A meta-analysis was conducted to assess the effect of glutathione peroxidase1 (GPX1) gene Pro200Leu (rs1050450) polymorphism on cancer risk. A comprehensive search was performed to identify all studies on the association of GPX1 gene Pro200Leu polymorphism with cancer risk. The fixed or random effect pooled measure was selected based on homogeneity test among studies. Heterogeneity among studies was evaluated using the I (2). Potential sources of between-study heterogeneity were explored by meta-regression and the sensitivity analysis. Publication bias was estimated using Egger's linear regression test. 35 published articles with 36 results were identified involving 16,920 cases and 19,946 controls. Results from the articles that both obeyed Hardy-Weinberg equilibrium in controls and met high quality design, showed no significant association of GPX1 gene Pro200Leu polymorphism with cancer risk in any of dominant (OR = 1.05, 95 %CI = 0.98-1.12), recessive (OR = 1.04 (0.95-1.13), and TT versus CC (OR = 1.05, 95 %CI = 0.97-1.15) models, and the findings were consistent considering the stratified analysis by cancer type. However, multivariate-adjusted ORs from articles that both obeyed Hardy-Weinberg equilibrium in controls and met high quality design, showed a significant association considering dominant (OR = 1.22, 95 %CI = 1.06-1.41), TT versus CC (OR = 1.16, 95 %CI = 1.02-1.32) models, and a marginally significant association was found considering TC versus CC (OR = 1.11, 95 %CI = 0.99-1.25) model. And compared with the CC genotype, the erythrocyte GPX activity was significantly lower for TT genotype: the standardized mean difference (SMD) = -0.37, 95 %CI = (-0.624, -0.118), and CT genotype: SMD = -0.19, 95 %CI = (-0.37, -0.002). The association of GPX1 gene Pro200Leu polymorphism with cancer risk might be influenced by confounders.