Weill Cornell Medicine researchers have developed a new AI-based method to evaluate patients with a blood cancer called myelodysplastic tumor. The method, published June 30 in the journal Leukemia, compares the physical location and morphometric (size and shape) characteristics of all hematopoietic cells in a patient’s bone marrow sample to those in healthy bone marrow, producing a score that reflects the severity of the disease.
Myelodysplastic tumors (MDS) are a type of blood cancer that usually affects older people. About one-third of patients progress to a more aggressive form of cancer called acute myeloid leukemia. MDS patients must undergo frequent biopsies to identify signs of remission or worsening of the disease.
It is basically a chronic and progressive disease. With the current methods pathologists use to evaluate MDS samples, there are some clear-cut cases and many gray areas. ”
Dr. Sanjay Patel, Clinical Director of Hematopathology and Associate Professor of Pathology and Laboratory Medicine, Weill Cornell University
more accurate diagnosis
Building on previous work using AI to create more detailed images of human bone marrow, Dr. Patel and his collaborator, Dr. David Redmond, research assistant professor of medical computational biology at Weill Cornell Medical College, developed a new method that uses completely anonymized patient samples. This could provide clearer information for MDS patients and their doctors.
“Using AI in observing the spatial structure of bone marrow using widely available clinical laboratory assays can improve our ability to assess patient prognosis and perhaps triage patients for precision treatment,” said Dr. Redmond.
The MDS-Microarchitectural Perturbation Score (MDS-MAPS) score ranks patient samples based on 82 characteristics associated with normal tissue and different genetic subtypes of MDS. This tool uses routinely collected samples, tissue staining protocols, and imaging techniques that can be implemented in most hospital pathology departments and laboratories.
“We can generate a patient’s MDS-MAPS value at diagnosis and track how it changes over time,” said Dr. Patel, who is also a hematopathologist at NewYork-Presbyterian/Weill Cornell Medical Center.
A lower score indicates that the sample is closer to healthy tissue, and a higher score indicates more disease-related changes. They then used computer modeling to show that this tool could be better at classifying a patient’s disease status than current methods.
“This takes something very complex and distills it down to numbers,” Dr. Patel says. “It makes it easier for patients to understand whether their scores are trending in the right direction, in the wrong direction, or if their disease is likely to be stable.”
The next step is to validate the tool in a larger patient sample cohort to determine whether its use improves patient care. Dr. Patel and Dr. Redmond will work with Dr. Pinkal Desai, associate professor of medicine at Weill Cornell Medicine and hematologist/oncologist at NewYork-Presbyterian/Weill Cornell Medical Center, to study how the tool works on samples from patients with other forms of MDS and the increasingly recognized prodromal symptoms.
“Patients with MDS and related prodromal symptoms have many mutations as part of their disease biology, and each patient’s molecular signature is different,” said Dr. Desai, who is also clinical director of the Englander Institute’s Precision Medicine and Molecular Aging Laboratory. “We have always wondered whether these mutations exhibit different spatial patterns and whether these patterns influence predictions of patient progression and response to treatment. Together, we are ready to leverage technology to answer real-world clinical questions.”
molecular clues
The study also provides new clues about how mutations in a tumor suppressor gene called TP53 contribute to the development and progression of MDS. Previous research by Dr. Redmond and Dr. Shahin Rafi, director of the Department of Regenerative Medicine at Weill Cornell Medical College, mapped the structure of healthy bone marrow. The study shows that the same basic structure exists in human bone marrow, and that bone marrow stem cells become increasingly misplaced as MDS progresses.
Dr. Redmond explained that in healthy tissue, blood vessels within the bone marrow help nurture bone marrow stem cells, which then grow into blood cells and immune system cells. However, the receptor that promotes these supportive interactions, CXCR4, is reduced in stem cells from MDS patients, particularly those with TP53 mutations.
Learning more about these interactions may lead to targeted therapies designed to alter gene expression and help restore normal bone marrow architecture and cell-cell interactions.
This information could also help pathologists further refine the methods they use to monitor patients with this and other similar bone marrow diseases, Dr. Patel said.
sauce:
Reference magazines:
Nachman, R. Others. (2026). Spatial remodeling of bone marrow architecture defines the histological hallmarks of disease activity and treatment response in myelodysplastic tumors. leukemia. DOI: 10.1038/s41375-026-03029-7. https://www.nature.com/articles/s41375-026-03029-7
