Outcome prediction in metabolic dysfunction-associated steatotic liver disease using stain-free digital pathological assessment.

Abstract

Computational quantification reduces observer-related variability in histological assessment of metabolic dysfunction-associated steatotic liver disease (MASLD). We undertook stain-free imaging using the SteatoSITE resource to generate tools directly predictive of clinical outcomes. Unstained liver biopsy sections (n = 452) were imaged using second-harmonic generation/two-photon excitation fluorescence (TPEF) microscopy, and all-cause mortality and hepatic decompensation indices constructed. The mortality index had greater predictive power for all-cause mortality (index >.14 vs. </=.14, HR 4.49, p = .003) than the non-alcoholic steatohepatitis-Clinical Research Network (NASH-CRN) (hazard ratio (HR) 3.41, 95% confidence intervals (CI) 1.43-8.15, p = .003) and qFibrosis stage (HR 3.07, 95% CI 1.30-7.26, p = .007). The decompensation index had greater predictive power for decompensation events (index >.31 vs. </=.31, HR 5.96, p < .001) than the NASH-CRN (HR 3.65, 95% CI 1.81-7.35, p < .001) or qFibrosis stage (HR 3.59, 95% CI 1.79-7.20, p < .001). These tools directly predict hard endpoints in MASLD, without relying on ordinal fibrosis scores as a surrogate, and demonstrate predictive value at least equivalent to traditional or computational ordinal fibrosis scores.