Association of single nucleotide polymorphism rs3803662 with the risk of breast cancer

Large scale association studies have identified the single nucleotide polymorphism rs3803662 associated with breast cancer risk. However, the sample size of most studies is too small. Here, we performed this meta-analysis to make the result more convincing. Relevant articles published up to 2016 were identified by searching the PubMed database. 13 studies, involving a total of 29405 participants, were included in the meta-analysis. Odds Ratios (ORs) with 95% confidence intervals (CIs) was calculated with random or fixed effects model. All data analyses were analyzed by Review Manger 5.3 software. In Caucasian subgroup: Dominant model (TT + CT vs CC): OR = 1.17 (1.06, 1.29), Recessive model (TT vs CT + CC): OR = 1.25 (1.13, 1.39) and Allele frequency (T vs C): OR = 1.15 (1.08, 1.22). The present meta-analysis suggests that rs3803662 polymorphism is significantly associated with breast cancer risk in Caucasian women, and we did not find the association in Asian women.

Scientific RepoRts | 6:29008 | DOI: 10.1038/srep29008 between rs3803662 and breast cancer risk or have sufficient data to calculate the ORs with 95% CIs. Afterwards, 4 articleswere excluded due to that the control group doesn't meet the Hardy-Weinberg Equilibrium (HWE), and 1 article was excluded because that the study population is male, not female. Study characteristics. The primary characteristics of the 13 studies are summarized in Table 1. A total of 29405 participants with 14306 cases and 15099 controls were included in this study. The studies were divided into two subgroups according to the ethnicity of their participants: East-Asian subgroup with 4 studies and Caucasian subgroup with 8 studies, and others subgroup with 1 study.
Association between rs3803662 polymorphism and breast cancer risk. The forest plot concerning the association between the rs3803662 polymorphism and the risk of breast cancer is shown in Fig. 2. Figure   A total of 95 articles were identified by the search strategy. Because don't meet the qualifications, 77 articles were removed, and 18 articles remained for further screening. Afterwards, since the control groups don't meet the Hardy-Weinberg Equilibrium (HWE), 4 articles were excluded. Besides, 1 article was excluded due to that the study population is male, not female. Finally, 13 studies were included in the meta-analysis.  Assessment of publication bias. Funnel plot were carried out to assess publication bias. There are many reasons leading to publication bias: low quality research with small sample, heterogeneity, and negative result. Figure 3A-D are funnel plots of dominant model, recessive model, additive model and Allele frequency, respectively. We found publication bias in Fig. 3B (recessive model). In this study, it mainly comes from few researches with small sample and heterogeneity.

Discussion
Our meta-analyses, including 14306 cases and 15099 controls group numbers from 13 case-control studies, explored the association between the rs3803662 polymorphism and the risk of breast cancer. The result indicates that rs3803662 is significant associated with breast cancer risk in Caucasian women. And we did not find the association Asian women.
The SNP rs3803662 is significant associated with breast cancer risk in Caucasian women which is consistent with results of previous studies [15][16][17][18][19][20][21][22] . But in Asian women, we get inconsistent result. That can be attributed to the following reasons. First, only 4 studies with 5237 cases and 5130 controls were included in the meta-analysis according to the study selection criteria 7,13,14,23 . The inadequate participant results in that rs3803662 is not significantly associated with breast cancer risk in Asian women statistically. The more the studies were included in meta-analysis, the more accurate the result will be. More independent studies concentrating on the association in Asian women should be added into this analysis and we will focus on the latest research. Second, we hypothesize that genetic background varies among different ethnic populations. Furthermore, this variation leads to change in susceptibility to some cancers. The hypothesis needs a huge amount of samples to validate whether it is true or not. The SNP rs3803662 lies 8 kb upstream of TNRC9 which is located on the chromosome 16q12 and consists of seven exons 25 . Rs3803662 has been confined to estrogen receptor-positive tumors 26 . Though we are not familiar with the function of TNRC9, this gene has been found to be relevant to bone metastasis in breast cancer 27 . TNRC9 is not only expressed in brain, but also expressed in breast with higher expression level in breast cancer compared to that in normal tissue 28,29 . And the minor allele of rs3803662 was associated with lower mRNA expression of TNRC9 gene 30 . James Owain Jones et al. found that TNRC9 maps to a known breast cancer susceptibility locus and hypothesized that TNRC9 could be a candidate tumor suppressor gene in 16q 31 .
There are some limitations in this study. First, we conducted subgroup analysis according to the ethnicity of participants. Compared with Caucasian subgroup, only 4 studies with 5237 cases and 5130 controls were included in the study. Second, the variation in genetic background may have effects on the susceptibility to breast cancer. More studies focusing on the same population should be included.
In conclusion, rs3803662 polymorphism is significantly associated with breast cancer risk in Caucasian women, and we did not find the association in Asian women.

Methods
Search strategy. A comprehensive literature searches for suitable studies published up to 2016 was conducted in the PubMed, EMbase and Web of Science database. Studies that investigated the rs3803662 and breast cancer were included in this meta-analysis. Studies should be published as a full paper. The search was conducted using the following keywords: "rs3803662" and "breast cancer". Study selection criteria. Two independent reviewers first screened the titles and abstracts to identify the relevant investigations. Then, full articles were read to include the eligible studies that met the following criteria: (1) used a case-control study design, (2) evaluated the association between rs3803662 and the risk of breast cancer, (3) provided the number of genotypes in case-controls groups for calculating ORs, (4) the control group has to satisfy HWE. Data extraction. For every eligible study, the following data were extracted by two independent reviewers: name of the first author, publication date, the ethnicity of study population, and the number of genotype in case-control group. In addition, the P value of HWE in control group was calculated.
Statistical analysis. All statistical analysis was conducted by STATA version 14.0 (STATA Corporation, College Station, TX, USA). The Odds Ratios (ORs) with 95% confidence intervals (CIs) was calculated to evaluate the association between the rs3803662 polymorphism and breast cancer risk.
The heterogeneity means the variation between different researches in systematic review. It has two major types: the clinical heterogeneity and the heterogeneity of methodology. The former comes from the difference of participants, interventions and outcome among studies. The latter derives from distinction of experimental Scientific RepoRts | 6:29008 | DOI: 10.1038/srep29008 design. We use Q and I 2 statistics to assess the heterogeneity. If I 2 < 50% or the P-value of heterogeneity > 0.10, we use fixed effects model. Otherwise, the random effects model was selected.
The Begg's rank correlation test and Egger's linear regression test were used to assess the publication bias. The genetic models we use here include dominant model, recessive mode, additive model and allele frequency 32,33 .