GE Healthcare Validates Novel AI Models for Predicting Patient Responses to Immunotherapies using Data

Friday, November 03, 2023

GE Healthcare, in collaboration with Vanderbilt University Medical Center (VUMC) and the University Medicine Essen (UME) in Germany, has successfully developed AI models capable of predicting patient responses to immunotherapies with a notable degree of accuracy. These findings will be presented at the Society for Immunotherapy of Cancer (SITC) conference in San Diego. Originally trained on a dataset of over 3,000 immunotherapy patients from VUMC, these AI models were further validated on a dataset of 4,000 patients from UME. These models demonstrate their ability to forecast treatment outcomes and the likelihood of adverse reactions for individual patients.

The potential applications of these AI models are substantial. They have the potential to facilitate personalized precision medicine by assisting healthcare professionals in selecting the most suitable treatment paths at an earlier stage, potentially reducing unnecessary side effects and healthcare costs. Immunotherapies, which leverage the immune system to target and combat cancer cells, have demonstrated their potential to be more effective than conventional treatments. Nonetheless, they often yield low response rates and severe side effects.

Beyond their use in clinical settings, these AI models may also play a pivotal role in drug development. With approximately 5,000 immunotherapies currently in development, the capability to identify patients with a higher likelihood of positive responses could significantly expedite clinical trials and increase their success rates.

The development of these AI models involved a retrospective analysis by GE Healthcare and VUMC, correlating the treatment responses of thousands of cancer patients at VUMC with de-identified demographic, genomic, tumor, cellular, proteomic, and imaging data. Significantly, these AI models rely solely on routinely acquired data from electronic health records (EHR), making them adaptable and scalable for various potential applications.

These AI models are a vital component of GE Healthcare's immuno-oncology development portfolio, which encompasses the creation of innovative PET tracers. Notably, GE Healthcare recently announced the initiation of a Phase I clinical trial involving the first fluorine-18 PET radiopharmaceutical specifically designed for CD8.



Harvard Medical School - Leadership in Medicine Southeast Asia47th IHF World Hospital CongressHealthcare Innovation & Transformation SummitHealthcare CNO SummitHealthcare CMO SummitThe Healthcare Patient Experience & Engagement Summit 2024