Integrative computational strategy for anticancer drug discovery: QSAR-ANN modeling, molecular docking, ADMET prediction, molecular dynamics and MM-PBSA simulations, and retrosynthetic analysis

Abstract

Breast cancer constitutes a primary cause of mortality among women. Existing therapeutic targets and treatment modalities are often confronted with drug resistance and the considerable financial burden associated with the development of new therapies, for which the results can be uncertain. Hormone therapy, which mainly focuses on inhibiting aromatase as a pivotal enzyme in estrogen biosynthesis, remains the preferred approach for treating this type of cancer while minimizing costs owing to advanced computer-aided drug design (CADD) methods. In this work, a combined strategy of 3D-QSAR, artificial neural networks (ANN), molecular docking, ADMET analysis, molecular dynamics (MD) simulations, and retrosynthesis was applied to design novel anti-breast cancer agents and study their interactions with aromatase to identify potential inhibitors. The predictive models underwent rigorous internal and external validations based on significant statistical parameters, confirming their robustness and reliability. As a result, 12 new drug candidates (L1–L12) were designed against breast cancer. Based on the results of virtual screening techniques, only one hit (L5) showed significant potential compared with the reference drug (exemestane) and previously designed drug candidates (ligand 5 and C2). Subsequent stability studies and pharmacokinetic evaluations reinforced the potential of L5 as an effective aromatase inhibitor. Retrosynthesis was used to optimize the proposed synthesis of this candidate, which required in vitro and in vivo validation.