Affordable AI Tool Predicts Cancer Immunotherapy Outcomes

Researchers from Memorial Sloan Kettering Cancer Center (MSK) and Mount Sinai’s Tisch Cancer Institute have developed an AI-powered model called SCORPIO to predict how cancer patients will respond to immune checkpoint inhibitors (immunotherapy) using routine blood tests and clinical data. This groundbreaking tool could make precision medicine more affordable, scalable, and globally accessible.

Checkpoint inhibitors are a powerful class of drugs that “unleash” the immune system to target and destroy cancer cells. However, they are expensive, come with potential side effects, and do not work for every patient. Existing predictive biomarkers, such as tumor mutational burden (TMB) and PD-L1 expression, require specialized testing that is often inaccessible in many healthcare settings.

SCORPIO eliminates these barriers by relying on widely available clinical information, including complete blood count (CBC) and comprehensive metabolic panels. According to Dr. Luc Morris, co-senior author and a surgeon at MSK, SCORPIO outperformed traditional FDA-approved tests, providing better accuracy in predicting treatment outcomes.

The AI model was trained using data from over 2,000 cancer patients at MSK, representing 17 cancer types. It was then validated across a diverse cohort of nearly 10,000 patients from 21 different cancer types across 10 global phase 3 clinical trials and Mount Sinai datasets. This extensive dataset marks one of the largest immunotherapy studies ever conducted.

SCORPIO uses ensemble machine learning techniques to identify patterns in blood tests and treatment outcomes. This method not only increases diagnostic accuracy but also offers a cost-effective solution for hospitals worldwide, regardless of their technological infrastructure.

Dr. Diego Chowell, co-senior author, emphasized the importance of developing tools that bridge the accessibility gap in cancer treatment. Future plans include collaborations with global healthcare providers to refine SCORPIO and develop an easy-to-use clinical interface for real-world implementation.

As immunotherapy continues to evolve, tools like SCORPIO represent a paradigm shift in personalized cancer care, ensuring patients receive the most appropriate treatments while reducing unnecessary costs and side effects.

This AI innovation signifies a major advancement in cancer research, promising more equitable healthcare outcomes worldwide.

More Information: Seong-Keun Yoo et al, Prediction of checkpoint inhibitor immunotherapy efficacy for cancer using routine blood tests and clinical data, Nature Medicine (2025). DOI: 10.1038/s41591-024-03398-5