Cervical cancer, predominantly induced by high-risk human papillomavirus (HPV) strains, remains a significant global health burden despite advancements in cancer therapeutics. Among viral oncoproteins, the HPV-16 E6 protein plays a pivotal role in carcinogenesis by targeting tumor suppressor pathways, making it an attractive therapeutic target. This study introduces an integrated computational framework for systematically repurposing FDA-approved anticancer drugs as potential E6 inhibitors. A library of 100 anticancer drugs was screened using machine learning models, with Gradient Boosting achieving optimal predictive performance (MCC = 1.0, AUC = 1.0), identifying 21 promising candidates. Molecular docking analyses further prioritized five lead compounds—Axitinib, Cabozantinib, Tivozanib, Deflazacort, and Cilostazol—with strong binding affinities (-8.13 to -9.24 kcal/mol) to the E6 protein. Quantum chemical calculations provided insights into the electronic properties and structure-activity relationships of these compounds, underscoring their potential inhibitory mechanisms. This study highlights the power of machine learning-driven approaches in drug discovery. It identifies clinically approved candidates for further experimental validation as HPV-16 E6 inhibitors, paving the way for innovative therapeutic strategies in HPV-associated malignancies. However, as the findings are based on computational screening, further in vitro and in vivo validation is required to confirm the inhibitory potential of the identified compounds against HPV-16 E6.
Ahmadi et al. (Tue,) studied this question.