RhoE downregulation leads to enhanced cholesterol biosynthesis and sorafenib resistance in hepatocellular carcinoma

Sorafenib remains the first-line systemic therapy for hepatocellular carcinoma (HCC), but its clinical efficacy is limited by acquired resistance. Here, we identified a significant association between acquired resistance to sorafenib and activation of the FAK signaling pathway. In vitro experiments confirmed that sorafenib suppresses Rho-related GTP-binding protein RhoE (RhoE) expression via Raf/MEK/ERK inhibition, thereby relieving its inhibitory effect on the transforming protein RhoA (RhoA) and Rho-associated protein kinase signaling pathway and ultimately leading to elevated phosphorylation of FAK at tyrosine 397 in HCC cells. Activated FAK subsequently promotes 3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR) expression via AKT signaling, thereby increasing intracellular cholesterol levels, which in turn upregulates glioma-associated oncogene homolog 1 (GLI1) expression and induces drug resistance. Genetic knockout of FAK or pharmacological inhibition using defactinib or PF-573228 effectively suppressed the sorafenib-induced upregulation of AKT and HMGCR. Notably, combination treatment with FAK inhibitors and sorafenib synergistically inhibited HCC cell viability, an effect reversed by HMGCR overexpression or exogenous cholesterol. In vivo and organoid experiments further demonstrated that combining sorafenib with defactinib significantly suppressed tumor growth and resistance signaling. Furthermore, bioinformatic analyses revealed that a gene signature related to cholesterol biosynthesis is significantly associated with poor prognosis in sorafenib-treated HCC patients, indicating its potential as a predictive biomarker. Collectively, this study identify systematically uncovers a new mechanism of sorafenib resistance in HCC: RhoE downregulation-mediated activation of the FAK/AKT-cholesterol-SHH/GLI1 axis as a key driver of sorafenib resistance and provide a rationale for combining FAK inhibitors with RAF-targeted therapies and for using cholesterol biosynthesis signatures to guide personalized treatment.