AI Model SCORPIO Enhances Prediction of Cancer Immunotherapy Outcomes​.

In a significant advancement for precision oncology, scientists at Memorial Sloan Kettering Cancer Center (MSK) and the Icahn School of Medicine at Mount Sinai have introduced SCORPIO, an artificial intelligence (AI) model designed to predict cancer patients' responses to immune checkpoint inhibitors (ICIs) using standard clinical data.​

Immune checkpoint inhibitors have revolutionized cancer treatment by enabling the immune system to target tumors more effectively. However, these therapies are not universally effective and can lead to severe side effects and high costs. Traditional biomarkers like tumor mutational burden (TMB) and PD-L1 immunohistochemistry require invasive procedures and are not always reliable predictors of treatment success.​

SCORPIO addresses these challenges by analyzing routine blood tests—such as complete blood counts and metabolic panels—alongside clinical variables like age, sex, and prior treatments. The model was trained on data from over 1,600 patients across 17 cancer types treated at MSK between 2014 and 2019. Subsequent validation involved nearly 10,000 patients from 21 cancer types, including data from Mount Sinai and 10 global Phase 3 clinical trials.​

In performance evaluations, SCORPIO achieved a median area under the curve (AUC) of approximately 0.76 for predicting overall survival at multiple time points, significantly outperforming TMB and PD-L1 tests. The model also demonstrated robust predictive capabilities across various cancer types and healthcare settings, highlighting its potential for broad clinical application.​

Dr. Luc Morris of MSK emphasized the model's accessibility: "Our goal was to develop a tool that leverages data already available in most clinical settings, making precision oncology more attainable worldwide." Dr. Diego Chowell from Mount Sinai added, "SCORPIO's use of routine clinical data allows for faster, more accurate, and more affordable predictions of immunotherapy success."​

While SCORPIO's initial results are promising, the researchers acknowledge the need for prospective clinical trials to further validate its efficacy. Plans are underway to collaborate with global healthcare institutions to refine the model and develop a user-friendly interface for clinicians.​

The development of SCORPIO marks a pivotal step toward personalized cancer treatment, offering a scalable and cost-effective method to identify patients most likely to benefit from ICIs, thereby optimizing therapeutic outcomes and resource allocation.

Source: https://www.the-scientist.com/an-ai-based-model-could-predict-cancer-immunotherapy-outcomes-72703

This is non-financial/medical advice and made using AI so could be wrong.