Objectives: Hepatocellular carcinoma (HCC) remains one of the leading causes of cancer-related mortality worldwide. Despite the availability of systemic therapies, treatment resistance and tumor progression remain major challenges in HCC management. Increasing evidence indicates that aberrant activation of the fibroblast growth factor receptor 4 (FGFR4) axis and the phosphoinositide 3-kinase (PI3K) signaling pathway plays a critical role in hepatocarcinogenesis, tumor proliferation, and therapeutic resistance. Drug repositioning strategies offer an efficient approach for identifying new therapeutic applications for widely used non-oncologic drugs. This study aimed to investigate the potential inhibitory interactions of commonly prescribed drugs—atorvastatin, metformin, and celecoxib—against FGFR4 and PI3Kα through molecular docking analysis and to compare their binding profiles with sorafenib, a reference drug used in HCC treatment.
Methods: The crystal structures of FGFR4 and PI3Kα were retrieved from the Protein Data Bank and prepared using AutoDockTools. The three-dimensional structures of atorvastatin, metformin, celecoxib, and sorafenib were obtained from the PubChem database. Molecular docking simulations were performed using AutoDock 4 employing the Lamarckian Genetic Algorithm. The docking protocol was validated by redocking the co-crystallized ligand into the ATP-binding pocket, and root mean square deviation (RMSD) values below 2.0 Å were considered acceptable. Binding energies (ΔG), ligand–protein interaction profiles, and pharmacokinetic properties were analyzed using SwissADME.
Results: Docking simulations revealed that atorvastatin exhibited the strongest binding affinity toward both FGFR4 (-9.94 kcal/mol) and PI3Kα (-9.10 kcal/mol), demonstrating binding energies comparable to or stronger than the reference inhibitor. Celecoxib also showed notable binding affinity toward PI3Kα (-8.79 kcal/mol), whereas sorafenib demonstrated moderate binding interactions. In contrast, metformin exhibited relatively weak binding energies for both targets. Interaction analysis revealed that atorvastatin formed stabilizing hydrogen bonds and hydrophobic contacts with key residues within the ATP-binding pockets of FGFR4 and PI3Kα. ADMET prediction indicated that all investigated compounds satisfied Lipinski’s rule of five and displayed generally acceptable pharmacokinetic properties.
Conclusion: These findings suggest that atorvastatin may interact strongly with both FGFR4 and PI3Kα signaling proteins, highlighting its potential as a dual-target modulator in hepatocellular carcinoma. The results provide preliminary in silico evidence supporting the repositioning of commonly prescribed drugs in HCC therapy, warranting further experimental and clinical validation.
Keywords: Hepatocellular carcinoma, FGFR4, PI3K, drug repositioning, molecular docking, Sorafenib