BACKGROUND: On the basis of preliminary results achieved with a high-dose cisplatin regimen including glutathione as chemoprotector, the efficacy and toxicity of the new regimen was further evaluated in a larger series of patients with advanced ovarian cancer (stage III and IV). PATIENTS AND METHODS: The study included patients with bulky or extensive residual disease after primary laparotomy or with bulky inoperable tumor masses. A total of 79 patients were treated with up to five courses of high-dose cisplatin (40 mg/m2 daily in normal saline, for four days) plus glutathione (2500 mg as a short-term infusion before cisplatin), together with cyclophosphamide (600 mg/m2 as an i.v. bolus on day 4). A standard i.v. hydration consisting of a total of 2000 ml of fluids without diuretics was employed. RESULTS: All eligible patients, who received a total of 345 courses, were assessed for response and toxicity and 52 received the planned five courses of the protocol. Forty-five patients (57%) achieved complete clinical responses and 20 (25%) had partial remissions for an overall response rate of 82%. The response rate was critically dependent on tumor size before chemotherapy. Thirty-eight of 45 patients who had complete clinical responses underwent second-look laparotomy, and 29 had pathological complete responses (37%). Seventeen of these 29 patients subsequently relapsed (median disease-free interval, 12 months; range, 6-45). With a median follow-up time of 44 months, the median survival for the 79 analyzed cases was 40 months. The toxicity of the regimen was moderate. Nausea/vomiting was the most severe acute toxicity. Myelotoxicity was acceptable, with severe leukopenia and thrombocytopenia (grade 4) occurring in 8% and 3% of patients, respectively. Nephrotoxicity was minimal with a transient increase (to < 2 mg/dL) in serum creatinine in only 6 patients (8%). Peripheral neurotoxicity and ototoxicity were the most significant long-term toxicities. The severity of these side effects (grade 3 WHO neurotoxicity occurred in only 4% of patients) was apparently less than has been reported with other high-dose cisplatin regimens. Neurotoxicity required discontinuation of therapy in three patients after four courses. Most affected patients had complete or partial recovery of symptoms with time. DISCUSSION: The efficacy and tolerability of the regimen confirm the feasibility of this new approach including glutathione in order to increase cisplatin dose intensity. The superiority of this regimen over standard induction therapy should be confirmed in randomized trials.