Investigation of non-invasive tests for fibrosis diagnosis and prognostication of events in metabolic dysfunction-associated steatotic liver disease: A prospective study cohort.
Abstract
INTRODUCTION AND OBJECTIVES
Risk stratification is important in the management of metabolic dysfunction-associated steatotic liver disease (MASLD) to prioritize monitoring and treatment resources. We evaluated whether non-invasive tests (NITs) can predict major adverse liver outcomes (MALOs) and diagnose advanced fibrosis in patients with MASLD.
PATIENTS AND METHODS
A prospective cohort of 270 patients with MASLD (median age 56 years; 44% female; 41% type 2 diabetes) was evaluated using nine NITs: FIB-4, LSM, Agile3+, Agile4, FAST, ADAPT, ELF, PRO-C3, and C1M. Cox proportional hazards models and ROC analyses were used to assess prognostic performance for MALOs and diagnosis of advanced fibrosis, respectively.
RESULTS
Over a four-year follow-up period, 25 patients experienced MALOs. All NITs were associated with MALOs, though confidence intervals were wide given the limited number of events. Most NITs demonstrated positive associations with risk (HR range 1.82-5.39; p < 0.05), whereas C1M was inversely associated (HR 0.63, 95% CI 0.49-0.82; p < 0.001), indicating reduced risk. As a secondary objective, we evaluated whether each NIT could diagnose advanced fibrosis. FIB-4 and LSM effectively ruled out advanced fibrosis, while ELF and the Agile scores showed the strongest overall discriminative performance. For confirming advanced fibrosis, ELF (81%/76%) and Agile3+ (85%/79%) achieved the best sensitivity-specificity balance.
CONCLUSIONS
Our findings suggest that NITs may hold both diagnostic and prognostic value, helping to identify advanced fibrosis and stratify future risk of MALOs in MASLD which is a clinically heterogenous but clinically relevant outcome. Additional research is needed to evaluate their performance in primary care settings.
IMPACT AND IMPLICATIONS
In MASLD, NITs have the potential for identification of at-risk patients. Identifying NITs that predict beneficial as well as detrimental outcomes could also add value to monitoring strategies although further evidence is needed to evaluate this potential.
