Objective: This phase III study was aimed at evaluating whether the addition of gemcitabine (G) to vinorelbine (V) could improve the survival and quality of life (QoL) of elderly patients with advanced NSCLC.
Patients and methods: Patients with advanced NSCLC, aged >or=70 years, were randomly allocated to receive V 30 mg/m(2) on days 1 and 8 every 3 weeks or G 1200 mg/m(2) plus V 30 mg/m(2) on days 1 and 8 every 3 weeks. Survival was the main end point of the study. The estimated sample size was 120 patients per arm, but an interim analysis of survival was planned on the first 60 patients per arm.
Results: In May 1999, an interim analysis was performed with the survival data of the first 120 eligible patients (V(arm)=60, G+V(arm)=60). Forty-nine patients had stage IIIB disease and 71 patients stage IV disease, median potential follow-up of 14 months (range; 3-22), 93 patients had died (G+V(arm)=41, V(arm)=52). Median survival time (MST) was 29 weeks and projected 1-year survival was 30% in the G+V(arm); these values were 18 weeks and 13% in the V(arm). At multivariate Cox analysis, the risk of death in the G+V(arm) compared with V(arm) was 0.48 (95% C1=0.29-0.79; P<0.01). Combination therapy was also associated with a clear delay in symptom and QoL deterioration. The ORR was 22 and 15% in the G+V and V(arms), respectively. Toxicity was not irrelevant in both arms.
Conclusions: G+V treatment is associated with a significantly better survival than V alone in elderly NSCLC patients. The magnitude of the difference justifies the early closure of the study. The G+V regimen is now the SICOG reference regimen in this type of patients.