Using computational tools, researchers from the Johns Hopkins Kimmel Cancer Center and the Johns Hopkins University School of Medicine have developed a method to predict which patients with a primary liver cancer called hepatocellular carcinoma (HCC) might benefit from specific immunotherapy combinations. The study, published in the journal Nature Cancer on July 15, 2026, analyzed tumor samples from 200 patients to train an AI model that identifies biomarkers for treatment response.
The AI model, named ImmunoPredict-HCC, was validated on an independent cohort of 150 patients from three clinical trials. It achieved an accuracy of 85% in predicting which patients would respond to combinations of immune checkpoint inhibitors and targeted therapies. The researchers emphasized that the tool is not yet ready for clinical use but could help design personalized treatment plans in the future.
Lead author Dr. Sarah Thompson, a computational biologist at Johns Hopkins, stated, 'This approach allows us to move beyond trial-and-error in immunotherapy selection. By leveraging AI, we can identify the most promising combinations for individual patients, potentially improving outcomes and reducing side effects.' The study was funded by the National Cancer Institute and the Lustgarten Foundation.
