Cisplatin combined with cyclophosphamide has been considered a very active treatment for advanced ovarian cancer. Unfortunately, cisplatin is associated with dose-limiting neurotoxicity, as well as possible neuropathy, ototoxicity, and occasional renal dysfunction. Carboplatin, a cisplatin analogue, is active against advanced ovarian cancer, with a presumed lower incidence of emesis, ototoxicity, neuropathy, and renal dysfunction. The Southwest Oncology Group initiated a phase III randomized trial, in which 342 patients with stage III (suboptimal disease) and stage IV ovarian cancer were randomly assigned to treatment with six courses of intravenous cisplatin 100 mg/m2 plus cyclophosphamide 600 mg/m2 or carboplatin 300 mg/m2 plus cyclophosphamide 600 mg/m2. The median survival for the cisplatin arm was 17.4 months; for the carboplatin arm, median survival was 20.0 months. The null hypothesis of a 30% survival superiority with the cisplatin arm was rejected at the p = 0.02 level. Clinical response rates were 52% for the cisplatin arm and 61% for the carboplatin arm. There was less thrombocytopenia in the cisplatin arm (p < 0.001); however, there was less nausea and emesis (p < 0.001 for courses one to five), renal toxicity (p < 0.001), anemia (p < 0.001), hearing loss (p < 0.001), and neuromuscular toxicity (p < 0.001) in the carboplatin arm. Carboplatin/cyclophosphamide proved to have a significantly better therapeutic index than cisplatin/cyclophosphamide in these patients with advanced ovarian cancer.