Quantitative digital pathology reveals reversal of MASH-related liver damage following bariatric surgery

Abstract Metabolic dysfunction-associated steatohepatitis (MASH) is a progressive liver disease characterized by steatosis, inflammation, hepatocellular injury, and fibrosis that may progress to cirrhosis or hepatocellular carcinoma. While laparoscopic sleeve gastrectomy (LSG) effectively treats severe obesity and improves liver histology, traditional pathological assessments rely on subjective evaluations. This study aimed to objectively assess LSG impact on MASH histological outcomes using digital pathology. We conducted a retrospective study analyzing paired liver biopsies from 44 patients who required a second bariatric intervention due to suboptimal outcomes following initial LSG. Wedge biopsies were obtained during both surgical procedures using standardized techniques, allowing comparisons between specimens obtained at the initial intervention and post-LSG. The liver tissue sections were digitized and analyzed using LiverExplore, an artificial intelligence (AI)-powered digital pathology tool to quantify hepatic cellular features, tissue architecture, and collagen characteristics. LSG resulted in significant improvements in key histopathological parameters, including reduced hepatic steatosis (p = 7.2×10⁻ 9 ) and hepatocellular ballooning (p = 2.6×10⁻⁴). While overall fibrosis scores remained unchanged, distinct fibrotic subtypes demonstrated marked improvement, with significant reductions in periportal fibrosis (p = 2.2×10⁻ 4 ), incomplete septal fibrosis (p = 6.9×10⁻⁶), and complete septal fibrosis (p = 6.6×10⁻⁴). The hepatic immune landscape showed comprehensive reorganization with increased neutrophil (p = 6.4×10⁻⁵) and eosinophil (p = 5.3×10⁻⁴) infiltration, alongside zonal redistribution of immune cells throughout the liver parenchyma. AI-enhanced digital pathology enables objective assessment of hepatic changes post-bariatric surgery. LSG improved key MASH features. Post-surgical increase in neutrophils and eosinophils along with zonal immune reorganization suggests immune-mediated tissue remodeling contributing to recovery. These findings support LSG efficacy for MASH treatment and demonstrate AI utility in capturing complex pathological changes.