ObjectiveGlasgow prognostic score (GPS), an inflammation-based scoring system, has been evaluated in various cancers. However, its clinical significance remains unclear in non-small cell lung cancer (NSCLC). Therefore, it is necessary to conduct a meta-analysis to explore the prognostic value of GPS in NSCLC patients.MethodsA quantitative meta-analysis was performed through a systematic search in PubMed, Web of Science, and the Cochrane Library. The pooled hazard ratios (HRs) of overall survival (OS) were calculated and compared.ResultsA total of 12 studies comprising 2669 patients were included in this meta-analysis. Compared with GPS 0 group, patients in GPS 1–2 group exhibited a reduced OS with a pooled HR of 1.89 (95 % CI 1.57–2.27, p < 0.001; I2 = 54 %). Six studies had sufficient data to calculate HRs of OS for GPS 1 and GPS 2 groups. Analysis revealed that GPS 2 group had a statistically significant reduced OS compared with GPS 1 group with a pooled HR of 1.87 (95 % CI 1.18–2.97, p = 0.008; I2 = 72 %). Study type (retrospective vs. prospective) and disease stage could partially explain the heterogeneity of each study by subgroup analysis.ConclusionPretreatment GPS could serve as a simple and reliable prognosis predictor for NSCLC. More well-designed studies that consider GPS as a stratification factor are warranted.