P014 | Azacitidine remodels hematopoietic stem/progenitor cell fate and restores erythroid differentiation in myelodysplastic/myeloproliferative neoplasms with neutrophilia via reactivation of the GATA switch

Abstract

Myelodysplastic/myeloproliferative neoplasm with neutrophilia (MDS/MPN-N) is a rare and heterogeneous hematologic malignancy characterized by overlapping myelodysplastic and myeloproliferative features and recurrent mutations in ASXL1, SETBP1, and ETNK1. Current therapeutic options remain limited, with hydroxyurea providing only transient benefit and allogeneic transplantation being feasible in only few cases. The hypomethylating agent azacitidine (aza) has shown efficacy in high-risk myelodysplastic syndromes, but its molecular effects in MDS/MPN-N are poorly understood. This study aimed to characterize clinical responses and molecular mechanisms of aza in MDS/MPN-N through integrated clinical and single-cell transcriptome analyses.
Three patients with MDS/MPN-N received subcutaneous aza (75 mg/m², days 1–7, every 28 days) for up to nine cycles. Longitudinal hematologic, morphologic, and targeted NGS analyses were performed at baseline and after cycles 3, 6, and 9. Bone marrow CD34⁺ cells collected at diagnosis and after three cycles underwent single-cell RNA sequencing to assess transcriptional and lineage changes. Complementary in vivo studies were conducted in a conditional SETBP1-G870S murine model treated with aza (5 mg/kg × 5 days).
Aza was well tolerated and achieved hematologic improvement in all patients, including normalization of leukocyte counts (A) and reduction in transfusion dependence (B). Bone marrow analyses revealed restored granulocytic maturation without disease acceleration. NGS monitoring showed heterogeneous clonal dynamics, indicating that clinical benefit was not strictly linked to clonal eradication.
Single-cell transcriptomics (C) demonstrated that aza profoundly remodels bone marrow architecture, reducing hematopoietic stem and early progenitor populations while expanding differentiating progenitors (D). Treated samples showed activation of cell cycle and differentiation programs (E) and repression of self-renewal drivers such as MECOM, EYA1, and FLT3 (F). Gene ontology and GSEA analyses revealed enrichment of pathways related to chromatin remodeling, histone modification, and hematopoietic differentiation. In the erythroid compartment, aza promoted reactivation of GATA1-dependent transcriptional programs and downmodulation of SPI1/PU.1 regulon activity (G), restoring the GATA2-to-GATA1 switch (H-I) and enhancing heme biosynthesis and proliferation of erythroid precursors. Consistent with human data, SETBP1-driven leukemic mice treated with aza showed expansion of early erythroid progenitors.

Aza promotes extensive epigenetic and transcriptional remodeling in MDS/MPN-N, promoting differentiation and erythropoietic recovery while limiting aberrant stem cell self-renewal. Collectively, these data position aza as a rational therapeutic approach capable of modulating hematopoietic stem cell fate and achieving durable hematologic improvement in MDS/MPN-N patients who are not eligible for transplantation.