A meta-analysis of the association between ESR1 genetic variants and the risk of breast cancer

Abstract

Background Single nucleotide polymorphisms (SNPs) in the estrogen receptor gene (ESR1) play critical roles in breast cancer (BC) susceptibility. Genome-wide association studies have reported that SNPs in ESR1 are associated with BC susceptibility; however, the results of recent studies have been inconsistent. Therefore, we performed this meta-analysis to obtain more accurate and credible results. Methods We pooled published literature from PubMed, EMBASE, and Web of Science and calculated odds ratios (ORs) with 95% confidence intervals (CIs) to assess the strength of associations using fixed effects models and random effects models. Twenty relevant case-control and cohort studies of the 3 related SNPs were identified. Results Three SNPs of the ESR1 gene, rs2077647:T>C, rs2228480:G>A and rs3798577:T>C, were not associated with increased BC risk in our overall meta-analysis. Stratified analysis by ethnicity showed that in Caucasians, the rs2228480 AA genotype was associated with a 26% decreased risk of BC compared with the GG genotype (OR = 0.740, 95% CI: 0.555- 0.987). The C allele of the rs3798577:T>C variant was associated with decreased BC risk in Asians (OR = 0.828, 95% CI: 0.730-0.939), while Caucasians with this allele were found to experience significantly increased BC risk (OR = 1.551, 95% CI: 1.037-2.321). A non-significant association between rs2077647 and BC risk was identified in all of the evaluated ethnic populations. Conclusion Rs3798577 was associated with an increased risk of BC in Caucasian populations but a decreased risk in Asians. Rs2228480 had a large protective effect in Caucasians, while rs2077647 was not associated with BC risk.