However, the present meta-analysis indicates that neither Arg nor Pro carriers may have a significant association with breast cancer risk. It is likely that TP53 codon 72 polymorphisms rarely affect the tumorigenesis and progression of breast carcinoma. Considering that the same polymorphism may play different roles in cancer susceptibility among different ethnic populations and the CBL-0137 frequencies of single nucleotide polymorphisms may be different ethnicity, we stratified the data by race into three groups concerning Asians, Caucasians or Africans, respectively. Ultimately, statistically similar results were obtained,

confirming nonassociation of TP53 codon 72 polymorphism with breast cancer risk. A well-known

risk factor, HPV infection, is thought to have an association P5091 solubility dmso with SB-715992 increased susceptibility to some cancers such as cervical [70] and oral cancer [71]. Evidence suggests that P53Arg72 protein may be more susceptible than P53Pro72 protein to HPV mediated degradation, thus increasing risk of HPV associated cancers [17]. Growing body of literature indicates HPV infection as a possible risk factor for breast cancer [72]. However, we did not further investigate the possible association of HPV infection with TP53 codon 72 polymorphism due to the insufficient data in the primary included studies. Heterogeneity is a potential problem when interpreting the results of meta-analysis [73]. In the present study, significant between-study heterogeneity existed in overall comparisons. Tobramycin Nevertheless, when the data were stratified by race, the heterogeneity was decreased or removed, suggesting that differences of genetic backgrounds and the environment existed among different ethnicities. In the present meta-analysis, we excluded the studies in which the control groups were deviate from HWE. Thus, the between-study heterogeneity might be reduced. Moreover, random-effect models

were used for combination of the data. Accordingly, the results may be credible and stable although the heterogeneity seemed evident. Some limitations might be included in this study. First, in this meta-analysis, most published studies and papers written in English or Chinese were searched. Moreover, although papers written in some other languages, cited by PubMed, were also searched, it is possible that some related published or unpublished studies that might meet the inclusion criteria were missed. Hence, some inevitable publication biases might exist in the results, though the Nfs0.05 showed no remarkable publication biases in the meta-analyses. Second, in the subgroup analysis, the number of studies regarding Africans was relatively limited. It may be underpowered to explore the real association. Thus, the results may be interpreted with caution.