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

Paper Details

Abstracts, authors, affiliations, and disclosures.

Sunday, December 13, 2026
09:30 a.m. - 11:00 a.m.
MCCNO - 291-292
Technology and Procedures, Assays, Molecular Testing, Bioinformatics, Gene Editing, Treatment Considerations, Non-Biological Therapies, Pharmacology, Biological Processes, Pathogenesis, CHIP, Genomics, Hematopoiesis, Immune Mechanism, Molecular Biology, Multi-Systemic Interactions, Diseases, Bone Marrow Failure Syndromes, Inherited Marrow Failure Syndromes, Genetic Disorders, Myeloid Malignancies, Acute Myeloid Malignancies, Chronic Myeloid Malignancies, Research, Fundamental Science, Translational Research, MDS, Clinical Research, Epidemiology, Real-World Evidence, Study Population, Human, Adult, Elderly, Embryo/Fetus, Animal Model
PAPER

481 Population scale disease risk revealed through cell fate biases of clonal hematopoiesis drivers in embryonic mosaicism

Jacey Rodriguez Pridgen, Sarah Waldvogel, PhD2, Phillip Burkhardt, PhD3, Yash Pershad, BS4, Lan Liao5, Alejandra Martell6, Anna Guzman7, Kerstin Grund8, Lilian Kaufmann9, Katrin Hinderhofer10, Nicola Dikow11, Linda Zhange12, Jennifer Padilla13, Adya Verma14, Chiraag D Kapadia, BA15, Anna Burkhalter16, Josephine De La Fuente, MS17, Robert Corty18, Caitlyn Vlasschaert, MD19, Chun-Wei Chen20, Ayala Tovy, PhD21, Jeffrey Steimle22, Ghayda Mirzaa23, Ludovica Marando, MD, PhD24, Jason Heaney25, James Martin26, Bekim Sadikovic, PhD27, Rachel E. Rau, MD28, Margaret Goodell, PhD29
1 Baylor College of Medicine, Houston, Texas, United States
2 Baylor College of Medicine, Houston, Texas, United States
3 Baylor College of Medicine, Houston, Texas, United States
4 Vanderbilt University Medical Center, Nashville, Tennessee, United States
5 Baylor College of Medicine, Houston, Texas, United States
6 Baylor College of Medicine, Houston, Texas, United States
7 Baylor College of Medicine, Houston, Texas, United States
8 , Heidelberg University, Heidelberg, Germany
9 , Heidelberg University, Heidelberg, Germany
10 , Heidelberg University, Heidelberg, Germany
11 , Heidelberg University, Heidelberg, Germany
12 Baylor College of Medicine, Houston, Texas, United States
13 Baylor College of Medicine, Houston, Texas, United States
14 Baylor College of Medicine, Houston, Texas, United States
15 Baylor College of Medicine, Houston, Texas, United States
16 Baylor College of Medicine, Houston, Texas, United States
17 Baylor College of Medicine, Houston, Texas, United States
18 Vanderbilt University Medical Center, Nashville, Tennessee, United States
19 Queen's University, Kingston, Ontario, Canada
20 Baylor College of Medicine, Houston, Texas, United States
21 Baylor College of Medicine, Houston, Texas, United States
22 Baylor College of Medicine, Houston, Texas, United States
23 University of Washington School of Medicine, Seattle Children's Hospital, Seattle, Washington, United States
24 Baylor College of Medicine, Houston, Texas, United States
25 Baylor College of Medicine, Houston, Texas, United States
26 Baylor College of Medicine, Houston, Texas, United States
27 London Health Sciences Centre, London, Ontario, Canada
28 Baylor College of Medicine, Houston, Texas, United States, University of Washington School of Medicine, Seattle Children's Hospital, Seattle, Washington, United States
29 Baylor College of Medicine, Houston, Texas, United States

Introduction: Clonal hematopoiesis (CH) is considered an age-associated phenomenon, with expanded DNMT3A, TET2, and other variant clones almost ubiquitously found in aged individuals. Deep clonal analyses have shown acquisition of the driver-variant can occur at any point throughout the lifespan, including during embryogenesis. Mutations acquired during embryogenesis may generate a mosaic individual, bearing a mixture of variant and non-variant cells that can seed multiple tissues including the germline. Whether early mutational events change variant cell distribution or CH dynamics is unknown. To examine this, we used DNMT3A as an exemplar as its somatic and germline consequences diverge sharply: somatic loss-of-function variants confer a strong selective advantage on hematopoietic stem cells (HSCs) and are the most common driver of CH, while germline loss causes multisystem Tatton-Brown-Rahman syndrome (TBRS; overgrowth, intellectual disability, cardiac defects). We asked how embryonic mosaic loss of DNMT3A shapes variant-cell distribution across an organism, and its implications for diagnosis, transmission, and disease. 

Results: We generated mosaic mice in which a single cell of a 4-cell mouse embryo contained a heterozygous Dnmt3a mutation, mimicking a potential human mosaic. Cre mRNA was injected into one cell of a 4-cell embryo, conditionally deleting a Dnmt3a allele (“3A”) and activating a tdTomato (tdT) reporter to trace variant cells; controls (WT) were Dnmt3a+/+. Despite inducing 1:4 cells, mosaic mice showed a wide range of mosaicism (2-95%), with no difference in average mosaicism between cohorts (WT 42.4%, 3A 41.9%). Variant cells contributed to all tissues examined, but single-nucleus RNAseq of brain, kidney, heart, and testis at 5 months revealed cell-type-specific bias: Dnmt3a variant cells were depleted in cerebellar granule, Leydig, and ventricular cardiomyocyte populations and enriched in cardiac and renal endothelium and atrial cardiomyocytes, orthogonally validated in vitro.

In peripheral blood (PB), Dnmt3a variant cells increased over the first 20 weeks in 3A but not WT mosaics, due to HSC expansion, exceeding the rate of aging-associated murine CH. In humans, we have identified 2 asymptomatic mosaic adults, both with ~40% variant allele frequency (VAF) in PB who had transmitted the variant to multiple offspring but exhibited low VAF in other tissues (one previously reported male, one female). We identified a third who presented with developmental delay and underwent diagnostic sequencing at multiple ages, until a TBRS-like diagnosis was made at age 26. Longitudinal analysis of PB samples showed a DNMT3A variant with a VAF that rose from 1% at age 2 to 30% at 18 and 42% at 26-27, with low but non-zero VAF in urine, buccal, and fibroblasts. The multi-germ-layer involvement of the DNMT3A variant cells confirmed embryonic origin, and this is the first documented rapid postnatal expansion of DNMT3A variant cells in an embryonic mosaic. Strikingly, a clinically validated methylation classifier (EpiSign) tracked this expansion: the age 2 PB sample matched healthy controls, while ages 18-27 samples progressively acquired the TBRS episignature, showing the diagnosis was detectable in PB well before the variant was identified.

This young-age, high-VAF blood signature, distinct from CH (lower VAF in older individuals), enabled population screening. Across 663,372 whole genomes (All of Us v8, n=413,334; BioVU, n=250,038), we found 30 individuals under age 40 with DNMT3A VAF ≥25% after excluding sequencing artifacts, germline TBRS, and hematologic malignancy, representing a prevalence of 1.37 per 10,000 (95% CI 0.93-1.96). Applying the same approach to TET2, the second most common CH driver, identified 24 comparable individuals (~1.10 per 10,000), indicating the signature generalizes across CH genes.

Conclusions: Embryonic mosaic driver variants produce cell-type-specific tissue biases and rapid postnatal blood expansion far exceeding age-related CH, detectable both in individuals and at population scale. Clinically, a high-VAF DNMT3A or TET2 variant in a young adult should not be assumed germline or malignant; evaluation for embryonic mosaicism is warranted, with cancer-surveillance and reproductive implications. Thousands of Mendelian syndromes likely have heritable mosaic counterparts, now tractable in the genomic era.

Jacey Rodriguez Pridgen: No relevant conflicts of interest to declare.