ABSTRACT: OBJECTIVE:To evaluate the efficacy and safety of gemcitabine (GEM) at 30 min standard-dose infusion (30 min-SDI) compared with prolonged low-dose infusion (P-LDI) in patients with advanced non-small-cell lung cancer (NSCLC). METHODS:Electronic databases including Pubmed, EMbase, Cochrane Library, CNKI, CBM, and VIP were searched using keywords "GEM", "P-LDI", and "NSCLC". Review Manager 5.3 was used to perform the meta-analysis. Primary endpoints were overall response rate (ORR) and 1-year survival rate (1-year SR). Secondary endpoints were grade 3/4 hematotoxicity and nausea/vomiting. In association. GRADE quality of evidence system was used to assess the results of meta-analysis. RESULTS:Six randomized controlled trials (RCTs) with a total of 637 patients were included and no statistical heterogeneity was found among the studies. The results showed that P-LDI was superior in ORR (RD = 0.09, 95% CI: 0.02 to 0.16, P = 0.02), but had a similar 1-year SR (RD = 0.05, 95% CI: -0.02 to 0.12, P = 0.18) as compared with 30 min-SDI. For grade 3/4 adverse events, there was no significant difference in anemia (RD = 0.02, 95% CI: -0.01 to 0.04, P = 0.27) and nausea/vomiting (RD = 0.01, 95% CI: -0.04 to 0.06, P = 0.64) between the two treatments. However, patients with P-LDI experienced less leukopenia (RD = -0.08, 95% CI: -0.15 to -0.01, P = 0.03) and thrombocytopenia ((RD = -0.05, 95% CI: -0.09 to -0.01, P = 0.006). The GRADE profile showed that the included RCTs had low quality of evidences. CONCLUSION:P-LDI was superior in terms of ORR, experienced less grade 3/4 thrombocytopenia and leukopenia compared with 30 min-SDI, and could be a viable treatment option for advanced NSCLC. However, the results need to be further verified by high quality trials and large samples owing to the low quality of evidences.