Multi-omics dissection of MASLD reveals divergent pathways to systemic diseases and targets for prevention.
TL;DR
The findings delineate etiologically and clinically distinct MASLD subtypes, reveal protein-mediated pathways to systemic diseases, and inform biologically guided prevention strategies.
Abstract
Background
Metabolic dysfunction-associated steatotic liver disease (MASLD) is a prevalent but heterogeneous condition. How its genetic diversity shapes systemic disease risk remains unclear. We aimed to identify genetically defined MASLD subtypes, assess their effects on extrahepatic diseases, and uncover potential protein-mediated mechanisms.
Methods
We performed a genome-wide association study of MRI-derived hepatic fat content (HFC) in 33,592 UK Biobank (UKBB) participants to identify MASLD-related SNPs. Clustering of associations between MASLD-SNPs and triglyceride-related traits identified distinct genetic subtypes. Phenotypic associations were evaluated in 427,030 UKBB European-ancestry participants and validated in two independent populations. Plasma proteomic profiling and Mendelian randomization/mediation analyses were used to identify protein mediators linking subtypes to disease outcomes.
Results
We identified three MASLD polygenic subtypes reflecting hepatic lipid retention, hepatic lipid synthesis, and systemic lipid metabolism. These subtypes exhibited distinct risk profiles across 22 incident diseases spanning multiorgan systems in the UKBB European-ancestry cohort, with partial replication in the UKBB non-European-ancestry populations and the Taizhou Longitudinal Study in China. All three subtypes consistently increased liver disease risk but showed divergent cardiometabolic effects. Proteomic profiling revealed subtype-specific signatures; for instance, systemic lipid metabolism was linked to elevated PCSK9 and higher cardiovascular risk, whereas hepatic lipid retention showed the opposite. Mediation and Mendelian randomization analyses identified both established (e.g., PCSK9, FURIN, APOE) and potentially novel (e.g., CCN1, ASGR1) protein mediators as putative targets for subtype-specific chronic disease prevention.
Conclusion
Our findings delineate etiologically and clinically distinct MASLD subtypes, reveal protein-mediated pathways to systemic diseases, and inform biologically guided prevention strategies.