2026 ASH®
Annual Meeting & Exposition December 12-15, 2026 ▪ New Orleans, Louisiana

Paper Details

Abstracts, authors, affiliations, and disclosures.

SESSION

602. Myeloid Oncogenesis: Basic: Poster II

Program: Poster Abstract Session
Sunday, December 13, 2026
06:00 p.m. - 08:00 p.m.
MCCNO - Hall F
Technology and Procedures, Assays, Imaging, Molecular Testing, Omics Technologies, Bioinformatics, Computational Biology, Emerging Technologies, Gene Editing, Profiling, Treatment Considerations, Combination Therapy, Biological Therapies, Gene Therapy, Biological Processes, Apoptosis, Pathogenesis, CHIP, Genomics, Hematopoiesis, Immune Mechanism, Metabolism, Molecular Biology, Diseases, Bone Marrow Failure Syndromes, Lymphoid Malignancies, Myeloid Malignancies, Acute Myeloid Malignancies, Chronic Myeloid Malignancies, Research, Fundamental Science, AML, Translational Research, MDS, Study Population, Human, Adult, Pediatric, Animal Model
PAPER

3200 A cell intrinsic IFN program driven by DNMT1 interacting RNA in MDS

Davide D'Onghia, PhD, Simone Ummarino, PhD2, Alexander Ebralidze, PhD3, Lucrezia Rinaldi Jr, PhD4, Ida Autiero, PhD5, Larysa Poluben Jr, MD,PhD6, Mailin Li, MD, PhD student7, Lida Xanthi Katopodi, PhD student8, Jerome Ritz, MD9, Gianluca Gaidano, MD10, Ioannis Vlachos, PhD11, Daniel Tenen, MD12, Rob S Welner, PhD13, Annalisa Di Ruscio, MD, PhD14
1 Beth Israel Deaconess Medical Center (BIDMC) - Harvard Medical School (HMS), Boston, Massachusetts, United States
2 Beth Israel Deaconess Medical Center (BIDMC) - Harvard Medical School (HMS), Boston, Massachusetts, United States, Tufts University, Medford, Massachusetts, United States
3 Beth Israel Deaconess Medical Center (BIDMC) - Harvard Medical School (HMS), Boston, Massachusetts, United States
4 Tufts University, Medford, Massachusetts, United States
5 CNR, Naples, NA, Italy
6 Beth Israel Deaconess Medical Center (BIDMC) - Harvard Medical School (HMS), Boston, Massachusetts, United States
7 Houston Methodist Hospital, Houston, Texas, United States
8 Beth Israel Deaconess Medical Center (BIDMC) - Harvard Medical School (HMS), Boston, Massachusetts, United States
9 Dana-Farber Cancer Institute, Boston, Massachusetts, United States
10 University of Eastern Piedmont, Novara, NO, Italy
11 Beth Israel Deaconess Medical Center (BIDMC) - Harvard Medical School (HMS), Boston, Massachusetts, United States
12 Beth Israel Deaconess Medical Center (BIDMC) - Harvard Medical School (HMS), Boston, Massachusetts, United States
13 O'Neal Comprehensive Cancer Center at The University of Alabama at Birmingham, Birmingham, Alabama, United States
14 Beth Israel Deaconess Medical Center (BIDMC) - Harvard Medical School (HMS), Boston, Massachusetts, United States

Myelodysplastic syndromes are clonal hematopoietic stem-cell disorders characterized by ineffective hematopoiesis, peripheral cytopenias, and a propensity to progress to acute myeloid leukemia. They are driven by aberrant DNA methylation (Figueroa et al., Blood 2009), yet also display a recurrent innate immune activation associated with adverse disease features (Gañán-Gómez et al., Leukemia 2015). How the epigenetic and immune phenotypes are mechanistically linked remains unknown. DNA methyltransferase 1 (DNMT1) is the maintenance DNA methyltransferase that propagates DNA methylation patterns during replication. Its activity is modulated by DNMT1-interacting RNAs (DiRs), transcripts that bind DNMT1 and locally inhibit its enzymatic activity, leading to local demethylation (Di Ruscio et al., Nature 2013). This RNA-directed layer of DNA methylation control has never been examined in primary MDS genome-wide.
We therefore asked whether an altered repertoire of DNMT1-bound RNAs in MDS predicts, locus by locus, which genes are de-repressed upon DNA methylation loss. Differential DNA methylation (whole-genome bisulfite sequencing), gene expression (RNA-seq) and DiRs (RNA immunoprecipitation of DNMT1) were profiled in primary bone marrow mononuclear cells (BMMCs) from 6 MDS patients and 2 healthy donor controls (study CE 154/15, 17 November 2015). First, we defined anticorrelating gene loci which display inverse DNA methylation levels and expression profiles in patients versus controls. Second, we overlapped the DiR profiles against the anticorrelating loci to identify those regulated by DiR.
Through this integration we defined DiR gain and DiR loss signatures. While promoter hypermethylation in MDS is well characterized, focal losses of DNA methylation also occur (Issa, Blood 2013) yet their regulation and functional output remain poorly defined. We therefore focused on the DiR-gain arm, where DiR mediated demethylation can be linked to gene activation.
Interestingly, we observed that DNA methylation loss was focal rather than global in MDS, overrepresented outside CpG islands at shores and first exons, with a loss of DNA methylation nearing 50%. Promoter DNA methylation and expression changes were inversely correlated across genes (Spearman ρ≈−0.6), consistent with DNA methylation-controlled transcription. Integration with DiR resolved a directed axis wherein DiR-gain loci (n=511) coincided with focal DNA methylation loss and transcriptional up-regulation in patients versus controls. Functionally, the DiR-gain arm was strongly enriched for type-I interferon and innate immune programs: interferon-α and interferon-γ responses, with Toll-like receptor and reactive oxygen species signaling. The interferon signature was coordinated and cell-intrinsic: canonical interferon-stimulated genes were among the most induced and showed increased DiR binding, while their CpG-island promoters remained unmethylated and the DNA methylation loss localized to adjacent shores and first exons; the program converged on the interferon-stimulated gene factor 3 node (STAT1/STAT2/IRF9).
To extend and validate this DiR gain signature, we examined 424 independent MDS patients across three distinct CD34+ cohorts: GSE114922 (RNA-seq, n=82), GSE58831 (microarray, n=159) and GSE19429 (microarray, n=183), by within-cohort z-scoring and cross-cohort meta-analysis (Stouffer, weighted by the square root of n), asking both whether the signature remained significant and whether its functional enrichment was reproduced. The DiR-gain program as a whole was significantly up-regulated versus cohort-matched healthy controls. Remarkably, the unbiased genome-wide enrichment across all 50 Hallmark programs ranked interferon-α first (normalized enrichment score 2.10, false discovery rate 5.4×10?8), with interferon-γ also enriched and the JAK-STAT pathway coordinately engaged, while generic inflammatory programs (TNFα/NF-κB, inflammatory response) were not.
These data define an RNA-directed epigenetic axis in MDS in which redistribution of DNMT1-bound RNA produces focal methylation loss and de-represses a coordinated, cell-intrinsic type-I interferon program, mechanistically linking the disease's epigenetic and immune phenotypes. We show that a signature derived from BMMCs remains significant in purified CD34+ cohorts, localizing the signature to the malignant clone and pointing to the DiR–DNMT1 axis as an intrinsic, targetable node to predict and modulate the disease course.

Davide D'Onghia, PhD: No relevant conflicts of interest to declare.