Smoking history as a predictive factor of treatment response in advanced non-small-cell lung cancer: a systematic review.

Recent trials in patients with advanced non-small-cell lung cancer (NSCLC) suggest that nonsmokers may benefit more from epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (TKI) therapy than will smokers. The aim of this systematic review was to assess smoking history as a predictive factor for treatment outcomes in patients with NSCLC. Relevant published literature was identified through systematic searches of databases (MEDLINE, EMBASE, Cochrane Library), oncology and thoracic journals, and abstracts from major oncology conferences using prespecified criteria. Articles reporting treatment outcomes (overall survival [OS], progression-free survival [PFS], and/or response rate) in smoking history subgroups from randomized controlled trials of targeted therapy and/or chemotherapy were reviewed. Data from 30 trials (32 articles, 4 abstracts) were included. Of these, 23 trials tested first-line therapy. Treatment arms included EGFR TKIs (13 trials), EGFR monoclonal antibodies (2 trials), non-EGFR targeted treatments (9 trials), chemotherapy (27 trials), and placebo or best supportive care only (3 trials). Smoking history definitions and analyses of its effect on treatment outcomes varied widely. Only 11 trials reported testing for a treatment-by-smoking history interaction. The available evidence supports but does not confirm smoking history as a predictive factor for the response to TKIs, particularly in previously treated patients. The evidence does not support smoking history as a predictor of response to non-EGFR-targeted therapies or chemotherapy. Smoking history and its effect on treatment response are inadequately reported. More rigorous collection, analysis, and reporting may clarify whether smoking history is a predictor of treatment response in advanced NSCLC.