Satraplatin is a novel oral platinum (IV) complex that shows activity against hormone-refractory prostate cancer (HRPC) in cisplatin-resistant human tumor lines in phase I and phase II trials. A randomized multicenter phase III trial with a target sample size of 380 patients was initiated in men with HRPC. After 50 randomized patients, the trial was closed to further accrual by the sponsoring company. An ad hoc analysis of all available data is reported here. Eligibility criteria included pathological proof of prostate cancer, documented progression despite prior hormonal manipulation, WHO PS 0-2, and no daily intake of narcotic analgesics. Patients were randomized between satraplatin 100 mg/m(2) for 5 days plus prednisone 10 mg orally BID or prednisone alone. Compliance was excellent. 48/50 patients have progressed and 42 have died, mostly due to prostate cancer. Median overall survival was 14.9 months (95% CI: 13.7-28.4) on the satraplatin plus prednisone arm and 11.9 months (95% CI: 8.4-23.1) on prednisone alone (hazard ratio, HR = 0.84, 95% CI: 0.46-1.55). A >50% decrease in prostrate specific antigen (PSA) was seen in 9/27 (33.3%) in the satraplatin plus prednisone arm vs. 2/23 (8.7%) on the prednisone alone arm. Progression-free survival was 5.2 months (95% CI: 2.8-13.7) on the satraplatin plus prednisone arm as compared to 2.5 months (95% CI: 2.1- 4.7) on the prednisone alone arm (HR = 0.50, 95% CI: 0.28-0.92). This difference is statistically significant (p = 0.023). Toxicity was generally minimal in both arms. This randomized comparison of a combination of satraplatin and prednisone versus prednisone alone supports the antitumor activity of the combination. Its role in the treatment of HPRC remains to be elucidated in an appropriate phase III setting.