University scientists develop artificial intelligence tool to personalize cancer therapies

Researchers at the Institute of Science of the data and Artificial Intelligence (DATAI) of the University of Navarra have developed artificial intelligence models to personalize immune therapies in oncology patients.

The study analyzes data of more than 3,000 patients with lung cancer and bladder cancer, two of the most frequently diagnosed cancers in Spain in 2024, according to the Spanish Society of Medical Oncology. Using machine learning models, the researchers have identified new genetic signatures specific to each stage of the disease and developed a system, called the "IFIT index" (index of "physical immunity"), which will allow therapies to be personalized, improving their effectiveness.

The IFIT index is a measure or score that evaluates a patient's 'aptitude immunological', which allows patients to be classified according to the risk they present at each stage of the disease. "This can help predict response to therapy based on the activity of their immune system at different stages of cancer treatment," explains Rubén Armañanzas, leader of DATAI's laboratory Digital Medicine and one of the lead authors of the study.

According to the expert, "immunotherapy represents one of the most promising frontiers in the fight against cancer, and by using artificial intelligence models, we can further fine-tune treatments based on each patient's immune profile ."

The University of Navarra study has been presented in Houston (USA), during the lecture of the Society for Immunotherapy of Cancer(SITC 2024). This meeting brings together international leaders from academia, regulatory and governmental agencies, as well as representatives from the pharmaceutical industry, to offer the latest advances in cancer immunotherapy.

IFIT index: system to personalize cancer treatments

The research, which has been nominated as one of the top 100 at lecture, is based on the analysis of the cancer immunity cycle (CIC), which determines how signals from the immune system influence the effectiveness of immunotherapy treatment. On this basis and using artificial intelligence tools, the researchers propose specific patterns of cellular activity according to the molecular stage of the disease and have created the IFIT "physical immunity" index. This breakthrough underscores the relevance of artificial intelligence in personalized medicine, offering new hope in the fight against cancer. The researchers note that the technique will continue to be refined in future joint programs of study with other types of cancer.

The work results from the research camp organized by the Roche Institute for Cancer Centers at network imCORE, an international network that brings together leading centers of excellence in immuno-oncology from around the world to collaborate in the search for innovative approaches to cancer. This global partnership , involving the Cancer Center Clínica Universidad de Navarra and other leading cancer institutions research from 10 countries around the world, underscores the collective effort in the search for innovative approaches to cancer. The Government of Navarra also supports some members of the study.

quotation from article:
Aghababazadeh FA, Alonso L, Lopez-de-Castro M, et al 1197 Harnessing the cancer immunity cycle via machine learning models to generate novel strategies for personalized cancer therapy.Journal for ImmunoTherapy of Cancer 2024;12: doi: 10.1136/jitc-2024-SITC2024.1197