Single-cell spatial mapping reveals dynamic bone marrow microarchitectural alterations and enhances clinical...

Area of Science:

• Hematology
• Computational Pathology
• Artificial Intelligence in Medicine

Background:

• Myelodysplastic neoplasms (MDS) are heterogeneous cancers of the blood with ineffective production and low cell counts.
• Current MDS diagnosis relies on subjective bone marrow biopsy assessment, potentially missing subtle changes.
• The clinical relevance of microarchitectural alterations in MDS remains largely unexplored.

Purpose of the Study:

• To apply an AI-driven spatial proteomic profiling method to bone marrow samples from MDS patients.
• To identify and characterize previously unrecognized microarchitectural changes in MDS.
• To correlate these changes with genetic mutations and clinical parameters for improved disease assessment.

Main Methods:

• Utilized a novel AI-driven, whole slide imaging-based single-cell spatial proteomic profiling technique.
• Analyzed 77 bone marrow tissue samples from MDS patients and age-matched controls, including longitudinal samples.
• Quantified progenitor cell frequencies, precursor morphologies, hematopoietic stem and progenitor cell (HSPC) localization, cell clustering, and erythroid island structures.

Main Results:

• MDS tissues exhibited significant alterations in progenitor cell frequencies, erythroid precursor and megakaryocyte morphology, HSPC displacement, abnormal progenitor clustering, and disrupted erythroid islands compared to controls.
• Observed microarchitectural changes showed stronger correlations with specific mutations (e.g., SF3B1, TP53) than with clinical risk scores (IPSS-M).
• Developed a composite 'MDS severity score' based on spatial tissue features, which correlated with clinical and genetic data across serial samples.

Conclusions:

• This study uncovers novel, genotype-linked microarchitectural alterations in MDS using AI-powered spatial proteomics.
• These findings suggest that quantitative assessment of subtle bone marrow changes can complement existing diagnostic and monitoring strategies for MDS.
• The developed MDS severity score shows potential for enhancing disease stratification and tracking treatment response.