Abstract

Objectives: A number of studies have evaluated the association of cyclooxygenase-2 (COX-2) polymorphisms with prostate cancer risk. However, the results still remain controversial. Methods: We carried out this meta-analysis to clarify the association of COX-2 polymorphisms and prostate cancer risk. An universal search in PubMed, Web of Knowledge and Google Scholar was performed to identify relevant studies up to January 2021. A total of 34 case-control studies including 11 studies with 13,248 cases and 14,768 controls were on -765G>C, 7 studies with 9,720 cases and 10,695 controls on -1195G>A, 9 studies with 11,476 cases and 11,761 controls on +202C>T, and 7 studies with 12,220 cases and 12,496 controls were on +8473T>C were selected. Results: Pooled data showed that the COX-2 +202C>T polymorphism (T vs. C: OR= 1.305, 95 CI: 1.849-9.490; p= 0.001; TT+TC vs. CC: OR= 0.781, 95 CI: 0.669-0.913; p=0.002) was associated with risk of prostate cancer, but not the -765G>C, -1195G>A and +8473T>C polymorphisms. Stratified analyses showed that the -765G>C, -1195G>A and +202C>T polymorphisms were associated with prostate cancer risk by ethnicity. Conclusion: To sum up, our results indicated that the COX-2 +202C>T polymorphism was associated with risk of prostate cancer, while the -765G>C, -1195G>A and +8473T>C polymorphisms were not associated.