ANSELMA: ANtiangiogneic SEcond line Lung cancer Meta-Analysis in non-small cell lung cancer (NSCLC)

Abstract

Background: Only 20% of NSCLC patients have a response to immune checkpoint inhibitors as second‐line treatment. Several randomized phase II and III trials have tested the efficacy of the combination of antiangiogenic drugs (AD) plus chemotherapy (CT) or erlotinib (E) with inconsistent benefit. We performed a meta‐analysis to validate the efficacy of these agents as strategy second‐line option in advanced NSCLC patients. Methods: Randomized trials of AD plus standard second‐line treatment, CT (docetaxel [Do], pemetrexed) or E compared to same standard treatment ending accrual before 2015 were included based on search of publication databases, abstract proceedings and trial registers. Data were extracted from publications. Random‐effect models, in case of significant heterogeneity (Het), fixed‐effect model otherwise, were used to compute pooled hazard ratios (HRs) for overall survival (OS, primary end‐point) and PFS and pooled odds ratios (ORs) for response rate (RR) and adverse events. Het was studied using Q‐test I2. Results: seventeen trials with 8,703 patients were included with 3 types of combinations: 5 trials monoclonal antibodies AD + CT (Do for 5), 7 on tyrosine kinase inhibitor (TKI) AD + CT (Do for 3) and 5 AD (TKI for 4) + E. Trial size ranged from 100 to 1391 patients. ADs evaluated were: cabozantinib (1 trial), aflibercept (1), nintedanib (2), ramucirumab (2), sorafenib (2), bevacizumab (3), vandetanib (3), sunitinib (3). Control arm was docetaxel (8 trials), pemetrexed (4), E (5). Compared with standard second‐line treatment alone, AD significantly prolonged OS (HR 0.94 [95% confidence interval 0.89‐0.99, random‐effect model; P = 0.03; P‐Het = 0.004), PFS (0.79; [0.73‐ 0.85], random effect model; P < 0.0001; P‐Het = 0.003), and improved RR (OR: 1.86 [1.63‐2.12], fixed‐effect model; P < 0.0001, P‐Het = 0.30). There was no difference in the benefit in OS, PFS and RR between the 3 types of combination tested. Conclusions: Antiangiogenic drugs significantly prolong OS and PFS when added to standard second‐line treatment in advanced NSCLC patients. Toxicity results will be presented during the congress.