OBJECTIVES: The aim of the study was to assess the effect of an immunotherapy for the treatment of advanced non–small-cell lung cancer (NSCLC). METHODS: Data from a phase III, multicenter, randomized, open-label trial evaluating the efficacy of one EGF-based cancer vaccine (CIMAvaxEGF) as switch maintenance in patients with advanced NSCLC, were used. Survival analysis using Kaplan Meier estimates was performed. Weighted log-rank was conducted to assess the later effect of the immunotherapy. Additionally, a finite mixture model to the primary endpoint (Overall Survival, OS) was fitted. Weibull distribution was assumed for the overall survival and a mixture model consists of one, two or three components was fitted.. All analysis was conducted using the NLMIXED procedure in SAS. The Akaike Information Criterion (AIC) was used for model selection. RESULTS: Intention-to-treat (ITT) analysis showed 1.44 months of OS benefit for vaccinated patients with confirmed delayed-separation phenomena (OS: Vaccine arm, 10.37 months vs. Control arm, 8.93 months; p= 0.043). The mixture model with the best goodness to fit to the data consists of two components (AIC=3097.7). The two mixture components represent short-term and long-term survival subpopulations. The proportions of the subpopulations are estimated to be equal to 0.89 and 0.11 respectively. The median OS was estimated to be equal to 9.59 and 60.32 for short- and long- term survival populations, respectively. 23.26% and 76.74% from the patients who were classified into long-term survival subpopulation were from the control and vaccinated group, respectively. From them, 7 (70%) patients in the control group and 25 (75%) patients in the treated group were still alive at the end of the study. CONCLUSIONS: The results confirm that the vaccination with CIMAvaxEGF prolongs the survival of the advanced NSCLC patients. Mixture models allow assessing the efficacy/effectiveness of vaccines and biological products in the presence of heterogeneous populations.