Electronic Health Record-Based Genome-Wide Meta-Analysis and Mendelian Randomization Identify Metabolic and Phenotypic Consequences of Non-Alcoholic Fatty Liver Disease

Non-alcoholic fatty liver disease (NAFLD) has been associated with several blood biomarkers and chronic diseases. Whether these associations underlie causal effects remains to be determined. We aimed at identifying blood metabolites, blood proteins and human diseases that are causally impacted by the presence of NAFLD using Mendelian randomization. We created a NAFLD genetic instrument from NAFLD loci (MTARC1, GCKR, LPL, TRIB1, LMO3, FTO, TM6SF2, APOE and PNPLA3) identified in a new electronic health record based-GWAS meta-analysis (6715 cases and 682,748 controls). We found a potentially causal effect of NAFLD on tyrosine metabolism as well as on blood levels of eight proteins that could potentially represent new early biomarkers of NAFLD. Using results from the UK Biobank, FinnGen and the COVID-19 Host Genetics Initiative, we found that NAFLD was not causally associated with diseases outside the spectrum of liver diseases, suggesting that the resolution of NAFLD might not prevent other diseases.