Multicenter Prospective Cohort Study on Clinical Outcomes and Fibrosis Patterns in Biopsy-Proven Steatotic Liver Disease Subtypes

Abstract

BACKGROUND & AIMS: Clinical outcomes and histologic characteristics of biopsy-proven steatotic liver disease (SLD) subtypes remain understudied. We investigated clinical outcomes and fibrosis patterns in metabolic dysfunction-associated steatotic liver disease (MASLD), metabolic dysfunction and alcohol-related liver disease (MetALD), and alcohol-associated liver disease (ALD). METHODS: This multicenter prospective cohort study enrolled 2551 individuals with biopsy-proven SLD or non-SLD controls between 2010 and 2023. Risks of all-cause mortality, liver-related events (LREs), cardiovascular disease, and extrahepatic malignancies were assessed using Fine-Gray models. Artificial intelligence (AI)-based quantitative fibrosis analysis using second harmonic generation imaging was performed on 88 liver biopsies (28 MASLD, 30 MetALD, and 30 ALD) to assess zone-specific collagen distribution patterns. RESULTS: During a median follow-up of 43.8 months, 166 deaths and 162 LREs occurred. Compared with non-SLD, Met-ALD (adjusted subdistribution hazard ratio [ASHR], 3.05; 95% confidence interval [CI], 1.08-8.58) and ALD (ASHR, 5.80; 95% CI, 2.26-14.87) demonstrated substantially increased all-cause mortality risk. MetALD (ASHR, 6.13; 95% CI, 1.33-28.20) and ALD (ASHR, 10.96; 95% CI, 2.55-47.14) also exhibited elevated LRE risk. In the advanced fibrosis (>= F3) subgroup, ALD demonstrated higher mortality and LRE risks than MASLD. AI-based analysis revealed that MetALD and ALD exhibited significantly higher collagen density in periportal and zone 2 regions compared with MASLD, despite comparable conventional fibrosis stages. CONCLUSIONS: MetALD and ALD are independently associated with substantially increased all-cause mortality and LRE risks compared with non-SLD. Distinct zone-specific collagen patterns identified by AI analysis suggest unique fibrotic progression pathways in MetALD and ALD, supporting the need for tailored risk stratification and targeted management strategies.