炎症因子与乳腺癌关系的孟德尔随机化研究  

Association between inflammatory factors and breast cancer:a Mendelian randomization study

作　　者：宋文富 关徐涛[1,2] 王冰 孙士玲[1,2] 李盈盈[1,2] SONG Wenfu;GUAN Xutao;WANG Bing;SUN Shiling;LI Yingying(Department of Hematological and Oncology,The First Affiliated Hospital of Henan University of Chinese Medicine,Zhengzhou,Henan 450000,China;First Clinical Medical School,Henan University of Chinese Medicine,Zhengzhou,Henan 450000,China)

机构地区：[1]河南中医药大学第一附属医院血液肿瘤科,河南郑州450000 [2]河南中医药大学第一临床医学院,河南郑州450000

出　　处：《预防医学》2024年第8期714-717,722,共5页China Preventive Medicine Journal

基　　金：河南省教育厅重点科研项目(24A360006);河南省高等学校青年骨干教师项目(2023GGJS083);河南省中医药管理局重点项目(2023ZY2004,2023ZY2015,20-21ZY1010);第七批全国老中医专家学术经验继承工作指导项目(20230522);河南省青苗人才指导项目(20210411)。

摘　　要：目的采用两样本孟德尔随机化(MR)方法探究炎症因子与乳腺癌的因果关系,为乳腺癌防治提供依据。方法通过公开数据库收集91种炎症因子(n=14824)和5种乳腺癌亚型(n=247173)的全基因组关联研究(GWAS)数据,选取与91种炎症因子相关的单核苷酸多态性(SNP)位点为工具变量。以炎症因子为暴露,乳腺癌为结局,采用逆方差加权法进行MR分析。采用FDR校正降低Ⅰ类错误风险和多重检验的影响。采用Steiger方向检验、MR-Egger回归法、MR-PRESSO检验和留一法验证结果的稳定性和可靠性。结果β神经生长因子、白介素-5、胱抑素D和C-X-C基序趋化因子1等23种炎症因子与乳腺癌存在统计学关联(均P<0.05);经FDR校正后,仅发现抑瘤素M丰度升高与Basal-like(三阴性)乳腺癌发病风险增加存在统计学关联(OR=1.186,95%CI:1.081~1.302,P=0.001,q=0.029),其他22种炎症因子与乳腺癌的关联发生Ⅰ类错误的风险较高(均q>0.1)。敏感性分析显示结果稳健,未发现对结果有强影响的工具变量,可排除异质性、水平多效性和反向因果对该结果产生的影响。结论抑瘤素M丰度升高可能增加Basal-like(三阴性)乳腺癌发病风险。Objective To examine the causal relationship between inflammatory factors and breast cancer using two-sample Mendelian randomization(MR)approach,so as to provide the basis for the prevention and treatment of breast cancer.Methods Data of 91 inflammatory cytokines(n=14824)and 5 subtypes of breast cancer(n=247173)were collected from genome-wide association studies(GWAS).Single nucleotide polymorphism(SNP)associated with 91 inflammatory factors were selected as instrumental variables.MR analyses were performed using the inverse-variance weighted(IVW)method with inflammatory factors as exposure factors and breast cancer as outcome variables.The risk of type I error and the effect of multiple testing were reduced using the FDR correction method.The stability and reliability of the results were verified using Steiger test of directionality,MR-Egger regression,MR-PRESSO test and leave-one out method.Results Twenty-three inflammatory factors,includingβnerve growth factor,interleukin-5,cystatin D and C-X C chemokine ligand 1 were statistically associated with breast cancer(all P<0.05).After FDR adjustment,only evaluated abundance of oncostatin-M was found to be statistically associated with an increased risk of Basal-like(triple-negative)breast cancer(OR=1.186,95%CI:1.081-1.302,P=0.001,q=0.029),and the other 22 inflammatory factors had a high risk of type I error(all q>0.1).The sensitivity analysis indicated that the results were robust.No instrumental variables were found to have a significant impact on the results,which could exclude the influence of heterogeneity,horizontal pleiotropy,and reverse causality on the outcome.Conclusion The increased abundance of oncostatin-M may increase the risk of Basal-like(triple-negative)breast cancer.

关 键 词：乳腺癌 炎症因子 抑瘤素M 孟德尔随机化 全基因组关联研究 因果关系 

分 类 号：R73[医药卫生—肿瘤]