Rational Design of Benzofuran‐Based mTOR Inhibitors for Head and Neck Cancer: A Synergistic 3D‐QSAR, Molecular Docking, and DFT Approach

ABSTRACT This work presents a novel computational pipeline for designing benzofuran‐derived mTOR inhibitors, combining advanced machine learning techniques with molecular modeling approaches. Using a carefully selected set of 52 benzofuran analogs, we established a highly predictive 3D‐QSAR model (R 2 = 0.94, Q 2 = 0.78) that identified crucial molecular features influencing inhibitory activity. Through systematic virtual screening, compound A7 emerged as the most promising candidate, exhibiting exceptional binding energy (‐8.49 kcal/mol) to the mTOR catalytic domain through specific interactions with Val2240 (1.86 Å hydrogen bond) and Tyr2225 (π‐stacking). Quantum mechanical analyses uncovered distinctive electronic properties of A7, including a small HOMO‐LUMO energy separation (2.0 eV) and favorable charge distribution patterns. Comprehensive pharmacokinetic evaluation revealed optimal drug‐like characteristics for A7, with excellent predicted oral bioavailability (96% absorption) and minimal toxicity concerns. Our integrated computational strategy demonstrates the potential of benzofuran derivatives as mTOR‐targeted therapeutics while providing a validated protocol for structure‐based drug discovery.