**3.13 Farnesoid X receptor agonist**

A crucial regulator of metabolic pathways, including glucose homeostasis, inflammation, and fibrosis, is the farnesoid X receptor (FXR). In patients with NASH, the level of hepatic FXR expression is closely associated with the disease's severity. The liver, kidneys, gut, and adrenal glands all express it. It controls the metabolism of lipoproteins and participates in the production and enterohepatic circulation of bile acids. Bile acid synthesis, hepatic lipogenesis, cholesterol synthesis, and glucose homeostasis are all directly impacted by FXR activation. In animal models, the treatment of FXR agonists has been shown to resolve steatohepatitis and fibrosis as well as prevent the development of NASH. For the treatment of NASH, several synthetic FXR agonists are currently being developed [38].

Bile acids are cholesterol metabolites that are produced in the liver and absorbed from dietary lipids. Type 2 diabetes mellitus raises their levels. In animal models, nor-ursodeoxycholic acid, a synthetic bile acid homolog, has been demonstrated to decrease liver enzymes, fibrosis, and inflammation. Obeticholic acid (OCA), a different modified bile acid, activates the FXR in individuals and raises insulin sensitivity. The FLINT trial, a double-blind placebo, controlled, randomized clinical trial assessed the effectiveness of OCA in patients with NASH without cirrhosis and NAFLD activity score (NAS) > 4 for 72 weeks. Most of these patients had a decrease in the indices of liver fibrosis and inflammation, but they also showed elevated levels of LDL-C, insulin, and a decrease in HDL-C. Because of this, the improvement in liver markers was offset by a worsening of the lipid profile, which required statin therapy. Another ongoing trial (REGENERATE) also demonstrated a slight reduction in fibrosis as compared to placebo, although it did not completely reverse NASH [39].

Tropifexor is another very potent non-bile acid agonist of FXR, which has been shown to be very effective in NASH in animal models and it is under evaluation [40, 41].
