BACKGROUND. The combination of oral estramustine and oral etoposide has generated response rates of 40ยฑ50% in patients with hormone refractory prostate cancer in single institution trials. This study tested this regimen in a multi-institutional setting. METHODS. Fifty-ยฎve patients were accrued over a period of 4 months between 1 March 1996 and 1 July 1996. Two patients were not analyzable and two patients were ineligible. They were given an oral regimen consisting of estramustine 15 mg/kg/day (capped at 1120 mg per day) and etoposide 50 mg/M 2 /day, days 1ยฑ21 every 28 days. Patients received a median of two cycles of therapy. RESULTS. Toxicities included 11 patients (20%) with grades 3 or 4 granulocytopenia, 5 patients (10%) with grades 3 or 4 edema, and 3 patients (6%) with a thrombotic event. There were two treatment-related deaths, one as a result of anemia and the other as a result of a myocardial infarction. Of the 32 men who received at least 2 cycles of therapy, 7 men (22%) demonstrated a partial response to this regimen as measured by prostate-speciยฎc antigen (PSA) criteria of a 50% decline from pretreatment values. CONCLUSIONS. This trial demonstrates the toxicity of estramustine delivered in high dose. It also illustrates the difยฎculty of conducting phase II trials in prostate cancer in the cooperative group setting where the experience and comfort level of oncologists with new agents is less