AI Predicts Potent Cancer-Killing TILs for Personalized Immunotherapies

Using artificial intelligence, Ludwig Cancer Research scientists have developed a powerful predictive model called TRTpred that can identify the most potent tumor-infiltrating lymphocytes (TILs)—immune cells capable of recognizing and attacking cancer cells. Combined with additional algorithms, this model enables personalized cancer treatments tailored to each patient's unique tumor makeup.

TRTpred utilizes machine learning to analyze gene expression patterns and distinguish tumor-reactive TILs from inactive ones. Trained on data from melanoma patients, it can reportedly predict with 90% accuracy which TILs will effectively target tumors across various cancer types.

To further refine the selection, the researchers applied a secondary algorithm to screen for TILs with high avidity, meaning they bind strongly to tumor antigens. A third filter maximizes recognition of diverse tumor antigens by selecting TILs with distinct antigen targets.

This combined approach, called MixTRTpred, identifies TILs that are tumor-reactive, have high avidity, and target multiple tumor antigens—key characteristics for effective cancer immunotherapy. In preclinical studies, the researchers used MixTRTpred to engineer T cells expressing the identified potent TIL receptors. These engineered T cells successfully eliminated tumors in mice, validating the approach's potential. 

"This method promises to overcome some of the shortcomings of current TIL-based therapy, especially for patients dealing with tumors not responding to such therapies today," said George Coukos, co-author of the study published in Nature Biotechnology, who plans to launch a clinical trial testing this technology in patients.