Metabolic stability of experimental chemotherapeutic agents in hepatocyte:tumor cell co-cultures.

A U.S. National Cancer Institute screening program for new anticancer drugs, based on the growth of primary human tumor cells in an in vitro soft agar colony formation assay, has resulted in the identification of a number of compounds that have cytotoxic activity against primary human tumor cells in vitro but are inactive in the conventional in vivo murine P388 leukemia animal model pre-screen. To investigate whether metabolic inactivation ov the compounds might be a factor in the lack of in vivo cytotoxicity we have co-cultured rat hepatocytes with A204 rhabdomyosarcoma and murine P388 leukemia cell lines in the soft agarose colony formation assay for 24 h during exposure to the compounds. Twenty compounds with a range of in vitro activities were studied. Thirteen compounds exhibited cytotoxicity against A204 cells in culture; nine of them were less active when co-cultured with hepatocytes, two were activated by hepatocyte co-culture, and two showed no effect of hepatocyte co-culture. P388 cells were more sensitive to the antiproliferative effects of the compounds than A204 cells. Two compounds that were not active against A204 cells exhibited cytotoxicity against P388 cells. One compound was inactivated by hepatocyte co-culture and one showed no effect. Five compounds showed no cytotoxicity toward either A204 cells or P388 cells. Two of the compounds showing hepatocyte inactivation in vitro possess activity in one or more in vivo tumor models. Thus, evidence for metabolic inactivation in hepatocyte co-culture is not always an indication for lack of in vivo antitumor activity. Hepatocyte co-culture methodology provides a simple and objective means, amenable to large-scale screening, of distinguishing metabolic activation or inactivation of a given compound from other pharmacokinetic and pharmacodynamic factors with a minimum of material.