A prospective randomized study was conducted in 51 patients with stage D hormone-resistant prostatic carcinoma, comparing a combination of doxorubicin and lomustine (DC) with cyclophosphamide and 5-FU (CF). Patients were assessed objectively (employing National Prostate Cancer Project criteria) and subjectively (using a numerical scoring scheme). Each regimen was well tolerated with acceptable levels of myelosuppression. The objective partial response rate was 57% for DC and 8% for CF. Objective stabilization occurred, respectively, in 14% and 44% of the patients. Similarly, DC demonstrated a significantly superior subjective response rate (partial plus complete) of 82%, compared to 48% for CF. Patients with poor initial performance status or liver involvement had significantly lower response rates and reduced survival. Overall, there was no significant difference in survival between the two arms, reflecting the similarity between DC and CF in total objective response rate (partial response plus stable disease). DC provided superior palliation and was well tolerated by an essentially geriatric population.