Abstract
BACKGROUND: Regorafenib is a multikinase inhibitor widely used in oncology, but it is associated with significant adverse events (AEs), particularly hepatotoxicity. Understanding the real-world safety profile and the molecular mechanisms underlying regorafenib-induced liver failure is critical for clinical risk management. METHODS: AE reports with regorafenib as the primary suspect drug were extracted from the FDA Adverse Event Reporting System (FAERS) from Q3 2012 to Q1 2025. Disproportionality analyses were conducted using reporting odds ratio (ROR), proportional reporting ratio (PRR), multi-item gamma Poisson shrinker (MGPS), and Bayesian confidence propagation neural network (BCPNN). Clinical prioritization was assessed via a semi-quantitative framework. Time-to-onset was analyzed using Weibull distribution. Additionally, network toxicology and molecular docking were employed to explore potential mechanisms and binding affinities between regorafenib and liver failure targets. RESULTS: A total of 9,442 AE reports were analyzed. Hepatobiliary disorders exhibited the strongest signal strength (ROR = 2.66, 95% CI: 2.50-2.83), with hepatic failure and fulminant hepatitis identified as high-priority AEs. The majority of AEs occurred within 30 days of treatment initiation. Subgroup analysis indicated that elderly patients (≥65 years) and males had higher risks of liver failure, whereas higher body weight appeared protective. Network analysis identified 63 overlapping targets, with MAPK and PI3K-Akt signaling pathways significantly enriched. CONCLUSION: This pharmacovigilance analysis identifies regorafenib-associated hepatotoxicity as a significant safety signal, with higher risks observed in elderly and male patients. Computational predictions suggest the involvement of MAPK and PI3K-Akt pathways, offering a theoretical basis for further mechanistic investigation and supporting the need for vigilant clinical monitoring.