Scientists stunned as new AI unlocks game-changing cancer treatment

Researchers at UCLA have developed an artificial intelligence system called MAESTRO that uses large language models to discover drug combinations that can overcome cancer treatment resistance. The platform analyzes existing scientific knowledge about drug mechanisms and cancer cell survival pathways, ranks potential drug pairings, and provides biological explanations for its predictions. Focusing on prostate cancer, the team used MAESTRO to screen 3,592 natural compounds for their ability to enhance the effectiveness of enzalutamide, a standard prostate cancer drug. The system identified 150 high-confidence candidates, and laboratory testing revealed that 23 compounds showed significant synergy with enzalutamide. Among them, berberine, a natural plant compound, demonstrated consistent synergy across four different prostate cancer models with varying resistance mechanisms. The researchers found that the combination works by increasing AMPK activity and suppressing mTORC1 signaling, leading to increased cellular stress and reduced cell growth. The team plans to further test this combination in patient-derived tumor models and animal studies, and to expand MAESTRO's capabilities for other cancers and drug combinations.

(Source：The Independent)