Researchers Classify AML Into 16 Epigenetic Subgroups From 1,563 Patients
Updated
Updated · Nature.com · Jul 8
Researchers Classify AML Into 16 Epigenetic Subgroups From 1,563 Patients
3 articles · Updated · Nature.com · Jul 8
Summary
ATAC-seq data from 1,563 newly diagnosed AML patients split the disease into 16 chromatin-accessibility subgroups, a framework the study says captures biology beyond mutation-based classifications.
Multiomics analyses showed the subgroups carry distinct driver mutations, differentiation states, gene expression, DNA methylation, super-enhancers and transcription-factor networks, and single-cell profiling found each subgroup shares a common epigenomic fingerprint.
Clinical differences also emerged: ATAC-defined groups had independent prognostic value versus genomic classification, with favourable outcomes in subgroups A, I, C and B and poorer survival in L, G, N and O.
Drug-response screening linked some subgroups to specific sensitivities, including FLT3 inhibitors in subgroup E, MEK inhibitors in C, F and H, and unexpected ABL-inhibitor sensitivity in subgroup K.
The authors say the classification could refine ELN risk stratification and support more personalized AML treatment, though they frame it as complementary to—not a replacement for—existing genetic systems.
This new cancer map redefines leukemia. How will it change patient treatment plans?
If cancer's 'operating system' is the real culprit, are we targeting the wrong thing?
Redefining AML: 16 Epigenetic Subgroups Unveiled for Precision Diagnosis and Personalized Therapy
Overview
A groundbreaking 2026 study by researchers from Karolinska Institutet and Japan redefined acute myeloid leukemia (AML) classification by analyzing 1,563 patients using advanced epigenetic profiling. By employing ATAC-seq and integrating multi-omics data, the team mapped the open chromatin landscape of AML cells, uncovering 16 distinct epigenetic subgroups. These subgroups revealed significant biological and clinical differences not seen with genetic classifications alone, meaning patients with similar genetic mutations could have very different disease characteristics and treatment responses. This new epigenetic classification promises more precise AML diagnosis and paves the way for personalized therapies.