乳腺癌与阿尔茨海默症的孟德尔随机化研究

The Gregor Mendel Randomized Study of Breast Cancer and Alzheimer’s Disease

摘要: 目的：通过两样本孟德尔随机化(mendelian randomization, MR)分析探究其中乳腺癌(breast cancer, BC)和阿尔茨海默症(Alzheimer’s disease, AD)的因果关系。方法：纳入阿尔茨海默症与乳腺癌作为暴露与结局变量，进行双向孟德尔随机化分析。通过R语言筛选与暴露因素(BC/AD)强关联的单核苷酸多态性(SNP)，排除连锁不平衡(LD，r² > 0.001且距离 > 10,000 kb)的SNP，合并暴露与结局数据后剔除效应方向矛盾的SNP。采用逆方差加权法(IVW)、MR-Egger回归等进行因果效应估计，并通过异质性检验(Cochran’s Q)、水平多效性检验(MR-Egger截距)及留一法(leave-one-out analysis)评估结果稳健性。结果：从IEU (https://gwas.mrcieu.ac.uk)数据库选择出了经过全基因组关联研究(Genome-Wide Association Studies, GWAS)处理的乳腺癌与阿尔茨海默症的SNP数据集，再通过孟德尔随机分析去除连锁不平衡且满足PVal < 5E−08的强相关SNP，整合暴露与结局数据后获得的SNP数据作为最终的有效工具变量。IVW法显示乳腺癌是阿尔茨海默症的风险因素，MR-Egger回归结果支持乳腺癌和阿尔茨海默症存在负相关联；根据IVW结果、MR-Egger回归截距、mr pleiotropy test函数和留一法分析法，进行异质性分析、多效性分析以及敏感性分析，进一步证明乳腺癌的发生与阿尔茨海默症存在因果关系，而阿尔茨海默症与乳腺癌没有因果关系。结论：乳腺癌是阿尔茨海默症的危险因素，乳腺癌的发生与阿尔茨海默症存在因果关系，为正向相关，而阿尔茨海默症的发生与乳腺癌不存在因果关系。

Abstract: Objective: The objective of this study is to explore the causal relationship between breast cancer (BC) and Alzheimer’s disease (AD) through two-sample mendelian randomization (MR) analysis. Methods: Incorporate Alzheimer’s disease and breast cancer as exposure and outcome variables for bidirectional Mendelian randomization analysis. Using R programming, select single nucleotide polymorphisms (SNPs) that are strongly associated with the exposure factors (BC/AD), excluding SNPs exhibiting linkage disequilibrium (LD, r² > 0.001 and distance > 10,000 kb). After merging the exposure and outcome data, eliminate SNPs with conflicting effect directions. Employ inverse variance weighting (IVW), MR-Egger regression, and others to estimate causal effects, and assess the robustness of the results through heterogeneity testing (Cochran’s Q), horizontal pleiotropy testing (MR-Egger intercept), and leave-one-out analysis. Result: SNP datasets of breast cancer and Alzheimer’s disease processed by Genome-wide Association Studies (GWAS) were selected from the IEU (https://gwas.mrcieu.ac.uk) database, and then the linkage imbalance was removed by Mendelian random analysis and PVal was satisfied < strongly correlated SNPs of 5E−08, the SNP data obtained after integrating exposure and outcome data were used as the final effective instrumental variables. IVW showed that breast cancer was a risk factor for Alzheimer’s disease, and MR-Egger regression results supported a negative association between breast cancer and Alzheimer’s disease. According to the IVW results, MR-Egger regression intercept, mr pleiotropy test function and retention method analysis, heterogeneity analysis, pleiotropy analysis and sensitivity analysis were carried out to further prove that there was a causal relationship between the occurrence of breast cancer and Alzheimer’s disease, while there was no causal relationship between Alzheimer’s disease and breast cancer. Conclusion: Breast cancer is a risk factor for Alzheimer’s disease, and there is a causal relationship between the occurrence of breast cancer and Alzheimer’s disease, which is positively correlated, while there is no causal relationship between the occurrence of Alzheimer’s disease and breast cancer.