The association between cyclooxygenase-2 (COX-2) -1195G>A (rs689466) polymorphism and cancer risk has been extensively explored. However, the results of previous studies remain controversial. To address this gap, we performed an updated meta-analysis of fifty-eight studies involving a total of 50,672 subjects. Searching of PubMed and Embase databases was performed for publications on the association between COX-2 -1195G>A polymorphism and the risk of cancer. Statistical correlation was identified between COX-2 -1195G>A variants and overall cancer risk in five genetic models. In a sub-group analysis based on cancer type, significant association between COX-2 -1195G>A polymorphism and increased risk of gastric cancer, pancreatic cancer, hepatocellular carcinoma and other cancers was found. In a sub-group analysis by ethnicity, increased cancer risk was observed among Asians instead of Caucasians, Africans and mixed populations. Furthermore, in a sub-group analysis based on cancer system, increased cancer risk was found in digestive system cancer and other system cancer. Non-parametric "trim-and-fill" method was harnessed as a sensitivity analysis method and the results suggested our findings reliable. In summary, the results of our meta-analysis highlight that COX-2 -1195G>A polymorphism may be a risk factor for cancer.
Keywords: Cancer; cyclooxygenase-2; gene polymorphism; meta-analysis.