Breakthrough Breast Cancer Research Centre, Institute of Cancer Research, London, UK.
J Natl Cancer Inst. 2011 Mar 2;103(5):425-35. doi: 10.1093/jnci/djq563. Epub 2011 Jan 24.
Genome-wide association studies have identified several common genetic variants associated with breast cancer risk. It is likely, however, that a substantial proportion of such loci have not yet been discovered.
We compared 296,114 tagging single-nucleotide polymorphisms in 1694 breast cancer case subjects (92% with two primary cancers or at least two affected first-degree relatives) and 2365 control subjects, with validation in three independent series totaling 11,880 case subjects and 12,487 control subjects. Odds ratios (ORs) and associated 95% confidence intervals (CIs) in each stage and all stages combined were calculated using unconditional logistic regression. Heterogeneity was evaluated with Cochran Q and I(2) statistics. All statistical tests were two-sided.
We identified a novel risk locus for breast cancer at 9q31.2 (rs865686: OR = 0.89, 95% CI = 0.85 to 0.92, P = 1.75 × 10(-10)). This single-nucleotide polymorphism maps to a gene desert, the nearest genes being Kruppel-like factor 4 (KLF4, 636 kb centromeric), RAD23 homolog B (RAD23B, 794 kb centromeric), and actin-like 7A (ACTL7A, 736 kb telomeric). We also identified two variants (rs3734805 and rs9383938) mapping to 6q25.1 estrogen receptor 1 (ESR1), which were associated with breast cancer in subjects of northern European ancestry (rs3734805: OR = 1.19, 95% CI = 1.11 to 1.27, P = 1.35 × 10(-7); rs9383938: OR = 1.18, 95% CI = 1.11 to 1.26, P = 1.41 × 10(-7)). A variant mapping to 10q26.13, approximately 300 kb telomeric to the established risk locus within the second intron of FGFR2, was also associated with breast cancer risk, although not at genome-wide statistical significance (rs10510102: OR = 1.12, 95% CI = 1.07 to 1.17, P = 1.58 × 10(-6)).
These findings provide further evidence on the role of genetic variation in the etiology of breast cancer. Fine mapping will be needed to identify causal variants and to determine their functional effects.
全基因组关联研究已经确定了一些与乳腺癌风险相关的常见遗传变异。然而，很可能还有相当一部分这样的基因座尚未被发现。
我们比较了 1694 例乳腺癌病例（92%为两个原发性癌症或至少两个一级亲属受累）和 2365 例对照的 296114 个标签单核苷酸多态性，在三个独立的系列中进行了验证，总共有 11880 例病例和 12487 例对照。每个阶段和所有阶段组合的优势比（OR）和相关的 95%置信区间（CI）均使用无条件逻辑回归进行计算。使用 Cochran Q 和 I(2)统计数据评估异质性。所有统计检验均为双侧检验。
我们在 9q31.2 处发现了一个新的乳腺癌风险位点(rs865686：OR=0.89，95%CI=0.85 至 0.92，P=1.75×10(-10))。这个单核苷酸多态性位于一个基因荒漠，最近的基因是 Kruppel 样因子 4（KLF4，636kb 着丝粒）、RAD23 同源物 B（RAD23B，794kb 着丝粒）和肌动蛋白样 7A（ACTL7A，736kb 端粒）。我们还发现了两个变体(rs3734805 和 rs9383938)，映射到 6q25.1 雌激素受体 1（ESR1），在北欧血统的受试者中与乳腺癌相关(rs3734805：OR=1.19，95%CI=1.11 至 1.27，P=1.35×10(-7)；rs9383938：OR=1.18，95%CI=1.11 至 1.26，P=1.41×10(-7))。位于第二内含子内 FGFR2 确立风险基因座的近 300kb 端粒处的 10q26.13 上的一个变体也与乳腺癌风险相关，尽管没有达到全基因组统计学意义(rs10510102：OR=1.12，95%CI=1.07 至 1.17，P=1.58×10(-6))。
这些发现为遗传变异在乳腺癌病因学中的作用提供了进一步的证据。需要精细定位以识别因果变异，并确定其功能影响。
