The accuracy of FibroScan, FIB-4, and nonalcoholic fatty liver disease fibrosis score in predicting biopsy-defined fibrosis and steatosis across all fibrosis stages in patients with metabolic dysfunction associated steatotic liver disease.

Liver biopsy is the gold standard for quantifying steatosis and fibrosis. It is unclear how noninvasive tests (NITs) accurately correlate to liver biopsy. The aim of this study was to characterize patients with metabolic-associated steatotic liver disease who underwent liver biopsy in South Texas and had at least 1 contemporaneous NIT result available to determine the accuracy of NITs as compared to liver biopsy in accurately staging fibrosis and steatosis. The study included 460 patients with liver biopsy and at least 1 NIT. Data captured was based on standard of care and was non-interventional in nature. Performance characteristics of the NITs were analyzed based on the degree of fibrosis defined by liver biopsy. The majority of patients were female (66.4%), middle-aged (51 years), and Hispanic/Latino (73.3%). In patients with F3/F4 fibrosis, FibroScan accurately identified only 45.9% with advanced fibrosis. Even when using society recommended results from fibrosis-4 combined with FibroScan, concordance with liver biopsy was only reported in 68.9% of patients. Patients with biopsy-defined advanced steatosis (S3) were identified as having advanced steatosis by FibroScan controlled attenuation parameter score in 89.7% of patients; however, controlled attenuation parameter score overly predicted advanced steatosis in 81.4% with biopsy-defined S1/S2 steatosis. NITs should be used as first line to assess steatosis and fibrosis; however, the optimal combination of these tests has not been elucidated and properly compared to liver biopsy to assess true diagnostic accuracy. Until the optimal set of NITs are found, any diagnostic or clinical inconsistencies should be resolved with liver biopsy to avoid staging errors, particularly underestimating fibrosis in those with advanced disease.