Back
Science

AI Tool Developed to Predict Complications After Stem Cell Transplants

View source

BIOPREVENT: AI Tool Targets Life-Threatening Transplant Complications

A new artificial intelligence (AI) tool, BIOPREVENT, has been developed to identify patients at higher risk for life-threatening complications following stem cell and bone marrow transplants. This innovative initiative was led by researchers from MUSC Hollings Cancer Center and the Center for International Blood and Marrow Transplant Research.

Understanding Chronic Graft-Versus-Host Disease (GVHD)

Chronic graft-versus-host disease (GVHD) is a serious complication where transplanted immune cells attack the patient's healthy tissues. It can affect multiple organs and cause long-term disability or death.

Current diagnosis often occurs months after biological changes have begun, highlighting a critical need for earlier detection and intervention.

How BIOPREVENT Predicts Risk

The BIOPREVENT tool utilizes machine learning applied to immune-related proteins and clinical information. It estimates a patient's future risk of developing chronic GVHD and dying from transplant-related causes. The study, published in the Journal of Clinical Investigation, combined immune biomarkers, clinical data, and machine learning for real-world risk prediction.

Comprehensive Data Analysis

Researchers analyzed data from 1,310 transplant recipients across four studies. Blood samples collected 90 to 100 days post-transplant were tested for seven immune proteins associated with inflammation, immune activation, regulation, and tissue injury. These biomarkers were integrated with nine clinical factors, including patient age, transplant type, primary disease, and prior complications.

Enhanced Predictive Power and Validation

The machine learning models that combined blood biomarkers with clinical data demonstrated improved accuracy in predicting patient outcomes, particularly for transplant-related mortality.

The tool was validated in an independent patient group, confirming its predictive reliability. BIOPREVENT successfully stratified patients into low- and high-risk groups, showing distinct outcomes up to 18 months later.

Accessible for Clinicians

The team has developed BIOPREVENT into a free, web-based application, allowing clinicians to input patient data and receive personalized risk estimates. This development aims to make the research accessible for testing and further improvement within the clinical community.

Future Impact and Personalized Medicine

Currently, BIOPREVENT is intended for risk assessment and clinical research, not for guiding treatment decisions.

Future steps include conducting clinical trials to evaluate whether acting on these early risk signals can improve long-term outcomes for patients. This initiative represents a move towards more personalized medicine in transplant care, providing clinicians with earlier, more informed decision-making support.