In a recent proof-of-concept study led by researchers at the National Institutes of Health (NIH), an artificial intelligence (AI) tool has been developed to predict whether a cancer patient will respond to immune checkpoint inhibitors, a type of immunotherapy drug. The machine-learning model utilizes routine clinical data, such as blood tests, and has the potential to assist doctors in determining the effectiveness of immunotherapy in treating cancer.

Published on June 3, 2024, in Nature Cancer, the study highlights the collaboration between researchers from the National Cancer Institute's (NCI) Center for Cancer Research and Memorial Sloan Kettering Cancer Center in New York. The researchers aimed to overcome the limitations of current predictive biomarkers, which sometimes fail to accurately determine the response to immune checkpoint inhibitors.

At present, the Food and Drug Administration has approved two predictive biomarkers for identifying candidates suitable for immunotherapy treatment. These include tumor mutational burden, which measures the number of mutations in cancer cell DNA, and PD-L1, a protein found in tumor cells that regulates immune response and is targeted by some immune checkpoint inhibitors. However, these biomarkers are not always reliable indicators of response.

Recent machine-learning models utilizing molecular sequencing data have shown promise in predicting response to immunotherapy. However, obtaining this type of data is expensive and not routinely collected. The new study addresses this issue by developing a machine-learning model that relies on five easily accessible clinical features: patient age, cancer type, history of systemic therapy, blood albumin level, and blood neutrophil-to-lymphocyte ratio, which indicates inflammation. Additionally, the model assesses tumor mutational burden through sequencing panels.

To evaluate the model, data from multiple independent sources comprising 2,881 patients treated with immune checkpoint inhibitors across 18 types of solid tumors were utilized. The model successfully predicted the likelihood of a patient's response to the immunotherapy drug, as well as their overall survival and time until disease recurrence. Significantly, it also identified patients with low tumor mutational burden who could still benefit from immunotherapy.

While the results are promising, the researchers acknowledge the need for larger prospective studies to assess the AI model's effectiveness in clinical settings. They have made the model, named Logistic Regression-Based Immunotherapy-Response Score (LORIS), publicly available for further research.

This groundbreaking study was co-led by Dr. Eytan Ruppin from NCI's Center for Cancer Research and Dr. Luc G. T. Morris from Memorial Sloan Kettering Cancer Center. Dr. Tiangen Chang and Dr. Yingying Cao from Dr. Ruppin's group at NCI's Center for Cancer Research spearheaded the research efforts.

The National Cancer Institute (NCI), a leading entity in cancer research and part of NIH, leads the National Cancer Program and strives to reduce cancer prevalence and improve patients' lives. Through grants and contracts, NCI supports a wide range of cancer research and training initiatives. The institute's intramural research program focuses on innovative research on cancer causes, prevention, risk prediction, early detection, and treatment. Additionally, NIH, a component of the U.S. Department of Health and Human Services, works to conduct and support medical research to uncover causes, treatments, and cures for various diseases.

This study's findings contribute to the ongoing efforts in precision medicine and personalized treatment options for cancer patients. As AI technology continues to advance, it holds significant potential for enhancing clinical decision-making and improving patient outcomes in the field of oncology.

For more information about cancer and NIH's initiatives, please visit the NCI website at cancer.gov or contact the NCI's contact center at 1-800-4-CANCER (1-800-422-6237). Additional details about NIH and its programs can be found at www.nih.gov.