New conflicting results obtained in the subgroup analysis of RAD51 135G>C polymorphism and breast cancer risk

Abstract

We read with great interest the article ‘‘RAD51 135G[C polymorphism and breast cancer risk: a metaanalysis’’ published online in the July 2010 issue of ‘‘Breast Cancer Research And Treatment’’. Zhou et al. [1] have reached an important conclusion that RAD51 variant 135C homozygote is associated with elevated breast cancer risk among BRCA2 mutation carriers, but not in subgroup analysis according to specific ethnicity. Nevertheless, close inspection of the studies analyzed by the authors revealed some methodological issues that are worth mentioning and clarifying. After we adopt the same search strategy and end-of-search date as Zhou et al. [1], six relevant case–control studies in pubmed with a total number of 1,228 breast cancer cases and 1,127 controls [2–7] was not included in this meta-analysis. Of note, except for one study of mixed ethnicities [2], other five case–control studies were performed on Caucasian populations. Sub-analysis on Caucasian and Mixed subjects demonstrated no significant association between RAD51 135G[C polymorphism and breast cancer susceptibility by Zhou et al. [1]. This may imply that the original odds ratio for the eligible studies [2–7] not included in the sub-analysis on Caucasian and Mixed subjects may significantly differ from that calculated by Zhou et al. [1]. As a result, the conclusion by Zhou et al. [1] was not entirely credible. Meanwhile, meta-analysis based on currently available evidence, was to derive a more precise assessment of the relationship, but this research brought a new conflicting result in the subgroup analysis. Zhou et al. [1] demonstrated a significant association was found among BRCA2 mutation carriers in additive (OR = 4.92; 95% CI = 1.11–21.83) and the recessive models (OR = 4.88; 95% CI = 1.10–21.67), but not in BRCA1 mutation carriers in all genetic models. It was different from the meta-analysis result published in the July 17, 2010 issue of ‘‘Breast Cancer Research And Treatment’’ [8]. Jakubowska et al. [3] observed a significantly reduced risk of breast cancer among Polish female carriers of RAD51 135C allele and BRCA1 founder mutations. Gao et al. [8] demonstrated that no significant association was found in stratified analyses according to the BRCA1 or BRCA2 mutation status in the dominant comparison model tested. It can be secondarily noted that this significant association among BRCA2 mutation carriers by Zhou et al. [1] was just based on one study by Antoniou et al. [9]. We also found one relevant case–control study in pubmed, which have not been included in this meta-analysis [2]. The number of studies involved in the meta-analysis for RAD51 135G[C polymorphism among BRCA2 mutation carriers was relatively small, thus performing the subgroup analysis was met with difficulty. The study not included was P.-H. Lu (&) W. Shen M.-Y. Wu G.-Q. Tao (&) Department of General Surgery, Wuxi People’s Hospital of Nanjing Medical University, No. 299, Qingyang Road, Wuxi 214023, Jiangsu, China e-mail: lphty1_1@yahoo.com.cn