Abstract 2266: Evaluations of more than 1500 blood metabolite biomarkers with risk of eight common cancers and their subtypes using genetic instruments

Abstract Background: Many epidemiologic studies have measured blood metabolites and evaluated their associations with risk of cancers. However, conventional epidemiologic studies may suffer from various biases, limiting their ability for causal inferences. We performed a large Mendelian randomization (MR) study to systematically evaluate blood metabolite biomarkers in relation to cancer risk, including large numbers of associations not previously investigated in either convention epidemiologic or MR studies. Methods: We systematically searched the literature and obtained data from 29 genome-wide association studies (GWAS) to identify blood metabolite-associated genetic loci (mQTLs) among 324, 771 individuals of European ancestry. After systematically evaluating all metabolites and mQTLs evaluated in these studies, we selected 1630 metabolite biomarkers for MR analyses using summary statistic data generated from GWAS conducted among 2, 068, 121 individuals of European ancestry for eight common cancers (breast, colorectal, endometrial, kidney, lung, ovarian, pancreatic, and prostate) and 19 subtypes of breast, kidney, lung, and ovarian cancers. The inverse variance weighted (IVW) method was used as the primary approach for MR analyses. Other methods, including MR-Egger, and MR-PRESSO, were used to evaluate possible pleiotropic effects and robustness of study findings. Statistical significance was judged using Bonferroni and FDR corrected P-value to reduce type 1 errors. Results: We identified 106 biomarker-cancer associations under the significant threshold after Bonferroni correction. Among these, 68 were associated with an increased risk of cancer, while 38 were associated with a decreased risk of cancer. Under an FDR-corrected P < 0.05 but not reaching the Bonferroni-adjusted significant threshold, we identified 119 additional associations, including 65 metabolite biomarkers as potential risk factors and 54 as potential protective factors. Conclusion: Our MR analysis uncovered large numbers of novel associations and provided strong evidence for causal inferences of blood metabolite biomarkers with cancer risk. Citation Format: Zheng Guo, Guochong Jia, Fangcheng Yuan, Lili Liu, Shuai Xu, Jirong Long, Wei Zheng. Evaluations of more than 1500 blood metabolite biomarkers with risk of eight common cancers and their subtypes using genetic instruments [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2266.