Highlight
- Latent class analysis of over 48,000 UK Biobank participants with MRI-defined hepatic steatosis reveals distinct MASLD clinical subtypes with strong sex-based differences.
- A novel four-protein biomarker panel (PS4), linked to lipid metabolism and immune pathways, predicts all-cause mortality more accurately than existing non-invasive scores.
- Proteomic components mediate 10-20% of the mortality risk attributable to MASLD phenotypic classes, highlighting key biological pathways driving systemic outcomes.
Study Background
Metabolic dysfunction-associated steatotic liver disease (MASLD) represents a broad and heterogeneous spectrum of liver pathology characterized by excess fat accumulation in hepatocytes, commonly accompanied by systemic cardiometabolic risk factors such as obesity, insulin resistance, dyslipidemia, and hypertension. Previously grouped as nonalcoholic fatty liver disease (NAFLD), MASLD nomenclature underscores the metabolic underpinnings of this condition. Its considerable heterogeneity in pathophysiology and clinical outcomes challenges risk stratification and prognostication, especially given overlapping metabolic comorbidities that affect patient management and long-term survival.
The need for molecular subclassification to unravel MASLD heterogeneity has grown as traditional clinical indices and liver-centric risk scores do not fully capture prognostic variability. Precisely characterizing distinct MASLD endotypes and linking them to systemic molecular signatures could enable personalized risk assessment and inform mechanistic understanding of how hepatic metabolic abnormalities impact broader health and mortality.
Study Design
This investigation leveraged data from the UK Biobank, a large, prospective population cohort, focusing on 48,806 individuals with MRI-derived proton density fat fraction (MRI-PDFF) measurements to quantify hepatic steatosis. Latent class analysis (LCA) was applied to define sub-phenotypes of MASLD based on clinical and imaging data during over 15 years of follow-up.
Large-scale plasma proteomic profiling was performed to elucidate molecular signatures associated with identified subtypes. Subsequently, a four-protein score (PS4), comprising FABP1, FTCD, ADGRG1, and GAST—proteins involved in hepatic lipid/one-carbon metabolism and immune/vascular integrity—was derived and validated in an extended cohort of 115,971 participants, including those without MRI data, for the prediction of all-cause mortality.
Mediation analysis assessed the degree to which the PS4 proteins contributed to associations between MASLD phenotypic classes and mortality outcomes, integrating molecular data with clinical phenotypes.
Key Findings
Molecular and Phenotypic Heterogeneity with Sexual Dimorphism
The latent class model identified distinct MASLD subtypes that differed markedly in clinical features, cardiometabolic risk profiles, and long-term mortality risk. Importantly, pronounced sexual dimorphism was observed, with subtypes characterized by differences in severity and proteomic signatures between men and women. These findings highlight the influence of biological sex on MASLD pathogenesis and prognosis.
PS4 Biomarker Score Outperforms Traditional Scores
The integrative proteomic score PS4 demonstrated superior performance compared to established non-invasive tests in predicting all-cause mortality across the large validation cohort. This suggests that circulating proteins reflective of hepatic metabolic and immune pathways provide added prognostic information beyond imaging and routine clinical markers.
Biological Significance of PS4 Components
FABP1 (Fatty Acid Binding Protein 1) and FTCD (Formiminotransferase Cyclodeaminase) are enzymes linked to hepatic lipid handling and one-carbon metabolism, respectively, underscoring metabolic disruptions in MASLD. Meanwhile, ADGRG1 (Adhesion G protein-coupled receptor G1) and GAST (Gastrin) are implicated in immune regulation and vascular integrity, suggesting that inflammatory and endothelial pathways contribute to systemic disease progression and mortality risk.
Mediation of Mortality Risk
Mediation analyses indicate that these four proteins collectively explain approximately 10-20% of the phenotypic class effect on mortality, bridging clinical characteristics and survival outcomes via biological mechanisms. This novel insight emphasizes that systemic molecular alterations are integral to MASLD-related morbidity and mortality.
Expert Commentary
This study represents a significant advance in MASLD research by integrating imaging, clinical phenotyping, and high-dimensional proteomics in a robust population-based cohort. The identification of sex-driven disease endotypes adds to mounting evidence that sex-specific mechanisms should be considered in MASLD management and research design.
The PS4 score offers a scalable, biologically informed biomarker panel for mortality risk stratification, which could be incorporated into clinical practice to better identify patients at higher systemic risk.
Limitations include reliance on a largely European ancestry cohort and the observational nature, which cannot definitively establish causality of proteomic changes. Moreover, while PS4 performance was validated in a large dataset, further studies are warranted to test its predictive utility across diverse populations and clinical settings.
Conclusions
This proteomic analysis clarifies the molecular heterogeneity underlying MASLD, revealing sex-dependent endotypes linked to distinct clinical trajectories and mortality risks. The novel PS4 protein score enhances prediction of systemic outcomes beyond traditional liver-centric metrics, highlighting the interconnectedness of hepatic metabolic dysfunction with systemic cardiometabolic and inflammatory pathways.
These insights pave the way for personalized medicine approaches in MASLD, where patient stratification can guide monitoring intensity and tailored interventions to reduce long-term morbidity and mortality.
Funding and Registration
No specific funding or clinical trial registration was reported for this analysis. It utilized existing UK Biobank data and proteomic platforms.
References
- Diambra L, Sookoian S, Pirola CJ. Proteomic resolution of the MASLD cardiometabolic spectrum identifies sex-driven endotypes and predicts systemic mortality. Gut. 2026 Jul 29. PMID: 42527119.
- Rinella ME. Nonalcoholic fatty liver disease: a systematic review. JAMA. 2015;313(22):2263-2273.
- Angulo P. Nonalcoholic fatty liver disease. N Engl J Med. 2002;346(16):1221-1231.
- Younossi ZM, et al. Global epidemiology of NAFLD-MASLD: Meta-analytic assessment of prevalence, incidence, and outcomes. Hepatology. 2016;64(1):73-84.
