Determination of potential doubling times in human melanoma cell cultures subjected to irradiation and/or hyperthermia by flow cytometry.

The proliferation of human melanoma cells in vitro after irradiation and/or hyperthermia was studied by means of two-parameter flow cytometry. Cultures were incubated with BrdU for 30 min and fixed either immediately or after a delay of several hours. Cells having synthesized DNA were identified with the help of an antibody against BrdU. DNA was stained quantitatively with propidium iodide. In this way the distribution of cells in the phases of cell cycle could be determined and the movement of labeled cells through the phases of the cycle could be analyzed. Experiments in which the cell cycle distribution was studied at 4-h intervals after treatment showed the following: (1) Irradiation (4 Gy X rays) causes the expected G2 block with a maximum after 12-16 h. The proportion of S-phase cells decreases continually during the first 48 h after treatment. (2) Hyperthermia (1 h, 43 degrees C) alone or in combination with irradiation causes a delay in S phase. The cells begin to move into G2 phase only after 12-16 h and accumulate there to some extent. From the progression of labeled cells through the cycle, the duration of S phase could be determined. Experiments and calculations of this kind were done 0, 24 and 48 h after treatment. The duration of S phase was increased only moderately (by 4 h) after irradiation, but a delay of about 30 h occurred after hyperthermia (alone or in combination with X rays). Smaller delays (up to 9 h) were observed 24 and 48 h after treatment. Two different methods were used to calculate potential doubling times. Both of them gave similar results, but a comparison with the actual population doubling times (determined by cell counting) showed that reasonable estimates could be achieved only for the untreated controls. With cultures subjected to irradiation and/or hyperthermia serious discrepancies were observed. This does not seem to be due to technical problems inasmuch as we are dealing with a whole set of data produced under well-defined in vitro conditions (in contrast to the clinical situation, where potential doubling times have to be estimated from single samples). Our results certainly do not encourage the extension of the method (which was originally intended for the prediction of unperturbed tumor growth) to a post-treatment setting.