Human blood metabolites have been closely linked to ankylosing spondylitis (AS) in observational studies, yet direct causal evidence remains limited. This study aims to use Mendelian randomization (MR) to pinpoint causal metabolites associated with AS and to predict potential side effects of metabolite interventions. Genetic instruments for exposure were sourced from a genome-wide association study of 1400 blood metabolites, while genome-wide association study data for AS outcomes were derived from the FinnGen cohort. The primary MR analysis was conducted using the inverse variance weighted method. Supplemental analyses were conducted using weighted median, MR-Egger, simple mode, and weighted mode methods, while sensitivity analyses were performed to evaluate heterogeneity and pleiotropy. A replication analysis using an additional the UK Biobank cohort was also performed to determine metabolites associated with AS. The Steiger test and linkage disequilibrium score regression were used to further strengthen causal inference. Lastly, a phenome-wide Mendelian randomization analysis was performed to investigate the potential on-target side effects of metabolite interventions. After comprehensive analyses, 3 metabolites (the 2′-deoxyuridine levels, the hate to mannose ratio, and the Uridine to 2′-deoxyuridine ratio) were identified as being genetically associated with AS. The phenome-wide Mendelian randomization analysis revealed that the hate to mannose ratio might have deleterious effects on 4 other diseases, while no significant associations were found for the 2′-deoxyuridine levels or the uridine to 2′-deoxyuridine ratio with other diseases. This systematic MR analysis unveiled the potential role of the 2′-deoxyuridine levels, hate to mannose ratio and uridine to 2′-deoxyuridine ratio as the causal mediator in the development of AS. Considering the advantages and disadvantages, 2′-deoxyuridine appears as the most promising prospective therapeutic target for the prevention of AS.
Li et al. (Fri,) studied this question.