Genome Wide Association Study: Variants Influencing Steroid Hormone Levels

描述(由申请人提供)： 有大量证据表明，循环性类固醇激素水平受到遗传因素的影响。性类固醇激素，如雌二醇和睾酮，与许多疾病的发展有关，包括乳腺癌。基因分型技术的最新进展降低了基因分型成本，使得能够进行全基因组关联扫描的基因分型平台得以开发。癌症易感性遗传标记(CGEMS)项目已经启动，利用Illumina Human Hap500微珠芯片进行全基因组关联研究，以检测易患乳腺癌的单核苷酸多态(SNPs)。护士健康研究(NHS)是这项合作努力的一部分，提供了2300名女性的DNA(1150名绝经后被诊断为浸润性乳腺癌的女性，以及1150名年龄匹配的未被诊断为乳腺癌的女性)。对近期未接受激素替代治疗的绝经后妇女的血浆中雌二醇、睾酮和SHBG水平进行了检测。在CGEMS项目中进行基因分型的样本中，约有430名患者和415名对照符合这些标准，并具有可用的雌二醇、睾酮和SHBG水平。这些相同的血浆激素水平已经在每个病例的第二个对照样本以及被诊断为原位癌的妇女的对照样本中进行了检测，但这些对照对象(n~1100)并未作为CGEMS项目的一部分进行基因分型。这项工作的目的是描述与血浆性类固醇激素水平相关的基因多态性。这将通过使用协方差分析对CGEMS项目中的550,000个SNPs进行基因分型与激素水平的相关性的初步筛选来完成。然后将在额外的对照样本中对这些SNPs的子集进行基因分型，对第一和第二阶段的联合分析将为SNP基因座与血浆激素水平之间的关联提供强有力的证据。作为产生二次分析的假说，我们还将检查拷贝数变异(CNV)与类固醇激素水平的关系。通过使用CGEMS项目中收集的基因数据和之前在NHS中分析的性类固醇激素水平，这项工作将提供一个独特的、经济有效的和强大的全基因组关联研究来描述预测性类固醇激素水平的多态。本研究收集的基因分型数据将与现有的性类固醇激素数据相结合，从这两个阶段的全基因组关联扫描中提供与性类固醇激素水平最相关的SNPs的有力联合分析。因此，在联合分析中发现的与性类固醇激素水平相关的基因多态将为进一步研究它们对性类固醇激素水平的影响提供非常有可能的候选，性类固醇激素水平是公认的许多疾病的预测因子，包括乳腺癌。

DESCRIPTION (provided by applicant): There is substantial evidence that circulating sex steroid hormone levels are influenced by genetic factors. Sex steroid hormones, such as estradiol and testosterone, are implicated in the development of many diseases, including breast cancer. Recent advances in genotyping technology have reduced genotyping costs, allowing for the development of genotyping platforms capable of performing genome wide association scans. The Cancer Genetic Markers of Susceptibility (CGEMS) project has been initiated to perform a genome wide association study to detect single nucleotide polymorphisms (SNPs) which predispose for breast cancer, using the Illumina HumanHap500 bead chip. The Nurses' Health Study (NHS) is part of this collaborative effort, providing DNA from 2300 women (1150 women diagnosed with invasive breast cancer after menopause, and 1150 age-matched women not diagnosed with breast cancer). Levels of estradiol, testosterone, and SHBG have been assayed in plasma collected prospectively from women who were postmenopausal at blood collection, not having recently taken hormone replacement therapy. Approximately 430 cases and 415 controls within the samples genotyped in the CGEMS project fit these criteria, and have estradiol, testosterone, and SHBG levels available. These same plasma hormone levels have been assayed on a second control sample for each case, as well as the control subjects for women diagnosed with in situ carcinoma, however these control subjects (n~1100) have not been genotyped as part of the CGEMS project. The objective of this work is to describe polymorphisms that are associated with plasma sex steroid hormone levels. This will be accomplished in an initial screen of the 550,000 SNPs genotyped in the CGEMS project for associations with hormone levels using analysis of covariance. A subset of these SNPs will then be genotyped in the additional control samples, and joint analyses of the initial and second stages will provide strong evidence for association between SNP loci and plasma hormone levels. As hypothesis generating secondary analysis, we will also examine copy number variants (CNVs) for association with steroid hormone levels. By using genotype data collected in the CGEMS project, and sex steroid hormone levels assayed previously in the NHS, this work will provide a unique, cost effective, and powerful genome wide association study to describe polymorphisms which predict sex steroid hormone levels. The genotyping data collected in the current study will be combined with already available sex steroid hormone data to provide well powered joint analyses of the SNPs most strongly associated with sex steroid hormone levels from this two stage genome wide association scan. Therefore, polymorphisms found to be associated with sex steroid hormone levels in this joint analysis will provide very likely candidates for further study with respect to their influence on sex steroid hormone levels, which are well recognized predictors of many diseases, including breast cancer.