Glioblastoma (GBM) remains the most aggressive and lethal form of primary brain tumor, characterized by a poor prognosis and limited therapeutic options. Among the molecular targets under investigation, the epidermal growth factor receptor (EGFR) plays a central role in GBM pathogenesis. However, current EGFR inhibitors are often limited by challenges such as drug resistance and inadequate penetration across the blood-brain barrier (BBB). This study aims to investigate Withanolide A and its AI-optimized derivatives as potential EGFR inhibitors for GBM using computational drug discovery approaches. The 3D structure of EGFR was retrieved and analyzed to predict active binding sites, followed by de novo drug design generating novel drug candidates from Withanolide A. These compounds were evaluated through molecular docking to predict binding affinity and key interactions, followed by protein-ligand interaction profiling (PLIP) and pharmacophore modeling to assess complex stability. ADMET profiling was performed to evaluate pharmacokinetic properties, BBB permeability, and toxicity, while molecular dynamics simulations assessed stability of ligand-protein complex over time, integrating all analyses to identify the most promising lead candidate. Molecular docking analysis revealed that the newly designed drug candidates, particularly an AI-optimized derivative of drug candidate B, exhibited highest binding energy (up to −9.8 kcal/mol) compared to the natural ligand. Protein–ligand interaction profiling (PLIP) highlighted key hydrogen bonding and hydrophobic interactions that contributed to complex stability. Pharmacophore modeling and ADMET profiling demonstrated favorable drug-likeness, blood-brain barrier permeability, and low toxicity profiles, showing the anti-cancer inhibitory effect of optimized drug candidate to overcome limitations of existing EGFR inhibitors. ADMET analysis of the lead compound indicated favorable pharmacokinetic and drug-likeness properties, including high blood-brain barrier penetration, optimal oral bioavailability, and Caco-2 permeability. Importantly, the compound showed low cardiotoxicity risk (hERG blockers: 0.104), no AMES mutagenicity (0.749), minimal skin sensitization, and acceptable hepatotoxicity (0.788), suggesting a low systemic toxicity profile. Metabolic profiling predicted good liver microsomal stability and moderate interaction with CYP3A4 and CYP2C19 enzymes, indicating a balanced metabolic clearance profile. AI-assisted bioisosteric optimization of Withanolide A produced a promising EGFR inhibitor with favorable pharmacokinetic and safety profiles. These findings highlight its potential as a promising lead compound for the development of targeted therapies against GBM and pave the way for further experimental validation.
Naveed et al. (Sun,) studied this question.