Novel digital pathology quantitative image analysis and AI method detects the treatment effects of NASH drug candidates with a performance that benchmarks imaging-based measurements

baseline fatigue score and the time to liver-related clinical events ( progression to histologic cirrhosis for F3, developing hepatic decompensation for F4).Results: There were 1679 advanced NASH patients (N = 802 F3 and 877 patients F4, age 58 ± 9 years, 40% male, 74% type 2 diabetes).During median follow-up of 16 months (IQR 14-18 month, max 26 months), 123 (15%) of F3 patients and 31 (3.5%)F4 patients experienced liver-related clinical events.The mean baseline CLDQ-NASH fatigue score was 4.77 ± 1.36 in F3 and 4.56 ± 1.44 in F4.In both F3 and F4 cohorts, patients with liver-related events in follow-up had higher baseline AST, lower platelet count, and elevated non-invasive test scores for fibrosis (APRI, FIB-4, ELF, Fibrotest, NAFLD Fibrosis Score) in comparison to patients who remained stable in follow-up (all p < 0.01).Their baseline fatigue scores were also significantly lower: 4.47 ± 1.36 in progressed F3 vs. 4.83 ± 1.35 in stable F3 ( p = 0.0091); 3.74 ± 1.31 in progressed F4 vs. 4.59 ± 1.43 in stable F4 ( p = 0.0011).In multivariate analysis, the association of lower fatigue scores (more fatigue) with the risk of liver-related events was significant: hazard ratio (HR) per 1 fatigue point = 0.84 (0.75-0.93), p = 0.0012, in all patients; HR = 0.85 (0.75-0.96), p = 0.0088, in F3; HR = 0.67 (0.53-0.85), p = 0.0014, in F4.These associations remained significant after adjustment for clinico-demographic confounders: adjusted HR = 0.79 (0.70-0.89), 0.85 (0.74-0.97), and 0.62 (0.48-0.81), respectively (all p < 0.02).Conclusion: Lower baseline fatigue scores were associated with adverse clinical events in patients with advanced NASH.A combination of baseline clinical and fatigue parameters can inform clinicians of which patients with advanced fibrosis due to NASH are at risk of adverse clinical outcomes.