Exploring molecular heterogeneity and clonal evolution in chronic myelomonocytic leukemia via single-cell transcriptome sequencing

Abstract Background: Chronic myelomonocytic leukemia (CMML) is a clonal hematopoietic stem cell disorder that combines clinical and biological characteristics of myelodysplastic neoplasms and myeloproliferative neoplasms. Current treatment choices are limited by incomplete understanding of its transcriptional programs and clonal evolution. Clinical heterogeneity necessitates single-cell resolution to resolve subclonal diversity and identify actionable nodes. We applied single-cell transcriptome sequencing to delineate cellular and molecular heterogeneity and to uncover clonal evolution features that may inform actionable targets. Methods: Bone marrow aspirates from 5 newly diagnosed CMML patients and 4 healthy donors underwent single-cell transcriptome sequencing. Comprehensive analysis included dimensionality reduction and clustering, gene ontology (GO) and KEGG pathway enrichment of significantly different genes (p<0.05 and fold changes > 1.5), intercellular communication inference (CellChat), regulatory network reconstruction (SCENIC), gene set variation analysis (GSVA), consensus non-negative matrix factorization, and construction of explainable gene ontology fingerprints. Results: 1. We identified 13 distinct cellular clusters with significant compositional shifts in CMML versus healthy controls; clusters 2, 5, 7, 9, and 10–13 showed the most pronounced differences. We also identified marker genes for each of the 13 cell clusters, and the top 10 marker genes in each cluster were identified based on log fold change and adjusted P-value, including ALPL, MAL, KLRF1, SASH1, SLC2A5, CRISP3, XCL2, E2F7, PAX5, ITGB3, and P2RY6. 2. A total of 1,866 genes were differentially expressed(1,584 upregulated, 282 downregulated) in CMML patients compared to healthy volunteers. 3. GO enrichment analysis was performed separately for biological process (BP), molecular function (MF), and cellular component (CC). In BP, upregulated genes were significantly enriched for neutrophil degranulation, whereas downregulated genes were overrepresented in SRP-dependent cotranslational protein targeting to the membrane. In CC, upregulated genes localized predominantly to the cytosol, while downregulated genes were concentrated in ribosomal structures. In MF, upregulated genes were chiefly associated with protein binding, whereas downregulated genes were enriched for functions related to the structural constituent of the ribosome. 4. KEGG enrichment analysis revealed significant enrichment in pathways such as B cell receptor signaling, microRNAs in cancer, ribosomes, and antigen processing and presentation. 5. Protein interaction analysis using the STRING database generated a PPI network of the top 25 differentially expressed genes, highlighting a central hub that includes NAMPT, SOD2, CXCL8, LYN, LTF, VMP1, PDE4D, FNDC3B, DEFA3, and ACSL1. Conclusion: This single-cell transcriptome sequencing study delineated disease-specific transcriptional and clonal evolution features in CMML versus healthy controls, providing mechanistic insight into CMML pathogenesis and identifying candidate molecular targets for therapeutic development.