A polyvalent melanoma vaccine prepared from shed antigens stimulates humoral and cellular immune responses and improves survival compared with historical controls. We conducted a double-blind, prospectively randomized, placebo-controlled trial to assess whether this vaccine could slow the progression of resected melanoma. Thirty-eight patients with resected melanoma metastatic to regional nodes (American Joint Committee on Cancer stage III) who had a particularly poor prognosis on the basis of the nodes being clinically positive or two or more histologically positive nodes were randomly assigned in a 2:1 ratio to treatment with 40 microg of melanoma or placebo (human albumin) vaccine, both of which were bound to alum as an adjuvant. Immunizations were given intradermally into the extremities every 3 weeks x 4, monthly x 3, every 3 months x 2, and then every 6 months for 5 years or until disease progression. Twenty-four patients were treated with the melanoma, and 14 patients were treated with the placebo vaccine. The groups were evenly balanced with respect to prognostic factors. Median length of observation was 2.5 years. There was no local or systemic toxicity. By Kaplan-Meier analysis, median time to disease progression was two and a half times longer in patients treated with melanoma vaccine compared with that in patients treated with placebo vaccine, i.e., 1.6 years (95% confidence interval, 1.0-3.0 years) compared with 0.6 year [95% confidence interval, 0.3-1.9 year(s)]. By Cox proportional hazards analysis, this difference was significant at P = 0.03. Overall survival was 40% longer in the melanoma vaccine-treated group (median overall survival of 3.8 years versus 2.7 years), but this difference was not statistically significant. In a double-blind and placebo-controlled trial, these results suggest that immunization with a melanoma vaccine may be able to slow the progression of melanoma. Although statistically significant, these results must be interpreted with caution because they are based on a small number of patients.