NIH Scientists Unveil Revolutionary AI Tool to Predict Cancer Patients’ Responses to Immunotherapy

0
25
NIH scientists develop AI tool to predict how cancer patients will respond to immunotherapy

NIH Scientists Develop AI Tool to Forecast Cancer Patients’ Response to Immunotherapy

Media Advisory

Monday, June 3, 2024

Introduction to a Revolutionary AI Tool

In an exciting breakthrough for cancer treatment, researchers at the National Institutes of Health (NIH) have unveiled a new artificial intelligence (AI) tool capable of predicting how individual patients with cancer will respond to immune checkpoint inhibitors, a type of immunotherapy. This innovative tool leverages routine clinical data, including information from simple blood tests, to enhance personalized treatment strategies.

The Importance of Accurate Predictions

Currently, treatment-related decisions for cancer patients often rely on two FDA-approved biomarkers: tumor mutational burden, which assesses the number of mutations in cancer DNA, and PD-L1, a protein that modulates immune response. However, these biomarkers can be unreliable in predicting treatment outcomes, leading researchers to seek alternative methods for improving accuracy in response prediction.

A Novel Machine-Learning Approach

The study, published on June 3, 2024, in Nature Cancer, details a novel machine-learning model designed to predict patient outcomes based on five easily collected clinical features: age, cancer type, history of systemic therapy, blood albumin levels, and blood neutrophil-to-lymphocyte ratio, which indicates inflammation levels. The model also integrates tumor mutational burden, measured via sequencing panels.

Study Scope and Methodology

This impressive model was built and validated using data from over 2,800 patients treated with immune checkpoint inhibitors across 18 different solid tumor types. By evaluating multiple independent datasets, the researchers ensured that their findings are robust and generalizable across diverse patient populations.

Key Findings

The results demonstrate the model’s remarkable ability to accurately predict both the likelihood of a patient’s response to immunotherapy and their expected survival duration. Notably, the AI tool can also identify patients with low tumor mutational burdens who may still respond favorably to treatment.

Future of the AI Tool

Given the promising results, the researchers emphasize the need for larger prospective studies to further validate the effectiveness of this AI model in clinical settings. In a key move toward accessibility, the AI model, named Logistic Regression-Based Immunotherapy-Response Score (LORIS), has been made publicly available at https://loris.ccr.cancer.gov. This resource will allow healthcare professionals to evaluate a patient’s potential response to immunotherapy using the defined clinical parameters.

Leading Voices Behind the Research

The study was co-led by esteemed researchers Eytan Ruppin, M.D., Ph.D., of NCI’s Center for Cancer Research, and Luc G. T. Morris, M.D., from Memorial Sloan Kettering Cancer Center. The investigation was driven by Tiangen Chang, Ph.D., and Yingying Cao, Ph.D., from Dr. Ruppin’s research team.

Clinical Implications

This innovative AI tool represents a significant leap forward in the personalization of cancer treatment. By accurately predicting responses to immunotherapy, it has the potential to enhance treatment efficacy, minimize unnecessary side effects, and ultimately improve patient outcomes.

Publication Details

The findings of this groundbreaking research are detailed in the study titled “LORIS robustly predicts patient outcomes with immune checkpoint blockade therapy using common clinical, pathologic and genomic features,” published in Nature Cancer. For those interested in the specifics of the study, it can be accessed at https://doi.org/10.1038/s43018-024-00772-7.

About the National Cancer Institute (NCI)

The National Cancer Institute leads the National Cancer Program, striving to significantly reduce cancer prevalence while enhancing the quality of life for patients. NCI supports a diverse array of cancer research and training, financing numerous studies through grants and contracts. Its intramural research program is committed to groundbreaking research across multiple disciplines, from basic to clinical research.

About the National Institutes of Health (NIH)

As the nation’s foremost medical research agency, NIH consists of 27 Institutes and Centers, operating under the U.S. Department of Health and Human Services. NIH is dedicated to advancing fundamental, clinical, and translational medical research and seeks to uncover the causes and treatments for a wide variety of diseases.

Conclusion

This novel AI tool represents a promising advancement in the field of cancer treatment, enhancing predictive capabilities regarding patient responses to immunotherapy. Ongoing research and validation will be essential in fully integrating this technology into clinical practice, offering hope for better-targeted treatments and improved health outcomes for cancer patients.

Frequently Asked Questions

1. What is the purpose of the new AI tool developed by NIH researchers?

The AI tool aims to predict how individual cancer patients will respond to immune checkpoint inhibitors based on routine clinical data.

2. What clinical features does the AI model consider?

The model considers a patient’s age, cancer type, history of systemic therapy, blood albumin level, neutrophil-to-lymphocyte ratio, and tumor mutational burden.

3. How was the effectiveness of the AI tool assessed?

The effectiveness of the model was assessed using data from 2,881 patients treated across 18 solid tumor types, ensuring its robust predictive capability.

4. Is the AI model publicly available for use?

Yes, the AI model, named LORIS, is publicly accessible at https://loris.ccr.cancer.gov.

5. Who were the main researchers involved in this study?

The study was co-led by Eytan Ruppin, M.D., Ph.D., and Luc G. T. Morris, M.D., with contributions from Tiangen Chang, Ph.D., and Yingying Cao, Ph.D.

source