Assessment of moderate coffee consumption and risk of epithelial ovarian cancer: a Mendelian randomization study
Assessment of moderate coffee consumption and risk of epithelial ovarian cancer: a Mendelian randomization study
复制标题
DOI：
10.1093/ije/dyx236
复制
发表时间：
2018-04-01
影响因子：
7.7
通讯作者：
MacGregor, Stuart
中科院分区：
文献类型：
--
作者：
Ong, Jue-Sheng;Hwang, Liang-Dar;MacGregor, Stuart
背景：观察性研究表明，咖啡的摄入量与各种健康结果有关。然而，其与上皮性卵巢癌(EOC)相关的证据并不一致，而且尚不清楚这些关联是否是因果关系。方法：我们使用与(I)咖啡和(Ii)咖啡因消费相关的单核苷酸多态对EOC风险进行孟德尔随机化(MR)。我们使用来自卵巢癌协会联合会(OCAC)的44062名欧洲血统个体的基因数据进行了两样本MR，并使用Wald-type比率估计器结合工具变量估计。结果：对于所有EOC病例，基因预测的每天多喝一杯咖啡的因果优势比(COR)为0.92[95%可信区间(CI)：0.79，1.06]。高级别浆液性卵巢癌的COR为0.90(95%CI：0.73，1.10)。基因预测额外摄入80毫克咖啡因对所有卵巢癌患者的COR为1.01(95%CI：0.92，1.11)，对于高度浆液性卵巢癌患者为0.90(95%CI：0.73，1.10)。结论：我们没有发现证据表明EOC风险与基因预测的咖啡或咖啡因水平有很强的相关性。然而，我们的估计与早期的观察性研究在统计上并不矛盾，我们不能排除小的保护性关联。
Background: Coffee consumption has been shown to be associated with various health outcomes in observational studies. However, evidence for its association with epithelial ovarian cancer (EOC) is inconsistent and it is unclear whether these associations are causal.Methods: We used single nucleotide polymorphisms associated with (i) coffee and (ii) caffeine consumption to perform Mendelian randomization (MR) on EOC risk. We conducted a two-sample MR using genetic data on 44 062 individuals of European ancestry from the Ovarian Cancer Association Consortium (OCAC), and combined instrumental variable estimates using a Wald-type ratio estimator.Results: For all EOC cases, the causal odds ratio (COR) for genetically predicted consumption of one additional cup of coffee per day was 0.92 [95% confidence interval (CI): 0.79, 1.06]. The COR was 0.90 (95% CI: 0.73, 1.10) for high-grade serous EOC. The COR for genetically predicted consumption of an additional 80 mg caffeine was 1.01 (95% CI: 0.92, 1.11) for all EOC cases and 0.90 (95% CI: 0.73, 1.10) for high-grade serous cases.Conclusions: We found no evidence indicative of a strong association between EOC risk and genetically predicted coffee or caffeine levels. However, our estimates were not statistically inconsistent with earlier observational studies and we were unable to rule out small protective associations.