This is a Southwest Oncology Group (SWOG) prospective randomized trial of cisplatin, vinblastine, and bleomycin (PVB) versus vinblastine, cisplatin, and etoposide (VP-16) (VPV) in the treatment of advanced germ cell tumors of the testis. The study objective was to determine what effect the replacement of bleomycin with VP-16 has on complete response (CR), survival, and drug toxicity. One hundred sixty-nine patients were registered and randomized. Of these patients, 160 were assessable for response. All had histologically confirmed disseminated germ cell neoplasms of testicular origin. Forty-six had minimal metastatic disease, and 114 had maximal disease. Seventy-seven were randomized to PVB and 83 to VPV chemotherapy. There was no significant difference in pretreatment characteristics between the two arms with regard to tumor burden, histologic type, and overall performance status. Patients received four courses of induction chemotherapy, either PVB (cisplatin 120 mg/m2 day 3, vinblastine 12 mg/m2 day 1, bleomycin 15 U/m2 twice per week) or VPV (vinblastine 8 mg/m2 day 1, cisplatin 120 mg/m2 day 3, VP-16 50 mg/m2 days 2 to 5). Chemotherapy was given every 3 weeks. Cytoreductive surgery was done postinduction if a chemotherapy CR was not achieved. There was no difference in the percentage of patients achieving a disease-free status between PVB (77%) and VPV (73%). The mean leukocyte nadir was similar for both treatments, but the mean platelet nadir was significantly lower (P = .003) in the VPV arm. All bleomycin-related toxicities (pulmonary, mucositis, skin) were avoided in the VPV arm. We conclude that bleomycin can be replaced in first-line therapy for advanced germ cell tumors without sacrificing efficacy and with the advantage of avoiding unnecessary drug toxicity.