This study aims to investigate the efficacy of targeted therapies in the treatment of non-small cell lung cancer (NSCLC) by using a network meta-analysis of clinical trials. PubMed, EMBASE, Cochrane Library, and Clinicaltrials.gov were searched by using keywords related to the topic on 19 September 2018. Two investigators independently selected relevant trials by pre-determined criteria. A pooled response ratio (RR) for overall response rate (ORR) and a hazard ratio (HR) for progression-free survival (PFS) were calculated based on both the Bayesian and frequentist approaches. A total of 128 clinical trials with 39,501 participants were included in the final analysis of 14 therapeutic groups. Compared with chemotherapy, both ORR and PFS were significantly improved for afatinib, alectinib, and crizotinib, while only PFS was significantly improved for cabozantinib, ceritinib, gefitinib, and osimertinib. Consistency was observed between the direct and indirect comparisons based on the Bayesian approach statistically and the frequentist approach visually. Cabozantinib and alectinib showed the highest probability for the first-line treatment ranking in ORR (62.5%) and PFS (87.5%), respectively. The current network meta-analysis showed the comprehensive evidence-based comparative efficacy of different types of targeted therapies, which would help clinicians use targeted therapies in clinical practice.
Keywords: network meta-analysis; non-small cell lung cancer; targeted therapy.