Abstract:Objective To evaluate the association between Cyclin D1 G870A polymorphism and risk of colorectal cancer. Methods Extensive searches of relevant studies on datebase like PubMed, EMCC and CNKI were performed. Case control studies involving unrelated subjects and genotype frequencies in control group consistent with HardyWeinberg equilibrium were included for the metaanalysis. Twentythree casecontrol studies with 6 344 cases and 9 018 controls were analyzed by the fixedeffect or randomeffect metaanalysis method. The metaanalysis was applied with RevMan 5.0 software for heterogeneity test. AA and GA were compared with those with GG genotype. Pooled ORvalue and 95% confidence interval(CI) were calculated. Results There were statistical differences between AA & GA and GG. The pooled ORvalue was 1.10 (95%CI：1.01~1.19, P=0.02). The values were analyzed hierarchically according to different populations. The pooled OR value of Asian was 1.11（95%CI：0.98~1.26, P=0.11）. The pooled OR value of American was 1.13（95%CI：0.97~1.32, P=0.12）. The pooled ORvalue of European was 1.06（95%CI：0.89~1.25, P=0.52）. The pooled ORvalue of Oceanian was 1.05（95%CI：0.80~1.38, P=0.73）. The values were analyzed hierarchically according to the comparison basis. The pooled ORvalue of hospitalbased was 1.07（95%CI：0.95~1.20,P=0.28）. The pooled ORvalue of populationbased was 1.13（95%CI：1.01~1.26, P=0.04）. Conclusion The Cyclin D1 G870A polymorphism is correlated with the susceptibility of colorectal cancer risk at the aggregate level analysis. Analysis by different methods: according to different populations, every group don′t support the correlation. According to comparison basis, there has no correlation in hospitalbased group, but there has correlation in populationbased group.

Key words:Colorectal neoplasms,Cyclin D1,G870A polymorphism,Meta anaiysis