Background: This randomised phase III study investigated if in responsive and stable disease (SD) stage IV patients after two courses of cisplatin and gemcitabine, single-agent gemcitabine (experimental arm) was not inferior in terms of overall survival (OS) to cisplatin-gemcitabine (standard arm).
Patients and methods: Noninferiority was defined as an increase in the hazard of death (HR) < or = 1.33 in the experimental arm. From January 2001 to February 2004, 340 patients were registered and 250 were randomised. Cisplatin was administered on day 1 at 75 mg/m2 and Gemcitabine on days 1 and 8 at 1250 mg/m2 every 3 weeks.
Results: Response rate after two courses was 29%. The 1-year progression-free survival was 13% in both arms. One-year survival was 52% in the standard and 42% in the experimental arm for an HR of 1.21 [90% confidence interval (CI) 0.97-1.51]. Postprogression survival was in favour of the standard arm (HR 1.30, 95% CI 0.99-1.70, P = 0.051). Grades 3-4 toxicity favoured in the experimental arm.
Conclusion: In responsive and SD patients with stage IV non-small-cell lung cancer it was not possible to demonstrate that three courses of gemcitabine alone are not inferior, in terms of OS, to the standard approach of three courses of cisplatin-gemcitabine.