目的:本研究旨在探讨糖尿病与前列腺癌之间的因果效应.通过孟德尔随机化方法,可以排除其他可能的干扰因素,以明确糖尿病与前列腺癌之间的因果效应,为临床预防和治疗这两种疾病提供依据.方法:1型、2型糖尿病与前列腺癌的全基因组关联研究(genome-wide association study,GWAS)综合数据基于欧洲人群数据.1型糖尿病的数据涵盖了3,428名糖尿病患者和3,428名健康对照个体.2型糖尿病的数据涵盖了148,726名糖尿病患者和965,732名健康对照个体.前列腺癌的数据包括30,945名前列腺疾病患者和368,725名健康人群作为对照.本研究采用1、2型糖尿病作为研究的暴露,并筛选出与之显著关联且独立的单核苷酸多态性(single nucleotide polymorphism,SNP)作为工具变量.使用两样本孟德尔随机化(two-sample Mendelian randomization,2SMR)方法来评估1、2型糖尿病与前列腺癌之间的因果效应.在分析过程中采用了逆方差加权法(IVW)、孟德尔随机化回归法(MR-Egger)、加权中值法(weighted median)、加权众数法(weighted mode)以及简单模式法(simple mode)五种方法,其中IVW为主要分析方法,并使用Cochran'Q检测异质性.结果:研究结果显示,1型糖尿病与前列腺癌之间不存在因果效应(IVW:OR=0.99,95%CI:0.99～1.001,P=0.18);2型糖尿病的存在能提高前列腺癌的发病率(IVW:OR=7.88,95%CI:6.95～8.93,P<0.05).结论:2型糖尿病与前列腺癌之间存在因果联系,2型糖尿病的存在会提高病人患前列腺癌的风险,而1型糖尿病与前列腺癌之间不存在显著的因果联系.
Analysis of the Impact of Diabetes on Prostate Cancer Using Two-Sample Mendelian Randomization
Objective This study aims to address the uncertainty regarding the causal relationship between diabetes mel-litus(DM)and prostate cancer.By employing Mendelian randomization,we can eliminate potential confounding factors and clarify the causal link between DM and prostate cancer,which will aid in the prevention and treatment of these two diseases in clinical practice.Methods We utilized data from genome-wide association studies(GWAS)focused on type 1 diabetes mellitus(T1DM),type 2 diabetes mellitus(T2DM),and prostate cancer,all based on European populations.The T1DM dataset com-prised 3,428 patients with diabetes and 3,428 healthy controls.The T2DM dataset included 148,726 patients with diabetes and 965,732 healthy controls.The prostate cancer dataset included 30,945 patients with prostate disease and 368,725 healthy individuals as controls.T2DM was used as the exposure variable,and we identified significantly associated and independent single nucleotide polymorphism(SNP)as instrumental variables.We employed a two-sample Mendelian randomization(2SMR)approach to evaluate the causal relationship between T1DM,T2DM,and prostate cancer.The analysis incorporated five meth-ods:inverse variance weighting(IVW),Mendelian randomization regression(MR-Egger),weighted median,weighted mode,and simple mode,with IVW serving as the primary analytical method.Cochran's Q statistic was utilized to assess heterogene-ity.Results The findings indicate that no causal effect was observed between T1DM and prostate cancer(IVW:OR=0.99;95%CI:0.99-1.001,P=0.18).In contrast,the presence of T2DM is associated with an increased incidence of prostate cancer in the overall population(IVW:OR=7.88,95%CI:6.95-8.93,P<0.05).Conclusion There is a causal relationship between T2DM and prostate cancer,and T2DM increases the risk of developing prostate cancer.However,no promotive relationship was found between T1DM and prostate cancer.
NETL
NSTL
万方数据
