{"title":"癌症研究中的孟德尔随机化:机遇与挑战。","authors":"Mengyao Tang, Lanlan Chen","doi":"10.1186/s13027-025-00672-0","DOIUrl":null,"url":null,"abstract":"<p><p>Mendelian Randomization (MR) is increasingly used in cancer research to infer causal relationships by leveraging genetic variants as instrumental variables. While the growth of genome-wide association studies and biobank data has expanded the utility of MR, this surge-particularly pronounced in China-raises concerns about methodological rigor. The widespread adoption may be partly driven by the Chinese translation of key MR literature. Recent advances such as multivariable MR, mediation analysis, and integration with AI and omics data have enhanced the robustness and biological interpretability of MR studies. However, challenges persist, including horizontal pleiotropy, weak instrument bias, and misinterpretation of biomarkers as causal exposures. To improve MR study credibility, frameworks like STROBE-MR and MR-GRADE are being adopted. This article reviews methodological improvements and persistent pitfalls in MR, especially within cancer epidemiology, and highlights strategies for ensuring validity in this rapidly evolving field.</p>","PeriodicalId":13568,"journal":{"name":"Infectious Agents and Cancer","volume":"20 1","pages":"37"},"PeriodicalIF":2.8000,"publicationDate":"2025-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12168377/pdf/","citationCount":"0","resultStr":"{\"title\":\"Mendelian randomization in cancer research: opportunities and challenges.\",\"authors\":\"Mengyao Tang, Lanlan Chen\",\"doi\":\"10.1186/s13027-025-00672-0\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>Mendelian Randomization (MR) is increasingly used in cancer research to infer causal relationships by leveraging genetic variants as instrumental variables. While the growth of genome-wide association studies and biobank data has expanded the utility of MR, this surge-particularly pronounced in China-raises concerns about methodological rigor. The widespread adoption may be partly driven by the Chinese translation of key MR literature. Recent advances such as multivariable MR, mediation analysis, and integration with AI and omics data have enhanced the robustness and biological interpretability of MR studies. However, challenges persist, including horizontal pleiotropy, weak instrument bias, and misinterpretation of biomarkers as causal exposures. To improve MR study credibility, frameworks like STROBE-MR and MR-GRADE are being adopted. This article reviews methodological improvements and persistent pitfalls in MR, especially within cancer epidemiology, and highlights strategies for ensuring validity in this rapidly evolving field.</p>\",\"PeriodicalId\":13568,\"journal\":{\"name\":\"Infectious Agents and Cancer\",\"volume\":\"20 1\",\"pages\":\"37\"},\"PeriodicalIF\":2.8000,\"publicationDate\":\"2025-06-15\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12168377/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Infectious Agents and Cancer\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1186/s13027-025-00672-0\",\"RegionNum\":2,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"IMMUNOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Infectious Agents and Cancer","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1186/s13027-025-00672-0","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"IMMUNOLOGY","Score":null,"Total":0}
Mendelian randomization in cancer research: opportunities and challenges.
Mendelian Randomization (MR) is increasingly used in cancer research to infer causal relationships by leveraging genetic variants as instrumental variables. While the growth of genome-wide association studies and biobank data has expanded the utility of MR, this surge-particularly pronounced in China-raises concerns about methodological rigor. The widespread adoption may be partly driven by the Chinese translation of key MR literature. Recent advances such as multivariable MR, mediation analysis, and integration with AI and omics data have enhanced the robustness and biological interpretability of MR studies. However, challenges persist, including horizontal pleiotropy, weak instrument bias, and misinterpretation of biomarkers as causal exposures. To improve MR study credibility, frameworks like STROBE-MR and MR-GRADE are being adopted. This article reviews methodological improvements and persistent pitfalls in MR, especially within cancer epidemiology, and highlights strategies for ensuring validity in this rapidly evolving field.
期刊介绍:
Infectious Agents and Cancer is an open access, peer-reviewed online journal that encompasses all aspects of basic, clinical, epidemiological and translational research providing an insight into the association between chronic infections and cancer.
The journal welcomes submissions in the pathogen-related cancer areas and other related topics, in particular:
• HPV and anogenital cancers, as well as head and neck cancers;
• EBV and Burkitt lymphoma;
• HCV/HBV and hepatocellular carcinoma as well as lymphoproliferative diseases;
• HHV8 and Kaposi sarcoma;
• HTLV and leukemia;
• Cancers in Low- and Middle-income countries.
The link between infection and cancer has become well established over the past 50 years, and infection-associated cancer contribute up to 16% of cancers in developed countries and 33% in less developed countries.
Preventive vaccines have been developed for only two cancer-causing viruses, highlighting both the opportunity to prevent infection-associated cancers by vaccination and the gaps that remain before vaccines can be developed for other cancer-causing agents. These gaps are due to incomplete understanding of the basic biology, natural history, epidemiology of many of the pathogens that cause cancer, the mechanisms they exploit to cause cancer, and how to interrupt progression to cancer in human populations. Early diagnosis or identification of lesions at high risk of progression represent the current most critical research area of the field supported by recent advances in genomics and proteomics technologies.