There has been a discrepancy between promising results of experimental chemotherapy in animal melanoma models and clinical response rates. This inconsistency seems to reflect weak points of the assays used so far to monitor the response of melanoma cells to chemotherapeutic agents. Therefore a less usual approach was chosen in the present study: Tumor cells were cultured in peritoneal cavity (B16 melanoma in inbred C57BL/6J mice and Cloudman S91 melanoma in inbred DBA2 mice) to maintain normal in vivo conditions; the animals were receiving the tested agents in i.p. injections and the prolongation of their life span was considered as the principle parameter of therapeutic efficiency of the compounds tested. Previously described therapeutic potency both of vitamins (C, alpha-tocopherol acid succinate) and some phenols (hydroquinone, 4-hydroxyanisole) was confirmed. Benzoate, spin trap N-butyl-alpha-phenyl-nitrone and ammonium chloride as a lysosomotropic agent failed to increase the survival of melanoma-bearing mice. Free radical scavenger methimazole exerted a therapeutic effect in mice with pigmented B16 melanoma. Only classic cytostatic agents--cisplatin and cyclophosphamide--proved its therapeutic effect in both melanoma models studied. These results are in accord with the known resistance of human melanoma to conventional chemotherapy. Measurement of serum activity of gamma-glutamyltransferase was shown to be useful for monitoring therapeutic effect.

2nd Department of Medical Chemistry and Biochemistry 1st Faculty of Medicine Charles University Prague Czech Republic