Castera, Laurent

Background: Noninvasive tests (NITs) are used to risk-stratify metabolic dysfunction–associated steatotic liver disease. The aim was to survey global patterns of real-world use of NITs. Methods: A 38-item survey was designed by the Global NASH Council. Providers were asked about risks for advanced fibrosis, which NITs (cutoff values) they use to risk-stratify liver disease, monitor progression, and which professional guidelines they follow. Results: A total of 321 participants from 43 countries completed the survey (54% hepatologists, 28% gastroenterologists, and 18% other). Of the respondents, 85% would risk-stratify patients with type 2 diabetes, obesity (82%), or abnormal liver enzymes (73%). Among NITs to rule out significant or advanced fibrosis, transient elastography (TE) and fibrosis-4 (FIB-4) were most used, followed by NAFLD Fibrosis Score, Enhanced Liver Fibrosis, and magnetic resonance elastography. The cutoffs for ruling out significant fibrosis varied considerably between practices and from guidelines, with only 50% using TE <8 kPa, 65% using FIB-4 <1.30 for age <65, and 41% using FIB-4 <2.00 for age ≥65. Similar variability was found for ruling in advanced fibrosis, where thresholds of FIB-4 ≥2.67 and TE ≥10 kPa were used by 20% and 17%, respectively. To establish advanced fibrosis, 48% would use 2 NITs while 23% would consider 1 NIT, and 17% would confirm with liver biopsy. TE was used by >75% to monitor, and 66% would monitor (intermediate or high risk) annually. Finally, 65% follow professional guideline recommendations regarding NITs. Conclusions: In clinical practice, there is variability in NIT use and their thresholds. Additionally, there is suboptimal adherence to professional societies’ guidelines.

INTRODUCTION: The recent US Food and Drug Administration approval of resmetirom for treating metabolic dysfunction-associated steatohepatitis in patients necessitates patient selection for significant fibrosis or higher (≥F2). No existing vibration-controlled transient elastography (VCTE) algorithm targets ≥F2. METHODS: The mAchine Learning ADvanceD fibrosis and rIsk metabolic dysfunction-associated steatohepatitis Novel predictor (ALADDIN) study addressed this gap by introducing a machine-learning-based web calculator that estimates the likelihood of significant fibrosis using routine laboratory parameters with and without VCTE. Our study included a training set of 827 patients, a testing set of 504 patients with biopsy-confirmed metabolic dysfunction-associated steatotic liver disease from 6 centers, and an external validation set of 1,299 patients from 9 centers. Five algorithms were compared using area under the curve (AUC) in the test set: ElasticNet, random forest, gradient boosting machines, XGBoost, and neural networks. The top 3 (random forest, gradient boosting machines, and XGBoost) formed an ensemble model. RESULTS: In the external validation set, the ALADDIN-F2-VCTE model, using routine laboratory parameters with VCTE (AUC 0.791, 95% confidence interval [CI]: 0.764–0.819), outperformed VCTE alone (0.745, 95% CI 0.717–0.772, P < 0.0001), FibroScan-aspartate aminotransferase (0.710, 0.679–0.748, P < 0.0001), and Agile-3 model (0.740, 0.710–0.770, P < 0.0001) regarding the AUC, decision curve analysis, and calibration. The ALADDIN-F2-Lab model, using routine laboratory parameters without VCTE, achieved an AUC of 0.706 (95% CI: 0.668–0.749) and outperformed Fibrosis-4, steatosis-associated fibrosis estimator, and LiverRisk scores. DISCUSSION: Along with the steatosis-associated fibrosis estimator model developed to target significant fibrosis or higher, ALADDIN-F2-VCTE (https://aihepatology.shinyapps.io/ALADDIN1) uniquely supports a refined noninvasive approach to patient selection for resmetirom without the need for liver biopsy. In addition, ALADDIN-F2-Lab (https://aihepatology.shinyapps.io/ALADDIN2) offers an effective alternative when VCTE is unavailable.

Abstract Cirrhosis, highly prevalent worldwide, develops after years of hepatic inflammation triggering progressive fibrosis. Currently, the main etiologies of cirrhosis are non‐alcoholic fatty liver disease and alcohol‐related liver disease, although chronic hepatitis B and C infections are still major etiological factors in some areas of the world. Recent studies have shown that liver fibrosis can be assessed with relatively high accuracy noninvasively by serological tests, transient elastography, and radiological methods. These modalities may be utilized for screening for liver fibrosis in at‐risk populations. Thus far, a limited number of population‐based studies using noninvasive tests in different areas of the world indicate that a significant percentage of subjects without known liver disease (around 5% in general populations and a higher rate −18% to 27%‐in populations with risk factors for liver disease) have significant undetected liver fibrosis or established cirrhosis. Larger international studies are required to show the harms and benefits before concluding that screening for liver fibrosis should be applied to populations at risk for chronic liver diseases. Screening for liver fibrosis has the potential for changing the current approach from diagnosing chronic liver diseases late when patients have already developed complications of cirrhosis to diagnosing liver fibrosis in asymptomatic subjects providing the opportunity of preventing disease progression.