicity assay according to COHEN et al. (1). They were incubated with unfractionated lymphocytes of the peripheral blood, purified T cells or adherent cells as effector cells. The cytotoxic effect was measured in cpm, i. e. the radioactivity released from the target cells into the supernatant. The cpm were adjusted by eliminating the variables of spontaneous release and for maximal incorporated radioactivity of the target cells. The cytotoxic activity of lymphocytes to colon carcinoma target cells was dose-dependent. It was demonstrated for lymphocytes from colon carcinoma patients as well as for lymphocytes from healthy controls. Although adjusted cpm values were used for statistical analysis lymphocytes from the 20 colon carcinoma patients did not show a higher cytotoxic effect than lymphocytes from the 20 healthy controls. However, when the results were compared separately for each experiment consisting of 2 colon carcinoma patients and 3 healthy controls, statistical significant specific lymphocytotoxicity was demonstrated for lymphocytes from colon carcinoma patients. The plot diagram of the results and the statistical evaluation by the multivariate analysis of variance (Manova) offer a clear and comprehensible representation of the results. Particularly for statistical reasons, this representation is superior to the usual calculation of percent specific lysis. It was demonstrated in this study that the cytotoxic activity in the absence of autologous serum is a function of T cells of the peripheral blood. Adherent cells have a high cytotoxic activity regardless whether the cells were obtained from colon carcinoma patients or from healthy controls.