AI Models Show Promise in Personalizing Cancer Immunotherapy Treatments

GE HealthCare artificial intelligence (AI) models were able to accurately forecast effectiveness and toxicities of cancer immunotherapies for patients, according to an article in the Journal of Clinical Oncology Clinical Cancer Informatics.

The AI model used only two manually collected data points – “smoking status and number of prior immune checkpoint inhibitor (ICI) drugs.” Otherwise, the model used only “routinely collected electronic health record (EHR) data” to make its predictions. The study authors specifically used this data with the hope that “these models would be able to be implemented in any clinical setting.”

The architects of the study “retrospectively analyzed and correlated the immunotherapy treatment response of thousands of…cancer patients, with their deidentified demographic, genomic, tumor, cellular, proteomic, and imaging data. The models were trained to predict efficacy outcomes and the likelihood of an individual patient developing an adverse reaction, providing information that may help clinicians select the most appropriate treatment pathway sooner while potentially sparing unnecessary side effects and cost.”

The study showed that these models “have demonstrated the ability to predict patients’ responses to immunotherapies with 70 to 80 percent accuracy based on a pan-cancer cohort.”

Traditionally, response rates to cancer immunotherapies have been low, and “side effects can be severe.” However, the promising results of this study show that these models “have the potential for wide deployment and adoption,” since they demonstrated the ability to “personalize predictions and provide decision support for the clinician in determining appropriate therapies.”

GE HealthCare’s website has the press release.