AI Boosts CAR T-Cell Therapy Target Discovery

Researchers at Penn Medicine have developed a human-in-the-loop AI framework to accelerate the discovery of targets for CAR T-cell therapy ^[1][2]. This system utilizes large language models to analyze vast biological datasets and identify viable antigens for treatment ^[2]. Through this process, the team successfully identified GPNMB, a protein that serves as a potential target across several types of cancer ^[1][3].

By combining machine learning with human expertise, the framework filters complex data more efficiently than traditional methods ^[2]. This streamlined approach allows scientists to pinpoint therapeutic targets that were previously difficult to isolate ^[3]. Experts believe this advancement could significantly improve the development of effective treatments for patients with solid tumors ^[1].