Effect of smoking habits on the efficacy of EGFR-TKI plus anti-angiogenic agent in advanced EGFR-mutant NSCLC

The types of epidermal growth factor receptor (EGFR)-mutant non-small-cell lung cancer (NSCLC) patients who could obtain significant clinical benefit from the dual inhibition of EGFR/vascular EGFR (VEGFR) pathways remain unclear. No consensus has been reached on the significance of smoking habits in clinical benefit obtained from EGFR-TKI plus anti-angiogenic agents.PubMed, EMBASE, and Cochrane databases for all phase II/III randomized clinical trials (RCTs) investigating the efficacy of EGFR-TKI combined with anti-angiogenic agents stratified by smoking habits (updated October 2021) were searched systematically. The primary outcomes were the pooled HRs for PFS/OS in smokers and non-smokers, and differences in efficacy of EGFR-TKI plus anti-angiogenic treatment between smokers and non-smokers, measured by difference in PFS and OS.Seven phase II/III RCTs involving 1452 patients were identified. The pooled analysis demonstrated that EGFR-TKI plus anti-angiogenic agent could decrease the risk of progression by 40% (HR, 0.60; 95%CI 0.48-0.75) in smokers when compared with EGFR-TKI alone, but not in non-smokers (HR, 0.92; 95%CI 0.68-1.25). The comparison analysis further demonstrated that EGFR-mutated NSCLC patients who smoked obtained greater progression-free survival (PFS) benefit from treatment with EGFR-TKI plus anti-angiogenic agents (HR, 0.68; 95%CI 0.51-0.91). Consistent with the results for PFS, smokers receiving EGFR-TKI plus anti-angiogenic agents appeared to exhibit better overall survival (OS) than non-smokers but not to a statistically significant degree (HR, 0.60; 95%CI 0.23-1.52). Meta-regression analysis revealed no significant effect of the line of treatment (P = 0.52), trial phase (P = 0.52), EGFR-TKI type (P = 0.13), or anti-angiogenic agent type (P = 0.50) on PFS effect sizes under multivariate models.Comprehensive analysis suggested that EGFR-TKI plus anti-angiogenic agents led to favorable PFS among smoking EGFR-mutant patients, comparable to nonsmokers, which might provide a useful guide for clinicians.