Blood biomarkers taken 90 days post-transplant predict chronic GVHD risk months ahead

By the time chronic graft-versus-host disease announces itself through skin rashes, dry eyes, or joint stiffness, the biological process driving it has often been underway for months. The disease - in which transplanted immune cells attack the recipient's own tissues - is one of the leading causes of illness and death after stem cell and bone marrow transplantation, affecting organs including the skin, eyes, mouth, and lungs, sometimes causing permanent disability.

A team at MUSC Hollings Cancer Center and the Medical College of Wisconsin has now developed a tool that can estimate a transplant patient's risk of developing chronic GVHD and dying from transplant-related causes - using a blood sample drawn just 90 to 100 days after transplant, months before symptoms typically appear.

How BIOPREVENT was built

The research team, led by Sophie Paczesny, MD, PhD, co-leader of the Cancer Biology and Immunology Research Program at MUSC Hollings, analyzed data from 1,310 stem cell and bone marrow transplant recipients enrolled across four large, multicenter studies. Blood samples collected at the 90-to-100-day post-transplant window were tested for seven immune proteins previously identified and validated by Paczesny's group. The biomarkers reflect different aspects of immune activity: inflammation, immune activation and regulation, and tissue injury and remodeling.

These seven biomarkers were combined with nine clinical factors - including patient age, transplant type, primary disease, and prior complications - drawn from the national transplant registry. U.S. transplant centers are required to submit detailed, standardized data to the Center for International Blood and Marrow Transplant Research, which meant the clinical inputs were consistent and high-quality across all four studies.

The team tested several machine-learning approaches. The best-performing model, based on Bayesian additive regression trees, became the foundation for BIOPREVENT. Models combining biomarkers with clinical data consistently outperformed models using clinical data alone, particularly in predicting transplant-related mortality.

What the tool can predict

BIOPREVENT successfully separated patients into low- and high-risk groups, with clear differences in outcomes up to 18 months post-transplant. Crucially, different biomarkers predicted different outcomes: one blood protein was closely linked to risk of death after transplant, while others were stronger signals for future chronic GVHD development. This finding suggests that chronic GVHD and transplant-related death are at least partially driven by distinct biological processes - reinforcing the case for a multi-marker approach rather than a single test.

The tool was further validated in an independent cohort of transplant recipients, confirming that it reliably predicted risk beyond the patients used to train the model - a necessary step before any risk tool can be considered clinically useful.

A free, web-based application

Rather than publishing the model as a theoretical algorithm, the team built BIOPREVENT into a free, web-based application. Clinicians can enter a patient's clinical details and biomarker values and receive personalized risk estimates over time. "Making BIOPREVENT freely available helps ensure that researchers and clinicians can test it, learn from it and, ultimately, improve care for transplant patients," Paczesny said.

What it cannot do yet

BIOPREVENT is currently intended for risk assessment and clinical research, not for guiding treatment decisions. It tells clinicians who is at higher risk - it does not yet tell them what to do differently for those patients. Determining whether acting on early risk signals, through closer monitoring or preventive therapies, actually improves outcomes will require prospective clinical trials. Those trials are the next step Paczesny identified.

The biomarker measurements themselves require laboratory infrastructure that not all transplant centers possess, particularly in lower-resource settings. Broader adoption depends on standardizing assay protocols and making testing economically viable across different healthcare systems.

The study was published in the Journal of Clinical Investigation.