Secondary hormonal manipulations are common following the failure of combined androgen blockade in patients with metastatic prostate cancer. Ketoconazole has been shown to have activity in this disease by inhibiting cytochrome P450 steroid hormone biosynthesis, thus inducing androgen deprivation. Gallium nitrate has been reported to target tumor tissue in vitro and some preliminary data suggests activity in patients with prostate cancer. Thus, we conducted a Phase II study of gallium nitrate in patients with androgen-independent prostate cancer. Two patients with progressive prostate cancer were removed from this study and subsequently placed on ketoconazole, as a palliative agent. Surprisingly, both of these patients had a greater than 50% decline in their prostate specific antigen (PSA) with this secondary endocrine maneuver. Based on this clinical observation, we conducted the following in vitro study to determine if there was a substantial additive effect of gallium nitrate followed by ketoconazole. Gallium nitrate or ketoconazole was added to the androgen-independent prostatic epithelial cell line, PC-3. One hundred and twenty hours (120 h) following the addition of one of the agents, the media was aspirated and the second agent was added to the wells. One plate was assayed every 24 h for cell viability using a non-isotopic cell proliferation assay kit. Cells treated with gallium nitrate followed by ketoconazole were 70-100% of control at the end of the gallium nitrate treatment; ketoconazole was then added and viability either remained constant or dropped steadily. Gallium nitrate by itself had a weak inhibitory effect on cell viability that only became apparent at the highest concentration evaluated. Ketoconazole, on the other hand, showed a substantial growth inhibition that was concentration-dependent. Cells treated with this agent alone showed a pronounced steady decrease in viability. Exposure to ketoconazole for 120 h followed by incubation in culture medium alone for 120 h caused a decrease in cell viability to 26.0% of control. Our in vitro results suggest that the combination of gallium nitrate and ketoconazole has no additive activity in the PC-3 cell line. Furthermore, this study confirms that ketoconazole added to prostate cancer cells has antiproliferative activity. The in vitro activity of ketoconazole has traditionally been thought to result from its inhibition of cytochrome P450-dependent enzymes responsible for steroidogenesis; however, an alternative hypothesis is necessary to explain the cytotoxic effect in the absence of adrenal and testicular androgen production as found in an in vitro system.