Machine learning liver histology scores correlate with portal hypertension assessments in nonalcoholic steatohepatitis cirrhosis

Abstract

Summary Background and Aims In cirrhotic nonalcoholic steatohepatitis (NASH) clinical trials, primary efficacy endpoints have been hepatic venous pressure gradient (HVPG), liver histology and clinical liver outcomes. Important histologic features, such as septa thickness, nodules features and fibrosis area have not been included in the histologic assessment and may have important clinical relevance. We assessed these features with a machine learning (ML) model. Methods NASH patients with compensated cirrhosis and HVPG ≥6 mm Hg (n = 143) from the Belapectin phase 2b trial were studied. Liver biopsies, HVPG measurements and upper endoscopies were performed at baseline and at end of treatment (EOT). A second harmonic generation/two‐photon excitation fluorescence provided an automated quantitative assessment of septa, nodules and fibrosis (SNOF). We created ML scores and tested their association with HVPG, clinically significant HVPG (≥10 mm Hg) and the presence of varices (SNOF‐V). Results We derived 448 histologic variables (243 related to septa, 21 related to nodules and 184 related to fibrosis). The SNOF score (≥11.78) reliably distinguished CSPH at baseline and in the validation cohort (baseline + EOT) [AUC = 0.85 and 0.74, respectively]. The SNOF‐V score (≥0.57) distinguished the presence of varices at baseline and in the same validation cohort [AUC = 0.86 and 0.73, respectively]. Finally, the SNOF‐C score differentiated those who had >20% change in HVPG against those who did not, with an AUROC of 0.89. Conclusion The ML algorithm accurately predicted HVPG, CSPH, the development of varices and HVPG changes in patients with NASH cirrhosis. The use of ML histology model in NASH cirrhosis trials may improve the assessment of key outcome changes.

Endpoints in NASH Clinical Trials: Are We Blind in One Eye?

“…These individuals with “at-risk NASH” should be followed with surrogate markers and algorithms predicting both liver fibrosis regression and progression to cirrhosis and other clinically relevant liver-related and cardiovascular outcomes [ 193 ]. Artificial intelligence technology might also help, and machine learning algorithms accurately predict the hepatic venous pressure gradient (HVPG), clinically significant portal hypertension, and development of esophageal varices and HVPG changes in patients with NASH-related cirrhosis [ 194 ].…”

Noninvasive Diagnostic Methods in Liver Cirrhosis

“…Additionally, a convolutional neural network model based on trichrome-stained liver biopsy slides has been developed to predict CSPH in patients with biopsy-proven NASH cirrhosis [57]. Another recent advancement involves an ML model based on histologic features, capable of accurately predicting HVPG, CSPH, the development of varices, and changes in HVPG in NASH cirrhosis [58]. Liu et al innovatively utilized ML to accurately extract liver capsule features from high-frequency ultrasound images, aiding in the diagnosis of cirrhosis [59].…”

Exploring Opportunities to Enhance the Screening and Surveillance of Hepatocellular Carcinoma in Non-Alcoholic Fatty Liver Disease (NAFLD) through Risk Stratification Algorithms Incorporating Ultrasound Elastography

“…Machine learning and the broader field of artificial intelligence could represent a roadmap to the future development of novel biomarkers, could predict liver-related events, including the development of HCC, and overall improve the assessment of key outcome changes in NAFLD [ 185 , 186 , 187 , 188 ].…”