Background,, The treatment of metastatic melanoma remains unsatisfactory despite encouraging results with biotherapy and combination chemotherapy. Combining these two modalities may improve outcomes for such patients. Methods: Eligible patients had metastatic melanoma, were in good medical condition, and had not been treated previously for metastatic disease. A 42-day treatment cycle consisted of: tamoxifen 10 mg p.o. b.i.d. days 1-42, carmustine 150 mg/m(2) i.v. on day three, dacarbazine 220 mg/m(2) and cisplatin 25 mg/m2 i.v. q.d. days 3-5 and 24-26, interferon-alpha 2b 6.0 MU/m(2) s.c. days 8-12 and 29-33, and interleukin-2 11 MU/m(2) s.c. days 8 10,12 and 29, 31, 33. In the absence of tumor progression, patients could receive up to six cycles of alternating treatment. Toxicity and tumor response was assessed following each treatment cycle; survival was determined from the first date of treatment. Results: The 28 melanoma patients included 21 men and 7 women, with a median age of 55 years with a range of 26-77. Fifty-four percent were asymptomatic when treatment was initiated. Eighty percent had soft tissue metastases, 32% lung, 28% liver, and 8% bone. There were nine patients with significant tumor responses (three complete, six partial)for a response rate of 32% (18-57% 95% CI) based on intent-to-treat analysis, and 38% (18-57%, 95% Cl)for the 24 patients who were evaluable for response. The months of duration of survival for responders were 38.9+, 27.2+, 22.8+, 16.3, 13.2, 9.4 7.5 6.5+, and 5.8. At a median follow-up of 31 months, the median duration of event-free survival was 4.6 months; median survival was 9.4 months. The survival rate one year from initiating treatment was 42%; 2-year survival was 14%. The most frequent toxicities were 96% nausea/vomiting, 80% fatigue, 73% thrombocytopenia, 60% neutropenia, and 44% fever. Two patients experienced early death that may have been treatment related; one died of cardiovascular complications and the other of a central nervous system event. Conclusion: This outpatient regimen was associated with significant toxicity including a 7% rate of possible treatment-related death. Tumor regression rates and survival were similar to results reported for chemotherapy alone, or inpatient IL-2-based therapy, but did not suggest an improvement in outcome.