Mendelian randomization analysis identifies blood tyrosine levels as a biomarker of non-alcoholic fatty liver disease

Non-alcoholic fatty liver disease (NAFLD) is a complex cardiometabolic disease associated with premature mortality. The diagnosis of NAFLD is challenging and the identification of biomarkers causally influenced by NAFLD may be clinically useful. We aimed at identifying blood metabolites causally impacted by NAFLD using two-sample Mendelian randomization (MR) with validation in a population-based biobank and a cohort of patients undergoing bariatric surgery. Our instrument for genetically-predicted NAFLD (the study exposure) included all independent genetic variants (n=7 SNPs) from a recent genome-wide association study on NAFLD. The study outcomes included 123 blood lipids, lipoproteins and metabolites measured in 24,925 individuals from 10 European cohorts. After correction for multiple testing, we identified a positive effect of NAFLD on plasma tyrosine levels but not on other metabolites. The association between NAFLD and tyrosine levels was consistent across MR methods and robust to outliers and pleiotropy. In observational analyses performed in the Estonian Biobank (10,809 individuals including 359 patients with NAFLD), after multivariable adjustment, tyrosine levels were positively associated with the presence of NAFLD (odds ratio per 1-SD increment = 1.23 (95% confidence interval = 1.12-1.36, p = 2.19e-05). In a sample of 138 patients undergoing bariatric surgery, compared to patients without NAFLD, blood tyrosine levels were higher in those with NAFLD, but were comparable among patients with or without non-alcoholic steatohepatitis. This analysis revealed a potentially causal effect of NAFLD on blood tyrosine levels, suggesting that blood tyrosine levels may represent a new biomarker of NAFLD. Graphical abstract