Clinical validation of the FLIP algorithm and the SAF score in patients with non-alcoholic fatty liver disease

• Published in: Journal of hepatology 2020-05, Vol.72 (5), p.828-838
• Main Authors: Nascimbeni, Fabio, Bedossa, Pierre, Fedchuk, Larysa, Pais, Raluca, Charlotte, Frédéric, Lebray, Pascal, Poynard, Thierry, Ratziu, Vlad
• Format: Article
• Language: English
• Subjects: Algorithms; Balloon treatment; Biopsy; Body mass index; Cirrhosis; Classification; Clinical trials; Fatty liver; Fibrosis; FLIP protein; Hemoglobin; Histology; Homeostasis; Insulin; Insulin resistance; Life Sciences; Liver cirrhosis; Liver diseases; Metabolism; Morphology; Non-alcoholic steatohepatitis; Risk factors; SAF score; Steatosis
• ISSN: 0168-8278; 1600-0641
• DOI: 10.1016/j.jhep.2019.12.008

Histological classifications used to diagnose/stage non-alcoholic fatty liver disease (NAFLD) are based on morphology, with undetermined clinical correlates and relevance. We assessed the clinical relevance of the fatty liver inhibition of progression (FLIP) algorithm and the steatosis, activity, and fibrosis (SAF) scoring system. One hundred and forty consecutive patients with suspected NAFLD and a separate validation cohort of 78 patients enrolled in a therapeutic trial, all with central reading of liver biopsy, were included. FLIP and SAF were used to categorize patients with non-alcoholic steatohepatitis (NASH), non-NASH NAFLD (NAFL), or non-NAFLD. The SAF activity score assessed hepatocyte ballooning and lobular inflammation; a histologically severe disease was defined as a SAF activity score of ≥3 and/or bridging fibrosis or cirrhosis. Clinical, biochemical, and metabolic data were analyzed in relation to histology. Patients with NASH according to the FLIP algorithm had a clinical profile distinct from those with NAFL, with a higher prevalence of metabolic risk factors (increased body mass index [BMI], central obesity, serum glucose, and glycated hemoglobin), more severe insulin resistance (fasting insulin and homeostasis model assessment for insulin resistance [HOMA-IR] values), and higher levels of aminotransferases. Similar findings were documented for patients with severe disease vs. those without. Positive linear trends existed between NASH or severe disease and increasing BMI and HOMA-IR. There was a strong association between liver fibrosis and NASH or SAF-defined scores of activity. Patients with either significant or bridging fibrosis overwhelmingly had NASH, and bridging fibrosis most often coexisted with severe activity. The FLIP algorithm/SAF score, although based on purely morphological grounds, are clinically relevant, as they identify patients with distinct clinical and biological profiles of disease severity. Disease activity in NAFLD is associated with fibrosis severity. The examination of liver tissue under the microscope (histology) serves to define the type and severity of non-alcoholic fatty liver disease morphologically, and is also used to determine improvement in therapeutic or natural history clinical trials. The FLIP algorithm/SAF classification is a new histological classification well validated on morphological but not clinical grounds. Here, we demonstrate that different disease categories defined by the FLIP/SAF classifi