Background: Results from previous studies concerning the association of ERCC4 rs1800067 polymorphism with risk of cancer were inconsistent. To explore the exact relation with susceptibility, we conducted the present meta-analysis. Materials and Methods: Literature of electronic databases including PubMed, Web of Science, EMBASE, Wanfang and Chinese National Knowledge Infrastructure (CNKI) were systematically searched. ORs and their 95%CIs were used to assess the strength of associations between ERCC4 polymorphism and cancer risk. Results: There was no significant association between ERCC4 rs1800067 AA or AG genotypes and overall risk of cancer (AA vs. GG: OR=0.998, 95%CI=0.670-1.486, P=0.992; AG vs. GG: OR=0.970, 95%CI=0.888-1.061, P=0.508). A dominant genetic model also did not demonstrate significant association of (AA+AG) genotype carriers with altered risk of overall cancer (OR=0.985, 95%CI=0.909-1.068, P=0.719). In addition, no significant association was observed between A allele of ERCC4 rs1800067 A/G polymorphism and altered cancer risk compared with G allele (OR=0.952, 95%CI=0.851-1.063, P=0.381). Subgroup analysis suggested that AA genotype carriers were significantly associated with decreased risk of glioma compared with wild-type GG genotype individuals (OR=0.523, 95%CI=0.275-0.993, P=0.048). For subgroup of lung cancer, A allele of ERCC4 rs1800067 A/G polymorphism was significantly associated with decreased risk of lung cancer compared with G allele (OR=0.806, 95%CI=0.697-0.931, P=0.003). Conclusions: This meta-analysis indicated that ERCC4 rs1800067 A/G polymorphism might not be associated with risk of overall cancer. However, individuals with the AA genotype were associated with significantly reduced risk of glioma compared with wild-type GG genotype; The A allele was associated with significantly reduced risk of lung cancer compared with G allele. Future large-scale studies performed in multiple populations are warranted to confirm our results.