Mapping bone marrow architecture provided a more accurate view of disease state in myelodysplastic neoplasms (MDS) than molecular or blast-based assessments, according to research findings published in Leukemia. The researchers created an AI-based score to assess disease status and changes over time.
“Using AI to assist us in looking at the spatial architecture of the bone marrow using widely available laboratory assays allows us to improve our capability to assess patients’ prognosis and perhaps triage them for precision therapies,” said co-corresponding and senior author David Redmond, PhD, MSc, Assistant Professor of Computational Biology Research in Medicine, Weill Cornell Medical College
Background and Study Methods
“[MDS is] fundamentally a chronic and progressive disease,” said co-corresponding and senior author Sanjay S. Patel, MD, MPH, MSc, Clinical Chief of Hematopathology, and an Associate Professor of Pathology and Laboratory Medicine at Weill Cornell Medicine. “With the current methods pathologists use to assess MDS samples, there are some clear-cut cases and a lot of grey area.”
Researchers completed whole-slide multiplex immunofluorescence imaging with single-cell phenotyping to map bone marrow microarchitecture in MDS samples. They analyzed diagnostic biopsies, longitudinal treatment samples, precursor states, and normal controls.
The work builds upon earlier research from Drs. Patel and Redmond to spatially map human bone marrow samples.
Sampled MDS marrow showed coordinated, genotype-imprinted architectural remodeling, such as altered progenitor composition and spatial patterning, disrupted erythroid island organization, and displacement of hematopoietic stem and progenitor cells from perivascular niches
The researchers created a composite Microarchitectural Perturbation Score (MDS-MAPS) based on 82 cellular and spatial features from diagnostic samples. They also conducted leave-one-patient-out cross-validation and found that the MDS-MAPS score was able to identify MDS samples in remission vs samples from patients with active disease with more accuracy than blast percentage (area under the curve [AUC] = 0.883 vs 0.660). Low-blast MDS was also discriminated from clonal cytopenia of undetermined significance (AUC = 0.815).
Mixed-effects modeling showed that the score decreased when patients were in remission independently from blast burden, and the marrow architecture normalized in remission. In relapse, architectural remodeling re-emerged.
“We can generate an MDS-MAPS value for a patient at diagnosis and track how it changes over time,” said Dr. Patel, who is also a practicing hematopathologist at NewYork-Presbyterian/Weill Cornell Medical Center
“It distills something very complex down into a numerical value,” Dr. Patel said. “It is easier for a patient to understand whether their score is trending in the right or wrong direction, or whether their disease is more likely stable.”
Additionally, the researchers found that CXCR4 effects on bone marrow stem cells are reduced in patients with MDS, especially in patients with TP53-mutated disease.
“Patients with MDS and related precursor conditions have many mutations as part of the disease biology and each patient’s molecular signature is different,” said study author Pinkal Desai, MD, Associate Professor of Medicine at Weill Cornell Medicine, a hematologist/oncologist at NewYork-Presbyterian/Weill Cornell Medical Center, and Clinical Director of the Englander Institute Precision Medicine Molecular Aging Institute. “We have always wondered if these mutations signal a different spatial pattern and whether these patterns have an impact in predicting how patients progress and respond to treatments. Together we are poised to harness technology to answer real-world clinical questions.”
Going forward, the researchers are planning to validate the tool in larger patient cohorts and assess its impact on patient care.
DISCLOSURES: This work was supported by the Department of Pathology and Laboratory Medicine, Weill Cornell Medical College (start-up funding to Dr. Patel). Tissue staining and imaging were performed in the Multiparametric In Situ Imaging (MISI) Laboratory of the Department of Pathology and Laboratory Medicine, Weill Cornell Medical College. For full disclosures of the study authors, visit nature.com.
