We conducted a randomized trial comparing a high (120 mg/m2 day 1) v a standard (60 mg/m2 day 1) dose of cisplatin in combination with etoposide (120 mg/m2 days 3, 5, and 7) in advanced non-small-cell lung carcinoma (NSCLC). Two hundred forty-one patients were evaluable for survival and 207 for response. We obtained a 25% objective response rate in the standard-dose arm and 29% in the high-dose arm; this difference was not statistically significant. There was no significant improvement in the overall survival or survival of responders with the high-dose regimen. However, toxicity (mainly myelosuppression) was significantly increased in the patients receiving the higher dose of cisplatin. An analysis of prognostic factors showed that disease progression, loss of body weight, performance status, and prior therapy were predictive parameters of survival.