A groundbreaking multiomics study has unraveled the intricate epigenetic and transcriptional landscapes of acute myeloid leukemia (AML) at single-cell resolution, offering unprecedented insights into disease heterogeneity and lineage-specific regulatory mechanisms. Researchers analysed over 280,000 mononuclear cells from 36 AML patient samples across 16 ATAC-defined subgroups alongside remission controls, integrating single-cell RNA sequencing (scRNA-seq) and assay for transposase-accessible chromatin using sequencing (scATAC-seq) to decode chromatin accessibility and gene expression simultaneously.
The study revealed that leukemic cells within each ATAC subgroup form distinct chromatin accessibility clusters, which coalesce despite originating from different patients, emphasizing shared epigenetic signatures that transcend individual variability. Notably, these subgroup-specific clusters were well-separated from residual normal cells from patients in remission, highlighting unique disease-associated chromatin states.
Delving deeper, the authors projected the single-cell expression profiles onto a reference human bone marrow differentiation pseudotime trajectory, illuminating unique lineage commitments and differentiation blocks inherent to each subgroup. For instance, HOX-related subgroups exhibited divergent maturation arrest points ranging from hematopoietic stem cells (HSCs) to granulocyte-monocyte progenitors. Meanwhile, TP53-associated subgroups demonstrated distinct lineage biases, with some arrested at early progenitor stages and others progressing toward erythroid or monocytic paths. Intriguingly, one subgroup showed a commitment to lymphoid lineages, a noteworthy deviation given AML’s myeloid classification.
The presence of leukemic stem cell-like populations was inferred from established stemness gene signatures, consistently enriched at early differentiation stages across all subgroups. However, the abundance of these leukemic stem cells (LSCs) appeared highly variable, suggesting differential potential for disease propagation and relapse among epigenetic subgroups.
A highlight of this research is the integration of chromatin accessibility and transcriptomic data using SCENIC+ software, which enabled the inference of transcription factor (TF) network activities within single cells. This approach uncovered key TFs uniquely activated in each subgroup that also exhibited distinct temporal activity profiles along the myeloid differentiation continuum. For example, HOXA9—a critical regulator in hematopoiesis—displayed differential peak activity depending on the subgroup, while other TFs like BCL11A, IRF8, and SPI1 showed stage-specific expression patterns, underscoring the dynamic epigenetic regulation driving subgroup-specific leukemogenesis.
This study exemplifies how the synergy between single-cell multiomics and advanced computational methods can deconvolute the molecular complexity of AML. By mapping epigenetic heterogeneity to cellular differentiation states and TF dynamics, it opens avenues for precision medicine strategies that target subgroup-specific vulnerabilities in this aggressive blood cancer. Future functional validations, especially regarding the role of leukemic stem cells in disease maintenance, will be pivotal in translating these discoveries into clinical interventions.
In sum, this comprehensive single-cell dissection of AML’s chromatin landscape propels our understanding of its epigenetic diversity, challenging the traditional bulk analyses and spotlighting the importance of temporal and lineage-specific regulatory mechanisms in cancer pathogenesis.
Subject of Research: Epigenetic landscape and heterogeneity of acute myeloid leukemia at the single-cell level
Article Title: Chromatin landscape and epigenetic heterogeneity of acute myeloid leukaemia
Article References:
Ochi, Y., Liew-Littorin, M., Nannya, Y. et al. Chromatin landscape and epigenetic heterogeneity of acute myeloid leukaemia. Nature (2026). https://doi.org/10.1038/s41586-026-10703-4
