Purpose: To evaluate whether cisplatin-based chemotherapy (gemcitabine, vinorelbine, and cisplatin [GVP]) prolongs overall survival in comparison to cisplatin-free chemotherapy (gemcitabine and vinorelbine [GV]) as first-line treatment in patients with advanced non-small-cell lung cancer (NSCLC). Patients and Methods: Between September 1999 and June 2001, 300 patients with NSCLC stage IIIB with malignant pleural effusion or stage IV disease were randomly assigned to receive GV (gemcitabine 1000 mg/m2 + vinorelbine 25 mg/m2 on days 1 and 8 every 3 weeks) or GVP (gemcitabine 1000 mg/m2 + vinorelbine 25 mg/m2 on days 1 and 8 + cisplatin 75 mg/m2 on day 2 every 3 weeks). Primary end point of the study was overall survival. Results: Two hundred eighty-seven patients (GV, 143 patients; GVP, 144 patients) were eligible for analysis. At the time of analysis, April 15, 2002, 209 patients (GV, 103 patients; GVP, 106 patients) of 287 patients had died (73%). No statistically significant difference was observed for overall survival (P = .73; median survival, 35.9 versus 32.4 weeks; 1-year survival rate, 33.6% versus 27.5%) as well as for event-free survival (P = .35; median time-to-event, 19.3 versus 22.3 weeks) between GV and GVP. Two hundred fourteen patients were assessable for best response. The overall response rates were 13.0% for GV versus 28.3% for GVP (P = .004; complete responders, 0% versus 3.8%; partial responders, 13.0% versus 24.5%). Hematologic and nonhematologic toxicity was significantly lower in the GV treatment arm compared with GVP. No statistically significant difference in quality of life was observed. Conclusion: In this phase III study, the cisplatin-based GVP regimen showed no survival benefit as first-line chemotherapy in advanced NSCLC when compared with the cisplatin-free GV regimen, which was substantially better tolerated.