Steatotic Liver Disease: Does Existing Non-Invasive Markers Fit the Updated Nomenclature?
摘要
To evaluate how the updated steatotic liver disease (SLD) nomenclature—metabolic dysfunction–associated steatotic liver disease (MASLD), metabolic–alcohol associated liver disease (MetALD), and alcohol-associated liver disease (ALD)—influences the diagnostic performance and clinical applicability of non-invasive tests (NITs) for fibrosis assessment, and to determine the need for phenotype-specific risk-stratification strategies.
Recent FindingsAcross studies published between 2023 and 2025, serum-based NITs such as FIB-4 and NAFLD Fibrosis Score (NFS) retain robust rule-out performance for advanced fibrosis (≥ F3) in MASLD (AUROC ~ 0.75–0.85) and MetALD (AUROC ~ 0.72–0.83), with high negative predictive values (> 95%). In contrast, diagnostic accuracy is consistently lower in ALD (AUROC ~ 0.65–0.75), largely due to alcohol-related transaminase variability and thrombocytopenia. Transient elastography (TE)–based composite scores and emerging machine learning–derived indices (FIB-9, FIB-11, FIB-12) demonstrate higher accuracy in MASLD (AUROC up to ~ 0.90–0.92) with reduced indeterminate zones, although validation in alcohol-predominant phenotypes remains limited.
SummaryUnder the revised SLD framework, non-invasive fibrosis assessment must be phenotype-adapted. A stepwise NIT-based strategy—using FIB-4 for initial risk stratification followed by TE-based or advanced composite algorithms in selected patients—appears most appropriate for MASLD and MetALD, whereas serum-only scores are less reliable in ALD.