icon-folder.gif   Conference Reports for NATAP  
  The International Liver Congress™
EASL - European Association for the
Study of the Liver
June 23-26 2021
Back grey_arrow_rt.gif
New FibroScan Score Improves
Advanced Fibrosis Detection in NAFLD

  EASL International Liver Congress, June 23-26, 2021
Mark Mascolini
A novel FibroScan-based score improved diagnosis of advanced fibrosis in an international group of 3761 people with nonalcoholic fatty liver disease (NAFLD) [1]. The new score improved sensitivity of ruling in advanced fibrosis and cut the rate of cases with indeterminate results.
Detecting advanced fibrosis (F3 or higher) and cirrhosis in people with NAFLD is important because only those with advanced fibrosis run a significantly higher risk of liver-related death. For that reason, international guidelines advise evaluating NAFLD patients for advanced fibrosis. Current methods-FIB-4 and liver stiffness measured by Vibration-Controlled Transient Elastography (VCTE) do well in excluding advanced fibrosis, but they have only moderate ability to rule in advanced fibrosis.
The recently introduced Agile 4 score combines liver stiffness with simple clinical measures to help rule in cirrhosis. This new study set out to develop and validate a score called Agile 3+ to identify advanced fibrosis in people with NAFLD by combining liver stiffness with standard clinical parameters.
The analysis included adults with a liver biopsy for suspected NAFLD, a liver stiffness measure by VCTE within 6 months of the liver biopsy, and blood samples collected within 1 month of VCTE. The study excluded people with nonmetabolic comorbidities that can cause liver disease (like viral hepatitis and HIV infection).
A pooled analysis involved 2134 people in 7 global cohorts, 54% of whom had F3 or greater fibrosis. Researchers divided the group into a training set of 1434 people (54% F3 or greater) and an internal validation set of 700 people (54% F3 or greater). A non-US validation group included 585 people from the Nonalcoholic Steatohepatitis Clinical Research Network (NASH CRN) (37% F3 or greater). A 3-site French validation set included 1042 people (38% F3 or greater).
With the training set the investigators considered 14 clinical variables to combine with liver stiffness. They used multivariate logistic regression to rule out or rule in advanced fibrosis, aiming to decrease numbers of indeterminate cases and increase positive predictive value in ruling in advanced fibrosis.
Age averaged in the mid-50s in the training set (55.0), the internal validation set (55.5), the NASH CRN validation set (54.0), and the French validation set (58.0). About half of participants in the training set (50.8%) and the internal validation set (51.3%) were men, compared with 37.4% in the NASH CRN validation set and 59.7% in the French validation set. Across those 4 groups, proportions with F3 fibrosis were 30.5%, 30.7%, 23.8%, and 25.6%-and with F4 fibrosis 23.4%, 23.6%, 12.8%, and 12.8%.
Logistic regression yielded an Agile 3+ score combining liver stiffness by VCTE with the variables AST/ALT ratio, platelet count, diabetes status, sex, and age. Receiver operating characteristic (ROC) curve analysis (with specificity on the x axis and sensitivity on the y axis) determined a slightly higher area under the curve (AUC) with Agile 3+ (0.90) than with FIB-4 (0.82), or liver stiffness (0.86). Adjusted positive predictive value was 0.81 with Agile 3+, 0.76 with FIB-4, and 0.78 with liver stiffness. Respective adjusted negative predictive values were 0.90, 0.87, and 0.89.
With Agile 3+ fewer than 20% of participants had indeterminate advanced fibrosis results, compared with about 30% with FIB-4 and 20% with liver stiffness. Rates of ruling in or ruling out advanced fibrosis were marginally better (just over 40%) with Agile 3+ than with FIB-4 or liver stiffness.
Calibration plots determined that Agile 3+ performed as well in the internal validation set (AUC 0.90) as in the training set, and virtually as well in the NASH CRN validation set (AUC 0.86) and the French validation set (AUC 0.87). As in the training set, rates of indeterminate advanced fibrosis results were always lower in all three validation sets with Agile 3+ than with FIB-4 or liver stiffness. In the validation sets, rates of ruling in or ruling out advanced fibrosis were generally similar with Agile 3+, FIB-4, and liver stiffness.
The researchers believe Agile 3+ will lower the need for confirmatory liver biopsy in people with NAFLD. The Agile 3+ score-and the Agile 4 score for identifying cirrhosis-are public and available on the myFibroScan app at Google Play and the Apple App Store [2].
1. Younossi ZM, Harrison SA, Newsome PN, et al. Development and validation of Agile 3+: novel FibroScan-based score for the diagnosis of advanced fibrosis in patients with non-alcoholic fatty liver disease. EASL International Liver Congress, June 23-26, 2021. Abstract OS-555.
2. myFibroScan-a smart application that streamlines your management of chronic liver diseases. YouTube. https://www.youtube.com/watch?v=BFsuE5Aj0_A