New York City-based Memorial Sloan Kettering Cancer Center and National Cancer Institute researchers created an artificial intelligence tool that uses routine clinical data to predict the best immunotherapy drug and how a patient's cancer will respond to it.
Currently, two predictive biomarkers are approved by the FDA to identify how immunotherapy treatment will impact cancer, however, both are not always accurate in predicting response to immune checkpoint inhibitors, are expensive to obtain and are not routinely collected, according to a June 3 NCI news release.
The AI model uses a patient's age, cancer type, history of systemic therapy, blood albumin level, blood neutrophil-to-lymphocyte ratio and tumor mutational burden. It compares the patient data to independent data sets that include 2,881 patients treated with immune checkpoint inhibitors across 18 solid tumor types.
The model accurately predicted a patient's likelihood to respond to an immune checkpoint inhibitor, how long they would live — both overall and before the disease returned — and was able to identify patients with low tumor mutational burden who could be treated effectively with immunotherapy.
The study was published June 3 in the peer-reviewed journal Nature Cancer.