ObjectiveTo systemically evaluate the efficacy and safety of cyclooxygenase-2 (COX-2) signal pathway inhibition in treating advanced non-small cell lung cancer (NSCLC). MethodsA systematic literature search in PubMed, EMbase, Cochrane Library, ASCO databases, CNKI and Wanfang database was conducted to identify relevant randomized controlled trials (RCTs) from the time of database establishment to June 2015. RCTs of COX-2 inhibitors treating advanced NSCLC were included. We assessed the methodology quality of the included studies by using Jadad's scale, and performed this meta-analysis by using stata12.0 software. ResultsTwelve RCTs involving three different COX-2 inhibitors with a total of 1 828 patients were identified including 8 studies of high quality and 4 studies of low quality. We found that COX-2 signal pathway inhibition could significantly increase overall response rate at RR=1.27 with 95%CI1.10 to 1.46 (P=0.001). While our present data could not confirm the efficacy of COX-2 inhibitors in improving progression-free survival (PFS) at HR=0.93 with 95%CI0.81 to 1.08 (P=0.334), overall survival (OS) at HR=0.95 with 95%CI0.84 to 1.08 (P=0.461), or one-year survival rate at RR=1.08 with 95%CI0.90 to 1.24 (P=0.29). As for toxicities, only increased risk of thrombocytopenia at RR=1.28 with 95%CI 1.03 to 1.85 (P=0.03) was observed in the patients treated with COX-2 inhibitors. ConclusionCOX-2 signal pathway inhibition is effective in improving the overall response rate of the patients with advanced NSCLC, and is well tolerated. Whether COX-2 signal pathway inhibition is effective in improving long-term survival of the patients with advanced NSCLC still needs to be confirmed via further clinical trials.