Overall survival in PEMVITASTART randomized trial comparing immediate vs. conventional strategies of vitamin supplementation in NSCLC patients on 1st line pemetrexed-platinum chemotherapy.

Abstract

e20538 Background: PEMVITASTART ( NCT02679443 ) was an open label phase II randomized trial [Singh N, et al. Cancer (DOI: 10.1002/cncr.32028)] involving locally advanced/metastatic non-squamous NSCLC patients treated with 1st line pemetrexed-platinum chemotherapy (CTx). The trial assessed the hematological toxicity profile in Delayed Arm (DA; conventional strategy) in whom CTx was initiated after 5-7 days of vitamin B12/folate supplementation (B12-FAS) vs. Immediate Arm (IA) in whom patients received B12-FAS simultaneously (≤24 hours) with CTx initiation. All outcomes were reported in modified ITT population (patients who received ≥1 cycle). A non-significant trend towards better radiological responses was observed in IA (PR 33% vs. 18%; p = 0.06). Herein, we report exploratory post-hoc analysis of overall survival (OS) for the mITT population (n = 150; 77 IA, 73 DA) of PEMVITASTART. Methods: OS was calculated from date of enrolment to date of death/last follow up. Survival cutoff date was 28 months after last patient enrolment. Median OS was calculated by Kaplan-Meier method and group differences analyzed by log-rank test. Factors affecting OS were assessed by Cox proportional hazards (CPH) analysis and hazard ratios [HRs] with 95% confidence intervals (CIs) calculated (from univariate and stepwise multivariate models). Results: Median OS was 12.7 m (95% CI 8.0–17.4) and did not differ between IA [15.0 m (10.5-19.4)] and DA [11.7 m (4.1-19.3)]. 1 yr and 2 yr survival rates were similar (IA 42% and 21% vs. DA 44% and 23%). On univariate CPH analysis, factors associated with better OS were female gender, ECOG PS 0, absence of metastatic disease and receipt of maintenance pemetrexed CTx (mPEM) while current/ex-smoker status, disease progression (as best response) and age ≥70 years were a/w worse OS. Baseline Hb and homocysteine (both assessed as continuous variables) and receipt of packed RBC transfusions/ESAs for anemia correction did not influence OS. On multivariate CPH analysis, current/ex-smokers HR 2.2 (95% CI 1.3-3.7; p < 0.01) and mPEM HR 0.5 (95% CI 0.3-0.9; p = 0.01) were both significant. Median OS for current/ex-smokers was 7.4m (95% CI 5.4-9.4) vs. 25.6m (95% CI 16.2-35.0) for non-smokers (log rank p < 0.001) while for patients receiving mPEM, it was 22.7m (95% CI 9.9-35.4) vs. 9.7m (95% CI 4.9-14.5) for those not receiving mPEM (log rank p = 0.02). Conclusions: Smoking status and mPEM were strong and independent prognostic factors for OS in the PEMVITASTART trial. Clinical trial information: NCT02679443.