Impact of smoking status on the relative efficacy of the EGFR TKI/angiogenesis inhibitor combination therapy in advanced NSCLC-a systematic review and meta-analysis.

BACKGROUND:The ETOP 10-16 BOOSTER trial failed to demonstrate a progression-free survival (PFS) benefit for adding bevacizumab to osimertinib in second line. An exploratory subgroup analysis, however, suggested a PFS benefit of the combination in patients with a smoking history and prompted us to do this study. METHODS:A systematic review and meta-analysis to evaluate the differential effect of smoking status on the benefit of adding an angiogenesis inhibitor to epidermal growth factor receptor (EGFR)-tyrosine kinase inhibitor therapy was carried out. All relevant randomized controlled trials appearing in main oncology congresses or in PubMed as of 1 November 2021 were used according to the Preferred Reporting Items for Systematic Review and Meta-Analyses statement. Primarily PFS according to smoking status, and secondarily overall survival (OS) were of interest. Pooled and interaction hazard ratios (HRs) were estimated by fixed or random effects models, depending on the detected degree of heterogeneity. Bias was assessed using the revised Cochrane tool for randomized controlled trials (RoB 2). RESULTS:Information by smoking was available for 1291 patients for PFS (seven studies) and 678 patients for OS (four studies). The risk of bias was low for all studies. Combination treatment significantly prolonged PFS for smokers [n = 502, HR = 0.55, 95% confidence interval (CI): 0.44-0.69] but not for nonsmokers (n = 789, HR = 0.92, 95% CI: 0.66-1.27; treatment-by-smoking interaction P = 0.02). Similarly, a significant OS benefit was found for smokers (n = 271, HR = 0.66, 95% CI: 0.47-0.93) but not for nonsmokers (n = 407, HR = 1.07, 95% CI: 0.82-1.42; treatment-by-smoking interaction P = 0.03). CONCLUSION:In advanced EGFR-non-small-cell lung cancer patients, the addition of an angiogenesis inhibitor to EGFR-tyrosine kinase inhibitor therapy provides a statistically significant PFS and OS benefit in smokers, but not in non-smokers. The biological basis for this observation should be pursued and could determine whether this might be due to a specific co-mutational pattern produced by tobacco exposure.