Deep Learning-Driven Anticancer Drug Discovery: Emodepside as a Potential Therapeutic Candidate for Triple-Negative...

Area of Science:

• Oncology
• Computational Biology
• Drug Discovery

Background:

• Triple-negative breast cancer (TNBC) is aggressive with limited treatment options.
• Drug repurposing accelerates discovery but faces biological complexity.
• Novel therapeutic strategies are crucial for TNBC management.

Purpose of the Study:

• Develop a deep learning framework for accelerated TNBC drug discovery.
• Identify novel therapeutic candidates and potential drug targets for TNBC.
• Validate drug efficacy and elucidate molecular mechanisms.

Main Methods:

• Deep neural network for predicting compound efficacy and toxicity.
• Screening of over 6,000 compounds from the Drug Repurposing Hub.
• In vitro and in vivo validation, transcriptomic sequencing, and proteomics.

Main Results:

• Identified emodepside, a structurally unique compound with potent anti-TNBC efficacy.
• Emodepside inhibited tumor growth in xenograft models.
• NAMPT identified as the primary molecular target via multiomics analysis.

Conclusions:

• Deep learning effectively identifies novel, structurally distinct anticancer agents.
• Emodepside is a promising therapeutic candidate for TNBC.
• Emodepside may exert anti-TNBC effects by inhibiting NAMPT and promoting apoptosis.