gene, as a tumor suppressor gene was involved in the development and progress of breast cancer (BC). However, the effect of polymorphisms on BC has rarely been reported. In the study, we aimed to evaluate the relation between variants and BC risk. Among 563 BC patients and 552 healthy controls, ten single-nucleotide polymorphisms (SNPs) in were genotyped by Agena MassARRAY. Odds ratios (ORs) and 95% confidence interval (CI) were calculated using logistic regression model. Our study demonstrated that rs1907168 polymorphism (heterozygous: OR = 0.71, = 0.017) was related to the reduced risk of BC in the overall. In stratified analyses by age, rs1907168 was associated with the decreased (heterozygous: OR = 0.53, = 0.002) while rs78205284 (homozygote: OR = 2.83, = 0.034) increased BC susceptibility among the population at age ≤51 years. Rs6551122 (recessive: OR = 0.51, = 0.028) and rs12635768 (homozygote, OR = 0.36, = 0.023) polymorphisms were related to the smaller BC tumor size (<2 cm). In addition, rs112276562 (heterozygote OR = 0.56, = 0.002), rs6551122 (heterozygote OR = 0.63, = 0.016), and rs73150416 (heterozygote OR = 0.57, = 0.005) variants contributed to the lower incidence of PR-positive BC. Moreover, rs6788033 was associated with a lower expression level of Ki-67 (log-additive: OR = 0.68, = 0.024). Furthermore, we found an association of 'GATT' haplotype with an increased risk for BC. In addition, gene was down-regulated in BC tumor and had a poor prognosis in BC in analysis. Our study firstly found SNPs contributed to the risk of BC, suggesting variants might help to predict BC progression.