Résumé : The present study was designed to analyse the cytotoxic effect of cisplatin in vitro as a function of various exposure times (up to 120 h), keeping constant the parameter C × T (product of the drug concentration per time). Intracellular drug concentrations were measured in parallel following analysis of cisplatin influx and efflux characteristics. A head and neck cancer cell line was selected to represent the spectrum of cisplatin antitumour activity. The IC50 values (μg/ml) for 1, 2, 11 and 121 h were, respectively 4.51, 2.73, 0.27 and 0.151. Reduction of the IC50 was clearly not linearly related to prolongation of the cisplatin exposure time. The kinetics of cisplatin incorporation into CAL 27 cells was investigated as a function of different cisplatin concentrations. A plateau was reached after 16 h of contact. For the extracellular cisplatin concentrations of 1, 2.5, 5 and 10 μg/ml, the average intracellular Pt concentrations at the plateau were, respectively (ng/106 cells): [mean (S.D.)] 12.8 (0.98), 31.11 (5.12), 71.38 (6.03) and 136.7 (16.5). Intracellular Pt concentrations were linearly related to the extracellular drug concentration (r = 0.99). The drug left the cells following a two-slope kinetics pattern with an α half-life of 1.29 h and a β half-life of 94.4 h. The cytotoxic effect for a given C × T clearly differed for the different cisplatin exposure times. The longest exposure time (121 h) gave the least pronounced cytotoxicity. The intracellular Pt concentrations were linearly related to the C × T values. Cisplatin levels were much lower after the 121 h exposure. These data may prove valuable in establishing a rationale which can aid in selection of optimal modes of clinical cisplatin administration. © 1992.