A randomized, prospective trial of adjuvant chemoimmunotherapy was compared in a group of 136 patients with melanoma after complete surgical resection of advanced regional metastasis (stage III) or isolated distant metastasis (stage IV). All patients received a 2-year course of nonspecific adjuvant immunotherapy of subcutaneous injections of Corynebacterium parvum (4 mg/m2 divided among the four extremities in 1- to 2-week cycles). Half of the patients also received a 6-month course of dimethyl triazeno imidazole carboxamide (DTIC) plus cyclophosphamide chemotherapy (each at 600 mg/m2 administered intravenously every 3 weeks for nine cycles) while the other half received no chemotherapy. Analysis of the data showed that adjuvant chemotherapy did not provide any demonstrable therapeutic effect, either in terms of disease-free survival or overall survival. No benefit was observed in subgroups of patients categorized by disease stage, site of metastasis, or sex. The drugs did cause a substantial rate of morbidity, however, since 42% of the patients had severe nausea and vomiting, while 13% had hair loss. The C. parvum was well tolerated in almost all patients. This is a particularly high-risk group for subsequent metastases since only 24% of the patients were free of disease for 1 year and 12% after 2 years. Adjuvant DTIC and cyclophosphamide had no observable therapeutic effect on this population of high-risk patients with melanoma.