肝癌电子杂志 ›› 2026, Vol. 13 ›› Issue (1): 33-38.

• 论著 • 上一篇    下一篇

氯吡格雷使用与原发性肝癌因果关系:基于孟德尔随机化研究

阳唯, 杨淑红*   

  1. 兰州理工大学生命科学与工程学院,甘肃兰州 730000
  • 收稿日期:2025-08-31 出版日期:2026-03-30 发布日期:2026-05-21
  • 通讯作者: * 杨淑红,E-mail: yangsh@lut.edu.cn

Clopidogrel use and primary liver cancer causality: a Mendelian randomisation-based study

Yang Wei, Yang Shuhong*   

  1. College of Life Sciences and Engineering, Lanzhou University of Technology, Lanzhou 730000, Gansu, China
  • Received:2025-08-31 Online:2026-03-30 Published:2026-05-21
  • Contact: * Yang Shuhong, E-mail: yangsh@lut.edu.cn

摘要: 目的:采用双样本孟德尔随机化法分析氯吡格雷使用与肝细胞癌(HCC)、肝内胆管癌(ICC)和混合型肝细胞癌-胆管癌(cHCC-CCA))的因果关系。
方法:在GWAS catalog数据库中提取2018—2021年使用氯吡格雷(n=462 933)、HCC(n=161 323)、ICC(n=476 091)及cHCC-CCA(n=218 488)数据集。以逆方差加权法为主要方法评估氯吡格雷使用与HCC、ICC及cHCC-CCA的因果关系,用MR-Egger Q检验和IVW Q检验进行异质性分析,以漏斗图评估是否存在离群值,用留一法评估结果可靠性,采用 MR-Egger 截距检验评估结果是否存在水平多效性。
结果:本研究筛选出8个与氯吡格雷使用强相关的单核苷酸多态性(SNP)作为工具变量(IV)。逆方差加权法与中位数加权法分析结果显示患者使用氯吡格雷与HCC发生存在负相关关系(OR<1,P<0.05),患者使用氯吡格雷与ICC及cHCC-CCA的发生均无关(均P>0.05)。异质性检验结果显示氯吡格雷使用与HCC的因果关系不存在异质性(P>0.05),MR-Egger截距检测结果表明未检测到氯吡格雷使用与HCC存在水平多效性(P>0.05)。留一法评估结果显示逐一删除单个SNP后对孟德尔随机化分析结果无显著影响。漏斗图分析结果显示所选择的IV呈对称性分布,提示结果无明显的偏倚。
结论:氯吡格雷使用与HCC的风险降低有关,但未降低ICC和cHCC-CCA的风险。

关键词: 氯吡格雷, 肝细胞癌, 肝内胆管癌, 混合型肝细胞癌, 两样本孟德尔随机化

Abstract: Objective: To investigate the potential causal relationship between clopidogrel usage and the incidence of primary liver cancer, including hepatocellular carcinoma (HCC), intrahepatic cholangiocarcinoma (ICC), and combined hepatocellular-cholangiocarcinoma (cHCC-CCA), using a two-sample Mendelian randomisation (MR) approach.
Methods: Data sets on clopidogrel use (n=462 933), HCC (n=161 323), ICC (n=476 091) and cHCC-CCA (n=218 488) were obtained from the GWAS catalog database from 2018 to 2021. The causal association of clopidogrel use with HCC, ICC and cHCC-CCA was assessed using inverse variance weighting as the primary method, heterogeneity analysed by MR-Egger Q test and IVW Q test, the presence of outliers assessed by funnel plots, reliability of the results assessed by leave-one-out rows, and horizontal multivariate validity tested by MR-Egger intercept test.
Results: In this study, eight SNPs strongly associated with clopidogrel use were screened as instrumental variables (IV). The results of the inverse variance weighted and median weighted analyses showed that there was a negative correlation between clopidogrel use in patients and the occurrence of HCC (OR<1, P<0.05), and there was no significant correlation between the use of clopidogrel in patients and the occurrence of both ICC and cHCC-CCA (P>0.05). The results of the heterogeneity test showed no heterogeneity in the causal relationship between clopidogrel use and HCC (P>0.05), and the results of the MR-Egger intercept test indicated that no horizontal pleiotropy was detected between clopidogrel use and HCC (P>0.05). The leave-one-out method of assessment showed no significant effect on the results of the Mendelian randomisation analysis after deleting individual SNPs one by one.The results of funnel plot analysis showed a symmetrical distribution of the selected IV, suggesting no significant bias in the results.
Conclusion: ClopidogrelusewasassociatedwithareducedriskofHCC, but did not reduce the risk of ICC and cHCC-CCA.

Key words: Clopidogrel, Hepatocellular carcinoma, Intrahepatic cholangiocarcinoma, Combined hepatocellular-cholangiocarcinoma, Two-sample Mendelian randomisation