Abstract
BACKGROUND: The risk factors for thyroiditis, an inflammatory disease with a complex etiology, remain poorly understood. Blood metabolites are known to change during thyroiditis development, suggesting a close relationship between blood metabolites and thyroiditis progression. However, evidence for a causal link is lacking. We employed Mendelian randomization (MR) methodology to systematically investigate the putative causal relationships between blood metabolite profiles and two clinically distinct thyroiditis phenotypes-subacute and autoimmune thyroiditis-providing insights into their metabolic underpinnings. METHODS: We analyzed genomic and health data from 88 million Finnish Biobank participants in the Genome-Wide Association Study (GWAS). The primary analytical method was random-effects inverse variance weighting (IVW), supplemented by the weighted median method (WME) and Mr-Egger. We implemented comprehensive sensitivity analyses encompassing Cochran's Q test, Mr-Egger intercept, leave-one-out analysis (LOO), and Mr-PRESSO to assess heterogeneity, pleiotropy, and outliers. Extended genetic investigations incorporated the linkage disequilibrium score regression (LDSC) method, multivariable Mr (MVMR), and metabolic pathway analyses to provide deeper mechanistic insights. RESULTS: Ten metabolites were significantly associated with autoimmune thyroiditis, and fifteen with subacute thyroiditis. Nonadecanoate (19:0) and 1-palmitoylglycerophosphoinositol* were found to directly affect subacute thyroiditis. MVMR analyses identified pelargonate (9:0), carnitine, and ADpSGEGDFXAEGGGVR* as having an independent and direct effect on autoimmune thyroiditis. Additionally, metabolic pathways such as neomycin, kanamycin, and gentamicin biosynthesis, histidine metabolism, and starch and sucrose metabolism were linked to autoimmune thyroiditis, while phenylalanine, tyrosine, tryptophan biosynthesis, phenylalanine metabolism, and arginine biosynthesis were associated with subacute thyroiditis. CONCLUSIONS: Our findings establish causal relationships between circulating metabolites and thyroiditis, revealing novel mechanistic insights through integrated genomic and metabolomic analyses. These results not only advance our understanding of thyroiditis pathogenesis but also suggest potential biomarkers for disease screening and therapeutic targets for intervention.