Abstract
Anthelmintic resistance (AR) poses an escalating threat to global soil-transmitted helminth control efforts, particularly in the context of mass drug administration (MDA) programs that rely heavily on benzimidazole drugs. To address the urgent need for novel therapeutics, a machine learning (ML) guided drug repurposing pipeline was developed to identify compounds effective against a multi-anthelmintic drug-resistant (MADR) isolate of Ancylostoma caninum. The computational strategy employed in this work involved training classifiers on curated datasets of anthelmintic actives and decoys, followed by screening all approved drugs from the online database, DrugBank. Selected compounds were tested using a tiered assay system comprising egg hatch assays (EHA), larval development assays (LDA), and adult survival assays. Among these, flutamide, a nonsteroidal antiandrogen approved for prostate cancer, demonstrated potent, egg- and larval-stage-specific activity against MADR hookworm. Flutamide exhibited a concentration-dependent inhibition of egg hatching and arrested larval development in MADR hookworm. These findings underscore the translational value of drug repurposing to accelerate therapeutic discovery against MADR helminths.