基于GWAS研究的遗传风险分值在乳腺癌筛查的潜在价值研究

Since 2007, genome-wide association studies (GWAS) of breast cancer, including five studies among Chinese women, have identified 93 single-nucleotide polymorphisms (SNPs) associated with breast cancer risk. With an increasing number of genetic susceptibility loci being identified for breast cancer, it is now important to examine whether genetic information could have utility in public health applications. Possible applications include risk stratified prevention and screening strategies targeted to susceptible subgroups of the population at elevated risk, or conversely defining subgroups at low risk that would benefit least from interventions. And the most important application could be the potential value of these SNPs for population-based screening..In this study, firstly, based on previous simulated 2 million Chinese women aged 35-69 and these candidate SNPs associated with increased risk of breast cancer, we will selected the top 8 SNPs which could be used for breast cancer screening according to the area under the curve (AUC) for receiver operating characteristic (ROC) , the integrated discrimination improvement (IDI), and the net reclassification improvement (NRI) (Model-based theory study). Secondly, based on the Tianjin cohort study of Chinese Multi-modality Independent Screening Trial (7911 females), these variants information of these targeted 8 SNPs will be tested for all women. After follow-up, these information will be used to develop the combined genetic risk scores (GRSs). Further analyses will be conducted to explore the potential value of GRS used in screening (Population-based validation study), including the improvement on accuracy (including sensitivity, specificity, et al) of traditional screening tools (clinical breast examination, B-mode ultrasound, mammography), detection rate of breast cancer, and the tumor characteristics (including cancer stage, lymph node statue, and histological type). Thirdly, after combining the traditional environmental risk factors collected in the baseline of the cohort, more effects will be used to evaluate the potential interaction between GRSs and environmental risk factors so as to develop a more feasible comprehensive risk-stratification method for breast cancer (Extended application study), which would be more useful for defining the targeted high-risk population in the future larger population-based breast cancer screening.

2007年以来，GWAS研究共发现了93个乳腺癌相关SNP。随着SNP发现个数的增加，探索这些SNP的公共卫生学价值，尤其是人群筛查价值，显得尤为重要。本课题首先将基于前期模拟的200万35-69岁中国女性人群，从候选SNP中采用ROC曲线下面积、整合区分指数、净再分类优化指数遴选出可用于乳腺癌筛查的前8位SNP(理论模型研究）。其次，基于课题组前期中国女性乳腺癌优化筛查方案天津队列研究人群(7911人)基线血样，检测8个目标SNP信息。结合进一步随访结果，构建遗传风险分值(GRS)，并评价在传统乳腺癌筛查方法基础上，引入GRS能否改善相应筛查方法的准确性、乳腺癌的检出率、筛查发现乳腺癌的肿瘤特征(人群验证研究)。最后，基于队列研究人群前期收集的基线环境因素暴露情况，构建更加完善的乳腺癌风险分层方法，进一步明确适合筛查的高危人群(应用扩展研究)，为今后更大人群的乳腺癌筛查提供政策支持。

首先，采用200万人群的数据模拟研究结果显示，从GWAS研究发现的与我国女性乳腺癌潜在相关的23个候选SNP中，初步证实12个SNP可用于构建乳腺癌的遗传风险预测。进一步分析提示在6个传统危险因素的基础上，加上目标SNP共同用于初筛乳腺癌高危人群时，可使乳腺癌检出率（229.00/10万）显著提高29.0%（P<0.001），同时高危人群中发生乳腺癌的风险将从3.32（95%CI: 2.45-4.49）上升到4.33（95%CI: 3.21-5.84）。其次，在由8594例乳腺癌患者及7397例健康对照组成的病例对照研究人群中，从22个候选SNP中，共发现有16个SNP的杂合突变或纯合突变基因型与我国女性乳腺癌的风险显著相关（P 值<0.05）。其中杂合突变基因型与乳腺癌的关联强度从1.11（rs7107217）到1.35（rs2046210）不等；纯合突变基因型与乳腺癌的关联强度从1.10（rs9485372）到1.53（rs2046210）不等。进一步分析提示共有21个SNP的等位基因均与我国女性乳腺癌的风险相关，关联强度从1.05（rs1432679）到1.27（rs2046210）不等（P<0.05）。根据得到验证的21个SNP构建的乳腺癌遗传风险得分的分布情况，可发现病例组乳腺癌遗传风险得分的平均分（23.19）显著高于对照组乳腺癌遗传风险得分的平均分（22.33）（P<0.001）。最后，分别采用乳腺癌传统风险得分，遗传风险得分，以及综合风险得分预测病例对照人群的乳腺癌患病情况的准确性，结果提示上述相应得分预测乳腺癌患病情况的ROC曲线下面积分别为69.56%（95%CI：68.66%-70.45%），59.67%（95%CI：58.70%-60.64%），和71.38%（95%CI：70.51%-72.26%）。进一步分析结果提示，在传统危险因素的基础上，加入21个SNP后，可进一步使传统风险得分的ROC曲线下面积显著提高1.83%（95%CI：1.42%-2.24%），同时病例对照人群风险再分类准确性均得到显著提升（P<0.001），其中整合区分改善和净再分类改善的大小分别为25.24%（95%CI: 19.93%-30.74%）和7.98%（95%CI：5.03%-10.92%）。