Chemosensitivity measurements of human tumour cells by soft agar assays are influenced by the culture conditions.

To investigate the influence of culture conditions on the in vitro responses of tumour cells to anticancer drugs, the sensitivities observed with the soft agar methods of Hamburger & Salmon (1977) (H-S) and of Courtenay & Mills (1978) (C-M) were compared. In all cases the ID50 values were determined from dose-response curves. Six human tumour cell lines exposed to 10 different agents, and 9 patients' melanomas exposed to 5 different agents, were examined. In the studies of cell lines the H-S method gave higher sensitivity values than the C-M method in 38 out of 52 cases, whereas in 14 cases the results were the same. In the patients' tumours the H-S method gave higher sensitivity in 21 of 35 cases, equal sensitivity in 11, and lower sensitivity in 3 cases. In many instances the ID50 values obtained with the two test systems differed by factors of 10 or more, both in the case of cell lines and tumour specimens. Systematic alterations in the culture conditions indicated that the presence or absence of rat erythrocytes is the most important factor responsible for the differences observed. Also, other factors, such as supplements (in the H-S method) and the use of different serum types, appeared to influence both colony growth and chemosensitivity.