Pathology In Depth

Library › Myeloid neoplasms: WHO 5th edition and ICC 2022 classification, myeloproliferative neoplasms, mastocytosis, MDS/MPN, myelodysplastic neoplasms, acute myeloid leukemia and related acute leukemias

Myeloid neoplasms: WHO 5th edition and ICC 2022 classification, myeloproliferative neoplasms, mastocytosis, MDS/MPN, myelodysplastic neoplasms, acute myeloid leukemia and related acute leukemias

Last reviewed 2026-11-17 by a board-certified pathologist. 159 references.

Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition

Definition. Myeloid neoplasms are clonal hematopoietic neoplasms classified by integrating morphology, immunophenotype, cytogenetics and molecular genetics (Khoury 2022). Two classifications were published in 2022: the 5th edition of the World Health Organization Classification of Haematolymphoid Tumours (WHO-HAEM5) and the International Consensus Classification (ICC) (Khoury 2022; Arber 2022). Both revise the revised 4th edition of the WHO classification (2017) (Arber and Orazi 2025; Xiao 2024).

Related page: Myeloproliferative Neoplasms: Chronic Myeloid Leukemia with Molecular Monitoring, Polycythemia Vera, Essential Thrombocythemia, Primary Myelofibrosis and Other Myeloproliferative Neoplasms (CML, PV, ET, PMF, CNL and CEL) Related page: Mastocytosis, Hypereosinophilic Syndrome and Myeloid/Lymphoid Neoplasms with Tyrosine Kinase Fusions, Chronic Myelomonocytic Leukemia and Other MDS/MPN (mastocytosis, clonal eosinophilia, CMML and other MDS/MPN) Related page: Myelodysplastic Neoplasms: WHO 5th Edition and ICC Types, Dysplasia Morphology, TP53-Mutated Myeloid Neoplasms and Prognostic Scoring (MDS types, dysplasia morphology and the TP53 framework) Related page: Acute Myeloid Leukemia with Defining Genetic Abnormalities, Acute Promyelocytic Leukemia and ELN 2022 Genetic Risk (AML blast thresholds, defining genetic abnormalities and ELN 2022 risk) Related page: AML, Myelodysplasia-Related, AML Defined by Differentiation, Myeloid Neoplasms Post Cytotoxic Therapy and of Down Syndrome, Myeloid Sarcoma, BPDCN and Acute Leukemias of Ambiguous Lineage (AML-MR, post-cytotoxic therapy neoplasms and MPAL) Related page: Bone Marrow Failure, Pure Red Cell Aplasia and the Laboratory Approach to Cytopenias and Leukocyte Disorders (Fanconi anemia, telomere biology disorders, Shwachman-Diamond syndrome, GATA2 deficiency and PNH clones in aplastic anemia) Related page: Hemolytic Anemias: Red Cell Membrane and Enzyme Defects and Paroxysmal Nocturnal Hemoglobinuria (PNH flow cytometry) Related page: Laboratory Evaluation of Inborn Errors of Immunity and Secondary Immunodeficiency (VEXAS syndrome among the phenocopies of inborn errors of immunity) Related page: Therapeutic Apheresis, ASFA Indications and Hematopoietic Progenitor Cell Products (CD34 enumeration) Related page: Flow Cytometry, FISH, Microbiology and Electron Microscopy on Cytology Specimens (flow cytometry on fine-needle aspiration material)

WHO 5th edition and ICC 2022: origin and structure of the two classifications

The WHO 3rd, 4th and revised 4th editions were produced with Clinical Advisory Committees organized by the Society for Hematopathology and the European Association for Haematopathology. For the 5th edition, the WHO ended that partnership. The societies continued the Clinical Advisory Committee process and produced the ICC (Arber 2022; Arber and Orazi 2025).

  • WHO-HAEM5 structure. Types are arranged by lineage, dominant clinical attribute and dominant biologic attribute. Gene fusions use the double colon (PML::RARA). Emerging entities are listed under "other defined genetic alterations" in place of provisional status. Family-level diagnoses are permitted when molecular testing is unavailable (Khoury 2022). Type names omit the cytogenetic notation and add "fusion" or "rearrangement" where appropriate; "rearrangement" is used when a dominant gene has many partners (KMT2A) (Khoury 2022; Xiao 2024).
  • ICC structure. The ICC defines entities by morphology, clinical features and genomic data, and makes extensive use of genomic testing (Arber 2022; Duncavage 2022). Prior therapy, an antecedent myeloid neoplasm and germline predisposition are qualifiers added to the diagnosis of AML (Weinberg 2023; Arber 2022).
  • CAP reporting. The CAP Myeloid and Mixed/Ambiguous Lineage Neoplasms template (version 1.1.0.2, September 2026) derives its final integrated diagnoses from WHO-HAEM5 (CAP 2026).

Key differences between WHO-HAEM5 and ICC 2022, with the WHO 2017 rule

Sources: Khoury 2022 (WHO-HAEM5); Arber 2022 and the cited companion reviews (ICC); previous rule as stated in each source.

Topic WHO-HAEM5 (2022) ICC 2022 WHO revised 4th edition (2017), previous rule Source
Name of the MDS category Myelodysplastic neoplasms (abbreviation MDS) Myelodysplastic syndromes Myelodysplastic syndromes Khoury 2022; Arber and Orazi 2025
Clonal hematopoiesis CHIP and CCUS defined as myeloid precursor lesions CCUS placed with MDS in a group of clonal cytopenias; some cases previously called MDS on cytogenetic grounds become CCUS, but complex karyotype, del(5q), −7 and del(7q) remain MDS-defining; VEXAS syndrome and PNH are clonal proliferations with cytopenia that are not MDS unless MDS morphologic criteria are met Not part of the classification Khoury 2022; Arber 2022; Hasserjian 2023; Xiao 2024
CML phases Chronic and blast phase; accelerated phase omitted; high-risk features listed Chronic, accelerated and blast phase Chronic, accelerated and blast phase Khoury 2022; Arber 2022; Apperley 2025; Xiao 2024
CNL leukocyte threshold ≥25 × 10⁹/L ≥13 × 10⁹/L with CSF3R T618I or another activating CSF3R mutation; ≥25 × 10⁹/L without ≥25 × 10⁹/L Arber and Orazi 2025; Gianelli 2023
Placement of JMML Myeloproliferative neoplasm Pediatric and/or germline mutation-associated disorders; JMML-like neoplasms added MDS/MPN Arber 2022; Arber and Orazi 2025; Xiao 2024
Rule for systemic mastocytosis Major criterion plus 1 minor criterion, or 3 minor criteria Major criterion alone, or 3 minor criteria Major criterion plus 1 minor criterion, or 3 minor criteria Li 2023; Valent 2021
SM with an associated neoplasm SM with an associated hematologic neoplasm (myeloid or lymphoid) SM with an associated myeloid neoplasm SM with an associated hematologic neoplasm Arber 2022; Li 2023; Leguit 2023
CMML absolute monocyte count ≥0.5 × 10⁹/L with ≥10% monocytes; at 0.5 to <1 × 10⁹/L, clonality and dysplasia required ≥0.5 × 10⁹/L with ≥10% monocytes; at 0.5 to <1 × 10⁹/L, clonality (VAF ≥10%) required; cytopenia and CMML-type marrow required in all cases ≥1.0 × 10⁹/L Khoury 2022; Arber 2022; Loghavi 2026
Atypical CML Renamed MDS/MPN with neutrophilia Atypical CML; absence of hypereosinophilia required Atypical CML Khoury 2022; Arber 2022; Arber and Orazi 2025
Ring sideroblasts with thrombocytosis MDS/MPN with SF3B1 mutation and thrombocytosis; ≥15% ring sideroblasts required MDS/MPN with thrombocytosis and SF3B1 mutation (ring sideroblasts not required); MDS/MPN-RS-T, NOS for SF3B1 wild type MDS/MPN with ring sideroblasts and thrombocytosis Arber 2022; Arber and Orazi 2025; Hazarika and Bain 2026
MDS with 10%–19% blasts in adults MDS-IB2 MDS/AML MDS-EB-2 Arber 2022; Arber and Orazi 2025; Xiao 2024
TP53-defined MDS MDS with biallelic TP53 inactivation at 0%–19% blasts; no VAF requirement MDS with mutated TP53 (multi-hit, <10% blasts); MDS/AML with mutated TP53 (any TP53 mutation, 10%–19% blasts); VAF ≥10% No TP53-defined MDS type Arber 2022; Arber and Orazi 2025; Shah 2025
SF3B1-mutated MDS MDS-SF3B1; ≥15% ring sideroblasts may substitute for the mutation MDS-SF3B1 with VAF ≥10%; SF3B1 wild-type cases with ring sideroblasts classified as MDS, NOS MDS with ring sideroblasts; ≥5% ring sideroblasts sufficient when SF3B1 is mutated Arber 2022; Arber and Orazi 2025; Lee 2024
Blast requirement for AML with defining genetic abnormalities None, except AML with BCR::ABL1 and AML with CEBPA mutation (≥20%) ≥10% for most recurrent genetic abnormalities, including NPM1 and CEBPA; ≥20% for AML with BCR::ABL1 20% blasts define AML; exceptions taught in Acute Myeloid Leukemia with Defining Genetic Abnormalities, Acute Promyelocytic Leukemia and ELN 2022 Genetic Risk Khoury 2022; Arber 2022; Weinberg 2023; Lee 2024; Platzbecker 2025
AML with myelodysplasia-related features AML-MR: MR cytogenetic abnormalities, mutation in any of 8 MR genes (ASXL1, BCOR, EZH2, SF3B1, SRSF2, STAG2, U2AF1, ZRSR2) or history of MDS or MDS/MPN; morphology alone removed AML with MR gene mutations (9 genes: ASXL1, BCOR, EZH2, RUNX1, SF3B1, SRSF2, STAG2, U2AF1, ZRSR2), AML with MR cytogenetic abnormalities (adds +8 and del(20q); omits del(11q) and −13/del(13q)) and AML with mutated TP53; history of MDS or MDS/MPN is a qualifier and not a criterion AML with myelodysplasia-related changes, including a morphologic criterion Khoury 2022; Arber 2022; Jung 2023; Xiao 2024; Platzbecker 2025
Neoplasms after cytotoxic therapy Myeloid neoplasm post cytotoxic therapy; PARP1 inhibitors added and methotrexate excluded Prior therapy recorded as a qualifier added to the AML diagnosis Therapy-related myeloid neoplasms Khoury 2022; Arber 2022; Weinberg 2023
Germline predisposition Myeloid neoplasms associated with germline predisposition: 3 subtypes; phenotype plus genotype naming Hematologic neoplasms with germline predisposition: 3 groups shared with myeloid neoplasms, a fourth group for ALL with germline predisposition (PAX5, IKZF1) and a provisional category of genes with growing evidence Recognized as a distinct category from 2016 Khoury 2022; Arber 2022; Rudelius 2023; Tesi 2025
Tyrosine kinase fusion neoplasms MLN-TK; JAK2 rearrangement, FLT3 rearrangement and ETV6::ABL1 added M/LN-eo with tyrosine kinase gene fusions; ETV6::ABL1 and FLT3 fusions added; PCM1::JAK2 and its variants made a formal entity PCM1::JAK2 provisional Khoury 2022; Arber 2022; Tzankov 2023
Chronic eosinophilic leukemia CEL (NOS dropped); hypereosinophilia over 4 weeks; clonality and abnormal marrow required CEL, NOS; abnormal marrow morphology is a criterion CEL, NOS; 6-month interval; increased blasts accepted in place of clonality Khoury 2022; Gianelli 2023
Childhood MDS with low blasts Childhood MDS with low blasts replaces refractory cytopenia of childhood Refractory cytopenia of childhood retained in the pediatric and/or germline category Refractory cytopenia of childhood, provisional Khoury 2022; Arber 2022; Arber and Orazi 2025

Bone marrow evaluation of a suspected myeloid neoplasm

  • Aspirate differential count. The differential count is performed on Wright-Giemsa or May-Grünwald-Giemsa stained aspirate smears. A 500-cell count is the traditional number, and 300-cell counts give similar results. Hemodiluted areas are avoided. When the aspirate is inadequate, the blast percentage is estimated from CD34 or another blast marker by immunohistochemistry (CAP 2026).

  • Blasts and blast equivalents. Myeloblasts, monoblasts and promonocytes are counted together as blasts and blast equivalents (Khoury 2022). The CAP template records the blast and blast-equivalent percentage in marrow and blood. When the blood percentage is higher, the blood value is reported as the more significant site (CAP 2026).

  • Blast equivalents in the ICC. The ICC counts promonocytes, and neoplastic promyelocytes in the appropriate pathologic setting, as blast equivalents toward the ≥10% blast requirement for AML with recurrent genetic abnormalities (Arber 2022).

  • Cellularity. Cellularity is the area of hematopoietic marrow relative to adipose tissue on the core biopsy or clot section. It varies with age and is lower in subcortical marrow (CAP 2026). The European consensus grades cellularity with an age-dependent reference (Thiele 2005).

  • Cytopenia thresholds. WHO-HAEM5 and the ICC use the same definitions for CCUS, MDS and MDS/MPN: hemoglobin <13 g/dL in males and <12 g/dL in females, absolute neutrophil count <1.8 × 10⁹/L and platelets <150 × 10⁹/L (Khoury 2022; Arber 2022; Xiao 2024). In the ICC, the cytopenia of CCUS persists for ≥4 months and is not explained by another condition (Arber 2022; Vanner 2026).

  • Dysplasia threshold. Dysplasia is counted as present at ≥10% of cells in a lineage (Khoury 2022; Arber 2022). Dysplasia morphology and its mimics are taught in Myelodysplastic Neoplasms: WHO 5th Edition and ICC Types, Dysplasia Morphology, TP53-Mutated Myeloid Neoplasms and Prognostic Scoring.

Reticulin and collagen fibrosis grading (European consensus, CAP template 2026)

Fiber density is assessed only in hematopoietic areas (CAP 2026; Thiele 2005; Ghosh 2023). When fibrosis is heterogeneous, the grade assigned is the highest grade present in at least 30% of the marrow. Trichrome stain is recommended at MF-2 and MF-3 (CAP 2026). The consensus uses four reproducible categories that separate reticulin from collagen (Thiele 2005; Ghosh 2023).

Grade Description Source
MF-0 Scattered linear reticulin with no intersections, corresponding to normal marrow CAP 2026; Ghosh 2023
MF-1 Loose network of reticulin with many intersections, especially in perivascular areas CAP 2026; Ghosh 2023
MF-2 Diffuse and dense increase in reticulin with extensive intersections, occasionally with focal bundles of thick fibers mostly consistent with collagen and/or focal osteosclerosis CAP 2026; Ghosh 2023
MF-3 Diffuse and dense increase in reticulin with extensive intersections and coarse bundles of thick collagen fibers, usually with osteosclerosis CAP 2026; Ghosh 2023

Cytogenetic and molecular workup of a suspected myeloid neoplasm

  • Karyotype and FISH. Conventional cytogenetics remains a comprehensive and unbiased method for chromosomal abnormalities. Targeted gene panels, whole-exome, whole-genome and RNA sequencing detect somatic mutations and germline variants. Whole-genome sequencing is a potential replacement for both karyotype and panel sequencing (Duncavage 2022).

  • Sensitivity for MPN drivers. The ICC recommends assays with a minimal sensitivity of 1% VAF for JAK2, CALR and MPL at diagnosis (Arber 2022; Gianelli 2023).

  • Germline variants. Sequencing of tumor DNA detects variants that may be germline, so patients are informed of this possibility before testing (Baliakas 2024; Duncavage 2022).

  • Genomic evaluation of AML at diagnosis. Complete genomic evaluation includes karyotype, an NGS panel, FLT3-ITD testing, and FISH or chromosomal microarray when needed to confirm cytogenetic findings or to obtain faster results. FLT3-ITD and FLT3-TKD are often tested separately by PCR and capillary electrophoresis. Classification-defining mutations are reported within 5 days when possible (Duncavage 2022).

CAP Myeloid and Mixed/Ambiguous Lineage Neoplasms template, version 1.1.0.2 (September 2026)

Element Content Source
Site of involvement Bone marrow (blast and blast-equivalent percentage and cellularity), blood (blast percentage), mediastinum, lymph node, skin, extranodal site CAP 2026
Final integrated diagnosis WHO-HAEM5 types; CHIP and CCUS optional and not required to complete the summary CAP 2026
Ring sideroblasts None, <15% or ≥15% on iron stain CAP 2026
Fibrosis grade MF-0 to MF-3 CAP 2026
Flow cytometry No aberrancy at the sensitivity of the assay, or abnormal population with immunophenotype; semiquantitative intensity (for example CD34 dim+, CD34++) CAP 2026
Cytogenetics and FISH Karyotype result; probes tested with normal and abnormal results CAP 2026
Molecular alterations detected Gene-by-gene list with a "Not performed" option added in version 1.1.0.2; FLT3 tyrosine kinase domain answer revised CAP 2026

Flow cytometric immunophenotyping of acute myeloid leukemia

  • Blast gate. The CD45-dim, side scatter low-to-intermediate region on a CD45/side scatter plot contains progenitor cells of several lineages in normal marrow. Myeloid progenitors predominate in this region in MDS (Porwit 2023).
  • Principle of lineage assignment. An antigen supports a lineage more strongly when its intensity and pattern resemble the most similar normal population. Coordinated expression of several antigens of one lineage is more specific than any single antigen (Khoury 2022).
  • Reporting. Expression is reported semiquantitatively so that heterogeneous and dim expression are recognizable (CAP 2026).

Lineage-associated markers in AML (WHO-HAEM5)

Lineage Markers Source
Myeloid MPO; CD13, CD33 and CD117; ≥2 myeloid-associated antigens required for AML with minimal differentiation Khoury 2022
Monocytic CD11c, CD14, CD36, CD64; nonspecific esterase; lysozyme Khoury 2022
Erythroid CD71 and CD235a (glycophorin A); CD117, CD105 and CD36 mark maturation steps Porwit 2023
Megakaryocytic CD41 (glycoprotein IIb), CD61 (glycoprotein IIIa), CD42b (glycoprotein Ib) Khoury 2022
Basophilic Metachromasia with toluidine blue; negative MPO, Sudan black B and NSE; no strong CD117 (excludes mast cell leukemia) Khoury 2022

Flow cytometry in myelodysplastic neoplasms: maturation patterns and the Ogata score

  • Myeloid progenitors. CD34+ myeloid progenitors above 2% of nucleated cells are a common and prognostic finding in MDS. In the ELN iMDS Flow multicenter study, >3% CD34+ cells was strongly indicative of MDS or MDS/MPN once acute leukemia was excluded (Porwit 2023; Veenstra 2025).
  • Maturing granulocytes. Decreased side scatter (hypogranularity), decreased CD45 relative to lymphocytes, aberrant CD33 and CD56, asynchronous CD34, CD117 or HLA-DR, and abnormal CD13/CD16/CD11b patterns indicate dysgranulopoiesis. Aberrant CD13/CD16 and CD13/CD11b patterns were present in 29% and 40% of lower-risk MDS, compared with 16% and 11% of hospital controls (Porwit 2023).
  • CD16 loss. Complete or partial loss of CD16, a GPI-anchored protein, raises the possibility of a PNH clone and prompts testing of other GPI-anchored antigens (Porwit 2023).
  • Monocytes. Aberrant CD56 and CD2 expression on monocytes was reported in 48% and 32% of MDS. Abnormal CD13 and HLA-DR occur in many controls and are less specific (Porwit 2023). Monocyte subset partitioning for CMML is taught in Mastocytosis, Hypereosinophilic Syndrome and Myeloid/Lymphoid Neoplasms with Tyrosine Kinase Fusions, Chronic Myelomonocytic Leukemia and Other MDS/MPN.
  • Erythroid lineage. CD71 and CD36 expression, including the coefficient of variation of each, is used to assess dyserythropoiesis (Porwit 2023; Araújo 2020; Veenstra 2025).

Ogata score

The four-parameter Ogata score is recommended by the ELN iMDS Flow group for preliminary assessment of flow cytometric dysplasia (Porwit 2023).

Parameter Abnormal result Source
CD34+ myeloid progenitors >2% of nucleated cells Porwit 2023
B-cell progenitors within CD34+ cells <5% Porwit 2023
CD45 expression of CD34+ myeloblasts relative to lymphocytes Abnormal; a lymphocyte-to-myeloblast CD45 MFI ratio ≤4 or ≥7.5 in a validation study Porwit 2023; Plander 2024
Granulocyte to lymphocyte side scatter ratio ≤6 Porwit 2023; Plander 2024
Interpretation 0 or 1 rules out MDS; 2 is inconclusive; 3 or 4 suggests MDS (Araújo 2020). A cutoff of ≥2 has diagnostic and prognostic value (Porwit 2023; Plander 2024) Araújo 2020; Porwit 2023; Plander 2024
  • Weight of parameters. Increased CD34+ myeloid progenitors and low granulocyte side scatter are more specific than a decreased B-progenitor fraction and abnormal CD45 (Porwit 2023).

  • Limits of the Ogata score. A score of 2 is frequent in non-clonal cytopenias (Porwit 2023). Hemodilution lowers the B-progenitor fraction (Araújo 2020). Small CD34+CD7+CD13-low populations (<0.1%) occur in regenerating marrow, and progenitor aberrancies can persist after hypomethylating agents or cytotoxic therapy (Porwit 2023). Laboratories validate their own cutoffs (Porwit 2023).

  • ELN iMDS Flow scoring system. Seventeen core markers are scored, and ≥3 aberrancies are considered positive for MDS. In a validation cohort of 180 MDS and 54 pathological controls, the sensitivity was 90% for this system and 66% for the Ogata score, and false-positive rates were 24% and 11% (Veenstra 2025).

Measurable residual disease in AML: flow cytometric and molecular methods

Measurable residual disease (MRD) in AML is used for prognostic, predictive, monitoring and efficacy-response assessment (Heuser 2021). Multiparameter flow cytometry and RT-qPCR were the two validated methods recognized by ELN in 2021. The validated molecular targets are NPM1 mutations and the core binding factor fusions RUNX1::RUNX1T1 and CBFB::MYH11 (Nachmias 2025).

ELN-DAVID 2025 AML MRD update (current rule)

  • Scope. The 2025 ELN-DAVID update is aligned with ELN 2022 genetic risk and tailors MRD recommendations to genetic subgroups (Cloos 2026; Alibrahim 2026).
  • Qualitative response category. A new qualitative MRD response category classifies the MRD burden as optimal, warning or high risk of treatment failure (Cloos 2026; Alibrahim 2026).
  • FLT3-ITD. Ultrahigh-sensitivity NGS-based MRD is recommended for FLT3-ITD-mutated AML after intensive chemotherapy and before allogeneic hematopoietic cell transplantation (Cloos 2026; Hourigan 2026).

ELN 2021 AML MRD recommendations (previous rule)

Topic ELN 2021 recommendation Source
Flow cytometric approach Leukemia-associated immunophenotype (LAIP) and different-from-normal (DfN) approaches used together Heuser 2021; Van Galen 2026
Sample First-pull marrow, processed undiluted within 3 days at ambient temperature; peripheral blood contamination assessed Heuser 2021; Van Galen 2026
Events >5 × 10⁵ CD45+ cells and ≥100 viable cells in the blast compartment before MRD negativity is concluded Heuser 2021; Van Galen 2026
Minimum marker set CD34, CD117, CD45, CD33, CD13, CD56, CD7, HLA-DR Heuser 2021; Van Galen 2026
Flow MRD positivity ≥0.1% of CD45+ cells with the target immunophenotype Heuser 2021; Van Galen 2026
Timing Repeat testing in 2–4 weeks recommended when regenerating marrow may alter the immunophenotype Heuser 2021; Van Galen 2026
qPCR MRD Positivity: cycle threshold <40 in ≥2 of 3 replicates. Negativity: cycle threshold ≥40 in ≥2 of 3 replicates with ≥10,000 (optimally ≥30,000) housekeeping gene copies Heuser 2021; Van Galen 2026
Molecular targets Mutant NPM1, CBF AML and APL (PML::RARA) monitored by qPCR or digital PCR; other AML monitored by flow cytometry Heuser 2021; Van Galen 2026
NGS Data insufficient for NGS as a stand-alone technique; germline variants (VAF about 50% in ANKRD26, CEBPA, DDX41, ETV6, GATA2, RUNX1, TP53) and age-related CH genes (DNMT3A, TET2, ASXL1) excluded as MRD markers; molecular assays reach a limit of detection ≤10⁻³; NGS-MRD positivity provisionally ≥0.1% VAF Heuser 2021; Van Galen 2026

Sensitivity of AML MRD methods

Method Sensitivity Blood validated Source
Multiparameter flow cytometry 10⁻³ to 10⁻⁴ (0.1%–0.01%); positivity usually ≥0.1% No Nachmias 2025; Van Galen 2026
RT-qPCR (NPM1, CBF fusions) 10⁻⁵ to 10⁻⁶; blood about 1 log lower than marrow Yes Nachmias 2025
Standard NGS About 1%–2% No Nachmias 2025
Error-corrected NGS 10⁻³ to 10⁻⁶ Yes Nachmias 2025
  • Molecular kinetics. In NPM1-mutated AML, a 4-log10 reduction in blood after induction predicted long-term relapse-free survival in the ALFA0702 trial. In RUNX1::RUNX1T1 AML, a 2.5-log10 reduction after 1 cycle or a 3-log10 reduction after 2 cycles was prognostic (Nachmias 2025).
  • CH in MRD samples. CH clones can predate AML and persist after successful induction, so a variant detected after therapy may represent CH (Heuser 2021; Van Galen 2026).

Clonal hematopoiesis, CHIP and clonal cytopenia of undetermined significance

Definitions of CH, CHIP, CCUS and ICUS (WHO-HAEM5 and ICC)

Term Definition Source
Clonal hematopoiesis (CH) Cells derived from a mutated multipotent stem or progenitor cell with a growth advantage, without unexplained cytopenia, hematologic cancer or another clonal disorder Khoury 2022
CHIP CH with somatic mutations in myeloid malignancy-associated genes in blood or marrow at VAF ≥2% (≥4% for X-linked genes in males), in a person without a diagnosed hematologic disorder or unexplained cytopenia Khoury 2022
CCUS CHIP with ≥1 persistent unexplained cytopenia, not meeting criteria for a myeloid neoplasm Khoury 2022
ICC CHIP Somatic mutation in a myeloid driver gene at VAF ≥2%, or a non-MDS-defining clonal cytogenetic abnormality, without a hematologic neoplasm or unexplained cytopenia Arber 2022; Xiao 2024
ICC CCUS exclusions Multi-hit TP53 (VAF ≥10%), SF3B1 (VAF ≥10%), complex karyotype (≥3 abnormalities, except −Y), del(5q), −7 or del(7q) mandate MDS in a cytopenic patient Xiao 2024
ICUS Unexplained cytopenia not meeting MDS morphologic criteria, with no clonal driver mutation on NGS and no abnormality on FISH or karyotype; absence of driver mutations has a high negative predictive value for MDS; in one series, 37 of 96 patients (39%) with unexplained cytopenia had ICUS, 9% CCUS and 9% clonal cytopenia and monocytosis of undetermined significance (CCMUS) Duncavage 2022; Erdogan Yucel 2026
  • ICC. The ICC groups CCUS and MDS as clonal cytopenias. Morphologic dysplasia separates MDS from other clonal cytopenias, except that SF3B1 and multi-hit TP53 mutations are MDS-defining in a cytopenic patient (Hasserjian 2023; Arber 2022).

  • Previous rule. CH was not part of the WHO 2017 classification. WHO-HAEM5 and the ICC were the first classifications to include CHIP and CCUS as myeloid precursor lesions (Khoury 2022; Xiao 2024).

  • Clinical significance of CH. CH increases with age and is associated with increased overall mortality, cardiovascular disease and myeloid malignancy (Khoury 2022; Duncavage 2022). The significance of variants below 2% VAF is unclear (Khoury 2022; Vanner 2026).

  • Genes. DNMT3A, TET2 and ASXL1 are the genes commonly mutated in age-related CH (Apperley 2025; Van Galen 2026; Duncavage 2022). Isolated DNMT3A, TET2 or ASXL1 mutations have less predictive value for myeloid transformation than spliceosome or RUNX1 mutations (Konstantinov 2026; Vanner 2026).

  • Clonal monocytosis. Clonal monocytosis of undetermined significance (CMUS) and clonal cytopenic monocytosis of undetermined significance (CCMUS) are clonal monocytoses that do not meet criteria for CMML. They carry a higher risk of myeloid neoplasm than CCUS without monocytosis (Loghavi 2026). The ICC recognizes CMUS as a CMML precursor condition (Arber 2022).

  • CH and cytotoxic therapy. Expansion of pre-existing clones under cytotoxic therapy contributes to myeloid neoplasms post cytotoxic therapy. CCUS is excluded from that category (Khoury 2022).

  • CH after cytotoxic therapy in the ICC. The ICC qualifies CHIP and non-MDS clonal cytopenias as therapy related when they follow marrow exposure to chemotherapy or radiation therapy (Arber 2022).

  • Risk of progression. CH carries a 0.5%–1% annual risk of progression to a myeloid neoplasm, and the rate for CCUS is more than 10-fold higher (Vanner 2026). Higher-risk features are mutations in TP53, ASXL1, JAK2, SF3B1, SRSF2, U2AF1 or IDH1/IDH2, more than 1 driver mutation and VAF >10% (Duncavage 2022; Vanner 2026). The Clonal Hematopoiesis Risk Score (CHRS) uses age ≥65 years, high-risk mutations, ≥2 mutations, VAF ≥20%, MCV ≥100 fL, red cell distribution width ≥15% and cytopenia (Vanner 2026).

  • Reporting. Reporting of CHIP or CCUS in the CAP template is optional (CAP 2026).

VEXAS syndrome: somatic UBA1 variants and marrow findings

VEXAS (vacuoles, E1 enzyme, X-linked, autoinflammatory, somatic) syndrome was described in 25 men with somatic variants affecting methionine-41 (p.Met41) of UBA1, the major E1 enzyme that initiates ubiquitylation. UBA1 is on the X chromosome (Beck 2020; Holden 2025).

  • Clinical and hematologic features. Fevers, cytopenias, neutrophilic cutaneous and pulmonary inflammation, chondritis and vasculitis begin in late adulthood. Characteristic vacuoles are present in myeloid and erythroid precursors, and the marrow is dysplastic (Beck 2020; Patel 2026). Many patients met criteria for relapsing polychondritis, Sweet syndrome, polyarteritis nodosa or giant cell arteritis, MDS or multiple myeloma (Beck 2020).

  • Cell distribution. The variants were present in more than half of hematopoietic stem cells and in blood myeloid cells. They were absent from lymphocytes and fibroblasts (Beck 2020; Patel 2026).

  • Mechanism. p.Met41 variants abolish the cytoplasmic isoform of UBA1 and produce a catalytically impaired isoform initiated at p.Met67 (Beck 2020; Holden 2025).

  • Variants. The common pathogenic variants are p.Met41Thr, p.Met41Val and p.Met41Leu (Konstantinov 2026; Holden 2025). p.Met41Val has the most severe phenotype. Non-Met41 variants occur with more co-mutations and with lymphomas and plasma cell neoplasms (Konstantinov 2026).

  • Women. In 12 women with VEXAS, acquired monosomy X was the predominant mechanism, found in 6 of 8 women with karyotypes (Konstantinov 2026).

  • MDS. MDS is reported in 18%–55% of patients, depending on the criteria (Konstantinov 2026), and about 30%–50% in a second review (Patel 2026). VEXAS-associated MDS carries mainly CH-type mutations (DNMT3A, TET2, ASXL1) and lacks the spliceosome and TP53 profile of high-risk MDS (Konstantinov 2026).

  • Classification. The ICC recognizes VEXAS syndrome as a clonal myeloid proliferation with cytopenia that is not MDS unless MDS morphologic criteria are met (Arber 2022; Xiao 2024).

  • Other mechanisms in women. Female cases also occur with constitutional 45,X, extreme X-inactivation skewing or loss of the wild-type allele (Holden 2025).

  • UBA1 in unselected MDS. About 1% of unselected patients with MDS carry UBA1 mutations (Patel 2026).

  • Plasma cells. Marrow plasma cells may be increased and are mostly polytypic; clonal plasma cells corresponding to monoclonal gammopathy, smoldering myeloma or myeloma occur in a small proportion of patients (Patel 2026).

Myeloid neoplasms with germline predisposition

Myeloid neoplasms with germline predisposition are named by coupling the myeloid diagnosis with the germline genotype, for example "MDS with low blasts associated with germline RUNX1 variant" (Khoury 2022). Genetic counseling and family history are part of the diagnostic evaluation (Khoury 2022; Arber 2022).

WHO-HAEM5 subtypes and ICC groups of germline predisposition

WHO-HAEM5 subtype Genes and conditions ICC group Source
Without a pre-existing platelet disorder or organ dysfunction Germline CEBPA (familial AML), DDX41, TP53 (Li-Fraumeni syndrome) Not associated with a constitutional disorder: CEBPA, DDX41, TP53 Khoury 2022; Arber 2022; Rudelius 2023; Elbaz Younes 2026
With a pre-existing platelet disorder Germline RUNX1 (familial platelet disorder with associated myeloid malignancy), ANKRD26 (thrombocytopenia 2), ETV6 (thrombocytopenia 5) Thrombocytopenia or platelet dysfunction: RUNX1, ANKRD26, ETV6 Khoury 2022; Arber 2022; Rudelius 2023; Elbaz Younes 2026
With potential organ dysfunction GATA2 deficiency; severe congenital neutropenia; Shwachman-Diamond syndrome; Fanconi anemia; telomere biology disorders; RASopathies; Down syndrome; SAMD9 (MIRAGE syndrome); SAMD9L (ataxia-pancytopenia syndrome); biallelic BLM (Bloom syndrome) Constitutional disorders affecting multiple organ systems: GATA2, SAMD9, SAMD9L, classic marrow failure syndromes, JMML-associated germline variants, Down syndrome Khoury 2022; Arber 2022; Rudelius 2023; Elbaz Younes 2026
No WHO-HAEM5 equivalent Genes with growing but incomplete evidence Provisional category of germline predisposition genes Rudelius 2023
No WHO-HAEM5 equivalent Germline PAX5 and IKZF1 Acute lymphoblastic leukemia with germline predisposition Arber 2022; Elbaz Younes 2026

Features of individual germline predisposition genes

Gene Features Source
CEBPA Autosomal dominant familial AML. Germline variants are usually N-terminal frameshift or nonsense variants; the AML usually carries an acquired somatic C-terminal (bZIP) mutation Elbaz Younes 2026; Hwang 2024
RUNX1 Familial platelet disorder with associated myeloid malignancy: autosomal dominant with incomplete penetrance; mild to moderate thrombocytopenia and/or platelet aggregation defects precede MDS or AML; T- and B-ALL are reported; sporadic RUNX1-mutated AML usually has a VAF <40% Elbaz Younes 2026
ANKRD26 Thrombocytopenia 2: autosomal dominant; variants cluster in the 5′ untranslated region (c.-116 to c.-134) and disrupt RUNX1 and FLI1 binding, which increases ANKRD26 expression in megakaryocytes Elbaz Younes 2026; Hwang 2024
ETV6 Thrombocytopenia 5: penetrance of thrombocytopenia approaches 90% and of hematologic malignancy about 20%–30%; AML, MDS and B-ALL are the most common malignancies Elbaz Younes 2026
  • Megakaryocytic dysplasia without MDS. Megakaryocytic dysplasia is common in the marrow of patients with germline RUNX1, ANKRD26 or ETV6 variants who have isolated thrombocytopenia and no MDS (Arber 2022).
  • MDS in a germline carrier (ICC). Development of MDS is generally associated with new or progressive cytopenia, often with rising marrow cellularity, overt multilineage dysplasia, increased blasts or acquired pathogenic alterations. Emergence of del(5q), −7/del(7q), complex karyotype, multi-hit TP53 (VAF ≥10%) or SF3B1 (VAF ≥10%) is MDS-defining (Arber 2022).

Recognition of a germline variant on a tumor panel

  • Variant allele fraction. A DDX41 variant at a VAF near the heterozygous level on a tumor-only panel prompts further investigation (Baliakas 2024). A VAF that remains high in complete remission without MRD supports germline origin, and serial NGS identified germline variants in 5.5% of 343 patients with AML (Ahn 2026).

  • Indications for germline testing. Germline testing is indicated for patients with ≥2 cancers, one of them hematologic, and for patients with a hematologic neoplasm and a positive family history. Young age is not required, because DDX41 predisposition presents at older ages. Variants in CEBPA, DDX41, GATA2, RUNX1 or TP53 at about 50% VAF prompt germline testing, particularly when they persist in remission (Duncavage 2022).

  • DDX41 variant origin. Of DDX41 variants in myeloid neoplasms, 85% are germline, and >95% of truncating DDX41 alleles are germline (Hwang 2024).

  • Saliva and buccal swabs. Leukocyte contamination complicates the interpretation of saliva and buccal swabs as germline sources (Hwang 2024).

  • Confirmation. DNA from cultured skin fibroblasts is recommended to confirm or exclude germline origin (Baliakas 2024; Duncavage 2022; Hwang 2024). In a Swedish prospective cohort, germline DNA came from cultured skin fibroblasts in 58 patients, T cells in 9 and remission blood in 4 (Tesi 2025). Skin is preferred for hematologic neoplasms, and buccal mucosa, hair bulbs, saliva and blood mononuclear cells are other listed sources (Elbaz Younes 2026; Hwang 2024; Duncavage 2022). A VAF of 40%–60% (heterozygous) or about 80% (homozygous) indicates a possible germline variant that requires confirmation (Elbaz Younes 2026).

  • Diagnostic yield. Applying the Nordic criteria, pathogenic or likely pathogenic germline variants were found in 35% of 85 patients. The yield was 52% when the criterion was a finding on the somatic panel and 6% when it was family history alone. DDX41 accounted for 43% of positive findings (Tesi 2025).

  • Somatic RUNX1 at high VAF. Of 10 patients included for RUNX1 variants at high VAF on somatic panels, 1 had a germline variant (Tesi 2025).

Germline DDX41 predisposition

  • Frequency. About 3%–5% of patients with MDS or AML carry a pathogenic or likely pathogenic germline DDX41 variant (Baliakas 2024).
  • Variants. Two-thirds of germline variants are truncating. About half of carriers acquire a somatic second hit, most often p.Arg525His (Baliakas 2024; Tesi 2025).
  • Penetrance. Penetrance for myeloid neoplasms is up to 10% by age 60 years and about 50% by age 90 years, higher in males. Onset is typically at the end of the seventh decade (Baliakas 2024).
  • Family testing. Adult first-degree relatives are offered testing at 50 years of age or 10 years before the earliest myeloid neoplasm in the family. A three-generation family history is recommended (Baliakas 2024).

Myeloproliferative Neoplasms: Chronic Myeloid Leukemia with Molecular Monitoring, Polycythemia Vera, Essential Thrombocythemia, Primary Myelofibrosis and Other Myeloproliferative Neoplasms

Definition. Myeloproliferative neoplasms (MPN) are clonal neoplasms with proliferation of one or more terminally differentiated myeloid lineages, sometimes with increased reticulin or collagen fibrosis (CAP 2026). Diagnosis integrates clinical features, molecular findings and, in most cases, the trephine biopsy (Khoury 2022; Arber 2022). Most patients present in chronic phase, which may progress to blast phase through secondary cytogenetic or molecular changes (Khoury 2022). The ICC MPN criteria keep marrow morphology central (Thiele 2023) and recommend assays with ≥1% VAF sensitivity for JAK2, CALR and MPL (Arber 2022; Gianelli 2023).

Related page: Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition (WHO-HAEM5 and ICC differences, marrow workup and fibrosis grading) Related page: Mastocytosis, Hypereosinophilic Syndrome and Myeloid/Lymphoid Neoplasms with Tyrosine Kinase Fusions, Chronic Myelomonocytic Leukemia and Other MDS/MPN (clonal eosinophilia workup, tyrosine kinase fusion neoplasms, CMML and JMML) Related page: Acute Myeloid Leukemia with Defining Genetic Abnormalities, Acute Promyelocytic Leukemia and ELN 2022 Genetic Risk (AML with BCR::ABL1 and myeloid blast phase of CML) Related page: Bone Marrow Failure, Pure Red Cell Aplasia and the Laboratory Approach to Cytopenias and Leukocyte Disorders (leukemoid reaction versus CML and the leukocyte alkaline phosphatase score) Related page: Thrombophilia Testing and Antiphospholipid Antibody Testing (JAK2, CALR and MPL testing after thrombosis at unusual sites) Related page: Von Willebrand Factor Testing and Platelet Function Testing (acquired von Willebrand syndrome with extreme thrombocytosis) Related page: Hemoglobin Variant Analysis and Laboratory Diagnosis of Thalassemia (hemoglobin variants)

MPN types in WHO-HAEM5 and ICC 2022

Type WHO-HAEM5 (2022) ICC 2022 Source
Chronic myeloid leukemia MPN; chronic and blast phase MPN (CML, BCR::ABL1-positive); chronic, accelerated and blast phase Khoury 2022; Arber 2022; Gianelli 2023; Arber and Orazi 2025
Polycythemia vera, essential thrombocythemia, primary myelofibrosis MPN MPN; primary myelofibrosis split into early/prefibrotic and overt Khoury 2022; Arber 2022; Gianelli 2023
Chronic neutrophilic leukemia MPN MPN Khoury 2022; Arber 2022; Gianelli 2023
Chronic eosinophilic leukemia MPN, named CEL MPN, named CEL, NOS Khoury 2022; Arber 2022; Gianelli 2023
Juvenile myelomonocytic leukemia MPN Pediatric and/or germline mutation-associated disorders Khoury 2022; Arber 2022; Arber and Orazi 2025
Unclassifiable MPN MPN, not otherwise specified (MPN-NOS) MPN, unclassifiable (MPN-U) Khoury 2022; Arber 2022; Gianelli 2023

Chronic myeloid leukemia: genetics, morphology and phases

BCR::ABL1 fusion and transcript types

  • Fusion. CML is defined by the BCR::ABL1 fusion from t(9;22)(q34;q11), the Philadelphia (Ph) chromosome (Khoury 2022). The t(9;22) is found in 85%–95% of patients (Apperley 2025), and variant translocations involving 9q34, 22q11 and other chromosomes in 5%–10% (Apperley 2025; Cross 2023).
  • Cryptic fusions. RT-PCR is required for the 1%–5% of patients with fusions not visible on karyotype and is recommended for all patients (Apperley 2025; Cross 2023).
  • Transcripts. e13a2 or e14a2 is expressed in 98% of patients, and e14a2 is the most frequent. Both are expressed in 10%. Atypical transcripts, most often e1a2, e6a2, e8a2, e19a2, e13a3 or e14a3, are expressed in 2% (Apperley 2025; Cross 2023).
  • Fusion proteins and phenotype. The usual protein is p210, encoded by both e13a2 and e14a2. About 1% of patients have p190 (e1a2) and 2%–5% a p210 variant or p230 (e19a2). Cases with p230 often show marked neutrophilic maturation and thrombocytosis. Cases with p190 may show increased mature monocytes that mimic CMML (Gianelli 2023; Cross 2023).
  • Asciminib and exon 2. Transcripts lacking ABL1 exon 2 (most often e13a3 and e14a3) are expected to be resistant to asciminib (Apperley 2025).

Blood and marrow morphology of chronic-phase CML

  • Blood. Leukocytosis (median 80 × 10⁹/L) with neutrophils at all stages of maturation, peaks of myelocytes and segmented neutrophils, no significant dysplasia, and absolute basophilia and eosinophilia. Some patients lack significant leukocytosis and present with sustained thrombocytosis that mimics ET (Gianelli 2023).
  • Marrow. The marrow is hypercellular with granulocytic proliferation and a left shift, reduced erythropoiesis, blasts usually <5% and small hypolobulated ("dwarf") megakaryocytes in about 40%–50%. Eosinophils and basophils are increased, and pseudo-Gaucher histiocytes may be present (Gianelli 2023).
  • Fibrosis. MF-1 fibrosis at diagnosis correlates with a lower rate of major molecular response in the first year of tyrosine kinase inhibitor (TKI) therapy (Arber 2022; Gianelli 2023).
  • Leukemoid reaction. The distinction from a leukemoid reaction is taught in Bone Marrow Failure, Pure Red Cell Aplasia and the Laboratory Approach to Cytopenias and Leukocyte Disorders.

Laboratory investigations at diagnosis of CML (ELN 2025)

Investigation Recommendation Source
Marrow morphology and conventional cytogenetics Recommended at diagnosis to stage disease and detect variant translocations, cryptic rearrangements and additional chromosomal abnormalities (ACA); ACA present in about 7% of chronic-phase patients; cytogenetics mandatory, with ≥20 metaphases analyzed Apperley 2025; Cross 2023
FISH Diagnostic screening, usually on blood; commercial probes do not identify ACA Apperley 2025; Cross 2023
RT-PCR Standard method; identifies the transcript type for later monitoring Apperley 2025; Cross 2023
NGS for somatic mutations Mutations in 20%–30%; not routine in chronic phase; recommended in blast phase Apperley 2025; Cross 2023
BCR::ABL1 kinase domain mutations Very rarely present at diagnosis in chronic phase; testing not recommended Apperley 2025; Cross 2023
Risk score EUTOS long-term survival (ELTS) score replaces Sokal, Euro and EUTOS scores; in 2949 patients ELTS discriminated long-term survival better than Sokal, and 56% of Sokal high-risk patients were not high risk by ELTS Apperley 2025; Hochhaus 2020; Pfirrmann 2020

ELTS score (ELN 2020). The ELTS score is 0.0025 × (age/10)³ + 0.0615 × spleen size + 0.1052 × peripheral blood blasts + 0.4104 × (platelet count/1000)^−0.5, using pre-treatment values: age in years, spleen in cm below the costal margin by palpation, blasts as a percentage of the blood differential and platelets in × 10⁹/L. Risk groups are low <1.5680, intermediate 1.5680–2.2185 and high >2.2185 (Hochhaus 2020). The score predicts the probability of dying of CML, and in its derivation it identified 61% of patients as low risk and 12% as high risk (Pfirrmann 2016).

Phases of CML by classification (WHO-HAEM5, ICC, WHO 2017 and ELN)

Phase WHO-HAEM5 (2022) ICC 2022 WHO revised 4th edition (2016/2017), previous rule ELN 2013 (previous ELN rule) Source
Chronic Blasts <20% in blood and marrow; high-risk features at diagnosis: high ELTS score, blasts 10%–19%, blood basophils ≥20%, ACA (3q26.2 rearrangement, −7, i(17q), complex karyotype), clusters of small megakaryocytes with fibrosis Blasts <10% Blasts <10% Blasts <15% Apperley 2025; Khoury 2022; CAP 2026
Accelerated Omitted; high-risk features on treatment: no complete hematologic response on first-line TKI, resistance to a second-generation TKI (unless due to a BCR::ABL1 mutation), new ACA, compound BCR::ABL1 mutations Blasts 10%–19%; blood basophils ≥20%; ACA at diagnosis or emerging on treatment (second Ph, +8, i(17q), +19, complex karyotype, 3q26.2 abnormalities) Blasts 10%–19%; basophils ≥20%; platelets <100 or >1,000 × 10⁹/L unrelated or unresponsive to treatment; splenomegaly or rising leukocyte count unresponsive to treatment; ACA at diagnosis or emerging on treatment Blasts 15%–29%; basophils ≥20%; platelets <100 × 10⁹/L unrelated to treatment; major-route ACA emerging on treatment Apperley 2025; Arber 2022; Gianelli 2023; Arber and Orazi 2025
Blast ≥20% myeloid blasts in blood or marrow; extramedullary blast proliferation; bona fide lymphoblasts in blood or marrow, even if <10% ≥20% blasts in blood or marrow; myeloid sarcoma; >5% lymphoblasts warrants consideration of lymphoid blast phase ≥20% blasts; extramedullary blast proliferation ≥30% blasts; extramedullary blast proliferation Apperley 2025; Arber 2022; Gianelli 2023; Khoury 2022; CAP 2026
  • Rationale for the WHO-HAEM5 change. With TKI therapy and monitoring, progression has decreased and 10-year overall survival is 80%–90%. Resistance from ABL1 kinase mutations, ACA and blast phase became the key disease attributes (Khoury 2022; CAP 2026).

  • ELN position. The ELN 2025 panel did not fully endorse either the WHO or the ICC phase classification (Apperley 2025).

  • Lymphoid blast phase. The optimal lymphoblast cutoff remains unclear in WHO-HAEM5 (Khoury 2022). Immunophenotyping is required to confirm lymphoid lineage (Arber 2022; Gianelli 2023). Blast phase is myeloid in about 60%, lymphoid in 30% and megakaryocytic or undifferentiated in 10% (Gianelli 2023).

  • ACA. Major-route ACA (+8, +Ph, i(17q), +19, +17, +21) occur in >5% of progressions and minor-route ACA (−7/7q−, 11q23, 3q26.2) in <5% (Apperley 2025). Among cases with additional changes at progression, +8 occurs in 34%, +Ph in 30% and i(17q) in 20% (Gianelli 2023).

  • AML with BCR::ABL1. The distinction from myeloid blast phase of CML is taught in Acute Myeloid Leukemia with Defining Genetic Abnormalities, Acute Promyelocytic Leukemia and ELN 2022 Genetic Risk.

  • Genetics of blast phase. ASXL1, RUNX1, BCOR/BCORL1 and IKZF1 mutations are each found in 15%–20% of blast phase cases. ASXL1 and TP53 mutations are associated with myeloid blast phase, and IKZF1 and CDKN2A/B mutations or deletions with lymphoid blast phase (Cross 2023).

  • Lymphoid blast phase versus Ph-positive ALL. Interphase FISH on blood granulocytes separates de novo lymphoid blast phase of CML (granulocytes BCR::ABL1-positive) from BCR::ABL1-positive acute lymphoblastic leukemia (granulocytes negative) (Cross 2023).

Molecular monitoring of CML: International Scale, milestones and kinase domain mutations

  • International Scale (IS). Molecular response is expressed as BCR::ABL1/control on the IS, with ABL1, BCR or GUSB as the reference gene, for patients with e13a2 or e14a2 transcripts (Apperley 2025; Cross 2023).

  • Assays. Laboratory-developed tests with conversion factors, IS-calibrated commercial kits and IS-calibrated whole systems are acceptable when validated locally. RT-qPCR is the most common method, and reverse transcriptase droplet digital PCR (RT-ddPCR) is an acceptable alternative (Apperley 2025; Cross 2023).

  • Frequency. Blood counts are checked every 2 weeks until complete hematologic response. BCR::ABL1 is measured at least every 3 months until major molecular response is achieved and confirmed, then every 4–6 months while the response is stable (Apperley 2025).

  • Previous monitoring rule. ELN 2020 required RT-qPCR at least every 3 months even after MMR was achieved and confirmed (Hochhaus 2020).

  • Atypical transcripts. The 2% of patients with atypical fusions are monitored by personalized RT-qPCR against their own baseline or by RT-ddPCR (Apperley 2025; Cross 2023). Karyotype is used for transcripts that cannot be measured by RT-qPCR (Apperley 2025).

  • Transcript type and RT-qPCR. Higher reported rates of MMR and MR4 with e14a2 partly reflect differences in RT-qPCR amplification efficiency between transcripts. RT-ddPCR is less sensitive to amplification kinetics (Apperley 2025; Cross 2023).

Molecular response levels (ELN 2025, unchanged from ELN 2020)

Deep molecular response (DMR) is MR4 or deeper, whether or not BCR::ABL1 is detected. Tests are optimized to detect MR4.5 routinely (Apperley 2025; Cross 2023).

Undetectable BCR::ABL1 with <10,000 ABL1 transcripts is not evaluable for DMR and is treated as a technical failure (Cross 2023).

Response BCR::ABL1 (IS) Minimum ABL1 control copies Minimum GUSB control copies Source
Equivalent to complete cytogenetic response ≤1% 10,000 24,000 Apperley 2025; Cross 2023
Major molecular response (MMR) ≤0.1% 10,000 24,000 Apperley 2025; Hochhaus 2020
MR4 (4-log reduction from the IRIS standardized baseline) ≤0.01% 10,000 24,000 Apperley 2025; Cross 2023
MR4.5 ≤0.0032% 32,000 77,000 Apperley 2025; Cross 2023
MR5 ≤0.001% 100,000 240,000 Apperley 2025; Cross 2023

CML response milestones (ELN 2025 and ELN 2020)

Time point ELN 2025 favorable ELN 2025 warning ELN 2025 unfavorable ELN 2020 optimal / warning / failure (previous rule) Source
Baseline Not applicable High-risk ACA, high-risk ELTS score Not applicable Warning: high-risk ACA, high-risk ELTS score Apperley 2025; Hochhaus 2020
3 months ≤10% >10% >10% if confirmed within 1–3 months ≤10% / >10% / >10% if confirmed within 1–3 months Apperley 2025; Hochhaus 2020
6 months ≤1% >1%–10% >10% (established resistance) ≤1% / >1%–10% / >10% Apperley 2025; Hochhaus 2020
12 months ≤0.1% >0.1%–1% >1% (1%–10% weighed with other considerations) ≤0.1% / >0.1%–1% / >1% Apperley 2025; Hochhaus 2020
Any time ≤0.1% >0.1%–1%; loss of MMR Loss of a previous response, resistant BCR::ABL1 mutations, high-risk ACA ≤0.1% / >0.1%–1% or loss of MMR / >1%, resistance mutations, high-risk ACA Apperley 2025; Hochhaus 2020
  • Terminology change. ELN 2025 replaces "optimal", "warning" and "failure" with "favorable" (switch unnecessary), "warning" (switch may become necessary) and "unfavorable" (switch preferred). The milestone values are unchanged (Apperley 2025).
  • Single values. No switch decision is made on a single BCR::ABL1 result. Steadily declining values in late responders may not require a switch. An unexplained rising RT-qPCR result triggers karyotype, kinase domain mutation analysis and a TKI switch (Apperley 2025).
  • Outcome data for the ELN 2025 terms. In the German CML IV study, 10-year survival of patients who would previously have met failure criteria at 3, 6 or 12 months was about 80%. Survival dropped to 55% when BCR::ABL1 was >10% at 12 months (Apperley 2025).
  • ELN 2020 TFR goal. ELN 2020 defined the optimal response for patients aiming at treatment-free remission as BCR::ABL1 ≤0.01% (MR4) at any time (Hochhaus 2020).

BCR::ABL1 kinase domain mutations

  • Indications for testing. Resistance or warning responses, presentation in or progression to blast phase, and relapse after allogeneic transplantation when a mutation was present before it. Testing is not recommended after loss of MMR following TKI discontinuation (Apperley 2025; Cross 2023).
  • Frequency of unfavorable responses. Unfavorable responses occur in about 15%–20% of patients on first-line therapy and up to 50% in later lines. Adherence is assessed whenever the response is unsatisfactory (Apperley 2025).
  • Methods. Sanger sequencing is unlikely to detect clones forming <20% of the leukemic burden (Apperley 2025); its limit of detection is a VAF of about 15%–20% (Cross 2023). cDNA-based targeted NGS is recommended and can be used when *BCR::ABL1* IS is >0.1% (Apperley 2025; Cross 2023). Commercial myeloid gene panels that include ABL1 are not recommended for this purpose (Apperley 2025).
Mutation TKIs recommended (ELN 2025) Source
T315I Ponatinib, asciminib Apperley 2025; Cross 2023
Y253H Dasatinib, bosutinib, ponatinib, asciminib Apperley 2025; Cross 2023
E255K/V Dasatinib, ponatinib, asciminib Apperley 2025; Cross 2023
V299L Nilotinib, ponatinib, asciminib Apperley 2025; Cross 2023
F317L/V/I/C, T315A Nilotinib, bosutinib, ponatinib, asciminib Apperley 2025; Cross 2023
F359V/I/C Dasatinib, ponatinib Apperley 2025; Cross 2023
M244V Nilotinib, dasatinib, bosutinib, ponatinib Apperley 2025; Cross 2023
Myristoyl pocket mutations (A337V/T, L340Q, A344P, A433D, G463D/S, P465S/Q, V468F, F497L, I502L/N, V506L/M) Any ATP-competitive TKI Apperley 2025; Cross 2023

Compound mutations are 2 mutations in cis in the same BCR::ABL1 molecule. Most are resistant to imatinib and second-generation TKIs (Apperley 2025; Cross 2023), and resistance is highest when T315I is included (Apperley 2025).

Treatment-free remission in CML (ELN 2025 and ELN 2020)

Requirement ELN 2025 ELN 2020 (previous rule) Source
Mandatory First chronic phase; motivated patient; high-quality IS monitoring with rapid turnaround; agreement to frequent monitoring First chronic phase; motivated patient; high-quality IS PCR with rapid turnaround; agreement to frequent monitoring Apperley 2025; Hochhaus 2020
Minimal (stop allowed) First line, or second line after intolerance or resistance due to a mutation sensitive to another TKI; e13a2 or e14a2 transcript (atypical transcripts in laboratories with high-standard quantification); TKI >5 years (>4 years for second-generation TKI); DMR (MR4 or better) >2 years First line, or second line if intolerance was the only reason for the change; e13a2 or e14a2 transcript; TKI >5 years (>4 years for second-generation TKI); DMR >2 years; no prior treatment failure Apperley 2025; Hochhaus 2020
Optimal (stop recommended for consideration) TKI >5 years; DMR >3 years if MR4; DMR >2 years if MR4.5 TKI >5 years; DMR >3 years if MR4; DMR >2 years if MR4.5 Apperley 2025; Hochhaus 2020
Monitoring after stopping Every 6–8 weeks for 6 months, every 2 months for months 6–12, then every 3–6 months; restart TKI on loss of MMR Monthly for 6 months, every 2 months for months 6–12, then every 3 months Apperley 2025; Hochhaus 2020
  • Outcomes. About 40%–50% of patients remain off treatment. More than 80% of molecular recurrences occur in the first 6–8 months, and 90%–95% of patients who restart regain DMR (Apperley 2025; Hochhaus 2020). Late loss of MMR occurred in up to 14% more than 2 years after stopping (Apperley 2025).
  • Withdrawal syndrome. Polymyalgia or arthralgia occurs in 20%–30% of patients after TKI cessation (Apperley 2025; Hochhaus 2020).

Polycythemia vera: diagnostic criteria and differential diagnosis of erythrocytosis

Diagnostic criteria for polycythemia vera (WHO-HAEM5, ICC 2022 and WHO 2017)

Diagnosis requires all 3 major criteria, or the first 2 major criteria plus the minor criterion, in both systems (CAP 2026; Gianelli 2023).

Criterion WHO-HAEM5 (2022) ICC 2022 WHO 2017 (previous rule) Source
Major 1 Hemoglobin >16.5 g/dL (men) or >16.0 g/dL (women), or hematocrit >49% (men) or >48% (women) Same hemoglobin and hematocrit thresholds, or red cell mass >25% above the mean normal predicted value Same thresholds; red cell mass was a criterion CAP 2026; Arber 2022; Gianelli 2023; Khoury 2022; Thiele 2025
Major 2 Age-adjusted hypercellularity with trilineage proliferation (panmyelosis) and pleomorphic, mature megakaryocytes Same, with megakaryocytes without atypia Same CAP 2026; Arber 2022; Gianelli 2023; Thiele 2025
Major 3 JAK2 V617F or JAK2 exon 12 mutation Same; highly sensitive V617F assay (<1%) and search for non-canonical JAK2 mutations in exons 12–15 when negative Same CAP 2026; Arber 2022; Gianelli 2023; Thiele 2025
Minor Subnormal serum erythropoietin Subnormal serum erythropoietin Same CAP 2026; Arber 2022; Gianelli 2023; Thiele 2025
  • ICC biopsy waiver. In ICC, a marrow biopsy may be omitted with hemoglobin >18.5 g/dL (men) or >16.5 g/dL (women) and hematocrit >55.5% (men) or >49.5% (women) together with a JAK2 mutation (Arber 2022; Gianelli 2023; Thiele 2025).

  • Change in WHO-HAEM5. Red cell mass measured with ⁵¹Cr-labeled red cells was removed as a criterion because it is rarely performed (Khoury 2022; Thiele 2025).

  • Morphology. The marrow shows panmyelosis with left-shifted erythropoiesis and granulopoiesis. Megakaryocytes vary markedly in size without significant atypia and form loose clusters. MF-1 is present in about 20% at diagnosis and is associated with progression to post-PV myelofibrosis (Gianelli 2023; Thiele 2025).

  • Genetics. JAK2 V617F is present in about 97% and other activating JAK2 mutations, including exon 12, in 3%. Patients with exon 12 mutations are younger (Gianelli 2023; Arber 2022), present with predominant erythropoiesis and subnormal erythropoietin (Gianelli 2023; Scott 2007), and have a prognosis similar to V617F (Gianelli 2023; Arber 2022). Variants other than JAK2 occur in about half of PV (Gianelli 2023; Tefferi and Barbui 2023), and ASXL1, SRSF2, IDH2 and EZH2 variants are associated with inferior survival (median 7.7 versus 16.9 years) (Gianelli 2023).

  • Masked PV. An international study defined masked PV as JAK2 V617F with PV marrow morphology and hemoglobin of 16.0–18.4 g/dL in men or 15.0–16.4 g/dL in women, below the WHO 2008 thresholds of 18.5 and 16.5 g/dL (Barbui 2014; Thiele 2025). A hematocrit of 49% in men and 48% in women was later adopted as the cutoff that separates JAK2-mutated ET from masked PV (Thiele 2025). These patients are often classified as ET. Iron deficiency, plasma volume expansion and sex-related differences in red cell mass obscure the erythrocytosis, and masked PV is consistent with an early stage of overt PV (Barbui 2026). The revised WHO 4th edition (2016) lowered the hemoglobin and hematocrit thresholds to capture masked PV (Barbui 2018; Barbui 2026), which reduced the number of unclassifiable cases (Gianelli 2023). WHO-HAEM5 avoids the term masked PV and recommends MPN-NOS with close follow-up for patients who meet the other PV criteria without sufficient erythrocytosis (Thiele 2025).

  • Clinical features. PV causes hypertension, hyperviscosity with microcirculatory symptoms, pruritus and venous or arterial thrombosis (Gianelli 2023; Tefferi and Barbui 2023).

  • Course. Thrombosis is the first manifestation in about 20%. Leukemic transformation is 2.3% at 10 years and 5.5% at 15 years. Post-PV myelofibrosis develops in 6%–14% at 15 years (Gianelli 2023).

Post-polycythemia vera myelofibrosis

Required: a previous diagnosis of PV and fibrosis grade 2–3. At least 2 additional criteria are required from: anemia or sustained loss of the need for phlebotomy or cytoreduction; leukoerythroblastosis; splenomegaly increased by >5 cm from baseline or newly palpable; and, as the fourth criterion, ≥2 constitutional symptoms (>10% weight loss in 6 months, night sweats, unexplained fever >37.5 °C) (CAP 2026; Arber 2022). The ICC MPN companion paper lists a lactate dehydrogenase above the reference range as the fourth criterion (Gianelli 2023).

Differential diagnosis of erythrocytosis

Category Mechanism and laboratory findings Source
Polycythemia vera JAK2 V617F or exon 12 mutation; subnormal erythropoietin; panmyelosis Gianelli 2023; McMullin 2021
Secondary erythrocytosis, hypoxic Hypoxia-related conditions and cardiopulmonary disease; erythropoietin elevated or inappropriately normal Giannella 2026; McMullin 2021
Secondary erythrocytosis, erythropoietin-secreting neoplasm Neoplasm-associated erythropoietin production Giannella 2026
High-oxygen-affinity hemoglobin Left-shifted oxygen dissociation curve (reduced P50) causes tissue hypoxia and erythropoietin production; autosomal dominant; the HBB variant Hb Kempsey was the only variant in one cohort associated with elevated erythropoietin Giannella 2026; McMullin 2021
Oxygen-sensing pathway variants VHL and EPAS1 (hypoxia-inducible factor 2α) variants cause congenital erythrocytosis Giannella 2026; McMullin 2021
Idiopathic erythrocytosis Diagnosis of exclusion after secondary causes and JAK2-mutated MPN; erythropoietin mostly within the reference interval Giannella 2026; McMullin 2021

Primary familial and congenital erythrocytosis is most often caused by germline truncating EPOR mutations that remove the negative regulatory domain of the erythropoietin receptor, and erythropoietin is below the reference range, as in PV (McMullin 2021). JAK2 exon 12 mutations also occur in patients previously diagnosed with idiopathic erythrocytosis (Scott 2007).

Splanchnic vein thrombosis and Budd–Chiari syndrome raise the possibility of PV (Gianelli 2023). The liver pathology of Budd–Chiari syndrome is taught in Hepatic Vascular Disorders and Liver Transplant Pathology: Outflow Obstruction, Porto-Sinusoidal Vascular Disorder, Rejection and Donor Biopsy. JAK2, CALR and MPL testing after thrombosis at unusual sites is taught in Thrombophilia Testing and Antiphospholipid Antibody Testing.

Essential thrombocythemia: diagnostic criteria and distinction from prefibrotic primary myelofibrosis

Diagnostic criteria for essential thrombocythemia (WHO-HAEM5 and ICC 2022)

Diagnosis requires all 4 major criteria, or the first 3 major criteria plus the minor criterion. The criteria did not change from WHO 2017 (Khoury 2022; CAP 2026; Gianelli 2023; Thiele 2025).

Criterion WHO-HAEM5 and ICC 2022 Source
Major 1 Platelet count ≥450 × 10⁹/L CAP 2026; Arber 2022; Gianelli 2023; Thiele 2025
Major 2 Proliferation mainly of megakaryocytes, with increased enlarged mature megakaryocytes with hyperlobulated (staghorn-like) nuclei; no significant increase or left shift of granulopoiesis or erythropoiesis; very rarely MF-1 CAP 2026; Arber 2022; Gianelli 2023; Thiele 2025
Major 3 Criteria for CML, PV, PMF or other myeloid neoplasms not met CAP 2026; Arber 2022; Gianelli 2023
Major 4 JAK2, CALR or MPL mutation CAP 2026; Arber 2022; Gianelli 2023; Thiele 2025
Minor Clonal marker, or no evidence of reactive thrombocytosis (iron deficiency, chronic infection or inflammation, medication, neoplasia, splenectomy) CAP 2026; Arber 2022; Gianelli 2023
  • ICC definition of a dense cluster. ≥3 megakaryocytes lying adjacent without other marrow cells in between. Clusters in ET are infrequent and small, usually with 6 or fewer megakaryocytes (Arber 2022; Gianelli 2023).

  • Epidemiology and thrombosis in ET. Incidence is 1.2–3.0 per 100,000 per year with a median age of 58 years, and more than 50% are asymptomatic. Fatal and nonfatal thrombosis occurred at 1.9% per patient-year in a large cohort (Gianelli 2023).

  • Genetics. JAK2 V617F is associated with higher thrombotic risk (Gianelli 2023; Tefferi 2024) and lower risk of post-ET myelofibrosis (Gianelli 2023). CALR-mutated patients are younger and more often male, with higher platelet counts and fewer thromboses (Gianelli 2023; Klampfl 2013; Nangalia 2013).

  • Driver type and fibrotic progression in ET. MPL and CALR type 1 mutations are associated with a higher risk of myelofibrotic transformation (Tefferi 2024; Thiele 2025).

  • Acquired von Willebrand syndrome. The association with extreme thrombocytosis is taught in Von Willebrand Factor Testing and Platelet Function Testing.

Essential thrombocythemia versus prefibrotic primary myelofibrosis

Feature Essential thrombocythemia Prefibrotic primary myelofibrosis Source
Cellularity Normocellular for age Hypercellular for age Gianelli 2023; Arber 2022
Granulopoiesis Not increased or left-shifted Pronounced proliferation and left shift; increased myeloid to erythroid ratio Gianelli 2023; Arber 2022
Megakaryocytes Large to giant, hyperlobulated, mature cytoplasm; loose clusters Atypical: increased nuclear to cytoplasmic ratio, abnormal chromatin clumping, bulbous hyperchromatic nuclei; large dense clusters (>6 cells) Gianelli 2023; Arber 2022
Reticulin MF-0; very rarely MF-1 MF-0 or MF-1 Gianelli 2023; Arber 2022
Clinical findings Usually normal leukocyte count, no anemia, normal lactate dehydrogenase, less often splenomegaly Anemia, leukocytosis ≥11 × 10⁹/L, splenomegaly, increased lactate dehydrogenase Gianelli 2023; Arber 2022
Survival Median 14.7 to about 21.8 years Median 11–17 years Gianelli 2023; Thiele 2025

Reclassification studies found prefibrotic PMF in 14%–18% of cases originally diagnosed as ET (Gianelli 2023).

In real-world practice at least 20%–30% of cases diagnosed as ET are prefibrotic PMF (Thiele 2025). ET carries a lower risk of major hemorrhage than prefibrotic PMF. At 10 years, post-ET myelofibrosis develops in 0.8%–4.5% and blast phase in 0.7%–1.9% of patients with ET (Arber 2022).

Post-essential thrombocythemia myelofibrosis

Required: a previous diagnosis of ET and fibrosis grade 2–3. Two additional criteria are required: anemia with a >2 g/dL decrease from baseline; leukoerythroblastosis; an increase in splenomegaly >5 cm from baseline or newly palpable splenomegaly; elevated lactate dehydrogenase; and ≥2 constitutional symptoms (CAP 2026; Arber 2022; Gianelli 2023).

Primary myelofibrosis: prefibrotic and overt stages

Diagnostic criteria for primary myelofibrosis (WHO-HAEM5 and ICC 2022)

Both stages require all 3 major criteria and at least 1 minor criterion confirmed in 2 consecutive determinations (Arber 2022; CAP 2026; Gianelli 2023; Shao 2025). WHO-HAEM5 and ICC criteria are the same except for the minor criteria shown in the table (CAP 2026; Thiele 2025).

Criterion Prefibrotic PMF Overt (fibrotic) PMF Source
Major 1 Megakaryocytic proliferation and atypia, reticulin fibrosis grade ≤1, increased age-adjusted cellularity, granulocytic proliferation and often decreased erythropoiesis Megakaryocytic proliferation and atypia with reticulin and/or collagen fibrosis grade 2 or 3 CAP 2026; Arber 2022; Gianelli 2023; Thiele 2025
Major 2 Criteria for CML, PV, ET, MDS or other myeloid neoplasms not met Same CAP 2026; Arber 2022; Gianelli 2023
Major 3 JAK2, CALR or MPL mutation; or another clonal marker (for example ASXL1, EZH2, TET2, IDH1, IDH2, SF3B1, SRSF2); or absence of minor reactive reticulin fibrosis Same, with absence of reactive fibrosis CAP 2026; Arber 2022; Gianelli 2023
Minor, ICC Anemia not attributed to a comorbid condition; leukocytosis ≥11 × 10⁹/L; palpable splenomegaly; lactate dehydrogenase above the reference range Same, plus leukoerythroblastosis Arber 2022; Gianelli 2023; Thiele 2025
Minor, WHO-HAEM5 Anemia not attributed to a comorbid condition; leukocytosis ≥11 × 10⁹/L; splenomegaly detected clinically and/or by imaging; lactate dehydrogenase above the upper limit of the reference range; leukoerythroblastosis Same CAP 2026; Thiele 2025
  • Reactive fibrosis. Minimal reticulin fibrosis secondary to infection, autoimmune disease, chronic inflammation, hairy cell leukemia or another lymphoid neoplasm, metastatic malignancy or toxic myelopathy does not count as a clonal criterion (Arber 2022; Gianelli 2023).
  • Megakaryocyte atypia. Small to giant megakaryocytes with severe maturation defects (cloud-like, hypolobulated and hyperchromatic nuclei) and large dense clusters of mostly >6 megakaryocytes lying strictly adjacent (Arber 2022; Gianelli 2023). Bulbous, cloud-like nuclei are rare in ET or PV (Shao 2025).
  • Overt stage morphology. Reticulin and collagen deposition lowers cellularity and erythropoiesis. Dilated sinusoids with intrasinusoidal hematopoiesis, increased microvessel density and osteosclerosis are common. Leukoerythroblastosis and anisopoikilocytosis with dacrocytes (teardrop red cells) correlate with fibrosis (Gianelli 2023; Thiele 2025). Fibrosis grading is taught in Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition.
  • Cytogenetics. Abnormalities are present in about 18% of prefibrotic PMF and 30%–40% of overt PMF (Gianelli 2023; Thiele 2025). Unfavorable abnormalities include complex karyotype, +8, −7/7q−, i(17q), −5/5q−, 12p−, inv(3) and 11q23 rearrangement (Gianelli 2023).
  • High molecular risk mutations. ASXL1, SRSF2, EZH2, IDH1 and IDH2 mutations define high molecular risk (Shao 2025; Gianelli 2023); some models add TP53 and U2AF1. ≥1 such mutation is associated with inferior overall and leukemia-free survival, and ≥2 with a dismal outcome (Shao 2025).
  • CALR types. Type 1 is a 52-bp deletion and type 2 a 5-bp insertion. Type 1-like mutations are associated with longer overall survival (Shao 2025; Tefferi 2023).
  • Course. Median survival is 11–17 years in prefibrotic PMF (Gianelli 2023; Thiele 2025) and about 7 years in overt PMF. Myeloid blast phase occurs in 11% at 5 years and 23% at 10 years in overt PMF (Gianelli 2023).

Prognostic scores in primary myelofibrosis: DIPSS and MIPSS70

Score Variables Risk categories Source
Dynamic International Prognostic Scoring System (DIPSS) Age >65 years; hemoglobin <100 g/L; leukocytes >25 × 10⁹/L; blood blasts ≥1%; constitutional symptoms; applicable at any time during the disease Low, intermediate-1, intermediate-2, high Passamonti 2010
MIPSS70 (transplant-age patients) Hemoglobin <100 g/L; leukocytes >25 × 10⁹/L; platelets <100 × 10⁹/L; blood blasts ≥2%; fibrosis grade ≥2; constitutional symptoms; absence of CALR type 1 mutation; ≥1 high molecular risk mutation (ASXL1, EZH2, SRSF2, IDH1/2); ≥2 high molecular risk mutations Low, intermediate, high; MIPSS70-plus adds cytogenetics and a very high category Guglielmelli 2018

Post-PV and post-ET myelofibrosis versus PMF

Post-PV and post-ET myelofibrosis cannot be reliably separated from overt PMF clinically or morphologically, and the diagnosis requires a documented prior PV or ET. A JAK2 exon 12 mutation indicates post-PV myelofibrosis because exon 12 mutations occur only in PV. Median survival was 73 months in post-ET myelofibrosis, 48 months in post-PV myelofibrosis and 45 months in PMF (Shao 2025).

Driver mutation frequencies in BCR::ABL1-negative MPN

Neoplasm JAK2 V617F JAK2 exon 12 CALR MPL Triple negative Source
Polycythemia vera About 96%–97% About 3%–4% Absent Not a PV criterion Rare; a JAK2 mutation is expected in almost all PV Gianelli 2023; Shao 2025; Klampfl 2013; Barbui 2018
Essential thrombocythemia About 50%–70% Not present About 20%–25% About 3%–8% About 10%–25% Gianelli 2023; Shao 2025; Thiele 2025
Primary myelofibrosis About 50%–60% Not present About 25%–30% About 5%–10% About 5%–14% Shao 2025; Gianelli 2023; Klampfl 2013
  • Triple-negative PMF. Triple-negative PMF has a poor prognosis, more thrombocytopenia and less splenomegaly (Shao 2025; Gianelli 2023), more trisomy 8 and more ASXL1 and SRSF2 co-mutations (Shao 2025).
  • Non-driver mutations. TET2, ASXL1 and DNMT3A mutations are found in over half of patients with MPN and are more frequent in PMF and advanced disease (Khoury 2022).

Chronic neutrophilic leukemia and chronic eosinophilic leukemia

Diagnostic criteria for chronic neutrophilic leukemia

Criterion WHO-HAEM5 (2022) ICC 2022 Source
Leukocytosis ≥25 × 10⁹/L ≥13 × 10⁹/L with CSF3R T618I or another activating CSF3R mutation; ≥25 × 10⁹/L without CAP 2026; Arber 2022; Gianelli 2023; Szuber 2024
Blood Segmented and band neutrophils ≥80% of leukocytes; precursors <10%; myeloblasts rare; monocytes <10%; no dysgranulopoiesis Segmented and band neutrophils ≥80%; precursors (promyelocytes, myelocytes and metamyelocytes) <10%; no significant dysgranulopoiesis; blasts rare; monocytes <10% CAP 2026; Arber 2022; Gianelli 2023
Marrow Hypercellular; neutrophils increased with normal maturation; myeloblasts <5% Hypercellular; neutrophils increased with normal maturation CAP 2026; Arber 2022; Gianelli 2023
Genetic or clinical CSF3R T618I or another activating CSF3R mutation, or persistent neutrophilia (≥3 months), splenomegaly and no reactive cause, including no plasma cell neoplasm unless myeloid clonality is shown Same CAP 2026; Arber 2022; Gianelli 2023
Exclusions CML, PV, ET, PMF; PDGFRA, PDGFRB, FGFR1 rearrangement or PCM1::JAK2 CML, PV, ET, PMF; tyrosine kinase fusion neoplasms CAP 2026; Arber 2022; Gianelli 2023
  • Previous rule. The leukocyte threshold before 2022 was ≥25 × 10⁹/L (Arber 2022; Arber and Orazi 2025).
  • Genetics. CSF3R mutations are present in >60% (Khoury 2022) to 80%–100% (Gianelli 2023) of CNL. CSF3R T618I was found in 83% of WHO-defined CNL and was absent in atypical CML (Pardanani 2013; Thomopoulos 2022). In the study that identified them, activating CSF3R mutations were present in 59% of patients with CNL or atypical CML (Maxson 2013). SETBP1 is mutated in about 41% of CNL, often with T618I (Thomopoulos 2022), and in 33% of WHO-defined CNL in another series (Pardanani 2013). The absence of a CSF3R mutation does not exclude CNL (Arber 2022). CSF3R T618I is associated with older age, higher leukocyte counts, lower hemoglobin and platelets, abnormal karyotype and lower overall survival than other CSF3R mutations. SETBP1, ASXL1 and SRSF2 mutations also occur (Gianelli 2023).
  • Morphology. The marrow is usually >90% cellular, with a myeloid to erythroid ratio that may exceed 20:1 (Gianelli 2023; Arber 2022). Monocytosis, basophilia, eosinophilia or dysgranulopoiesis prompts review for MDS/MPN (Gianelli 2023; Arber 2022).
  • Phases. Blasts of 10%–19% define accelerated phase and ≥20% blast phase (Arber 2022; Gianelli 2023).
  • Reactive neutrophilia. Leukemoid reaction and serum G-CSF are taught in Bone Marrow Failure, Pure Red Cell Aplasia and the Laboratory Approach to Cytopenias and Leukocyte Disorders.

Diagnostic criteria for chronic eosinophilic leukemia

Criterion WHO-HAEM5 (2022): CEL ICC 2022: CEL, NOS WHO 2017 (previous rule) Source
Hypereosinophilia >1.5 × 10⁹/L on ≥2 occasions at least 4 weeks apart ≥1.5 × 10⁹/L and ≥10% of leukocytes Sustained over 6 months CAP 2026; Arber 2022; Gianelli 2023; Khoury 2022
Clonality Required; CHIP excluded Clonal cytogenetic abnormality and/or somatic mutation; marrow findings suffice with persistent eosinophilia when clonality and increased blasts are absent and other causes are excluded Clonality or increased blasts (≥2% in blood or 5%–19% in marrow) CAP 2026; Arber 2022; Gianelli 2023; Khoury 2022
Marrow morphology Abnormal, usually hypercellular with dysplastic megakaryocytes and increased eosinophils Increased cellularity with dysplastic megakaryocytes, often fibrosis, with an eosinophilic infiltrate, or blasts ≥5% in marrow and/or ≥2% in blood Not required CAP 2026; Arber 2022; Gianelli 2023; Khoury 2022
  • Exclusions. WHO-HAEM5 excludes other MPN, MDS/MPN, MDS, MLN-TK, mastocytosis and AML (CAP 2026). ICC requires blasts <20% in blood and marrow and excludes AML, including AML with recurrent genetic abnormalities and <20% blasts, tyrosine kinase fusions including BCR::ABL1, other MPN, CMML and SM (Arber 2022; Gianelli 2023).

The stepwise clonal eosinophilia workup and the tyrosine kinase fusion neoplasms are taught in Mastocytosis, Hypereosinophilic Syndrome and Myeloid/Lymphoid Neoplasms with Tyrosine Kinase Fusions, Chronic Myelomonocytic Leukemia and Other MDS/MPN.

Myeloproliferative neoplasm, not otherwise specified (unclassifiable)

  • Definition. An MPN with clinical, laboratory, morphologic and molecular features of MPN that lacks the criteria of any specific type or overlaps between types (Khoury 2022; Arber 2022). It accounts for 5%–10% of MPN (Gianelli 2023).
  • Groups. Early presentations without fully developed features; advanced stages in which fibrosis, osteosclerosis or transformation obscures the underlying type; and MPN obscured by a coexisting neoplastic or inflammatory disorder (CAP 2026; Arber 2022; Gianelli 2023). Some cases presenting with portal or splanchnic vein thrombosis belong to the first group (CAP 2026; Arber 2022). About 50% of patients with MPN presenting with splanchnic vein thrombosis are classified as MPN-U (Gianelli 2023).
  • ICC criteria. All 3 required: clinical and hematologic features of MPN; a JAK2, CALR or MPL mutation or another clonal marker; and criteria for another MPN, MDS, MDS/MPN or CML not met. Reactive causes of fibrosis are excluded (Arber 2022; Gianelli 2023).
  • Exclusions. The diagnosis is not made in the presence of BCR::ABL1 or a tyrosine kinase fusion (Gianelli 2023).

Accelerated and blast phase of BCR::ABL1-negative MPN

WHO-HAEM5 retains accelerated phase for MPN other than CML (Khoury 2022; Arber and Orazi 2025). Accelerated phase is defined by 10%–19% blasts and blast phase by ≥20% blasts in blood or marrow (Khoury 2022; Gianelli 2023; Shao 2025). CD34 immunohistochemistry identifies increased blasts, and paratrabecular CD34+ progenitor clusters indicate early progression (Gianelli 2023; Shao 2025).

Leukemic transformation is more frequent in PMF than in PV or ET, and leukemia-free survival is shorter in fibrotic than in prefibrotic PMF (Khoury 2022).

Mastocytosis, Hypereosinophilic Syndrome and Myeloid/Lymphoid Neoplasms with Tyrosine Kinase Fusions, Chronic Myelomonocytic Leukemia and Other MDS/MPN

Definition. Mastocytosis is a neoplastic accumulation of abnormal mast cells in one or more organs, usually caused by constitutive activation of KIT (Khoury 2022; Valent 2021; Li 2023). Myeloid/lymphoid neoplasms with eosinophilia and tyrosine kinase gene fusions (MLN-TK) are myeloid or lymphoid neoplasms caused by fusions that constitutively activate a tyrosine kinase domain (Khoury 2022; CAP 2026). Myelodysplastic/myeloproliferative neoplasms (MDS/MPN) combine pathologic and molecular features of MDS and MPN, usually with cytopenias and cytoses together (Khoury 2022; Arber 2022). Chronic myelomonocytic leukemia (CMML) is the most common MDS/MPN (Khoury 2022; Selimoglu-Buet 2015).

Related page: Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition (WHO-HAEM5 and ICC differences, flow cytometry, CMUS and germline predisposition) Related page: Myeloproliferative Neoplasms: Chronic Myeloid Leukemia with Molecular Monitoring, Polycythemia Vera, Essential Thrombocythemia, Primary Myelofibrosis and Other Myeloproliferative Neoplasms (chronic eosinophilic leukemia and chronic neutrophilic leukemia criteria) Related page: Myelodysplastic Neoplasms: WHO 5th Edition and ICC Types, Dysplasia Morphology, TP53-Mutated Myeloid Neoplasms and Prognostic Scoring (MDS with SF3B1 mutation and dysplasia morphology) Related page: Bone Marrow Failure, Pure Red Cell Aplasia and the Laboratory Approach to Cytopenias and Leukocyte Disorders (definitions of eosinophilia and hypereosinophilia, the reactive eosinophilia workup, lymphocyte-variant hypereosinophilia and monocytosis) Related page: Allergy and Hypersensitivity Testing: Total and Specific IgE, Component Diagnostics, Tryptase and Basophil Activation (tryptase methods, the 20% + 2 formula and hereditary α-tryptasemia) Related page: Small Intestinal Mucosal Disease: Celiac Disease and Its Mimics, Infections, Drug Injury and Immune Enteropathies (systemic mastocytosis in duodenal biopsies) Related page: Neuroendocrine, Mesenchymal and Lymphoid Tumors of the Esophagus, Stomach and Small Intestine (mast cell infiltrates in upper gastrointestinal biopsies)

Systemic mastocytosis: diagnostic criteria in WHO-HAEM5 and ICC 2022

  • KIT mutations. A somatic KIT codon 816 mutation is present in >90% of SM. Other activating KIT mutations in the extracellular (codon 419 deletion, A502_Y503dup), transmembrane (F522C) and juxtamembrane (V560G) domains occur in <1% of advanced SM and are enriched in indolent SM (Khoury 2022; Li 2023).
  • Co-mutations. Most patients with advanced SM and KIT D816V have additional TET2, SRSF2, ASXL1, RUNX1 or JAK2 mutations, and an associated, usually myeloid, neoplasm may be present (Khoury 2022; Li 2023).

Major and minor criteria for systemic mastocytosis

Criterion WHO-HAEM5 (2022) ICC 2022 WHO 2016/2017 (previous rule) Source
Major Multifocal dense infiltrates of mast cells (≥15 mast cells in aggregates) in marrow or another extracutaneous organ Multifocal dense infiltrates of tryptase- and/or CD117-positive mast cells (≥15 in aggregates) Same aggregate criterion Li 2023; Valent 2021; Arber 2022
Morphology ≥25% of mast cells atypical (type I or II) on marrow smears or spindle-shaped in infiltrates >25% of mast cells spindle-shaped or atypical immature in sections Same morphologic criterion Li 2023; Valent 2021; Arber 2022
KIT Activating KIT point mutation at codon 816 or in other critical regions KIT D816V or another activating KIT mutation KIT codon 816 mutation; other activating KIT mutations added in 2022 Li 2023; Khoury 2022; Arber 2022
Aberrant phenotype CD2 and/or CD25 and/or CD30 CD25, CD2 and/or CD30 CD2 and/or CD25; CD30 added in 2022 Li 2023; Khoury 2022; Arber 2022
Tryptase Baseline >20 ng/mL; not counted with an associated myeloid neoplasm; adjusted for hereditary α-tryptasemia Persistently >20 ng/mL; not counted in SM with an associated myeloid neoplasm >20 ng/mL Li 2023; Valent 2021; CAP 2026; Arber 2022
Diagnostic rule Major plus ≥1 minor, or ≥3 minor Major alone, or ≥3 minor Major plus ≥1 minor, or ≥3 minor Li 2023; Valent 2021; Arber 2022
  • Spindle-shaped cells that do not count. Spindled mast cells lining vessels, fat cells, nerves or the endosteum do not count. Atypical mast cells within or adjacent to marrow particles on smears do not count (Valent 2021).
  • Activating KIT variants. Variants in codons 417, 501–509, 522, 557–560, 642, 654, 799, 816, 820 and 822 are listed as qualifying when transforming activity is proven (Valent 2021).
  • Tryptase adjustment for hereditary α-tryptasemia. One proposed method divides the basal tryptase by 1 plus the number of extra α-tryptase gene copies; 30 ng/mL with 2 extra copies becomes 10 ng/mL (Valent 2021). Tryptase methods and hereditary α-tryptasemia are taught in Allergy and Hypersensitivity Testing: Total and Specific IgE, Component Diagnostics, Tryptase and Basophil Activation.
  • CD30. CD30 is expressed by neoplastic mast cells in most patients with indolent and advanced SM and is not found in mast cells of other myeloid neoplasms. CD30 does not grade SM (Valent 2021). The ICC added CD30 to the aberrant phenotype criterion (Arber 2022).
  • CD25 and CD2. CD25 is expressed more often than CD2 and is the most specific neoplastic mast cell marker. CD25+ mast cells also occur in chronic inflammation, after chemotherapy and in MLN-TK with PDGFRA rearrangement. Mast cell flow cytometry is a rare-event analysis, ideally with ≥1 million events (Li 2023).
  • Discordant KIT D816V and CD25 results. In 15 patients with discordant SM test results, Sanger sequencing and NGS did not detect KIT D816V that allele-specific or quantitative PCR detected, and PCR assays disagreed when the VAF was near the limit of detection. Marrow core decalcification caused false-negative CD25 immunohistochemistry that was positive on the clot section (Boggs 2023). The ICC recommends high-sensitivity PCR for KIT D816V and, if it is negative, testing for other KIT variants in suspected SM (Boggs 2023; Arber 2022).

Well-differentiated systemic mastocytosis

Well-differentiated SM (WDSM) is a morphologic pattern in any SM subtype, with round, well-granulated mast cells (Khoury 2022; Li 2023). Most cases lack KIT codon 816 mutations and CD2 and CD25, and express CD30. KIT K509I or F522C may be present, and several of these variants are imatinib-sensitive (Valent 2021; Li 2023). The ICC notes CD25-negative, CD2-negative, CD30-positive mast cells and a higher response rate to tyrosine kinase inhibitors in this pattern (Arber 2022).

Mastocytosis types, B-findings, C-findings and subtypes

Types of mastocytosis

Type WHO-HAEM5 (2022) ICC 2022 Source
Cutaneous mastocytosis Urticaria pigmentosa/maculopapular (monomorphic, polymorphic); diffuse; cutaneous mastocytoma (isolated, multilocalized) Maculopapular (monomorphic, polymorphic); diffuse; mastocytoma of skin (≥4 lesions favor maculopapular) Khoury 2022; Li 2023
Systemic mastocytosis, non-advanced Bone marrow mastocytosis, indolent SM, smoldering SM Indolent SM (bone marrow mastocytosis as a variant), smoldering SM Khoury 2022; Li 2023; Arber 2022
Systemic mastocytosis, advanced Aggressive SM, SM with an associated hematologic neoplasm, mast cell leukemia Aggressive SM, SM with an associated myeloid neoplasm, mast cell leukemia Khoury 2022; Li 2023; Arber 2022
Mast cell sarcoma Mast cell sarcoma Mast cell sarcoma Khoury 2022; Li 2023

B-findings and C-findings in systemic mastocytosis

Finding WHO-HAEM5 (2022) ICC 2022 Source
B: mast cell burden Mast cell infiltration ≥30% on histology, and/or tryptase ≥200 ng/mL, and/or KIT D816V VAF ≥10% in marrow or blood leukocytes Mast cell aggregates >30% of cellularity and tryptase >200 ng/mL; KIT VAF not a B-finding Li 2023; Valent 2021
B: myeloproliferation or dysplasia Hypercellular marrow with prominent myelopoiesis, left shift and eosinophilia, or discrete dysplasia (<10% of cells), not meeting criteria for MPN, MDS or MDS/MPN Cytopenia not meeting C-finding criteria, or cytosis, after reactive causes and other myeloid neoplasms are excluded Li 2023; Valent 2021
B: organomegaly Hepatomegaly without ascites or organ damage, splenomegaly without hypersplenism or weight loss, or lymphadenopathy >2 cm Hepatomegaly without liver dysfunction, splenomegaly without hypersplenism, or lymphadenopathy >1 cm Li 2023; Valent 2021
C: cytopenia ANC <1 × 10⁹/L, hemoglobin <10 g/dL or platelets <100 × 10⁹/L Same thresholds, without an obvious non-mast cell hematopoietic malignancy Li 2023; Valent 2021
C: liver Ascites and elevated liver enzymes, with or without hepatomegaly, cirrhosis or portal hypertension Palpable hepatomegaly with impaired liver function, ascites and/or portal hypertension Li 2023; Valent 2021
C: spleen Palpable splenomegaly with hypersplenism, with or without weight loss or hypoalbuminemia Palpable splenomegaly with hypersplenism Li 2023; Valent 2021
C: gastrointestinal Malabsorption with hypoalbuminemia, with or without weight loss Malabsorption with weight loss from gastrointestinal mast cell infiltrates Li 2023; Valent 2021
C: bone Large osteolysis (≥2 cm) with pathologic fracture, with or without bone pain Large osteolytic lesions and/or pathologic fractures Li 2023; Valent 2021
  • Previous rule. ICC C-findings are unchanged from WHO 2016 (Li 2023). In the 2021 consensus adopted by WHO-HAEM5, weight loss is a supporting feature of malabsorption and splenomegaly, and osteoporosis and small osteolyses are not C-findings (Valent 2021).
  • Cause of organ damage. Organ damage caused by an associated neoplasm, infection or therapy does not count as a C-finding, and a biopsy documents the mast cell infiltrate whenever possible (Valent 2021).

Subtypes of systemic mastocytosis

Subtype Criteria Source
Bone marrow mastocytosis SM criteria; no skin lesions; no B-findings; basal tryptase <125 ng/mL; no dense infiltrates in an extramedullary organ; no C-findings, mast cell leukemia or associated neoplasm (WHO-HAEM5 subtype; ICC variant of indolent SM) Valent 2021; Li 2023
Indolent SM SM criteria; typical skin lesions; ≤1 B-finding; no C-finding. Without skin lesions, indolent SM is diagnosed when there is 1 B-finding, tryptase ≥125 ng/mL or dense extramedullary infiltrates Valent 2021
Smoldering SM SM criteria; ≥2 B-findings; no C-finding; no mast cell leukemia or associated neoplasm Valent 2021
Aggressive SM SM criteria; ≥1 C-finding; mast cells <20% on smears; 5%–19% mast cells on marrow smears defines ASM in transformation Valent 2021
SM with an associated neoplasm SM criteria plus criteria for another hematologic neoplasm (WHO-HAEM5, myeloid or lymphoid) or a myeloid neoplasm (ICC); CMML and other MDS/MPN are most frequent Li 2023; CAP 2026
Mast cell leukemia SM criteria and ≥20% mast cells on marrow aspirate smears; WHO-HAEM5 separates leukemic (≥10% in blood) from aleukemic (<10%) and acute (C-findings) from chronic forms; ICC requires immature mast cell morphology and allows diagnosis on a biopsy with diffuse immature infiltration when the aspirate is a dry tap, without a leukemic or aleukemic subcategory CAP 2026; Li 2023; Aperna 2025; Valent 2021; Arber 2022
Mast cell sarcoma Localized destructive tumor of atypical mast cells; SM criteria not met; KIT D816V uncommon in de novo cases Valent 2021; CAP 2026
  • Previous rule. Bone marrow mastocytosis was not a separate SM subtype before WHO-HAEM5 (Khoury 2022). SM with an associated hematologic neoplasm was the WHO 2016 term, renamed SM with an associated myeloid neoplasm in the ICC (Leguit 2023; Arber 2022).
  • Marrow patterns. Indolent SM usually involves <5%–10% of the section area with compact mixed infiltrates containing eosinophils, lymphocytes and histiocytes. Smoldering and aggressive SM show diffuse-compact infiltrates of spindled mast cells with fibrosis and osteosclerosis. Mast cell leukemia usually involves >50% of the section with round hypogranulated mast cells (CAP 2026).
  • Associated neoplasm workup. Immunostains for mast cells (tryptase, CD117, CD25, CD30) and for myeloid neoplasms (CD14, CD34, CD42b, CD71) are applied in suspected SM with an associated neoplasm (CAP 2026). All SM criteria are reapplied after cytoreduction in KIT D816V-mutated myeloid neoplasms (Valent 2021).
  • Survival. In 910 patients, median survival was 24.3 years in indolent or smoldering SM and 5.8–6 years in aggressive SM. ICC-defined mast cell leukemia had a median survival of 0.08 years, compared with 1.8 years for WHO-defined mast cell leukemia. SM with an associated lymphoid neoplasm had a median survival of 8.1 years and SM with an associated myeloid neoplasm 2 years. Tryptase ≥125 ng/mL was an independent adverse factor (Aperna 2025).
  • Gastrointestinal biopsies. Mast cell infiltrates in gastrointestinal biopsies are taught in Small Intestinal Mucosal Disease: Celiac Disease and Its Mimics, Infections, Drug Injury and Immune Enteropathies and Neuroendocrine, Mesenchymal and Lymphoid Tumors of the Esophagus, Stomach and Small Intestine.

Hypereosinophilia: the stepwise clonal workup and idiopathic hypereosinophilic syndrome

Definitions of eosinophilia, hypereosinophilia and tissue hypereosinophilia, and the reactive workup, are taught in Bone Marrow Failure, Pure Red Cell Aplasia and the Laboratory Approach to Cytopenias and Leukocyte Disorders. Hypereosinophilic syndrome (HES) is hypereosinophilia with eosinophil-mediated organ damage (Eletrebi 2026; Morales-Camacho 2024).

Stepwise workup of persistent unexplained hypereosinophilia

Step Investigation Finding sought Source
1 Exclusion of reactive causes; assessment of organ damage Secondary eosinophilia; HES Eletrebi 2026; Lübke 2024
2 Marrow aspirate and trephine biopsy Myeloproliferation, dysplastic megakaryocytes, fibrosis, increased blasts, abnormal mast cell infiltrates Eletrebi 2026; CAP 2026; Morales-Camacho 2024
3 FISH or RT-PCR for FIP1L1::PDGFRA Cryptic 4q12 deletion not visible on karyotype Khoury 2022; CAP 2026; Lübke 2024
4 Karyotype and FISH for PDGFRB, FGFR1 and PCM1::JAK2 Translocations involving 5q32 (PDGFRB), 8p11 (FGFR1), 9p24 (JAK2; PCM1::JAK2 from t(8;9)(p22;p24.1)) or 13q12.2 (FLT3) that point to a tyrosine kinase gene and guide break-apart probes Khoury 2022; CAP 2026; Eletrebi 2026; Morales-Camacho 2024; Lübke 2024
5 BCR::ABL1 testing and KIT D816V testing CML; systemic mastocytosis with eosinophilia CAP 2026; Li 2023; Lübke 2024
6 Myeloid NGS panel Somatic mutations supporting CEL; interpreted against CHIP Eletrebi 2026; Lübke 2024
7 Evaluation for lymphoid hypereosinophilia (methods in Bone Marrow Failure, Pure Red Cell Aplasia and the Laboratory Approach to Cytopenias and Leukocyte Disorders) Lymphoid HES, the final diagnosis in 3 of 38 patients in one series Eletrebi 2026; Lübke 2024
  • Diagnostic yield of the hypereosinophilia workup. Among 38 patients with hypereosinophilia of persistent unknown cause who had marrow evaluation, integrated workup classified 28.9% as reactive, 36.8% as idiopathic, 28.9% as clonal and 5.3% as reactive versus idiopathic. Cytogenetics and FISH were largely noncontributory but detected FIP1L1::PDGFRA in 2 patients. Myeloid NGS found mutations in 7 of 15 tested patients, including TP53, BCOR, SF3B1 and TET2 (Eletrebi 2026).
  • CHIP-type variants. Age-related CHIP variants may be present without a neoplasm and may not explain the eosinophilia (Eletrebi 2026). CHIP-associated mutations in eosinophilia are usually single DNMT3A, TET2 or ASXL1 mutations at low VAF (Lübke 2024).
  • Idiopathic categories. Idiopathic HES is hypereosinophilia with organ damage and no reactive or clonal cause. Hypereosinophilia of undetermined significance is idiopathic hypereosinophilia without organ damage (Eletrebi 2026; Morales-Camacho 2024; Lübke 2024).
  • ICC. Marrow morphology is a diagnostic criterion of both CEL, NOS and idiopathic hypereosinophilia/HES in the ICC, which separates a myeloid neoplasm with eosinophilia from hypereosinophilia of unknown cause (Tzankov 2023; Arber 2022).
  • Lymphoid neoplasms with eosinophilia. B-lymphoblastic leukemia/lymphoma with t(5;14)(q31;q32) IGH::IL3 presents with marked eosinophilia, and T-lymphoblastic lymphoma with eosinophilia has been described (Eletrebi 2026; Morales-Camacho 2024).
  • Chronic eosinophilic leukemia. WHO-HAEM5 and ICC criteria for CEL are taught in Myeloproliferative Neoplasms: Chronic Myeloid Leukemia with Molecular Monitoring, Polycythemia Vera, Essential Thrombocythemia, Primary Myelofibrosis and Other Myeloproliferative Neoplasms.

ICC criteria for idiopathic hypereosinophilic syndrome

Criterion ICC 2022 requirement Source
1 Persistent blood hypereosinophilia: eosinophils ≥1.5 × 10⁹/L and ≥10% of leukocytes Arber 2022
2 Organ damage or dysfunction attributable to a tissue eosinophilic infiltrate Arber 2022
3 No reactive, well-defined autoimmune or neoplastic condition underlying the hypereosinophilia Arber 2022
4 Lymphocyte-variant HES excluded Arber 2022
5 Bone marrow morphologically normal except for increased eosinophils Arber 2022; Lübke 2024
6 No clonal molecular genetic abnormality, with the caveat of CHIP Arber 2022
  • Number of criteria. All 6 criteria are required for idiopathic HES in the ICC (Arber 2022).
  • Serum tryptase. Serum tryptase is usually normal in idiopathic HES and is often increased in MLN-TK with PDGFRA or PDGFRB fusions (Lübke 2024).

Myeloid/lymphoid neoplasms with eosinophilia and tyrosine kinase gene fusions

MLN-TK include MPN, MDS, MDS/MPN, AML, mixed-phenotype acute leukemia and B- or T-lymphoblastic leukemia/lymphoma presentations. Extramedullary disease is common, and eosinophilia may be absent. The diagnosis supersedes other myeloid and lymphoid types and SM, and MLN-TK is excluded before CEL is diagnosed. The defining fusion may be acquired during an MDS, MDS/MPN or MPN (Khoury 2022; Arber 2022). Blast phase or AML is diagnosed only when blasts are ≥20% in blood or marrow (CAP 2026).

Type Genetics Features Response to TKI Source
PDGFRA rearrangement Most often cryptic 4q12 deletion producing FIP1L1::PDGFRA; other partners occur MPN with prominent eosinophilia and splenomegaly; marked elevation of serum vitamin B12; increased tryptase and marrow mast cells Imatinib-sensitive Khoury 2022; CAP 2026; Arber 2022; Lübke 2024
PDGFRB rearrangement Most often t(5;12)(q32;p13.2) ETV6::PDGFRB; >30 partners Prominent eosinophilia with variable neutrophilia or monocytosis Imatinib-sensitive Khoury 2022; CAP 2026; Arber 2022; Lübke 2024
FGFR1 rearrangement 8p11 aberrations, for example t(8;13)(p11.2;q12.1) ZMYM2::FGFR1 Chronic myeloid neoplasm or blast phase of B, T, myeloid or mixed lineage, usually with eosinophilia Variable; long-term survival usually requires allogeneic transplantation Khoury 2022; CAP 2026; Arber 2022
JAK2 rearrangement t(8;9)(p22;p24.1) PCM1::JAK2; variants t(9;12)(p24.1;p13.2) ETV6::JAK2 and t(9;22)(p24.1;q11.2) BCR::JAK2 Myeloid or lymphoid neoplasm with eosinophilia; prominent erythroblastic islands in marrow and extramedullary sites Variable Khoury 2022; CAP 2026; Morales-Camacho 2024; Tzankov 2023; Arber 2022
FLT3 rearrangement Rearrangements of 13q12.2, most often t(12;13)(p13.2;q12.2) ETV6::FLT3 Myeloid sarcoma with MPN features in marrow, or T-ALL with eosinophilia Variable Khoury 2022; Arber 2022
ETV6::ABL1 t(9;12)(q34.1;p13.2) ETV6::ABL1 Chronic-phase hematopoietic neoplasm, CML-like with frequent eosinophilia; separated from B-ALL with ETV6::ABL1 Variable Khoury 2022; CAP 2026; Arber 2022
Other tyrosine kinase fusions ETV6::FGFR2, ETV6::LYN, ETV6::NTRK3, RANBP2::ALK, BCR::RET, FGFR1OP::RET Listed as subtypes until more data accrue Not defined Khoury 2022; CAP 2026
  • ICC. The ICC renames the category myeloid/lymphoid neoplasms with eosinophilia and tyrosine kinase gene fusions, adds ETV6::ABL1 and FLT3 fusions, and makes PCM1::JAK2 with ETV6::JAK2 and BCR::JAK2 a formal entity (Tzankov 2023; Arber 2022; Morales-Camacho 2024).

  • Previous rule. The category was introduced in the WHO 4th edition for PDGFRA, PDGFRB and FGFR1 abnormalities (Arber and Orazi 2025; Morales-Camacho 2024), and PCM1::JAK2 was provisional in WHO 2016 (Tzankov 2023; Morales-Camacho 2024).

  • Mast cells in MLN-TK. Spindled, clustered and CD25+ mast cells occur in these neoplasms, which are excluded before mast cell neoplasms are diagnosed (CAP 2026). Mast cells in FIP1L1::PDGFRA neoplasms are scattered or in loose clusters and usually lack CD2 (Li 2023). Under ICC, a neoplasm with both a tyrosine kinase fusion and a KIT mutation is classified as SM with an associated myeloid neoplasm (Li 2023; Arber 2022).

  • Ph-like B-ALL. BCR::ABL1-like B-ALL with PDGFRB or JAK2 fusion and no associated myeloid neoplasm is excluded from these types (CAP 2026; Arber 2022).

  • Targeted therapy by fusion. Imatinib produces durable complete hematologic, cytogenetic and molecular remissions in >90% of patients with PDGFRA or PDGFRB fusions (Lübke 2024). The FGFR1–3 inhibitor pemigatinib is approved for MLN-TK with FGFR1 fusion; responses are more frequent and durable in chronic phase than in blast phase (Arber 2022; Lübke 2024). Ruxolitinib gives limited and often transient responses in JAK2-rearranged neoplasms and is used mainly as interim therapy before allogeneic transplantation (Arber 2022; Lübke 2024). ETV6::ABL1 neoplasms respond variably to second-generation TKIs (Arber 2022).

Chronic myelomonocytic leukemia

Diagnostic criteria for CMML (WHO-HAEM5, ICC 2022 and WHO 2017)

Element WHO-HAEM5 (2022) ICC 2022 WHO 2017 (previous rule) Source
Monocytosis Persistent ≥0.5 × 10⁹/L and ≥10% of leukocytes Sustained ≥0.5 × 10⁹/L and ≥10% of leukocytes ≥1.0 × 10⁹/L and ≥10% Khoury 2022; Loghavi 2026; Arber 2022
Blasts and promonocytes <20% in blood and marrow <20% in blood and marrow <20% Khoury 2022; Loghavi 2026; Arber 2022
Supporting criteria Dysplasia in ≥1 lineage (≥10% of cells); acquired clonal cytogenetic or molecular abnormality; abnormal monocyte subset partitioning (classical monocytes >94%) Cytopenia at MDS thresholds (borderline or absent in a few early cases) and marrow consistent with CMML (hypercellular myeloid proliferation, often with increased monocytes) in all cases; clonality, shown by an abnormal karyotype or a myeloid neoplasm-associated mutation at VAF ≥10% Not structured as prerequisite and supporting criteria Khoury 2022; Loghavi 2026; Arber 2022
Requirement at ≥1 × 10⁹/L ≥1 supporting criterion Clonality; without clonality, monocytes ≥1 × 10⁹/L and >10% plus increased blasts (≥5% marrow or ≥2% blood), dysplasia or an abnormal immunophenotype consistent with CMML Not structured as supporting criteria Khoury 2022; Arber 2022
Requirement at 0.5 to <1 × 10⁹/L Dysplasia and a clonal abnormality Clonality with VAF ≥10%, cytopenia and CMML-type marrow; dysplasia not required Not CMML Khoury 2022; Loghavi 2026; Arber 2022
Blast subgroups CMML-1: <5% blood and <10% marrow; CMML-2: 5%–19% blood and 10%–19% marrow CMML-1: <5% blood and <10% marrow; CMML-2: 5%–19% blood, 10%–19% marrow or Auer rods; CMML-0 eliminated CMML-0: <2% blood and <5% marrow; CMML-1; CMML-2 Khoury 2022; Loghavi 2026; Arber 2022; Patnaik 2024
  • Exclusions. WHO-HAEM5 excludes CML, other MPN and MLN-TK, and the MLN-TK criteria are specifically excluded when eosinophilia is present (Khoury 2022). Absence of BCR::ABL1 and other tyrosine kinase fusions is required in both systems (Loghavi 2026; Arber 2022).

  • Subtypes. WHO-HAEM5 separates myelodysplastic CMML (WBC <13 × 10⁹/L) from myeloproliferative CMML (WBC ≥13 × 10⁹/L) (Khoury 2022). The 13 × 10⁹/L split was described in WHO 2008 (Loghavi 2026) and formally adopted as MD-CMML and MP-CMML in WHO-HAEM5 (Khoury 2022).

  • ICC subtypes. The ICC text also recognizes myeloproliferative CMML, defined there as WBC >13 × 10⁹/L and associated with RAS pathway (NRAS, KRAS, CBL), JAK2 V617F and SETBP1 mutations (Arber 2022).

  • Blast equivalents. Blast equivalents are defined in Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition.

  • MPN with monocytosis. A documented history of MPN excludes CMML. MPN marrow features or a high JAK2, CALR or MPL allele burden support MPN with monocytosis (Khoury 2022).

  • NPM1 mutation. A CMML-like presentation with NPM1 mutation meets WHO-HAEM5 criteria for AML with NPM1 mutation (CAP 2026).

  • NPM1 in the ICC. NPM1 mutation occurs in 3%–5% of CMML and is associated with rapid progression to AML. An NPM1 mutation acquired in known CMML does not define de novo AML in the ICC (Arber 2022).

  • Oligomonocytic CMML. Cases with 0.5 to <1 × 10⁹/L monocytes progressed to overt CMML in 38%–47% of cases (Wagner-Ballon 2026). Cases with biallelic TET2 inactivation or TET2 with SRSF2 resemble overt CMML, and cases with SF3B1, biallelic TP53 or del(5q) resemble MDS (Loghavi 2026).

Monocyte subset partitioning by flow cytometry

  • Subsets. Classical monocytes (CD14++CD16−) are about 85% of blood monocytes, intermediate monocytes (CD14++CD16+) about 5% and nonclassical monocytes (CD14-low/negative CD16+) about 10% (Wagner-Ballon 2026; Selimoglu-Buet 2015; Porwit 2023).
  • CMML signature. Classical monocytes >94% of blood monocytes had a sensitivity of 90.6% and specificity of 95.1% in the original French cohort (Selimoglu-Buet 2015). Real-world sensitivity was 73%–75% in some series (Loghavi 2026).
  • Inflammation. Active autoimmune disease or systemic inflammation is an exclusion for the >94% criterion (Khoury 2022; CAP 2026). A slan+ nonclassical monocyte fraction ≤1.7% supports CMML in inflammatory disease (Loghavi 2026; Porwit 2023).
  • Acquisition. At least 10,000 classical monocyte events are acquired, with exclusion gating of NK cells (Porwit 2023).

Genetics and prognosis of CMML

  • Mutations. Over 90% of patients carry ≥1 somatic mutation (Loghavi 2026; Arber 2022), most often TET2 (about 60%), SRSF2 (about 50%) and ASXL1 (about 40%) (Loghavi 2026; Patnaik 2024). Biallelic TET2 inactivation (2 mutations, 1 mutation with 4q24 loss, or VAF ≥55%) correlates most strongly with the CMML phenotype (Loghavi 2026). RUNX1 (10%–15%) is associated with AML progression (Loghavi 2026; Patnaik 2024), and TP53 (3%–5%) with the shortest survival (Loghavi 2026).
  • Myeloproliferative CMML. RAS pathway mutations, mainly NRAS G12D, occur in 46% of myeloproliferative CMML and 25% of myelodysplastic CMML (Loghavi 2026). Myeloproliferative CMML has adverse outcomes (Khoury 2022; Arber 2022).
  • Cytogenetics. Abnormalities occur in 20%–30%: trisomy 8, monosomy 7/del(7q), del(20q), complex karyotype and −Y (Loghavi 2026; Patnaik 2024).
  • CPSS-Mol. A genetic score built from cytogenetics and RUNX1, NRAS, SETBP1 and ASXL1 mutations is combined with red cell transfusion dependence, leukocyte count and marrow blasts. Four risk groups had median survival from >144 to 18 months and 4-year leukemic evolution from 0% to 48% (Elena 2016).
  • Associated findings. Clonal nodules of mature plasmacytoid dendritic cells occur in about 20% of CMML and confer a higher incidence of AML transformation (Loghavi 2026). SM with CMML is a common SM with an associated neoplasm (Li 2023). Autoimmune and inflammatory manifestations occur in 15%–20% (Loghavi 2026), and in 20%–30% in another review (Patnaik 2024).

Other MDS/MPN: MDS/MPN with neutrophilia, MDS/MPN with SF3B1 mutation and thrombocytosis, and MDS/MPN, NOS

Names of MDS/MPN types by classification

WHO-HAEM5 (2022) ICC 2022 WHO 2017 (previous rule) Source
Chronic myelomonocytic leukemia Chronic myelomonocytic leukemia; clonal monocytosis of undetermined significance defined separately Chronic myelomonocytic leukemia Arber and Orazi 2025; Khoury 2022; Arber 2022
MDS/MPN with neutrophilia Atypical CML Atypical CML Arber and Orazi 2025; Khoury 2022; Arber 2022
MDS/MPN with SF3B1 mutation and thrombocytosis MDS/MPN with thrombocytosis and SF3B1 mutation; MDS/MPN with ring sideroblasts and thrombocytosis, NOS MDS/MPN with ring sideroblasts and thrombocytosis Arber and Orazi 2025; Khoury 2022; Arber 2022
MDS/MPN, not otherwise specified MDS/MPN, NOS, with MDS/MPN with isolated i(17q) as a provisional subentity MDS/MPN, unclassifiable Arber and Orazi 2025; Khoury 2022; Arber 2022
Juvenile myelomonocytic leukemia listed under MPN JMML under pediatric and/or germline mutation-associated disorders Juvenile myelomonocytic leukemia under MDS/MPN Arber and Orazi 2025; Khoury 2022; Arber 2022

Diagnostic criteria for MDS/MPN with neutrophilia (WHO-HAEM5) and atypical CML (ICC 2022)

Element WHO-HAEM5 (2022): MDS/MPN with neutrophilia ICC 2022: atypical CML Source
Leukocytes ≥13 × 10⁹/L with neutrophilia; promyelocytes, myelocytes and metamyelocytes ≥10% of leukocytes ≥13 × 10⁹/L from neutrophils and precursors; promyelocytes, myelocytes and metamyelocytes ≥10% of leukocytes CAP 2026; Arber 2022; Iurlo 2025
Dysgranulopoiesis Neutrophilic dysplasia, which may include abnormal chromatin clumping Abnormal hyposegmented and/or hypersegmented neutrophils, with or without abnormal chromatin clumping CAP 2026; Arber 2022; Iurlo 2025
Cytopenia Not a stated criterion ≥1 cytopenia at MDS thresholds Arber 2022; Iurlo 2025
Blasts <20% in blood and marrow <20% in blood and marrow CAP 2026; Arber 2022
Monocytes No or minimal absolute monocytosis; monocytes <10% of leukocytes No or minimal absolute monocytosis; monocytes <10% of leukocytes Iurlo 2025; Arber 2022
Basophils No or minimal absolute basophilia; basophils usually <2% of leukocytes Not a stated criterion Iurlo 2025; Arber 2022
Eosinophils Not a stated criterion Eosinophils <10% of leukocytes; hypereosinophilia excludes the diagnosis Arber 2022; Iurlo 2025
Marrow Hypercellular with granulocytic predominance and granulocytic dysplasia, with or without megakaryocytic and erythroid dysplasia Hypercellular with granulocytic proliferation and dysplasia, with or without erythroid and megakaryocytic dysplasia CAP 2026; Arber 2022
Exclusions BCR::ABL1 and other MPN, MLN-TK, CMML and MDS/MPN with SF3B1 mutation and thrombocytosis BCR::ABL1 and the tyrosine kinase fusions of MLN-TK CAP 2026; Arber 2022
Supporting genetics SETBP1 and/or ETNK1 mutation; absence of JAK2, CALR, MPL and CSF3R mutations SETBP1 mutation, often with ASXL1; absence of JAK2, CALR and MPL mutations CAP 2026; Arber 2022; Iurlo 2025
  • Previous rule. WHO-HAEM5 renamed atypical CML to MDS/MPN with neutrophilia; the criteria are otherwise unchanged from WHO 2017 (Khoury 2022; CAP 2026). The ICC keeps the name atypical CML and drops "BCR::ABL1-negative" from it (Arber 2022).

  • Mutations. ASXL1 is mutated in over 40% of patients, and SETBP1, SRSF2, TET2 and EZH2 are also frequent (Iurlo 2025). A CSF3R mutation is uncommon and prompts morphologic review for chronic neutrophilic leukemia (Arber 2022).

  • Distinction from MPN and CMML. A history of MPN or an MPN driver mutation (JAK2, CALR, MPL) tends to exclude atypical CML (Arber 2022). Monocytes <10% of leukocytes separate it from CMML (Iurlo 2025).

  • Prognosis. Median overall survival is 10 to 29 months, and up to 30%–40% of patients transform to AML within 12–18 months of diagnosis (Iurlo 2025).

MDS/MPN with SF3B1 mutation and thrombocytosis

  • WHO-HAEM5 criteria. Anemia with dyserythropoiesis and ≥15% ring sideroblasts; persistent platelets ≥450 × 10⁹/L; SF3B1 mutation with concurrent JAK2 V617F or, without these, biologically similar spliceosome or signaling mutations (for example MPL or CBL); criteria for MDS, MPN, CMML, AML with MECOM rearrangement and MLN-TK not met (CAP 2026). The term MDS/MPN with ring sideroblasts and thrombocytosis is acceptable for SF3B1 wild-type cases with ≥15% ring sideroblasts (Khoury 2022).

  • ICC. Ring sideroblasts are common but not required when SF3B1 is mutated. A patient with SF3B1-mutated thrombocytosis and no ring sideroblasts is classified as MDS/MPN, NOS in WHO-HAEM5 and as MDS/MPN with thrombocytosis and SF3B1 mutation in ICC (Hazarika and Bain 2026; Arber 2022).

  • ICC requirements. The ICC requires platelets ≥450 × 10⁹/L, anemia at MDS thresholds, blasts <1% in blood and <5% in marrow, and an SF3B1 mutation at VAF ≥10%. It excludes BCR::ABL1, MLN-TK fusions, inv(3)/t(3;3) and, in cases meeting MDS with del(5q) criteria, del(5q), as well as a history of MPN, MDS or other MDS/MPN (Arber 2022).

  • JAK2 co-mutation. The SF3B1 mutation usually accompanies JAK2 V617F; CALR or MPL W515 co-mutation occurs in <10%, and none is required in the ICC (Arber 2022).

MDS/MPN, not otherwise specified

Blood shows cytopenias with proliferative features, and the marrow shows dysplasia with proliferative features. Molecular alterations of both proliferative and dysplastic myeloid neoplasms are present. Myeloid neoplasms post cytotoxic therapy, disease-defining fusions and the specific MDS/MPN types are excluded (CAP 2026).

  • ICC criteria. The ICC requires mixed myeloproliferative and myelodysplastic features, cytopenia at MDS thresholds, blasts <20%, platelets ≥450 × 10⁹/L and/or WBC ≥13 × 10⁹/L, and clonality; without demonstrable clonality, the findings must persist and other causes must be excluded. BCR::ABL1 and MLN-TK fusions are excluded, as are inv(3)/t(3;3) and del(5q) in cases that otherwise meet MDS criteria (Arber 2022).
  • MPN and CEL mimics. A history of MPN or an MPN driver mutation, particularly at high VAF, tends to exclude MDS/MPN, NOS, and hypereosinophilia favors CEL (Arber 2022).
  • MDS/MPN with isolated i(17q). This provisional ICC subentity of MDS/MPN, NOS requires leukocytosis ≥13 × 10⁹/L, cytopenia, blasts <20%, dysgranulopoiesis with nonsegmented or pseudo-Pelger–Huët neutrophils and i(17q) alone or with one other abnormality other than −7/del(7q). It is aggressive (Arber 2022).

Juvenile myelomonocytic leukemia

JMML is a hematopoietic stem cell-derived neoplasm of early childhood with RAS pathway activation in at least 90% of cases (Khoury 2022; Monika 2026).

Diagnostic criteria for JMML (WHO-HAEM5 and ICC 2022)

Criterion WHO-HAEM5 (2022) ICC 2022 Source
Blood monocytes ≥1 × 10⁹/L ≥1 × 10⁹/L CAP 2026; Monika 2026
Blasts and promonocytes <20% in blood and marrow <20% in blood and marrow CAP 2026; Monika 2026
Organ infiltration Clinical evidence, most commonly splenomegaly Splenomegaly CAP 2026; Monika 2026
Exclusions No Ph chromosome or BCR::ABL1; no KMT2A rearrangement BCR::ABL1 and KMT2A rearrangement; viral infection, infantile malignant osteopetrosis, Wiskott-Aldrich syndrome and GATA2, SAMD9 or SAMD9L disorders CAP 2026; Monika 2026
Genetic criteria (≥1) Somatic PTPN11, KRAS or NRAS; somatic or germline NF1 with loss of heterozygosity or compound heterozygosity; somatic or germline CBL with loss of heterozygosity; noncanonical RAS pathway variants or fusions upstream of RAS (ALK, PDGFRB, ROS1) PTPN11, KRAS, NRAS or RRAS; germline NF1 with loss of heterozygosity or clinical NF1; germline CBL with loss of heterozygosity; JMML-like neoplasm when canonical RAS mutations are absent CAP 2026; Monika 2026
Without a genetic criterion Increased hemoglobin F for age; myeloid and erythroid precursors in blood; thrombocytopenia with hypercellular marrow; GM-CSF hypersensitivity or STAT5 phosphorylation ≥2 of circulating myeloid or erythroid precursors, increased hemoglobin F for age, thrombocytopenia with hypercellular marrow, GM-CSF hypersensitivity CAP 2026; Monika 2026
  • Previous rule. WHO-HAEM5 added exclusion of KMT2A rearrangement, eliminated monosomy 7 as a cytogenetic criterion and emphasized molecular demonstration of RAS pathway activation (Khoury 2022).
  • Clinical and laboratory features. Hepatosplenomegaly is present in 80%–90%, with leukocytosis, monocytosis ≥1 × 10⁹/L, anemia, thrombocytopenia and elevated hemoglobin F for age. The marrow is hypercellular with myelomonocytic proliferation without an expansion of blasts and promonocytes (Monika 2026).
  • Genes. PTPN11 is mutated in 35%–38%, NRAS in 15%–20%, KRAS in 8%–15%, NF1 in 10%–15% and CBL in 10%–17% (Monika 2026). Somatic PTPN11 and NF1-associated JMML are the most aggressive types, and some JMML with germline CBL variants regress spontaneously (Khoury 2022). KRAS-mutated JMML is associated with monosomy 7 (Monika 2026).
  • Noonan syndrome. Germline PTPN11 variants cause Noonan syndrome-associated transient myeloproliferative disorder in the neonatal period or early infancy, with about 10% progressing to JMML. In 641 patients with germline PTPN11 variants, 5.6% had myeloproliferative features and 3% met JMML criteria. Constitutional tissue testing separates germline from somatic PTPN11 variants (Monika 2026).
  • Germline predisposition. Neurofibromatosis type 1, CBL syndrome and Noonan syndrome are the RASopathies that predispose to JMML (Monika 2026; Khoury 2022). The germline predisposition framework is taught in Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition.

Differential diagnosis of monocytosis and eosinophilia in myeloid neoplasms

Differential diagnosis of persistent monocytosis

Condition Distinguishing findings Source
Reactive or inflammatory monocytosis Active autoimmune or inflammatory disease excludes use of the classical monocyte >94% criterion; causes taught in Bone Marrow Failure, Pure Red Cell Aplasia and the Laboratory Approach to Cytopenias and Leukocyte Disorders Khoury 2022; Loghavi 2026
CMML Monocytes ≥0.5 × 10⁹/L and ≥10%; classical monocytes >94%; TET2, SRSF2, ASXL1 Khoury 2022; Loghavi 2026
Clonal monocytosis of undetermined significance Clonal monocytosis not meeting CMML criteria Loghavi 2026; Arber 2022
MPN with monocytosis History of MPN; MPN marrow features; high JAK2, CALR or MPL allele burden Khoury 2022; Gianelli 2023
CML variant with monocytosis Often carries the minor breakpoint (p190); BCR::ABL1 Monika 2026
JMML Early childhood; monocytes ≥1 × 10⁹/L; RAS pathway mutation; elevated hemoglobin F CAP 2026; Monika 2026
AML with KMT2A rearrangement or NUP98 rearrangement in children Monocytosis and myelomonocytic differentiation; KMT2A or NUP98 fusion Monika 2026
RAS-associated autoimmune leukoproliferative disorder Somatic KRAS or NRAS mutations with monocytosis and an indolent nonmalignant course Monika 2026

Differential diagnosis of eosinophilia in myeloid and lymphoid neoplasms

Condition Distinguishing findings Source
Reactive eosinophilia Secondary cause identified; workup in Bone Marrow Failure, Pure Red Cell Aplasia and the Laboratory Approach to Cytopenias and Leukocyte Disorders Eletrebi 2026
MLN-TK PDGFRA, PDGFRB, FGFR1, JAK2, FLT3 or ETV6::ABL1 fusion Khoury 2022; Arber 2022
Chronic eosinophilic leukemia Clonality and abnormal marrow morphology; MLN-TK excluded Khoury 2022; Arber 2022
SM with eosinophilia SM criteria with KIT D816V; tyrosine kinase fusions excluded Li 2023; Arber 2022
Idiopathic HES and hypereosinophilia of undetermined significance No reactive or clonal cause; organ damage present in HES and absent in hypereosinophilia of undetermined significance Eletrebi 2026; Arber 2022
Lymphoid hypereosinophilia Lymphoid HES on integrated workup Eletrebi 2026; Lübke 2024
B-lymphoblastic leukemia with IGH::IL3 t(5;14)(q31;q32); IL-3 overexpression Eletrebi 2026; Morales-Camacho 2024

Myelodysplastic Neoplasms: WHO 5th Edition and ICC Types, Dysplasia Morphology, TP53-Mutated Myeloid Neoplasms and Prognostic Scoring

Definition. Myelodysplastic neoplasm (WHO 5th edition, WHO-HAEM5) / myelodysplastic syndrome (International Consensus Classification, ICC 2022), abbreviated MDS in both systems, is a clonal hematopoietic neoplasm defined by cytopenia and morphologic dysplasia (Khoury 2022; Arber 2022). The ICC adds a propensity to progress to bone marrow (BM) failure or acute myeloid leukemia (AML) (Arber 2022). No fixed duration of cytopenia is required. The blood count abnormality is expected to be chronic, typically 4 months or longer, and unexplained by a drug, toxin or comorbid condition (Arber 2022). WHO-HAEM5 renamed the disease "neoplasm" to stress its neoplastic nature and to match myeloproliferative neoplasm (MPN) terminology (Khoury 2022; Xiao 2024). The ICC kept "syndrome" (Xiao 2024).

Cytopenia thresholds. The ICC uses the WHO-HAEM5 thresholds for anemia, neutropenia and thrombocytopenia in MDS, clonal cytopenia of undetermined significance (CCUS) and MDS/myeloproliferative neoplasm (MDS/MPN), given in Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition (Khoury 2022; Arber 2022).

Clonality. A somatic genetic aberration is found by targeted next-generation sequencing (NGS) in about 90% of MDS (Arber 2022; Haferlach 2014) and by conventional karyotype in about 50% (Arber 2022). MDS can be diagnosed without proven clonality when qualifying dysplasia and persistent cytopenia are present (Arber 2022). The adult classification applies from age 18 years. Patients younger than 18 years are classified under childhood MDS (Khoury 2022; Xiao 2024).

Recurrent mutations and patient profile. At least 1 oncogenic genomic alteration was mapped in 94% of the 2,957-patient IPSS-M cohort (Bernard 2022). TET2, SF3B1, ASXL1, SRSF2, DNMT3A and RUNX1 are each mutated in more than 10% of MDS (Haferlach 2014). Among 738 patients with MDS or closely related neoplasms, SF3B1 was the most frequently mutated gene (24%), followed by TET2 (22%) and SRSF2 (14%). Median leukemia-free survival fell from 49 months with 1 oncogenic mutation or cytogenetic lesion to 4 months with 6 or more (Papaemmanuil 2013). In the 7,012-patient IPSS-R cohort of untreated primary MDS, the median age was 71 years and the male-to-female ratio was 1.5:1 (Greenberg 2012).

Related page: Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition (WHO-HAEM5 and ICC frame, marrow workup, flow cytometry in MDS, CHIP, CCUS, VEXAS syndrome and germline predisposition) Related page: Acute Myeloid Leukemia with Defining Genetic Abnormalities, Acute Promyelocytic Leukemia and ELN 2022 Genetic Risk (AML-defining genetic lesions at the MDS/AML border and ELN 2022 risk) Related page: AML, Myelodysplasia-Related, AML Defined by Differentiation, Myeloid Neoplasms Post Cytotoxic Therapy and of Down Syndrome, Myeloid Sarcoma, BPDCN and Acute Leukemias of Ambiguous Lineage (AML, myelodysplasia-related, acute erythroid leukemia and post-cytotoxic therapy neoplasms) Related page: Mastocytosis, Hypereosinophilic Syndrome and Myeloid/Lymphoid Neoplasms with Tyrosine Kinase Fusions, Chronic Myelomonocytic Leukemia and Other MDS/MPN (CMML and other MDS/MPN) Related page: Bone Marrow Failure, Pure Red Cell Aplasia and the Laboratory Approach to Cytopenias and Leukocyte Disorders (aplastic anemia criteria, inherited marrow failure and the hypocellular marrow differential) Related page: Hemolytic Anemias: Red Cell Membrane and Enzyme Defects and Paroxysmal Nocturnal Hemoglobinuria (PNH flow cytometry and PNH clones in marrow failure)

MDS types by WHO-HAEM5 and ICC 2022, with WHO 2017 as the previous rule

WHO-HAEM5 groups MDS into types with defining genetic abnormalities and morphologically defined types (Khoury 2022). The ICC defines three lower-risk types (MDS-SF3B1, MDS-del(5q) and MDS, not otherwise specified [NOS]), one excess-blast type, MDS/AML and MDS with mutated TP53 (Arber 2022). The revised WHO 4th edition (WHO 2017) defined types by the number of dysplastic lineages, ring sideroblasts (RS) and blast counts, with del(5q) the only genetically defined type (Malcovati 2020; Hasserjian 2023).

The ICC reduced the 8 WHO 2017 entities, 1 of them genetically defined, to 7 entities, 3 of them genetically defined (Hasserjian 2023).

Table 1. MDS types in adults: WHO-HAEM5 (Khoury 2022; Xiao 2024), ICC 2022 (Arber 2022; Xiao 2024) and WHO 2017 (Malcovati 2020, adapted from the WHO 2016 revision). BM, bone marrow; PB, peripheral blood; VAF, variant allele frequency.

Defining feature WHO-HAEM5 ICC 2022 WHO 2017 (previous rule)
Isolated del(5q) MDS with low blasts and isolated 5q deletion (MDS-5q): BM <5%, PB <2%; del(5q) alone or with 1 other abnormality other than monosomy 7 or del(7q); dysplasia in ≥10% of cells of ≥1 lineage required MDS with del(5q): BM <5%, PB <2%; del(5q) with up to 1 additional abnormality except −7/del(7q); no multi-hit TP53; dysplasia not required; thrombocytosis allowed MDS with isolated del(5q): BM <5%, PB <1%, no Auer rods; del(5q) alone or with 1 additional abnormality except −7 or del(7q)
SF3B1 mutation MDS with low blasts and SF3B1 mutation (MDS-SF3B1): BM <5%, PB <2%; SF3B1 VAF ≥5% (Xiao 2024); ≥15% RS may substitute for the mutation; no del(5q), monosomy 7 or complex karyotype MDS with mutated SF3B1: BM <5%, PB <2%; SF3B1 VAF ≥10%; no multi-hit TP53 or RUNX1 mutation; excludes isolated del(5q), −7/del(7q), abnormal 3q26.2 and complex karyotype; RS not required MDS with RS (MDS-RS-SLD or MDS-RS-MLD): RS ≥15%, or ≥5% when SF3B1 is mutated; BM <5%, PB <1%
TP53 MDS with biallelic TP53 inactivation (MDS-biTP53): <20% BM and PB blasts; ≥2 TP53 mutations, or 1 mutation with TP53 copy number loss or copy-neutral loss of heterozygosity (cnLOH); no VAF minimum MDS with mutated TP53 (0%–9% blasts): multi-hit TP53 with each VAF ≥10%; MDS/AML with mutated TP53 (10%–19% blasts): any TP53 mutation with VAF ≥10% No TP53-defined type
Low blasts without a defining genetic lesion MDS with low blasts (MDS-LB): BM <5%, PB <2%; single-lineage versus multilineage dysplasia optional; "MDS with low blasts and ring sideroblasts" accepted for SF3B1-wild-type cases with ≥15% RS MDS, NOS without dysplasia (−7/del(7q) or complex karyotype); MDS, NOS with single-lineage dysplasia; MDS, NOS with multilineage dysplasia MDS with single-lineage dysplasia (MDS-SLD); MDS with multilineage dysplasia (MDS-MLD)
Hypocellular marrow MDS, hypoplastic (MDS-h): ≤25% cellularity, age adjusted Not a separate type Not a separate type
Increased blasts MDS-IB1: BM 5%–9% or PB 2%–4%; MDS-IB2: BM 10%–19% or PB 5%–19% or Auer rods; MDS with fibrosis (MDS-f): BM 5%–19%, PB 2%–19% with grade 2–3 fibrosis MDS with excess blasts (MDS-EB): BM 5%–9%, PB 2%–9% (or 1% PB on 2 occasions); MDS/AML: 10%–19% BM or PB MDS-EB-1: BM 5%–9% or PB 2%–4%, no Auer rods; MDS-EB-2: BM 10%–19% or PB 5%–19% or Auer rods
Unclassifiable cases MDS, unclassifiable removed MDS-U removed MDS-U: PB 1% blasts on 2 occasions, single-lineage dysplasia with pancytopenia, or an MDS-defining cytogenetic abnormality without dysplasia

Correspondence of blast-based types. WHO MDS-IB1 corresponds mostly to ICC MDS-EB, and WHO MDS-IB2 mostly to ICC MDS/AML. Cases with Auer rods and <10% blasts, and cases with 5%–9% PB blasts, are MDS-IB2 in WHO-HAEM5 and MDS-EB in the ICC (Xiao 2024; Lee 2024).

Precedence. MDS-biTP53 supersedes MDS-5q and MDS-SF3B1 in WHO-HAEM5 (Khoury 2022). In the ICC, excess blasts supersede the lower-risk types, and MDS with mutated TP53 supersedes excess blasts (Arber 2022).

MDS-defining genetic lesions without dysplasia, and features that exclude MDS

  • ICC. In a patient with unexplained cytopenia, a complex karyotype (≥3 independent acquired abnormalities, excluding −Y), −7/del(7q), del(5q), SF3B1 mutation or multi-hit TP53 defines MDS without ≥10% dysplasia (Xiao 2024; Hasserjian 2023). The SF3B1 and TP53 mutations must each have a VAF of at least 10% (Xiao 2024). The other cytogenetic abnormalities that were MDS-defining in WHO 2017 now indicate CCUS when dysplasia is absent (Arber 2022; Hasserjian 2023).
  • WHO-HAEM5. Dysplasia in ≥10% of cells of at least 1 lineage is required for MDS in all cases (Xiao 2024).
  • Dysplasia accompanying these lesions. These lesions are almost always accompanied by dysplasia. Absent dysplasia in their presence more often reflects a suboptimal sample (Xiao 2024).
  • Exclusions (ICC). A BCR::ABL1 rearrangement or a rearrangement of a myeloid/lymphoid neoplasm with eosinophilia and tyrosine kinase gene fusion excludes MDS, even with cytopenia. A sustained WBC ≥13 × 10⁹/L, monocytosis (≥0.5 × 10⁹/L and ≥10% of leukocytes) or platelets ≥450 × 10⁹/L at diagnosis indicates MDS/MPN or MPN. Thrombocytosis is allowed in MDS with del(5q) and in MDS with inv(3) or t(3;3) (Arber 2022).
  • Progression (ICC). Leukocytosis, thrombocytosis or monocytosis arising in established MDS is reported as MDS with neutrophilic, thrombocytotic or monocytic progression. MDS-SF3B1 that later develops thrombocytosis, with or without JAK2 mutation, is no longer reclassified as MDS/MPN (Arber 2022).

Dysplasia morphology, counting rules and the 10% threshold

Threshold. Both systems set the dysplasia threshold at 10% of cells for every lineage (Khoury 2022; Arber 2022). Micromegakaryocytes are the most specific indicator of MDS, and a higher threshold may be warranted when other forms of dysmegakaryopoiesis are counted (Arber 2022).

Counting. Differential counts follow Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition. At least 100 erythroblasts, 100 granulocytic cells and 30 megakaryocytes are evaluated for dysplasia (Invernizzi 2015). Blasts are counted visually. The percentage of CD34-positive cells by flow cytometry does not replace the morphologic count, because not all blasts express CD34 and aspirates are diluted with blood (Invernizzi 2015; Bennett 2009). Megakaryoblasts are counted as blasts together with the blast equivalents listed in Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition. Dysplastic megakaryocytes and proerythroblasts are not, except in pure erythroid leukemia (Invernizzi 2015).

Specimen. Blood and marrow are collected before definitive therapy, and MDS is not reclassified while the patient receives growth factor therapy. Smears for dysplasia are made from fresh specimens, because prolonged anticoagulant exposure produces artifacts (Invernizzi 2015). Stainable iron and RS are evaluated on Prussian blue-stained aspirate smears or touch preparations. Decalcification reduces stainable iron in the core biopsy (CAP 2026).

Ring sideroblasts. A ring sideroblast is an erythroblast with at least 5 siderotic granules covering at least one third of the nuclear circumference (Mufti 2008; Invernizzi 2015). RS are expressed as a percentage of marrow erythroid precursors (Malcovati 2020).

Table 2. Dysplastic features by lineage (Invernizzi 2015).

Lineage Blood Bone marrow
Erythroid Anisocytosis, poikilocytosis, basophilic stippling Nuclear budding, internuclear bridging, karyorrhexis, multinuclearity, nuclear hyperlobation, megaloblastic change; ring sideroblasts, cytoplasmic vacuolization, PAS positivity, incomplete hemoglobinization
Granulocytic Nuclear hypolobation (pseudo-Pelger–Huët), hypogranulation or degranulation, blasts Anisocytosis, pseudo-Pelger–Huët nuclei, irregular hypersegmentation, bizarre nuclear shapes, hypogranularity or agranularity, pseudo-Chédiak–Higashi granules, Auer rods
Megakaryocytic Platelet anisocytosis, giant platelets Micromegakaryocytes, nuclear hypolobation, small binucleated forms, widely separated nuclei, degranulation

Defining individual features. A dysplastic neutrophil shows at least a two-thirds reduction in granule content, pseudo-Pelger nuclear hypolobation, other abnormal segmentation, macropolycytes or abnormal chromatin clumping. A micromegakaryocyte is about the size of the surrounding myeloid cells, with scanty granular cytoplasm. A medium-sized megakaryocyte with a single, ovoid, eccentric nucleus is characteristic of the 5q− syndrome (Invernizzi 2015).

Limits of morphology. Among 120 healthy marrow donors, 46% showed dysgranulopoiesis >10% and 26% showed multilineage dysplasia. Pseudo-Pelger cells and micromegakaryocytes did not exceed 10%, and total dysmegakaryopoiesis did not exceed 40% (Invernizzi 2015). No single morphologic finding is diagnostic of MDS. Granulocytic and megakaryocytic abnormalities are more specific than erythroid changes (Invernizzi 2015).

Morphology and genotype. In 1,150 patients with peripheral blood cytopenia, a minimal morphological score for marrow dysplasia had sensitivity and specificity above 90%. The severity of granulocytic and megakaryocytic dysplasia affected survival. Ring sideroblasts were closely associated with SF3B1 mutation, and severe granulocytic dysplasia with ASXL1, RUNX1, TP53 and SRSF2 mutations (Della Porta 2015).

Reactive and inherited mimics of MDS

Table 3. Conditions that mimic MDS dysplasia (Invernizzi 2015; D'Angelo 2016; Wahab 2020; Khan 2023; Arber 2022; CAP 2026).

Mimic Distinguishing points Source or related page
Vitamin B12 or folate deficiency Megaloblastic dyserythropoiesis; exclusion of both deficiencies is an essential criterion for MDS-LB CAP 2026; Macrocytic Anemias and Vitamin B12 and Folate Testing (also transcobalamin deficiency)
Copper deficiency, including zinc excess Anemia and neutropenia; vacuoles in myeloid and erythroid precursors, ring sideroblasts, fewer granulocytic precursors, iron-containing plasma cells; risk factors are gastric or bariatric surgery, zinc or penicillamine therapy and protein-losing enteropathy D'Angelo 2016; Wahab 2020; Toxic Metals and Trace Elements: Lead, Arsenic, Mercury, Implant Metals, Copper and Wilson Disease
Excess alcohol intake; heavy metals (lead, arsenic); infections (HIV, Epstein–Barr virus, hepatitis C virus, parvovirus, Leishmania) Listed causes of reactive dysplasia Invernizzi 2015
Methotrexate Pancytopenia with hypoplastic, dysplastic marrow that improved after the drug was withheld and folinic acid was given (case report) Khan 2023; Macrocytic Anemias and Vitamin B12 and Folate Testing
Drug-induced cytopenia; growth factor therapy MDS is not reclassified during growth factor therapy Invernizzi 2015
Hemophagocytic lymphohistiocytosis, anemia of chronic disease, autoimmune cytopenia, liver or kidney failure Listed nonclonal causes of cytopenia with dysplasia Invernizzi 2015; Bone Marrow Failure, Pure Red Cell Aplasia and the Laboratory Approach to Cytopenias and Leukocyte Disorders
Germline predisposition (RUNX1, ANKRD26, ETV6, GATA2, SAMD9/SAMD9L and others) Baseline dyspoiesis, often megakaryocytic dysplasia with isolated thrombocytopenia; MDS is diagnosed with new or progressive cytopenia, rising cellularity, overt multilineage dysplasia, increased blasts or acquired MDS-defining lesions Arber 2022; Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition
VEXAS syndrome (somatic UBA1) Kept separate from MDS unless morphologic MDS criteria are met Arber 2022; Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition
Congenital and acquired sideroblastic anemias Ring sideroblasts without a myeloid neoplasm Microcytic Anemias and Iron Studies

Germline context (ICC). In a patient with germline predisposition, emergence of del(5q), −7/del(7q), complex karyotype, multi-hit TP53 (VAF ≥10%) or SF3B1 mutation (VAF ≥10%) is MDS-defining. In Shwachman–Diamond syndrome, small stable clones with monoallelic TP53 mutation are frequent and do not by themselves indicate MDS. Biallelic TP53 alterations in this setting are associated with myeloid malignancy (Arber 2022).

MDS with low blasts and isolated del(5q)

  • Criteria (CAP 2026, WHO-HAEM5). Anemia with or without other cytopenias or thrombocytosis; megakaryocytic dysplasia, often micromegakaryocytes, with or without dysplasia of other lineages; BM blasts <5% and PB blasts <2%; del(5q) isolated or with 1 other abnormality other than monosomy 7 or del(7q); criteria for AML, MDS-biTP53, MDS-IB and MDS/MPN not met.
  • Unchanged cytogenetic rule. The cytogenetic definition is the same in WHO-HAEM5, the ICC and WHO 2017 (Khoury 2022; Arber 2022; Malcovati 2020; Gangat 2024). The ICC shortened the name to MDS with del(5q) (Arber 2022; Lee 2024).
  • Megakaryocytes. Medium-sized megakaryocytes with a single, ovoid, eccentric nucleus are characteristic (Invernizzi 2015).
  • TP53 caveat. An SF3B1 mutation or a single monoallelic TP53 mutation does not override the diagnosis of MDS-5q (Khoury 2022). TP53 mutations occur in 18%–20% of MDS with del(5q), compared with 5%–10% of other MDS, and are more frequent in the leukemic phase. A TP53 VAF of ≥20% or ≥22% identified the subset with poor survival in published series (Shah 2025; Gangat 2024). Monoallelic TP53 mutations with VAF <22% had survival and AML progression risks comparable to TP53-wild-type del(5q) MDS (Shah 2025). Multi-hit TP53 reassigns the case to MDS-biTP53 (WHO-HAEM5) or MDS with mutated TP53 (ICC) (Khoury 2022; Arber 2022).
  • Treatment relevance. NCCN guidelines list lenalidomide as first- or second-line therapy for transfusion-dependent MDS-del(5q), typically with no more than 1 other cytogenetic abnormality and none involving chromosome 7 (Brunner 2022).

TP53 mutations can emerge or expand during lenalidomide therapy of MDS with del(5q), and NGS at diagnosis and periodically during lenalidomide treatment has been recommended (Gangat 2024).

MDS with SF3B1 mutation and ring sideroblasts

  • Distinct type. SF3B1 mutation defines a more homogeneous group than RS in MDS without excess blasts. SF3B1-unmutated MDS-RS cases behave like MDS with single-lineage or multilineage dysplasia (Arber 2022). MDS-SF3B1 includes over 90% of MDS with ≥5% RS (Khoury 2022).
  • Criteria (CAP 2026, WHO-HAEM5). Cytopenia of 1 or more lineages without thrombocytosis; erythroid dysplasia; BM blasts <5% and PB blasts <2%; SF3B1 mutation, or ≥15% RS of erythroid precursors when mutation analysis is unavailable; no del(5q), monosomy 7/del(7q) or complex karyotype; criteria for AML, MDS-5q, MDS-biTP53, MDS-IB and MDS/MPN not met.
  • System differences. The ICC requires SF3B1 VAF ≥10% and excludes cases with abnormal 3q26.2 or RUNX1 mutation. WHO-HAEM5 requires VAF ≥5% (Xiao 2024). In the ICC, SF3B1-wild-type cases with RS are MDS, NOS, irrespective of the RS count. WHO-HAEM5 accepts the term MDS with low blasts and ring sideroblasts for SF3B1-wild-type cases with ≥15% RS (Khoury 2022; Arber 2022).
  • Previous rule (WHO 2017). MDS-RS required ≥15% RS, or ≥5% RS when an SF3B1 mutation was present (Malcovati 2020).
  • Biology. SF3B1 mutations are usually heterozygous, with a median VAF of about 40%, and are often the initiating lesion (Malcovati 2020; Malcovati 2015). SF3B1-mutated MDS has favorable survival within very low- and low-risk IPSS-R categories. Coexisting inv(3) or t(3;3) marks a subset with markedly shorter survival (Malcovati 2020).
  • Treatment relevance. Luspatercept is used for anemia in lower-risk MDS with RS or SF3B1 mutation after erythropoiesis-stimulating agent failure (Brunner 2022).

Evidence for the genetic definition. SF3B1 mutations were found in 20% of 354 patients with MDS and in 65% of those whose disease showed ring sideroblasts (Papaemmanuil 2011). In a later series, the mutation was present in 90% of refractory anemia with RS and 70% of refractory cytopenia with multilineage dysplasia and RS. In MDS with RS, SF3B1-mutated patients had better overall survival (hazard ratio 0.37) and less disease progression (hazard ratio 0.31) than SF3B1-unmutated patients in multivariable analysis. Among SF3B1-mutated patients, ≥15% versus <15% RS had no significant effect on overall survival or progression. SF3B1 allele burden correlated with the RS percentage (Malcovati 2015).

TP53-mutated myeloid neoplasms: WHO-HAEM5 and ICC frameworks

Frequency and setting. TP53 alterations occur in 8%–10% of MDS and 10%–12% of AML. They are enriched in older patients, in carriers of germline pathogenic variants (Li–Fraumeni syndrome, BRCA1, BRCA2) and after cytotoxic or immunosuppressive therapy, including poly(ADP-ribose) polymerase (PARP) inhibitors (Shah 2025). In MDS, about two thirds of patients with TP53 alterations have multiple hits (Khoury 2022; Bernard 2020).

Biallelic (multi-hit) status. Multi-hit status is a clone without residual wild-type p53. It results from 2 or more TP53 mutations, which usually affect both alleles, or from 1 mutation with deletion or cnLOH of the other allele (Khoury 2022).

Table 4. Myeloid neoplasms with TP53 alterations by system (Khoury 2022; Arber 2022; Shah 2025; Xiao 2024; Lee 2024).

Feature WHO-HAEM5 ICC 2022
MDS category MDS-biTP53: <20% BM and PB blasts MDS with mutated TP53: 0%–9% BM and PB blasts
Genetic requirement in MDS ≥2 TP53 mutations, or 1 mutation with copy number loss or cnLOH 2 distinct TP53 mutations, each VAF ≥10%; or 1 mutation (VAF ≥10%) with 17p deletion, VAF >50% or cnLOH at the 17p TP53 locus
VAF minimum None >10%
Complex karyotype without LOH data Does not establish biallelic status A single TP53 mutation (VAF ≥10%) with complex karyotype counts as multi-hit equivalent
10%–19% blasts MDS-biTP53 when biallelic; a monoallelic mutation does not define a TP53 type MDS/AML with mutated TP53: any somatic TP53 mutation, VAF ≥10%, monoallelic allowed
≥20% blasts No TP53-defined AML type; most cases have a complex karyotype and qualify as AML, myelodysplasia-related; acute erythroid leukemia is a separate type AML with mutated TP53: any somatic TP53 mutation, VAF ≥10%; includes pure erythroid leukemia
Precedence MDS-biTP53 supersedes MDS-5q and MDS-SF3B1 Supersedes excess blasts and the other MDS types

VAF threshold. The ICC TP53 and SF3B1 VAF threshold is ≥10% (Arber 2022; Shah 2025).

Exclusions from ICC MDS with mutated TP53. A complex karyotype without a TP53 mutation, even with 17p deletion, does not qualify, because such cases have a better prognosis than TP53-mutated MDS. Monoallelic TP53 mutation in MDS with <10% blasts has a less adverse effect and is excluded from the MDS entity. Monoallelic mutations are allowed in MDS/AML and AML with mutated TP53, because monoallelic TP53-mutated AML has a poor prognosis (Arber 2022).

Detection. Sequencing covers at least exons 4 to 11. Copy number status is assessed by FISH for the TP53 locus at 17p13.1, by comparative genomic hybridization or SNP array, or inferred from NGS. cnLOH requires SNP array, a specialized NGS panel or whole-genome sequencing. A TP53 VAF ≥50% is presumptive evidence of loss of the other allele when a constitutional variant has been excluded. It does not prove the loss (Khoury 2022; Shah 2025).

Up to 6% of multi-hit TP53 lesions have a VAF <50% and can escape detection by the VAF >50% and 17p-deletion criteria. Copy number probes or SNP array-based karyotyping are recommended to detect cnLOH (Duncavage 2022). Mutant p53 protein has a prolonged half-life and is detectable by immunohistochemistry (Cha 2014).

Karyotype and behavior. Over 90% of MDS-biTP53 cases have a complex karyotype, mostly with >3 abnormalities, and are very high risk by IPSS-R (Khoury 2022; Lee 2024). Prognosis of multi-hit TP53 MDS does not depend on blast percentage (Arber 2022). WHO-HAEM5 states that MDS-biTP53 may be regarded as AML-equivalent for therapeutic decisions (Khoury 2022). The ICC groups MDS, MDS/AML and AML with mutated TP53 for a unified treatment approach across the blast spectrum (Arber 2022).

Allelic state and outcome. In 3,324 patients with MDS, complex karyotype, few co-occurring mutations, high-risk presentation and poor outcome were specific to multi-hit TP53. Multi-hit status predicted death and leukemic transformation independently of IPSS-R. Patients with monoallelic mutations did not differ from TP53-wild-type patients in outcome or response to therapy (Bernard 2020).

Related entities. Acute erythroid leukemia and AML, myelodysplasia-related are described in AML, Myelodysplasia-Related, AML Defined by Differentiation, Myeloid Neoplasms Post Cytotoxic Therapy and of Down Syndrome, Myeloid Sarcoma, BPDCN and Acute Leukemias of Ambiguous Lineage. TP53 in ELN 2022 risk is described in Acute Myeloid Leukemia with Defining Genetic Abnormalities, Acute Promyelocytic Leukemia and ELN 2022 Genetic Risk.

Hypoplastic MDS and its distinction from aplastic anemia

  • Definition (WHO-HAEM5). MDS-h is a morphologically defined type with cytopenia, a hypocellular marrow assessed on the trephine biopsy and adjusted for age, dysplasia of myeloid and/or megakaryocytic lineages, BM blasts <5% and PB blasts <2%. Hypocellularity due to drugs, toxins or nutritional deficiency is excluded (CAP 2026). WHO-HAEM5 defines MDS-h as ≤25% bone marrow cellularity, age adjusted (Khoury 2022). Published series have used thresholds between 20% and 30%, including <30% under 70 years and <20% over 70 years (Fattizzo 2021), and <30% at 60 years or younger and <20% over 60 years (Cha 2014). Neither classification adopts these series cutoffs (Khoury 2022; Arber 2022).
  • System difference. The ICC has no hypoplastic type and sets no cellularity threshold. Such cases are classified by genetics, blasts and lineages of dysplasia (Xiao 2024; Arber 2022). WHO 2017 had no hypoplastic type, and WHO-HAEM5 lists MDS-h as a distinct type for the first time (Khoury 2022; Malcovati 2020).
  • Pathogenesis. MDS-h shows a T-cell-mediated attack on stem and progenitor cells with oligoclonal expansion of CD8-positive cytotoxic T cells producing interferon-γ and/or tumor necrosis factor-α. It overlaps with aplastic anemia (AA) and paroxysmal nocturnal hemoglobinuria (PNH), and many patients respond to antithymocyte globulin (Khoury 2022).
  • Germline exclusion. Germline variants in GATA2, DDX41, Fanconi anemia genes or telomerase complex genes produce hypoplastic marrows that evolve to MDS or AML and do not respond to immunosuppressive therapy (Khoury 2022).
  • Frequency and outcome. MDS-h accounts for 10%–15% of MDS (Fattizzo 2021; Calabretto 2022; Bennett 2009). Progression to AML at 5 years is about 10%–40%, between AA and normocellular or hypercellular MDS. Median survival is 33–58 months, compared with 19–28 months in non-hypoplastic MDS (Fattizzo 2021).

Table 5. Hypoplastic MDS compared with aplastic anemia (Fattizzo 2021; Arber 2022; Bennett 2009).

Finding Hypoplastic MDS Aplastic anemia
Dysmegakaryopoiesis and dysgranulopoiesis Present; micromegakaryocytes are the most specific indicator of MDS Not a feature
Mild isolated dyserythropoiesis Present Common; does not separate the two
Ring sideroblasts or clusters of blasts Support MDS Not a feature
Monosomy 7 or del(7q) Aids the distinction from AA in a large hypoplastic MDS series Also reported in MDS evolving from AA
PNH clones About 40%; MDS overall 10%–20%, usually <10% of granulocytes Frequent; frequency and monitoring in Bone Marrow Failure, Pure Red Cell Aplasia and the Laboratory Approach to Cytopenias and Leukocyte Disorders
Reduced CD19-positive B cells Present Present

Aplastic anemia criteria and severity grading are described in Bone Marrow Failure, Pure Red Cell Aplasia and the Laboratory Approach to Cytopenias and Leukocyte Disorders. FISH panels for chromosome 3, 5 and 7 abnormalities are used when a hypocellular aspirate yields too few metaphases (Fattizzo 2021).

Trephine findings that favor hypoplastic MDS. Easily identifiable megakaryocytes in an architecturally disorganized marrow and reticulin fibrosis favor MDS over aplastic anemia. An increased percentage of CD34-positive cells and their tendency to form aggregates separate hypoplastic myeloid neoplasms from aplastic anemia (Bennett 2009). CD34-positive and p53-positive marrow cells were more numerous in hypoplastic MDS than in aplastic anemia, and increased CD34-positive cells predicted poorer survival in hypoplastic MDS (Cha 2014).

MDS with fibrosis

  • Definition (WHO-HAEM5). MDS-f has BM blasts 5%–19% and/or PB blasts 2%–19% with significant fibrosis, defined as grade 2 or 3 (CAP 2026). It corresponds to ICC MDS-EB and MDS/AML with grade 2–3 fibrosis (Xiao 2024).
  • Grading. Fibrosis is graded MF-0 to MF-3 (EUMNET); grading rules and the table are in Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition. MDS-f requires grade 2 or 3 (CAP 2026).
  • Significance. Marrow fibrosis is an independent risk factor for poor survival in MDS and is not included in IPSS-R. Collagen fibrosis is more strongly linked to abnormal counts and poor outcome than reticulin fibrosis and is less often reversed by therapy. Monosomy 7 is a reported cytogenetic feature of fibrotic MDS (Jain 2022).

Fibrotic MDS compared with other myeloid neoplasms. In myeloid neoplasms with fibrosis, multilineage dysplasia, hypolobulated or multinucleated megakaryocytes and increased CD34-positive progenitors without JAK2, MPL or CALR mutation were associated with a myelodysplastic phenotype (Della Porta 2015). In a 2024 WHO-HAEM5 cohort, MDS-f had a median BM blast percentage similar to MDS-IB1 and survival similar to MDS-IB2 (Lee 2024).

MDS with increased blasts and ICC MDS/AML

  • WHO-HAEM5. MDS-IB requires cytopenia, dysplasia in ≥10% of cells of at least 1 lineage, and ≥5% BM blasts and/or ≥2% PB blasts. MDS-IB1 has 5%–9% BM or 2%–4% PB blasts without significant fibrosis. MDS-IB2 has 10%–19% BM or 5%–19% PB blasts without significant fibrosis, or Auer rods (CAP 2026). WHO-HAEM5 kept a 20% blast boundary between MDS and AML and states that MDS-IB2 may be regarded as AML-equivalent for therapeutic decisions and trial design (Khoury 2022).
  • ICC. MDS-EB requires ≥5% BM blasts or ≥2% PB blasts, or 1% PB blasts documented on 2 occasions. One percent PB blasts on a single occasion is accepted in any non-excess-blast type (Arber 2022).
  • ICC MDS/AML. The former MDS-EB-2 range in adults is renamed MDS/AML, a cytopenic myeloid neoplasm with 10%–19% blasts in blood or marrow on the continuum between MDS and AML (Arber 2022; Hasserjian 2023). MDS/AML is subclassified as MDS/AML with mutated TP53, with myelodysplasia-related gene mutations, with myelodysplasia-related cytogenetic abnormalities, or NOS. Patients are eligible for MDS and AML trials (Arber 2022). AML-defining genetic lesions are classified as AML and excluded from MDS/AML (Xiao 2024). BCR::ABL1-positive cases are not classified as MDS/AML (Arber 2022).
  • Previous rule (WHO 2017). MDS-EB-1 had BM 5%–9% or PB 2%–4% blasts without Auer rods. MDS-EB-2 had BM 10%–19% or PB 5%–19% blasts or Auer rods (Malcovati 2020).

In a 2024 WHO-HAEM5 cohort, MDS-IB2 had shorter leukemia-free and overall survival than MDS-IB1, and MDS-biTP53 (median overall survival 4.5 months) had shorter survival than MDS-f, MDS-IB2 and MDS-IB1 (Lee 2024).

Childhood MDS

  • WHO-HAEM5. Childhood MDS occurs before 18 years, has an annual incidence of 1–2 per million children, and presents with increased blasts in 10%–25%. Juvenile myelomonocytic leukemia, myeloid proliferations of Down syndrome and MDS post cytotoxic therapy are excluded (Khoury 2022).
  • Childhood MDS with low blasts (cMDS-LB). It replaces refractory cytopenia of childhood (RCC) and has hypocellular and NOS subtypes. About 80% of cases have a hypocellular marrow resembling severe AA (Khoury 2022; Drexler 2025). Monosomy 7, del(7q) and complex karyotype carry an increased risk of progression to AML. Normal karyotype and trisomy 8 can follow an indolent course (Khoury 2022). Criteria are cytopenia, dysplasia in ≥10% of cells of ≥1 lineage, BM blasts <5% and PB blasts <2%, and a clonal abnormality or exclusion of other causes of cytopenia (CAP 2026).
  • Childhood MDS with increased blasts (cMDS-IB). It has 5%–19% BM blasts and/or 2%–19% PB blasts. Acquired cytogenetic abnormalities and RAS-pathway mutations are more common than in cMDS-LB (Khoury 2022; CAP 2026).
  • ICC. RCC is kept in a section of pediatric and germline disorders. It requires dysplasia in at least 2 lineages or in ≥10% of cells of 1 lineage, BM blasts <5% and PB blasts <2%, no fibrosis and no prior cytotoxic chemotherapy or radiation therapy. Immunohistochemistry for megakaryocyte markers is required (Arber 2022, Box 1). Pediatric MDS-EB includes 10%–19% blasts, and MDS/AML does not apply (Arber 2022).
  • Genetics. Pediatric MDS lacks the epigenetic and RNA splicing mutations of adult clonal hematopoiesis. Somatic SETBP1, ASXL1, RUNX1 and RAS/MAPK pathway mutations predominate. RCC can represent progression of Fanconi anemia, dyskeratosis congenita, Shwachman–Diamond syndrome, GATA2 deficiency or SAMD9/SAMD9L syndromes (Arber 2022; Drexler 2025), described in Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition and Bone Marrow Failure, Pure Red Cell Aplasia and the Laboratory Approach to Cytopenias and Leukocyte Disorders.

RCC histology and management. The RCC trephine shows patchy hematopoiesis with left-shifted erythropoiesis and increased mitoses, and markedly decreased granulopoiesis and megakaryopoiesis (Arber 2022; Drexler 2025). Monosomy 7 and del(7q) are the most frequent MDS-defining abnormalities in RCC (Arber 2022). In the risk-adapted EWOG-MDS approach, transfusion dependence, absolute neutrophil count (ANC) <1 × 10⁹/L, monosomy 7, del(7q) or ≥2 chromosomal aberrations indicated early hematopoietic stem cell transplantation or, in selected cases, immunosuppressive therapy. Observation was safe in selected patients with a normal karyotype after a germline predisposition syndrome was excluded (Drexler 2025).

Prognostic scoring: IPSS-R and IPSS-M

IPSS-R. The Revised International Prognostic Scoring System (IPSS-R) was derived from 7,012 patients. It scores 5 variables: cytogenetic category, BM blast percentage, hemoglobin, platelet count and absolute neutrophil count (ANC) (Greenberg 2012).

Table 6. IPSS-R cytogenetic categories (Greenberg 2012).

Category Abnormalities Median survival (years)
Very good −Y, del(11q) 5.4
Good Normal; del(5q), del(12p), del(20q); double abnormality including del(5q) 4.8
Intermediate del(7q), +8, +19, i(17q); any other single or double independent clone 2.7
Poor −7; inv(3), t(3q) or del(3q); double abnormality including −7/del(7q); complex karyotype with 3 abnormalities 1.5
Very poor Complex karyotype with >3 abnormalities 0.7

Table 7. IPSS-R score values (Greenberg 2012). Hemoglobin in g/dL; platelets and ANC in × 10⁹/L.

Variable 0 0.5 1 1.5 2 3 4
Cytogenetics Very good Not used Good Not used Intermediate Poor Very poor
BM blasts (%) ≤2 Not used >2 to <5 Not used 5–10 >10 Not used
Hemoglobin ≥10 Not used 8 to <10 <8 Not used Not used Not used
Platelets ≥100 50 to <100 <50 Not used Not used Not used Not used
ANC ≥0.8 <0.8 Not used Not used Not used Not used Not used

Table 8. IPSS-R risk categories (Greenberg 2012).

Risk category Score Patients (%) Median survival (years)
Very low ≤1.5 19 8.8
Low >1.5 to 3 38 5.3
Intermediate >3 to 4.5 20 3.0
High >4.5 to 6 13 1.6
Very high >6 10 0.8

The IPSS-R categories describe scores for 70-year-old patients (Greenberg 2012).

Age adjustment (IPSS-RA). The age-adjusted score adds (age in years − 70) × [0.05 − (IPSS-R score × 0.005)] to the sum of the 5 variables. Age affected survival and had no effect on AML evolution (Greenberg 2012).

IPSS-M. The Molecular International Prognostic Scoring System (IPSS-M) was developed from 2,957 patients profiled for mutations in 152 genes and validated in 754 Japanese patients. It combines hematologic parameters, cytogenetic abnormalities and mutations in 31 genes into an individual risk score with 6 categories. TP53 multi-hit status, FLT3 mutations and KMT2A partial tandem duplication (MLL-PTD) were the strongest adverse genetic predictors. SF3B1 mutations were favorable, modulated by comutation patterns. Compared with IPSS-R, the IPSS-M improved discrimination for leukemia-free survival, leukemic transformation and overall survival and restratified 46% of patients. It applies to primary and secondary or therapy-related MDS, and an open-access web calculator accounts for missing values (Bernard 2022).

IPSS-M variables and validation. Cytogenetics enters by IPSS-R category. TP53 is scored as not mutated, monoallelic or multi-hit. SF3B1 is split into SF3B1-5q (with isolated del(5q), or del(5q) with 1 additional aberration other than −7/del(7q)) and SF3B1-α (without comutation in BCOR, BCORL1, RUNX1, NRAS, STAG2 or SRSF2 and without del(5q)). The 6 categories are very low, low, moderate low, moderate high, high and very high. In a 2,876-patient validation cohort, concordance was 0.81 for IPSS-M and 0.74 for IPSS-R for overall survival. A minimum set of 15 genes (ASXL1, CBL, DNMT3A, ETV6, EZH2, FLT3, IDH2, MLL-PTD, NPM1, NRAS, RUNX1, SF3B1, SRSF2, TP53 multi-hit and U2AF1) maintained 70%–80% accuracy (Sauta 2023). IPSS-M improved prediction of relapse and survival after allogeneic transplantation. It did not predict response to hypomethylating agents (Sauta 2023).

Genetic workup and CAP reporting elements for MDS

  • Testing. Conventional karyotype detects del(5q), −7/del(7q) and complex karyotype. A myeloid NGS panel detects SF3B1, TP53 and the 31 IPSS-M genes. TP53 copy number is assessed by FISH or arrays when a TP53 mutation is found (Arber 2022; Khoury 2022; Bernard 2022; Shah 2025).
  • CAP case summary (version 1.1.0.2, September 2026). Elements are site and percent blasts or blast equivalents, marrow cellularity, final integrated diagnosis by WHO-HAEM5 type, dysplasia by lineage (erythroid, granulocytic, megakaryocytic; significant dysplasia is ≥10% of cells of a lineage), ring sideroblasts on iron stain (<15% or ≥15%), fibrosis grade (MF-0 to MF-3), flow cytometry, conventional cytogenetics, FISH and molecular alterations detected (CAP 2026).

FISH panels for −5, del(5q) and −7 are not required when metaphase cytogenetic studies are adequate (Duncavage 2022).

IPSS-R calculator for myelodysplastic neoplasms

Acute Myeloid Leukemia with Defining Genetic Abnormalities, Acute Promyelocytic Leukemia and ELN 2022 Genetic Risk

Definition. Acute myeloid leukemia (AML) with defining genetic abnormalities is the WHO 5th edition (WHO-HAEM5) family of AML types named by a fusion, rearrangement or mutation. WHO-HAEM5 separates this family from AML defined by differentiation and drops the term AML, not otherwise specified (NOS) (Khoury 2022; CAP 2026). The International Consensus Classification (ICC 2022) lists AML with recurrent genetic abnormalities in a hierarchy and keeps AML, NOS for cases without a defining lesion (Arber 2022; El Chaer 2023). Both systems apply prior cytotoxic therapy and germline predisposition as qualifiers to the genetic diagnosis (Xiao 2024). The ICC also applies progression from MDS or MDS/MPN as a qualifier (Arber 2022; Xiao 2024).

Integration of morphology and genetics. Removing a blast threshold requires correlation of morphology with the genetic result, to confirm that the defining abnormality accounts for the disease (Khoury 2022; Xiao 2024).

Related page: Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition (marrow workup, blast counting, flow cytometry of acute leukemia and AML measurable residual disease methods) Related page: Myelodysplastic Neoplasms: WHO 5th Edition and ICC Types, Dysplasia Morphology, TP53-Mutated Myeloid Neoplasms and Prognostic Scoring (MDS/AML border and TP53-mutated myeloid neoplasms) Related page: AML, Myelodysplasia-Related, AML Defined by Differentiation, Myeloid Neoplasms Post Cytotoxic Therapy and of Down Syndrome, Myeloid Sarcoma, BPDCN and Acute Leukemias of Ambiguous Lineage (AML, myelodysplasia-related, AML defined by differentiation, post-cytotoxic therapy and Down syndrome neoplasms, myeloid sarcoma and mixed-phenotype acute leukemia) Related page: Myeloproliferative Neoplasms: Chronic Myeloid Leukemia with Molecular Monitoring, Polycythemia Vera, Essential Thrombocythemia, Primary Myelofibrosis and Other Myeloproliferative Neoplasms (chronic myeloid leukemia phases and BCR::ABL1 monitoring) Related page: DIC, Thrombotic Microangiopathy and Viscoelastic Testing (coagulopathy of acute promyelocytic leukemia)

AML families and blast thresholds by WHO-HAEM5, ICC 2022 and WHO 2017

Blasts and blast equivalents. Blast and blast-equivalent counting is described in Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition. Megakaryoblasts are also counted as blasts, and abnormal monocytes are not counted as blast equivalents (Döhner 2017). The ICC also counts neoplastic promyelocytes as blast equivalents in the appropriate setting (Arber 2022).

Table 1. Blast requirement for AML with defining or recurrent genetic abnormalities (Khoury 2022; Arber 2022; Xiao 2024; El Chaer 2023; CAP 2026; Umeda 2025; Döhner 2017 for WHO 2017). "No threshold" and "increased blasts" mean that no fixed cutoff applies and morphology is correlated with the genetic result.

AML type WHO-HAEM5 ICC 2022 WHO 2017 (previous rule)
APL with PML::RARA Increased blasts or atypical promyelocytes; may be <20% ≥10% Any count
APL with other RARA rearrangements Not a separate type ≥10% Reported as AML with the variant translocation
AML with RUNX1::RUNX1T1 No threshold ≥10% Any count
AML with CBFB::MYH11 No threshold ≥10% Any count
AML with KMT2A rearrangement (ICC lists t(9;11) MLLT3::KMT2A separately) Increased blasts; <20% acceptable ≥10% ≥20%
AML with DEK::NUP214 Increased blasts ≥10% ≥20%
AML with MECOM rearrangement (ICC lists inv(3)/t(3;3) separately) Increased blasts; <20% acceptable ≥10% ≥20%
AML with NUP98 rearrangement Increased blasts; <20% acceptable ≥10% ≥20%
AML with RBM15::MRTFA Increased blasts ≥10% (listed with other rare recurring translocations) ≥20%
AML with NPM1 mutation Increased blasts ≥10% ≥20%
AML with CEBPA mutation ≥20%; biallelic or single bZIP mutation ≥10%; in-frame bZIP mutation only ≥20%; biallelic mutation
AML with BCR::ABL1 ≥20% ≥20%; MDS/AML not used ≥20% (provisional entity)
AML with other defined genetic alterations (WHO) or other rare recurring translocations (ICC) ≥20% ≥10% ≥20%
MDS with 10%–19% blasts and no defining lesion MDS-IB2 MDS/AML MDS-EB-2

Blast requirement for genetically defined AML in practice. In WHO-HAEM5, any increase in blasts with an AML-defining lesion establishes AML. Increased blasts are typically ≥5% in marrow or ≥2% in blood, but no specific cutoff is defined (Xiao 2024). With BCR::ABL1 and 10%–19% blasts, the ICC classifies the case as chronic myeloid leukemia (CML), accelerated phase, described in Myeloproliferative Neoplasms: Chronic Myeloid Leukemia with Molecular Monitoring, Polycythemia Vera, Essential Thrombocythemia, Primary Myelofibrosis and Other Myeloproliferative Neoplasms (Xiao 2024).

AML with RUNX1 mutation. WHO-HAEM5 does not recognize it as a type, because RUNX1 mutations overlap with many defining lesions (Khoury 2022). WHO 2017 listed AML with mutated RUNX1 as a provisional entity (Döhner 2017). The ICC places RUNX1 among the myelodysplasia-related gene mutations, described in AML, Myelodysplasia-Related, AML Defined by Differentiation, Myeloid Neoplasms Post Cytotoxic Therapy and of Down Syndrome, Myeloid Sarcoma, BPDCN and Acute Leukemias of Ambiguous Lineage (Arber 2022).

Hierarchy (ICC and ELN 2022). Single-gene mutation and fusion categories take precedence over myelodysplasia-related gene mutations and cytogenetic abnormalities (Arber 2022; El Chaer 2023). A TP53 mutation coexisting with another defining lesion is reported with it because it predicts a worse prognosis (Arber 2022).

Acute promyelocytic leukemia with PML::RARA

Frequency and urgency. APL accounts for 5%–10% of AML (Shimony 2025). Suspicion of APL is a medical emergency, because it requires immediate all-trans retinoic acid (ATRA), prompt genetic confirmation and management of the coagulopathy (Sanz 2019; de Figueiredo-Pontes 2024). The coagulopathy, with coagulation activation, hyperfibrinolysis and thrombocytopenia, is described in DIC, Thrombotic Microangiopathy and Viscoelastic Testing.

Morphology.

  • Hypergranular (typical) form. Abnormal promyelocytes have eccentric, often bilobed nuclei with folded contours and prominent nucleoli, irregular azurophilic granules and Auer rods. Cells with bundles of Auer rods are faggot cells. About 70% of patients present with pancytopenia (de Figueiredo-Pontes 2024).
  • Microgranular (hypogranular) variant. Granules are scant or absent and nuclei are bilobed. This variant presents more often with leukocytosis and commonly expresses CD2 (de Figueiredo-Pontes 2024; Gupta 2021).
  • Marrow. The aspirate is usually hypercellular. Blasts and abnormal promyelocytes exceed 20% in most cases, but APL is diagnosed with t(15;17)(q24.1;q21.2) or PML::RARA even when they are <10% (de Figueiredo-Pontes 2024).

Immunophenotype. Abnormal promyelocytes have high side scatter and intermediate CD45. They express CD13, CD33 (bright) and CD117 (variable) and are low or negative for CD34, CD11b, HLA-DR and CD15 (de Figueiredo-Pontes 2024; Gupta 2021). AML with NPM1 mutation also has HLA-DR-negative blasts, so every HLA-DR-negative leukemia requires molecular exclusion of PML::RARA (de Figueiredo-Pontes 2024). Side scatter, a monocytic population, CD34, HLA-DR, CD2, CD11c and CD15 separate APL from NPM1-mutated AML by flow cytometry (Gupta 2021).

Genetic confirmation.

  • Karyotype. G-banding is mandatory and shows t(15;17)(q24.1;q21.2) in most cases. Additional abnormalities, such as trisomy 8 and del(7q), occur in more than 40% and do not affect prognosis (de Figueiredo-Pontes 2024).
  • Rapid tests. FISH, RT-PCR, real-time quantitative PCR and immunostaining with an anti-PML monoclonal antibody are all used for rapid diagnosis (Sanz 2019). FISH for PML::RARA on blood or marrow returns a result in under 24 hours (de Figueiredo-Pontes 2024).
  • RT-PCR. FISH and PML immunostaining do not substitute for reverse transcription PCR, which is run in parallel because it alone defines the PML::RARA isoform and quantifies the transcript for measurable residual disease (MRD) (Sanz 2019; de Figueiredo-Pontes 2024). Breakpoints in PML produce long (bcr1, intron 6), variant (bcr2, exon 6) and short (bcr3, exon 3) isoforms (de Figueiredo-Pontes 2024).

Sensitivity of APL genetic tests. Karyotype, FISH, PCR and PML immunostaining are equally specific for APL, and karyotyping is the most prone to false-negative results (Sanz 2019).

Variant RARA fusions. The ICC lists APL with other RARA rearrangements: t(1;17)(q42.3;q21.2)/IRF2BP2::RARA, t(5;17)(q35.1;q21.2)/NPM1::RARA, t(11;17)(q23.2;q21.2)/ZBTB16::RARA, cryptic inv(17q) or del(17)(q21.2q21.2)/STAT5B::RARA and STAT3::RARA, and rare partners TBL1XR1, FIP1L1 and BCOR (Arber 2022). APL with t(11;17)/ZBTB16::RARA or STAT5B::RARA responds poorly to ATRA and arsenic trioxide (ATO) (Sanz 2019; Zhang 2021; de Figueiredo-Pontes 2024).

Frequency and treatment of variant RARA fusions. PML::RARA is present in 98% of APL, and ZBTB16::RARA, the most common variant fusion, accounts for about 1% (Zhang 2021). APL with an ATRA-resistant variant fusion is treated with AML-type therapy (Sanz 2019).

Risk and treatment.

  • Risk group. A white blood cell (WBC) count >10 × 10⁹/L defines high-risk APL. Risk-based treatment now uses the WBC alone (Sanz 2019; de Figueiredo-Pontes 2024).
  • ATRA on suspicion. ATRA is started at the first suspicion of APL, without waiting for molecular confirmation, because it improves the coagulopathy. ATRA is stopped if genetic testing does not confirm APL (Sanz 2019; de Figueiredo-Pontes 2024).
  • Non-high-risk APL. ATRA with ATO is the standard of care (Sanz 2019; de Figueiredo-Pontes 2024).
  • High-risk APL. ATRA with chemotherapy, or ATRA and ATO with cytoreductive chemotherapy or gemtuzumab ozogamicin, is used. Hyperleukocytosis at presentation is treated with idarubicin or daunorubicin together with ATRA and ATO, without waiting for molecular results (Sanz 2019; de Figueiredo-Pontes 2024).

Differentiation syndrome. ATRA or ATO causes differentiation syndrome in 2.5%–37% of patients (de Figueiredo-Pontes 2024). It is a systemic inflammatory response with capillary leak: dyspnea, pulmonary infiltrates, fever, pleural or pericardial effusion, acute kidney injury, hypotension, weight gain and edema (Montesinos 2011; de Figueiredo-Pontes 2024). Median onset is 7–12 days after differentiation therapy is started (Montesinos 2011). WBC >5–10 × 10⁹/L, elevated creatinine and coagulopathy are risk factors (Montesinos 2011; de Figueiredo-Pontes 2024). Treatment is dexamethasone 10 mg intravenously every 12 hours at the first suspicion, with temporary interruption of ATRA and ATO only in severe cases (Sanz 2019; de Figueiredo-Pontes 2024).

MRD and relapse. The key MRD end point is PCR negativity for PML::RARA in marrow at the end of consolidation (Sanz 2019; Heuser 2021). Conversion from undetectable to detectable PML::RARA, or a ≥1 log₁₀ rise in high-risk patients, confirmed in a second sample, indicates imminent relapse (Heuser 2021). Molecular relapse is 2 successive RT-PCR assays with stable or rising PML::RARA in independent samples. Hematologic relapse is >20% blasts or abnormal promyelocytes in 1 marrow sample, or >5% in 2 samples 2 weeks apart (de Figueiredo-Pontes 2024).

Core-binding factor AML: RUNX1::RUNX1T1 and CBFB::MYH11

Shared features. Core-binding factor (CBF) AML comprises t(8;21)(q22;q22.1)/RUNX1::RUNX1T1 and inv(16)(p13.1q22) or t(16;16)(p13.1;q22)/CBFB::MYH11. Together they account for about 25% of pediatric AML (Quessada 2021). Each fusion blocks myeloid differentiation but is insufficient alone. Additional cytogenetic abnormalities and/or somatic mutations are present in all cases, with a median of 7 mutations per case, mostly in tyrosine kinase signaling genes such as KIT, NRAS, KRAS and FLT3 (Quessada 2021). WHO-HAEM5 kept the established diagnostic criteria and added cytogenetic features and post-induction MRD to KIT mutation as prognostic factors (Khoury 2022). Both are favorable risk in ELN 2017 and ELN 2022, and KIT or FLT3 mutations do not change the ELN 2022 category (Döhner 2017; Shimony 2025).

Table 2. Core-binding factor AML (Quessada 2021; Papadakis 2024; Döhner 2017; Heuser 2021; CAP 2026; El Achkar 2026).

Feature AML with RUNX1::RUNX1T1 AML with CBFB::MYH11
Karyotype t(8;21)(q22;q22.1) inv(16)(p13.1q22) or t(16;16)(p13.1;q22)
Morphology and phenotype Maturing myeloid blasts (FAB M2 in 80% of pediatric cases); single, thin Auer rods; dysgranulopoiesis; aberrant CD19 and CD56, useful for flow MRD Myelomonocytic leukemia (FAB M4Eo) with abnormal eosinophils bearing large, irregular basophilic (purple) granules, mostly at the late promyelocyte and myelocyte stages; unlike normal eosinophils, they stain with chloroacetate esterase and PAS (Papadakis 2024)
Detection pitfalls Variant or cryptic forms in about 10%, confirmed by FISH Inversion can be subtle on karyotype; FISH or PCR confirms it, especially in M4Eo or with trisomy 22
Additional cytogenetic abnormalities Loss of X or Y (46%), del(9q) (12%), trisomy 8, abnormal 7q, trisomy 4 Trisomy 22 (characteristic), del(7q), trisomy 8
Prognostic modifiers Del(9q) associated with a lower complete remission rate in children Trisomy 8, twice as frequent as in t(8;21), had the strongest negative prognostic impact in a large adult CBF AML study
MRD Molecular MRD by qPCR or digital PCR Molecular MRD by qPCR or digital PCR

Morphology and KIT in core-binding factor AML. Auer rods and salmon-pink granulation are the typical marrow aspirate cytomorphology of AML with t(8;21) (El Achkar 2026). In a series of CBF AML, high-risk additional cytogenetic abnormalities, mainly complex karyotypes, were associated with loss of this typical morphology and with more relapses (50% versus 18%) (El Achkar 2026). In AML with CBFB::MYH11, marrow eosinophils usually amount to ≥5% of nonerythroid cells, and blood eosinophils are not increased (Papadakis 2024). In t(8;21) AML, KIT mutations, especially at higher mutant levels, have been associated with a poorer prognosis, and absence of MRD abrogates their effect (Döhner 2017).

MRD in CBF AML (ELN 2021). MRD is assessed in blood after 2 cycles of chemotherapy, in marrow at the end of consolidation and in blood every 4–6 weeks for 24 months. Failure of fusion transcripts to reach a 3- to 4-log reduction after consolidation, measured in the same tissue as at diagnosis, prompts individualized treatment or conditioning strategies, preferably in trials. A single positive MRD test is not the sole basis for clinical action, and stable low-level molecular MRD (MRD-LL) does not by itself require a treatment change (Heuser 2021).

AML with NPM1 mutation

  • Frequency. NPM1 mutations occur in about 30% of adult AML and in 50%–60% of AML with a normal karyotype (Forghieri 2020; Falini 2020; Gupta 2021). More than 55 mutations are described, mainly in exon 12, and they are stable at relapse (Forghieri 2020).
  • Morphology and phenotype. Blasts often show cup-like nuclei with prominent nuclear invaginations; in 83 NPM1-mutated normal-karyotype AML, cup-like blasts were present in 67% (Jakovic 2024). Blasts are often CD34-negative, and a monocytic component is common (Gupta 2021; Jakovic 2024). Blasts are HLA-DR-negative in many cases, which mimics APL (de Figueiredo-Pontes 2024). A fraction of 20%–28% express CD34 and/or HLA-DR on at least a subpopulation (Gupta 2021). Multilineage dysplasia does not change the diagnosis (Döhner 2017).
  • Cytoplasmic nucleophosmin. Immunohistochemistry detects aberrant cytoplasmic NPM1 in formalin-fixed tissue (Duncavage 2022; Falini 2020). It is useful in myeloid sarcoma, for mutations outside the usual hotspot and where molecular testing is unavailable (Duncavage 2022). In the original series, cytoplasmic NPM1 was found in 35.2% of 591 de novo AML trephines and in no secondary AML or other neoplasm (Forghieri 2020).
  • Blast requirement. WHO-HAEM5 cites rapid progression of NPM1-mutated MDS, MDS/MPN and clonal hematopoiesis to AML as the reason for accepting any increase in blasts (Khoury 2022; CAP 2026). Cases with <20% blasts are mostly high-risk MDS or chronic myelomonocytic leukemia with a normal karyotype and progress to AML, usually within 12 months (Forghieri 2020).
  • Prior therapy. WHO-HAEM5 requires no history of cytotoxic therapy and assigns NPM1-mutated AML after cytotoxic therapy to myeloid neoplasm post cytotoxic therapy, with the specific type stated in the report (Khoury 2022; CAP 2026).
  • Risk. In ELN 2022, mutated NPM1 without FLT3-ITD is favorable, NPM1 with FLT3-ITD is intermediate, and NPM1 with adverse-risk cytogenetics is adverse (Shimony 2025; El Chaer 2023). In ELN 2017, NPM1 without FLT3-ITD or with low allelic ratio (<0.5) FLT3-ITD was favorable and NPM1 with high allelic ratio FLT3-ITD was intermediate (Döhner 2017).
  • MRD (ELN 2021). Molecular MRD uses qPCR or digital PCR, preferably in blood after 2 cycles, in marrow at the end of consolidation and then in marrow every 3 months for 24 months. MRD positivity is ≥2% NPM1 mutant copies per ABL1 copies in marrow after consolidation. MRD-LL is <2% and is associated with a very low relapse risk at the end of consolidation (Heuser 2021). NPM1-negative relapse is reported, so other mutations are also tracked by NGS-MRD (Heuser 2021).

AML with CEBPA mutation

  • Definition. WHO-HAEM5 accepts biallelic CEBPA mutations or a single mutation in the basic leucine zipper (bZIP) region and requires ≥20% blasts. The ICC requires an in-frame bZIP mutation and ≥10% blasts. WHO 2017 required biallelic mutations (Xiao 2024; Khoury 2022; Arber 2022; El Chaer 2023; Döhner 2017).
  • Rationale. The favorable prognosis is determined by an in-frame bZIP mutation, with or without a second mutation, and was confirmed in children and adults up to 70 years (Khoury 2022; Arber 2022; Taube 2022; Tarlock 2021).
  • Risk. In ELN 2022 only in-frame bZIP mutations, monoallelic or biallelic, are favorable. ELN 2017 classified biallelic mutated CEBPA as favorable (Shimony 2025; El Chaer 2023; Döhner 2017). Patients with in-frame bZIP CEBPA mutation had a 5-year overall survival of 60% in a validation cohort (Rausch 2023).
  • Germline CEBPA. Germline CEBPA variants cause familial AML and are a germline predisposition category without platelet disorder or organ dysfunction in both systems and in WHO 2017 (Khoury 2022; Arber 2022; Döhner 2017), described in Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition.

AML with KMT2A rearrangement

  • Definition. WHO-HAEM5 replaced AML with t(9;11)(p22;q23); KMT2A-MLLT3 by AML with KMT2A rearrangement. More than 80 partners are described, most commonly MLLT3, AFDN, ELL and MLLT10 (Khoury 2022; Meyer 2023). Identifying the partner is desirable for prognosis and monitoring (Khoury 2022; Meyer 2023; CAP 2026). The ICC lists AML with t(9;11)(p21.3;q23.3)/MLLT3::KMT2A separately from other KMT2A rearrangements (Arber 2022).
  • Morphology. Adults usually have high blast counts with monocytic differentiation. In children, KMT2A::MLLT3 and KMT2A::MLLT10 AML can show megakaryoblastic differentiation and low aspirate blast counts (Khoury 2022; Umeda 2024). KMT2A rearrangements can be cryptic on karyotype (Khoury 2022).
  • Pediatric data. KMT2A rearrangements occur in 15%–20% of pediatric AML and are more frequent in infants (Quessada 2021; Bolouri 2018). They are associated with monocytic AML (FAB M4 or M5) in 73% (Quessada 2021).
  • Risk. In ELN 2022 and ELN 2017, t(9;11)/MLLT3::KMT2A is intermediate risk and other KMT2A rearrangements are adverse. ELN 2022 excludes KMT2A partial tandem duplication from the adverse rearranged group, and t(9;11) takes precedence over rare concurrent adverse-risk mutations (Shimony 2025). The intermediate/adverse split is unchanged from ELN 2017 (Döhner 2017).
  • Exclusions (CAP 2026). Blasts express a myeloid immunophenotype and do not meet mixed-phenotype acute leukemia criteria; post cytotoxic therapy cases are qualified as such.

Table 3. Selected KMT2A fusions in AML (Arber 2022; Quessada 2021; Khoury 2022; Meyer 2023). Percentages of all KMT2A-rearranged AML are from 1,116 cases (Meyer 2023).

Translocation Fusion Notes
t(9;11)(p21.3;q23.3) MLLT3::KMT2A Most common (30.4% of KMT2A-rearranged AML); 6%–9% of pediatric AML; ELN intermediate
t(10;11)(p12.3;q23.3) MLLT10::KMT2A 18.7% of KMT2A-rearranged AML; about 16% of pediatric KMT2A-rearranged AML; prevalent in infants; adverse in children
t(6;11)(q27;q23.3) AFDN::KMT2A 8.1% of KMT2A-rearranged AML; about 6% of pediatric KMT2A-rearranged AML; adverse in children
t(11;19)(q23.3;p13.1) KMT2A::ELL 10.1% of KMT2A-rearranged AML; among the 4 most common partners
t(11;19)(q23.3;p13.3) KMT2A::MLLT1 4.2% of KMT2A-rearranged AML; the ICC notes that it occurs predominantly in infants and children
t(4;11)(q21.3;q23.3) AFF1::KMT2A 0.8% of KMT2A-rearranged AML; 56.5% of KMT2A-rearranged acute lymphoblastic leukemia
t(10;11)(q21.3;q23.3) TET1::KMT2A Listed by the ICC

Other AML types with defining genetic abnormalities and rare fusions

Table 4. Other genetically defined AML (Khoury 2022; Arber 2022; Quessada 2021; Döhner 2017; Malcovati 2020; CAP 2026; Shimony 2025; Papadakis 2024; Sandahl 2014; Gröschel 2014; Hollink 2011; Masetti 2019; Umeda 2025).

Type Genetics Clinical and morphologic features ELN 2022 risk
AML with DEK::NUP214 t(6;9)(p22.3;q34.1) Maturing or myelomonocytic AML with multilineage dysplasia; marrow basophilia in 42%–62% and FLT3-ITD in about 75% (Papadakis 2024); FLT3-ITD in 42% of 62 pediatric cases (Sandahl 2014); 1%–2% of pediatric AML, no infant cases (Quessada 2021) Adverse
AML with MECOM rearrangement inv(3)(q21.3q26.2) or t(3;3)(q21.3;q26.2), repositioning a GATA2 enhancer to overexpress MECOM (EVI1) with GATA2 haploinsufficiency (Gröschel 2014; Döhner 2017); also t(3;21), t(3;12) and other 3q26.2 partners (Arber 2022) Dysmyelopoiesis and platelet abnormalities (Quessada 2021); thrombocytosis is allowed in MDS with inv(3) or t(3;3) (Arber 2022); SF3B1-mutated MDS with inv(3) or t(3;3) had a median overall survival of 27 versus 60 months without 3q26 abnormalities in 13 patients, a difference that was not statistically significant (Malcovati 2020); WHO-HAEM5 requires no history of myeloproliferative neoplasm (CAP 2026) Adverse
AML with NUP98 rearrangement Often cryptic; NUP98::NSD1 t(5;11)(q35;p15) in about 75% of pediatric NUP98-rearranged cases; NUP98::KDM5A t(11;12)(p15;p13) in megakaryoblastic AML NUP98::NSD1 in 8%–16% of pediatric normal-karyotype AML, with FLT3-ITD in about 80% and WT1 mutation in about 50%; high induction failure; detected by NUP98 break-apart FISH (Quessada 2021). In a screen of >1,000 cases, NUP98::NSD1 was found in 16.1% of pediatric and 2.3% of adult normal-karyotype AML, with 4-year event-free survival <10% (Hollink 2011) Not specifically listed (intermediate by default; ELN applies to adults)
AML with RBM15::MRTFA t(1;22)(p13.3;q13.1) (formerly RBM15::MKL1) Infants and toddlers (median 0.7 years); megakaryoblastic; hepatosplenomegaly and marrow fibrosis; FISH preferred when fibrosis limits aspiration (Quessada 2021) Not specifically listed (intermediate by default; ELN applies to adults)
AML with CBFA2T3::GLIS2 (WHO other defined genetic alteration) Cryptic inv(16)(p13.3q24.3) 27%–31% of non-Down syndrome pediatric megakaryoblastic AML; RAM immunophenotype (bright CD56, dim-to-negative CD45 and CD38, HLA-DR negative) (Masetti 2019); poor prognosis (Masetti 2019; Bolouri 2018) Not specifically listed (intermediate by default; ELN applies to adults)
AML with KAT6A::CREBBP (WHO other defined genetic alteration) t(8;16)(p11;p13) Peak in infants; disseminated intravascular coagulation (DIC); monocytic AML with erythrophagocytosis; spontaneous remission in some neonates Adverse (added in 2022)
Other WHO-HAEM5 defined alterations FUS::ERG, MNX1::ETV6, NPM1::MLF1 ≥20% blasts required; the ICC places these fusions, NUP98 rearrangements and RBM15::MRTFA among 12 rare recurring translocations with a ≥10% threshold (Umeda 2025; CAP 2026) Not specifically listed (intermediate by default; ELN applies to adults)

Previous rule. WHO 2017 named AML with inv(3)/t(3;3) GATA2, MECOM and AML (megakaryoblastic) with t(1;22) RBM15-MKL1 as recurrent genetic types with a ≥20% blast requirement (Döhner 2017). In ELN 2017, inv(3)/t(3;3) and t(6;9) were adverse, and t(8;16) and other 3q26.2 rearrangements were not listed (Döhner 2017; Rausch 2023).

AML with BCR::ABL1 and myeloid blast phase of chronic myeloid leukemia

  • Criteria. AML with BCR::ABL1 requires ≥20% myeloid blasts in blood or marrow in WHO-HAEM5 and the ICC, detection of BCR::ABL1 at diagnosis and no features of CML before, at or after diagnosis (Khoury 2022; Arber 2022; Xiao 2024; El Chaer 2023; CAP 2026). The 20% threshold avoids overlap with CML (Khoury 2022).
  • Distinction from CML blast phase. Deletion of immunoglobulin heavy chain and T-cell receptor genes, IKZF1 or CDKN2A may support AML (Döhner 2017). Transcript type and a baseline BCR::ABL1 level are recommended for monitoring (CAP 2026).
  • Therapy. Patients receive a tyrosine kinase inhibitor (Döhner 2017; Quessada 2021).
  • Risk. Adverse in ELN 2017 and ELN 2022 (Döhner 2017; Shimony 2025).
  • Previous rule. WHO 2017 listed AML with BCR-ABL1 as a provisional entity (Döhner 2017). CML phases are described in Myeloproliferative Neoplasms: Chronic Myeloid Leukemia with Molecular Monitoring, Polycythemia Vera, Essential Thrombocythemia, Primary Myelofibrosis and Other Myeloproliferative Neoplasms.

ELN 2022 genetic risk classification, with ELN 2017 as the previous rule

The European LeukemiaNet (ELN) 2022 genetic risk classification applies to patients treated with intensive chemotherapy (Shimony 2025; Döhner 2024a). Lesions not classified as favorable or adverse are intermediate by default (Shimony 2025).

Table 5. ELN 2022 genetic risk (Döhner 2022, as reproduced in Shimony 2025) beside ELN 2017 (Döhner 2017).

Risk ELN 2022 ELN 2017 (previous rule)
Favorable t(8;21)(q22;q22.1)/RUNX1::RUNX1T1; inv(16)(p13.1q22) or t(16;16)(p13.1;q22)/CBFB::MYH11; mutated NPM1 without FLT3-ITD; bZIP in-frame mutated CEBPA t(8;21)/RUNX1-RUNX1T1; inv(16) or t(16;16)/CBFB-MYH11; mutated NPM1 without FLT3-ITD or with low-ratio FLT3-ITD; biallelic mutated CEBPA
Intermediate FLT3-ITD irrespective of allelic ratio or NPM1 mutation; t(9;11)(p21.3;q23.3)/MLLT3::KMT2A; cytogenetic and/or molecular abnormalities not classified as favorable or adverse Mutated NPM1 with high-ratio FLT3-ITD; wild-type NPM1 without FLT3-ITD or with low-ratio FLT3-ITD (without adverse lesions); t(9;11)/MLLT3-KMT2A; cytogenetic abnormalities not classified as favorable or adverse
Adverse t(6;9)(p23;q34.1)/DEK::NUP214; t(v;11q23.3)/KMT2A-rearranged (excluding KMT2A-PTD); t(9;22)(q34.1;q11.2)/BCR::ABL1; t(8;16)(p11;p13)/KAT6A::CREBBP; inv(3)(q21.3q26.2) or t(3;3)(q21.3;q26.2)/GATA2, MECOM(EVI1); t(3q26.2;v)/MECOM(EVI1)-rearranged; −5 or del(5q), −7, −17/abn(17p); complex karyotype; monosomal karyotype; mutated ASXL1, BCOR, EZH2, RUNX1, SF3B1, SRSF2, STAG2, U2AF1 or ZRSR2; mutated TP53 (VAF ≥10%) t(6;9)/DEK-NUP214; t(v;11q23.3)/KMT2A rearranged; t(9;22)/BCR-ABL1; inv(3) or t(3;3)/GATA2, MECOM(EVI1); −5 or del(5q), −7, −17/abn(17p); complex karyotype; monosomal karyotype; wild-type NPM1 with high-ratio FLT3-ITD; mutated RUNX1; mutated ASXL1; mutated TP53

Footnotes to ELN 2022 (Shimony 2025). Complex karyotype is ≥3 unrelated chromosome abnormalities in the absence of class-defining recurring genetic abnormalities and excludes hyperdiploid karyotypes with ≥3 trisomies or polysomies without structural abnormalities. Monosomal karyotype is ≥2 distinct monosomies (excluding X or Y), or 1 autosomal monosomy with at least 1 structural abnormality (excluding CBF AML). The myelodysplasia-related gene mutations do not confer adverse risk when they occur with favorable-risk lesions.

ELN 2017 allelic ratio. A low ratio was <0.5 and a high ratio ≥0.5, measured by DNA fragment analysis as the area under the curve of FLT3-ITD divided by FLT3 wild type (Döhner 2017).

Changes from ELN 2017 to ELN 2022.

  • FLT3-ITD allelic ratio is no longer used. FLT3-ITD without CBF fusions or adverse-risk markers is intermediate, irrespective of NPM1 status, reflecting FLT3 inhibitor therapy and MRD-guided decisions (Rausch 2023; Shimony 2025).
  • KAT6A::CREBBP, other MECOM rearrangements and mutations in BCOR, EZH2, SF3B1, SRSF2, STAG2, U2AF1 or ZRSR2 were added to adverse risk (Rausch 2023).
  • Hyperdiploid karyotypes with only numerical changes no longer count as complex (Shimony 2025).
  • TP53 adverse status requires VAF ≥10% (Shimony 2025).

Validation. In 1,118 intensively treated patients, ELN 2022 reclassified 15% compared with ELN 2017: 3% to a more favorable group and 12% to a less favorable group, mainly by the added myelodysplasia-related mutations. Five-year relapse-free survival was 52%, 32% and 16% for favorable, intermediate and adverse ELN 2022 risk (Rausch 2023).

ELN 2024 classification for less-intensive therapy. The ELN 2017 and 2022 genetic risk classifications were developed in younger adults receiving intensive chemotherapy (Döhner 2024a). In 279 patients treated with venetoclax and azacitidine, ELN 2017 and ELN 2022 classified most patients as adverse risk (60.2% and 72.8%) and did not provide clinically meaningful stratification of overall survival. TP53, FLT3-ITD, NRAS and KRAS mutation status separated groups with median overall survival of 26.5, 12.1 and 5.5 months (Döhner 2024b). The 2024 ELN classification for adults receiving less-intensive therapy is shown in Table 6. It does not apply to patients with prior hypomethylating agent exposure (Döhner 2024a, as reproduced in Shimony 2025).

Table 6. ELN 2024 genetic risk for less-intensive therapy (Döhner 2024a, as reproduced in Shimony 2025). wt, wild type.

Risk Genetic abnormality
Favorable Mutated NPM1 (FLT3-ITD wt, NRAS wt, KRAS wt, TP53 wt); mutated IDH2 (FLT3-ITD wt, NRAS wt, KRAS wt, TP53 wt); mutated IDH1 (TP53 wt)ᵃ; mutated DDX41ᵇ; other cytogenetic and/or molecular abnormalitiesᶜ (FLT3-ITD wt, NRAS wt, KRAS wt, TP53 wt)
Intermediate Other cytogenetic and/or molecular abnormalitiesᶜ (FLT3-ITD positive and/or NRAS mutated and/or KRAS mutated, TP53 wt)
Adverse Mutated TP53

ᵃ Favorable risk applies specifically to patients treated with azacitidine plus ivosidenib, irrespective of activating signaling gene mutations. ᵇ A DDX41 mutation at near-heterozygous allele frequency prompts consideration of a germline DDX41 mutation. ᶜ For many abnormalities no outcome data exist; they are tentatively categorized as favorable or intermediate by the absence or presence of activating signaling gene mutations.

Targetable and risk markers.

  • FLT3. ITD and tyrosine kinase domain (TKD) mutations at codons D835 and I836 are both tested (Döhner 2017). FLT3-ITD and FLT3-TKD are often tested separately by PCR and capillary electrophoresis because of NGS limitations (Duncavage 2022).
  • IDH1 and IDH2. Mutations occur in 15%–20% of AML and produce R-2-hydroxyglutarate; mutant-specific inhibitors exist (Shimony 2025). Ivosidenib with azacitidine improved overall survival over azacitidine alone in the AGILE trial (El Chaer 2023). Rapid annotation of IDH1, IDH2, FLT3-ITD and FLT3-TKD guides targeted therapy (Duncavage 2022).
  • TP53. The ICC category AML with mutated TP53 and the WHO-HAEM5 handling of TP53 are described in Myelodysplastic Neoplasms: WHO 5th Edition and ICC Types, Dysplasia Morphology, TP53-Mutated Myeloid Neoplasms and Prognostic Scoring.

Response criteria. ELN 2022 retained the ELN 2017 definitions of complete remission (CR), CR with incomplete hematologic recovery (CRi), partial remission and morphologic leukemia-free state. It added CR with partial hematologic recovery (CRh): marrow blasts <5%, no blood blasts or extramedullary disease, ANC ≥0.5 × 10⁹/L and platelets ≥50 × 10⁹/L. Responses are annotated with MRD status, for example CR without MRD (Shimony 2025).

MRD framework. MRD methods, the ELN 2021 flow and NGS thresholds and the 2025 ELN-DAVID update are described in Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition. Molecular MRD is recommended for NPM1-mutated, CBF and APL cases (Heuser 2021).

Genetic workup at diagnosis and turnaround

  • Complete genomic evaluation. Cytogenetics, an NGS panel, FLT3-ITD testing and FISH or chromosomal microarray where needed identify the defining abnormalities and the ELN 2022 risk markers (Duncavage 2022; Arber 2017).
  • Rapid fusion testing. qPCR, FISH or NGS fusion panels for PML::RARA, RUNX1::RUNX1T1 and CBFB::MYH11 supplement the karyotype. FISH also detects KMT2A and MECOM rearrangements and loss of 5q, 7q or 17p when cytogenetics fails (Duncavage 2022; Döhner 2017; Quessada 2021).
  • Turnaround. Mutations needed for classification are reported within 5 days when possible (Duncavage 2022). ELN 2017 recommended NPM1 and FLT3 results within 48–72 hours and cytogenetics within 5–7 days. At least 20 marrow metaphases define a normal karyotype (Döhner 2017).
  • Immunohistochemistry. Hotspot IDH1 and IDH2 mutations and p53 accumulation or null expression can also be detected on formalin-fixed tissue (Duncavage 2022).
  • History. Prior therapy, antecedent myeloid neoplasm and family history are recorded, because they qualify the diagnosis and identify possible germline predisposition (Döhner 2017; Arber 2022).

CAP reporting elements. The CAP case summary (version 1.1.0.2, September 2026) records site and percent blasts or blast equivalents, marrow cellularity, the final integrated diagnosis by WHO-HAEM5 type, flow cytometry, conventional cytogenetics, FISH and molecular alterations detected, including PML::RARA, RUNX1::RUNX1T1, CBFB::MYH11, DEK::NUP214, KMT2A, MECOM and NUP98 rearrangements, NPM1, monoallelic CEBPA, FLT3-ITD and TKD, IDH1 and IDH2 (CAP 2026).

Definitions. Acute myeloid leukemia (AML), myelodysplasia-related (AML-MR) is the WHO 5th edition (WHO-HAEM5) type defined by myelodysplasia-related (MR) cytogenetics, MR gene mutations or a history of myelodysplastic neoplasm (MDS) or MDS/myeloproliferative neoplasm (MDS/MPN). AML defined by differentiation comprises cases with no defining genetic abnormality (Khoury 2022). The International Consensus Classification (ICC 2022) divides these cases into AML with mutated TP53, AML with MR gene mutations, AML with MR cytogenetic abnormalities and AML, not otherwise specified (NOS) (Arber 2022). Prior cytotoxic therapy, Down syndrome, extramedullary presentation, plasmacytoid dendritic cell differentiation and mixed or absent lineage commitment define the other entities.

Related page: Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition (flow cytometry of acute leukemia, lineage markers, germline predisposition) Related page: Myelodysplastic Neoplasms: WHO 5th Edition and ICC Types, Dysplasia Morphology, TP53-Mutated Myeloid Neoplasms and Prognostic Scoring (TP53 framework, MDS/AML and AML with mutated TP53) Related page: Acute Myeloid Leukemia with Defining Genetic Abnormalities, Acute Promyelocytic Leukemia and ELN 2022 Genetic Risk (AML with defining genetic abnormalities and ELN 2022 risk) Related page: Pattern Approach to Lymph Node Cytology (myeloid sarcoma in lymph node aspirates) Related page: Round Cell, Spindle Cell and Pleomorphic Sarcomas on FNA (myeloid sarcoma among round cell tumors on aspiration) Related page: Flow Cytometry, FISH, Microbiology and Electron Microscopy on Cytology Specimens (flow cytometry on fine-needle aspiration material) Related page: Lymphoblastic Lymphoma in Cytology (precursor lymphoid neoplasms on aspiration)

WHO-HAEM5 definition. AML-MR is a neoplasm with ≥20% blasts of myeloid immunophenotype carrying specific MDS-associated cytogenetic or molecular abnormalities, arising de novo or after a known history of MDS or MDS/MPN (Khoury 2022). Criteria for other AML types with defined genetic alterations and for myeloid neoplasm post cytotoxic therapy are not met (CAP 2026). Three changes were made: morphology alone no longer establishes the diagnosis, the defining cytogenetic list was updated, and 8 MR genes were added. A mutation in any of the 8 genes was >95% specific for AML arising after an antecedent myeloid neoplasm in a study of 194 patients with secondary or therapy-related AML and 105 unselected patients (Khoury 2022; Lindsley 2015). AML arising from MPN stays in the MPN category. AML arising from MDS or MDS/MPN is AML-MR (Khoury 2022).

ICC categories. The ICC eliminated AML with myelodysplasia-related changes (AML-MRC) and created AML with mutated TP53, AML with MR gene mutations and AML with MR cytogenetic abnormalities, each as MDS/AML with 10%–19% blasts or AML with ≥20% blasts (Arber 2022; Weinberg 2023b). Mutations in ASXL1, BCOR, EZH2, RUNX1, SF3B1, SRSF2, STAG2, U2AF1 or ZRSR2 define the MR gene category, which includes cases of the WHO 2017 provisional entity AML with mutated RUNX1. MR cytogenetic abnormalities define a case only when no MR gene mutation, TP53 mutation or other defining lesion is present. History of MDS or MDS/MPN becomes a qualifier ("progressed from MDS") (Arber 2022; Weinberg 2023b).

Rationale. Multilineage dysplasia was an imprecise surrogate for MR cytogenetics. It also occurs in AML with low-risk lesions such as NPM1 or biallelic CEBPA mutation, and it overlapped with therapy-related AML (Arber 2022). Morphologic dysplasia was removed from both systems because of poor interobserver reproducibility (Xiao 2024).

Prognostic basis of the MR gene category. MR gene mutations confer a similarly adverse prognosis in cases presenting as de novo AML that previously fell into AML, NOS (Arber 2022). In therapy-related and elderly de novo AML, these mutations defined a subgroup with a lower complete remission rate, more frequent reinduction and shorter event-free survival (Lindsley 2015).

Table 1. Myelodysplasia-related AML by WHO-HAEM5 (Khoury 2022), ICC 2022 (Arber 2022; hyperdiploid exclusion from Xiao 2024) and WHO 2017 (Döhner 2017).

Criterion WHO-HAEM5 AML-MR ICC 2022 WHO 2017 AML-MRC (previous rule)
Blasts ≥20% ≥20% (AML) or 10%–19% (MDS/AML) ≥20%
Morphology Not a criterion Not a criterion Multilineage dysplasia sufficient
History of MDS or MDS/MPN Defining criterion Diagnostic qualifier Defining criterion
MR gene mutations ASXL1, BCOR, EZH2, SF3B1, SRSF2, STAG2, U2AF1, ZRSR2 Same 8 genes plus RUNX1 Not used
Complex karyotype ≥3 abnormalities ≥3 unrelated clonal abnormalities without other class-defining lesions; excludes hyperdiploid karyotypes with ≥3 trisomies or polysomies without structural change ≥3 abnormalities without a WHO recurring translocation or inversion
Chromosome 5 5q deletion or loss of 5q by unbalanced translocation del(5q), t(5q) or add(5q) −5 or del(5q)
Chromosome 7 Monosomy 7, 7q deletion or loss of 7q by unbalanced translocation −7 or del(7q) −7 or del(7q)
Other unbalanced abnormalities del(11q); 12p deletion or loss; monosomy 13 or 13q deletion; 17p deletion or loss; i(17q); idic(X)(q13) +8; del(12p), t(12p) or add(12p); i(17q); −17, add(17p) or del(17p); del(20q); idic(X)(q13) del(11q); del(12p) or t(12p); −13 or del(13q); i(17q) or t(17p); idic(X)(q13)
Balanced abnormalities None listed None listed t(11;16)(q23.3;p13.3), t(3;21)(q26.2;q22.1), t(1;3)(p36.3;q21.2), t(2;11)(p21;q23.3), t(5;12)(q32;p13.2), t(5;7)(q32;q11.2), t(5;17)(q32;p13.2), t(5;10)(q32;q21.2), t(3;5)(q25.3;q35.1)
TP53 No separate type; most TP53-mutated AML has a complex karyotype and is AML-MR AML with mutated TP53 (VAF ≥10%) takes precedence Not used

Differences in the cytogenetic lists. WHO-HAEM5 includes del(11q) and −13/del(13q). The ICC includes +8 and del(20q) (Khoury 2022; Arber 2022; Xiao 2024).

Precedence. In WHO-HAEM5, acute erythroid leukemia supersedes AML-MR (Khoury 2022; CAP 2026). In the ICC, single-gene and fusion categories take precedence over the MR gene and MR cytogenetic categories (Arber 2022). AML with mutated TP53 and the TP53 framework are described in Myelodysplastic Neoplasms: WHO 5th Edition and ICC Types, Dysplasia Morphology, TP53-Mutated Myeloid Neoplasms and Prognostic Scoring.

Risk. In a validation cohort, patients moved to adverse ELN 2022 risk by MR gene mutations had better relapse-free and overall survival than other adverse-risk patients (Rausch 2023). ELN risk is described in Acute Myeloid Leukemia with Defining Genetic Abnormalities, Acute Promyelocytic Leukemia and ELN 2022 Genetic Risk.

AML defined by differentiation (WHO-HAEM5) and AML, NOS (ICC)

Shared criteria (WHO-HAEM5). Each type requires ≥20% blasts in marrow and/or blood, except acute erythroid leukemia. Criteria for AML with defined genetic alterations, mixed-phenotype acute leukemia (relevant to AML with minimal differentiation) and myeloid neoplasm post cytotoxic therapy are not met, and there is no prior MPN (Khoury 2022; CAP 2026).

ICC. Cases without any defining genetic category are AML, NOS (or MDS/AML, NOS with 10%–19% blasts). Morphologic and cytochemical subtypes have limited prognostic value, and subclassification is optional (Arber 2022). WHO-HAEM5 discarded the term AML, NOS (Khoury 2022).

Table 2. AML defined by differentiation (Khoury 2022; CAP 2026). MPO, myeloperoxidase; SBB, Sudan black B; NSE, nonspecific esterase.

Type Diagnostic criteria
AML with minimal differentiation Blasts negative (<3%) for MPO and SBB by cytochemistry; ≥2 myeloid antigens such as CD13, CD33 and CD117
AML without maturation ≥3% blasts positive for MPO (immunophenotype or cytochemistry) or SBB and negative for NSE; maturing granulocytic cells <10% of nucleated marrow cells; ≥2 myeloid antigens
AML with maturation ≥3% blasts positive for MPO or SBB; maturing granulocytic cells ≥10% of nucleated marrow cells; monocytic cells <20% of marrow cells; ≥2 myeloid antigens
Acute basophilic leukemia Blasts and immature or mature basophils with metachromasia on toluidine blue; blasts negative for MPO, SBB and NSE; no strong CD117 (excludes mast cell leukemia); desirable: blasts positive for CD123, CD11b, CD9 and/or CD203c and negative for HLA-DR
Acute myelomonocytic leukemia ≥20% monocytes and precursors; ≥20% maturing granulocytic cells; ≥3% of blasts MPO-positive
Acute monocytic leukemia ≥80% monocytes and/or precursors (monoblasts, promonocytes); <20% maturing granulocytic cells; blasts and promonocytes express ≥2 of CD11c, CD14, CD36 and CD64, or are NSE-positive
Acute erythroid leukemia Erythroid predominance, usually ≥80% of marrow elements, with ≥30% proerythroblasts; no 20% blast requirement
Acute megakaryoblastic leukemia ≥20% blasts expressing ≥1 platelet glycoprotein: CD41 (glycoprotein IIb), CD61 (glycoprotein IIIa) or CD42b (glycoprotein Ib)

Lineage markers and MPO thresholds. Blast-equivalent rules, lineage-associated markers and MPO thresholds are described in Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition. ELN 2017 lists CD65 and cytoplasmic MPO as granulocytic markers. AML with minimal differentiation is MPO-negative with myeloid markers, which separates it from acute undifferentiated leukemia (Döhner 2017; Fuda 2025).

Previous rule (WHO 2017). AML, NOS included the same differentiation subtypes plus pure erythroid leukemia and acute panmyelosis with myelofibrosis (Döhner 2017). Acute panmyelosis with myelofibrosis is not among the WHO-HAEM5 types defined by differentiation (Khoury 2022).

Acute erythroid leukemia

  • Definition (WHO-HAEM5). Acute erythroid leukemia (AEL), previously pure erythroid leukemia (an acceptable term), meets the Table 2 criteria. Cases with <80% erythroid cells but the same clinicopathologic features are recognized. Biallelic TP53 alteration is highly prevalent, and AEL supersedes AML-MR (Khoury 2022; CAP 2026). Evidence of TP53 mutation is a desirable criterion (CAP 2026).
  • ICC. Pure erythroid leukemia is classified within AML with mutated TP53 (Arber 2022; Xiao 2024).
  • Previous rule (WHO 2017). The erythroid/myeloid subtype (≥50% marrow erythroid precursors with ≥20% myeloblasts among nonerythroid cells) was removed, and myeloblasts were counted as a percentage of all marrow cells. Pure erythroid leukemia required >80% immature erythroid precursors with ≥30% proerythroblasts (Döhner 2017).
  • Mayo series of 41 cases (Reichard 2022). Cases were de novo (14), therapy-related (14) or secondary to a prior myeloid neoplasm (12). All had biallelic TP53 alterations and a complex karyotype, and 90% had a monosomal karyotype. Aspirates were often suboptimal from dry tap, hemodilution or broken cells (56%). Dysplastic megakaryocytes were present in 80%. All patients died, at a median of 1.8 months.
  • Immunophenotype (Reichard 2022). By immunohistochemistry, pronormoblasts were CD71-positive (96%) and E-cadherin-positive (88%). Glycophorin A and hemoglobin stained only small subsets, CD117 was positive in 68%, CD34 in 16% and MPO in none. Keratins AE1/AE3 and OSCAR were negative.

AEL after cytotoxic therapy. In 25 TP53-mutated cases, de novo and post-therapy AEL had comparable clinical findings and overall survival, and the TP53 mutation was in the dominant clone in all cases. After cytotoxic therapy, a single TP53 mutation (89% versus 62.5%) and monosomy 17 or del(17p) (71.4% versus 26.6%) were more frequent, and hotspot mutations were less frequent (22.2% versus 56.2%) (Tashakori 2022).

Acute megakaryoblastic leukemia

  • Definition. Acute megakaryoblastic leukemia (AMKL) has ≥20% blasts with megakaryocytic differentiation, expressing CD41, CD61 or CD42b. Criteria for other defined AML types are not met, there is no prior MPN, and Down syndrome is evaluated (CAP 2026). Immunophenotyping is required for the diagnosis (Khoury 2022).
  • Clinical groups. AMKL arises in children with Down syndrome, children without Down syndrome and adults (Khoury 2022).
  • Morphology. Megakaryoblasts often resemble lymphoblasts, and AMKL is associated with extensive marrow fibrosis (Kosmidou 2023; Bhatnagar 2016). Fibrosis can impair cytogenetic and PCR analysis, and FISH is used for primary lesions in infant AMKL (Quessada 2021).
  • Flow cytometry (EuroFlow, 72 cases). Blasts expressed CD42a/CD61 in 92%, CD36 in 92%, CD42b in 81%, CD41 in 72% and CD7 in 70%, and generally lacked CD13 and HLA-DR. A CD13−/CD34+/HLA-DR−/CD42a.CD61+ profile had 40% sensitivity and 100% specificity for AMKL (Brouwer 2022).
  • Non-Down syndrome pediatric AMKL. It accounts for about 10% of pediatric AML, mainly in infants (median about 1.5 years). Cryptic inv(16)(p13.3q24.3)/CBFA2T3::GLIS2 defines an aggressive subtype (Gruber 2012). NUP98::KDM5A, KMT2A rearrangements and monosomy 7 also predict poor outcome. t(1;22)/RBM15::MRTFA, hyperdiploidy and a normal karyotype had better outcomes (Quessada 2021). Sequencing divides pediatric non-Down syndrome AMKL into subgroups defined by these fusions, with cooperating mutations in epigenetic and kinase signaling genes and varying outcomes (de Rooij 2017).
  • Classification of genetic subsets. RBM15::MRTFA AML is a defining genetic type, and CBFA2T3::GLIS2 AML is an AML with other defined genetic alterations in WHO-HAEM5 (Khoury 2022). Both, and the RAM immunophenotype of CBFA2T3::GLIS2 AML, are described in Acute Myeloid Leukemia with Defining Genetic Abnormalities, Acute Promyelocytic Leukemia and ELN 2022 Genetic Risk.

Myeloid proliferations associated with Down syndrome

Pathogenesis. Transient abnormal myelopoiesis (TAM) and myeloid leukemia of Down syndrome (ML-DS) arise in children with constitutional trisomy 21 and are typically associated with a somatic exon 2 or 3 GATA1 mutation (Khoury 2022). Most mutations are insertions, deletions or duplications in exon 2 that produce a short isoform (GATA1s) lacking the amino-terminal activation domain (Kosmidou 2023; Wechsler 2002). About 97% are in exon 2 and the remainder in exon 3, and the mutation type does not predict progression (Bhatnagar 2016). GATA1 mutations of this type occur only with trisomy 21, constitutional or acquired (Kosmidou 2023). Progression from TAM to ML-DS requires additional mutations, in cohesin complex genes (STAG2 and others), JAK kinases, RAS pathway genes and epigenetic regulators (EZH2, KANSL1 and others) (Kosmidou 2023; CAP 2026; Bhatnagar 2016).

Table 3. TAM and ML-DS (Khoury 2022; CAP 2026; Kosmidou 2023; Grimm 2021; Bhatnagar 2016).

Feature Transient abnormal myelopoiesis Myeloid leukemia of Down syndrome
Age Neonatal; confined to the first 6 months of life Develops within the first 4 years of life in 20%–30% of children with TAM
Frequency 10%–30% of neonates with Down syndrome: about 10%–15% with clinical TAM and blasts >10%, and a further 10%–15% with clinically silent GATA1-mutant clones Incidence 46- to 83-fold, or about 150-fold, higher in children with Down syndrome than in other children, by study
Essential criteria (WHO-HAEM5) Constitutional trisomy 21; blood leukocytosis with increased blasts; exon 2/3 GATA1 mutation Constitutional trisomy 21; persistent increased blood and/or marrow blasts (may be <20%); exon 2/3 GATA1 mutation
Testing rule GATA1 exon 2/3 sequencing in all cases with blood blasts >10% Mutation profiling for cohesin, EZH2, KANSL1 and JAK3 mutations is desirable
Clinical features Hepatomegaly or hepatopathy, pleural and pericardial effusions, skin rash, splenomegaly (30%), hyperleukocytosis, coagulopathy; some cases asymptomatic Megakaryoblastic leukemia with myelofibrosis; preceded by TAM
Course Spontaneous resolution within 1–2 months in most cases (Kosmidou 2023) and within 3 months in more than 80% (Bhatnagar 2016) 5-year overall survival 89%–93%

Blood film timing. A blood film is examined in the first week of life, because circulating blast counts fall rapidly afterward. In the Oxford Imperial Down Syndrome Cohort Study of 200 neonates, >10% blood blasts in the first week together with clinical features identified all neonates with TAM (Kosmidou 2023).

Silent TAM and the blast threshold. In the same cohort, 195 of 200 neonates (97.5%) had circulating blasts. Sanger sequencing detected GATA1 mutations in 17 (8.5%), all with blasts >10%, and next-generation sequencing detected low-abundance GATA1-mutant clones in 18 of 88 (20.4%) neonates without a Sanger-detectable mutation. These neonates had no distinguishing clinical or hematologic features and were termed "silent TAM" (Roberts 2013). About 25% of neonates with blasts >10% have no GATA1 mutation, so the diagnosis of TAM requires both increased blasts and the mutation (Bhatnagar 2016).

Natural history of TAM and ML-DS. Overall mortality of TAM is about 20%, half of it directly attributable to TAM, usually hepatic failure from fibrosis and blast infiltration. Retrospective studies estimate progression to ML-DS at 20%–30% of TAM; counting silent TAM, the risk is estimated at 5%–10%. ML-DS presents at a median age of 1–1.8 years and is rare after 4 years. It often follows months of an indolent myelodysplastic phase with thrombocytopenia and a low leukocyte count, and marrow fibrosis often causes a dry tap, so a trephine biopsy may be required. The blasts express CD34, CD117, CD33, CD41, CD42b, CD36, glycophorin A and CD7. t(8;21), t(15;17), t(9;11), inv(16) and the AMKL-associated t(1;22) do not occur in ML-DS (Bhatnagar 2016).

Classification. WHO-HAEM5 keeps TAM and ML-DS as the 2 clonal conditions of Down syndrome (Khoury 2022). The ICC left myeloid neoplasms associated with Down syndrome unchanged and lists Down syndrome among germline predisposition disorders (Arber 2022). WHO 2017 listed myeloid proliferations related to Down syndrome (TAM and ML-DS) as a separate group (Döhner 2017).

Myeloid neoplasms post cytotoxic therapy

WHO-HAEM5 definition. Myeloid neoplasm post cytotoxic therapy (MN-pCT) is any MDS, MDS/MPN or AML in a patient with documented prior cytotoxic (DNA-damaging) chemotherapy or large-field radiation therapy for an unrelated neoplasm. MPN after cytotoxic therapy is excluded (CAP 2026). Clonal cytopenia of undetermined significance is excluded because it lacks dysplasia. Cases with a de novo molecular signature, such as NPM1 mutation or a core-binding factor fusion, are still MN-pCT, because the designation is based on history. The specific type is stated, for example "AML with KMT2A rearrangement post cytotoxic therapy" (Khoury 2022).

Changes in WHO-HAEM5. "Post cytotoxic therapy" replaces "therapy-related". Exposure to PARP1 inhibitors was added as a qualifying criterion, and methotrexate was excluded (Khoury 2022). Prior therapy does not necessarily imply causation, which motivated the change of term (Xiao 2024).

ICC. "Therapy-related" is a qualifier after the MDS, MDS/AML or AML diagnosis and covers prior chemotherapy, radiotherapy and immune interventions. Therapy-related CHIP and CCUS are also qualified. Classification by morphology and genetics comes first (Arber 2022; Weinberg 2023b).

Previous rule (WHO 2017). Therapy-related myeloid neoplasms were a stand-alone category, and cases were reported with the related genetic abnormality (Döhner 2017; Arber 2022).

Table 4. Clinicopathologic types of therapy-related myeloid neoplasms (Fianchi 2018; Döhner 2017).

Feature Alkylating agents or radiation Topoisomerase II inhibitors
Median latency 4–7 years (Fianchi 2018); 5–7 years (Döhner 2017) 2–3 years (Fianchi 2018); 1–3 years (Döhner 2017)
Presentation Dysplasia of 1 or more lineages, often an MDS phase before overt AML Overt AML; antecedent MDS rare
Genetics −5/del(5q) and −7/del(7q); complex karyotype; frequent TP53 mutation Balanced translocations of 11q23 (KMT2A), 17q21 (including PML::RARA) and/or 21q22 (RUNX1)
Share of therapy-related AML About 75% Minority

Genetics and outcome. TP53 mutations are enriched in myeloid neoplasms after cytotoxic therapy (13%–48%, average 31%, versus 2%–18%, average 8%, de novo) (Shah 2025). TP53 was the most frequently mutated gene (33.3%) in 111 therapy-related AML and MDS, and DNMT3A and NPM1 mutations were less common than in de novo AML (Wong 2015). Outcomes are worse with multi-hit TP53, and PPM1D and DNA damage response gene mutations may prompt germline evaluation (Khoury 2022). Preexisting TP53-mutated clones expand during PARP inhibitor therapy (Shah 2025). Therapy-related neoplasms account for 10%–20% of myeloid neoplasms, and 5-year overall survival is <10% (Fianchi 2018). The prognosis is worse than that of matched de novo neoplasms (CAP 2026). Recurrent balanced translocations carry a better prognosis, as in de novo disease (Fianchi 2018). APL after chemotherapy for another cancer still has a good prognosis (Shimony 2025).

Clonal selection and germline predisposition. Chemotherapy is not thought to induce TP53 mutations directly. In 4 of 7 therapy-related cases tested, the diagnostic TP53 mutation was present at low frequency in blood or marrow banked 3–8 years earlier, in 2 cases before any chemotherapy. Low-frequency TP53 mutations were also found in 9 of 19 cancer-free elderly donors (Wong 2015). TP53-mutant progenitors resistant to chemotherapy are selected and expand after treatment (Wong 2015; Khoury 2022). In breast cancer survivors with therapy-related AML, 10 of 41 (21%) carried germline BRCA1, BRCA2, TP53 or CHEK2 mutations (Döhner 2017). The distinction between the alkylator and topoisomerase II types in Table 4 has become less evident with multiagent chemotherapy combined with radiotherapy (Döhner 2017).

PARP inhibitors. In a meta-analysis of 28 randomized trials, PARP inhibitors were associated with a 2.6-fold higher risk of therapy-related myeloid neoplasm (0.73% versus 0.47%), after a median exposure of 9.8 months and a median latency of 20.3 months. TP53 mutations were present in 50%–75% of myeloid neoplasms after PARP inhibitors (Shah 2025).

Myeloid sarcoma

Definition. Myeloid sarcoma is a tumor mass at a site other than the bone marrow that effaces tissue architecture and is composed of myeloid blasts or blast equivalents, with or without maturation. It may precede, accompany or follow marrow leukemia. Non-effacing, non-tumor-forming extramedullary blastic proliferations are not myeloid sarcoma. The report specifies the associated myeloid neoplasm, for example myeloid sarcoma with AML with RUNX1::RUNX1T1 fusion (CAP 2026). Myeloid sarcoma is a tissue manifestation of AML or of transformed MDS, MDS/MPN or MPN. De novo cases are investigated with cytogenetic and molecular studies for classification (Khoury 2022). The ICC lists myeloid sarcoma unchanged (Arber 2022).

Clinical features. Isolated myeloid sarcoma with a normal marrow is usually followed by AML. Myeloid sarcoma can represent blast transformation of MPN, MDS or MDS/MPN and relapse after allogeneic transplantation, where 5%–12% of patients develop it and about 70% of cases are isolated. Common sites are skin (leukemia cutis), lymph nodes, genitals, breast, gastrointestinal tract, peritoneum, bone and central nervous system (Loscocco 2023).

Frequency relative to marrow AML. At AML diagnosis, synchronous myeloid sarcoma is reported in 0.2%–2.8% of patients and isolated myeloid sarcoma in 0.6%–0.8%. In a US registry of 94,185 AML cases, 0.8% had myeloid sarcoma, and skin or breast accounted for 12.3% of presentation sites (Loscocco 2023).

Morphology and differential diagnosis. Blasts may show granulocytic maturation, in which eosinophilic precursors support the diagnosis, and many cases show myelomonocytic or monoblastic morphology (Loscocco 2023). In 92 adult cases, 50% were of blastic type and 43.5% monoblastic or myelomonocytic (Pileri 2007). Misdiagnoses include Hodgkin and other lymphomas, Ewing sarcoma, round blue cell tumors, poorly differentiated carcinoma and blastic plasmacytoid dendritic cell neoplasm (Loscocco 2023).

Table 5. Immunohistochemistry of myeloid sarcoma (ranges from Loscocco 2023, which include the series of Pileri 2007).

Antibody Positive cases (%)
CD68 (KP1) 90–100
CD33 55–94
MPO 50–88
CD117 55–80
CD68 (PG-M1) 51–53
Lysozyme 23–92
CD34 27–44
CD43 9–100

In 92 adult cases, CD99 (54.3%), TdT (31.5%) and CD56 (13%) were also expressed, and CD4 was rare (1.1%) (Pileri 2007).

Genetics. Clonal cytogenetic abnormalities occur in 54%–70%. RUNX1::RUNX1T1 is found in only 2%–3% in larger series (Loscocco 2023; Pileri 2007). NPM1 mutations occur in about 20%–50% and RAS pathway mutations in 30%–85% (Loscocco 2023).

In the 92-case series, monosomy 7 (10.8%), trisomy 8 (10.4%) and KMT2A (MLL) splitting (8.5%) were the commonest abnormalities. Outcome was poor regardless of presentation, age, sex, phenotype or cytogenetics, and most long-term survivors had received bone marrow transplantation (Pileri 2007).

Mutations in myeloid sarcoma and concurrent marrow disease are concordant in about 70% of patients (Khoury 2022). Cytoplasmic NPM1 immunohistochemistry is useful in myeloid sarcoma (Duncavage 2022). Cytologic diagnosis on aspirates is described in Pattern Approach to Lymph Node Cytology and Round Cell, Spindle Cell and Pleomorphic Sarcomas on FNA.

Blastic plasmacytoid dendritic cell neoplasm

Definition and placement. Blastic plasmacytoid dendritic cell neoplasm (BPDCN) consists of immature cells with plasmacytoid dendritic cell (pDC) differentiation, predominantly of myeloid origin. A preceding or associated myeloid neoplasm, typically MDS or MDS/MPN, is present in 20%–30% of cases (CAP 2026). WHO-HAEM5 places BPDCN with the histiocytic and dendritic cell neoplasms under plasmacytoid dendritic cell neoplasms, beside mature plasmacytoid dendritic cell proliferation associated with myeloid neoplasm (Khoury 2022). The ICC lists BPDCN unchanged (Arber 2022). WHO 2017 listed BPDCN as a separate category after AML (Döhner 2017).

Clinical features. BPDCN mainly affects older males (male-to-female ratio 3.3:1) in the seventh or eighth decade, with a median age of 68 years, and can occur in children (Sapienza 2019; Mehra 2024). The skin is involved in 60%–100% and is the first site in about 90%, with erythematous-violaceous plaques or nodules. Marrow and blood are involved in 60%–90% and lymph nodes in 40%–50% (Sapienza 2019).

Morphology. A diffuse, monomorphous dermal infiltrate of medium-sized blasts spares the epidermis. Nuclei are slightly irregular with fine chromatin and small nucleoli, and the narrow cytoplasm is gray-blue and agranular on Giemsa stain. Ki-67 ranges from 20% to 80% (Sapienza 2019).

Table 6. Immunophenotypic criteria for BPDCN (Khoury 2022; CAP 2026; Sapienza 2019; Garnache-Ottou 2019).

Category Markers
Expected positive pDC markers CD123, TCF4, TCL1, CD303, CD304
Expected positive, shared with other cell types CD4, CD56
Expected negative CD3, CD14, CD19, CD34, lysozyme, MPO
Diagnostic rule (WHO-HAEM5) CD123 plus 1 other pDC marker plus CD4 and/or CD56; or any 3 pDC markers with all expected negative markers absent

About 8% of cases lack CD4 or CD56 (Sapienza 2019). TCF4 is a master regulator of the BPDCN program and a reliable diagnostic marker (Sapienza 2019; Sukswai 2019).

Dual TCF4/CD123 immunohistochemistry. A dual-color TCF4/CD123 stain on paraffin sections showed coexpression in all 48 BPDCN and in 1 of 464 non-BPDCN cases (a B-lymphoblastic leukemia/lymphoma), giving 100% sensitivity and 99.8% specificity (Sukswai 2019). In a French series of 86 patients, BPDCN expressed HLA-DR (high), CD303, CD304 and cytoplasmic TCL1 with CD4 and CD56, often with isolated myeloid, B- or T-lineage markers, and lacked MPO, CD14, cytoplasmic CD3, CD19 and cytoplasmic CD22 (Garnache-Ottou 2019).

Differential diagnosis of BPDCN.

  • Cutaneous myeloid sarcoma (leukemia cutis). Lysozyme and MPO are expected negative in BPDCN (Khoury 2022) and are expressed in a large proportion of myeloid sarcomas (Loscocco 2023). Gene expression and array comparative genomic hybridization separate BPDCN from cutaneous myelomonocytic leukemia (Sapienza 2019).
  • AML with pDC expansion. Clonally expanded pDCs in AML share the mutations of the CD34-positive blasts and are frequently associated with RUNX1 mutation (Khoury 2022).
  • Mature plasmacytoid dendritic cell proliferation. Clonal pDCs with low-grade morphology accumulate in the marrow in myeloproliferative chronic myelomonocytic leukemia with RAS pathway mutations (Khoury 2022). Plasmacytoid dendritic cell nodules in CMML are described in Mastocytosis, Hypereosinophilic Syndrome and Myeloid/Lymphoid Neoplasms with Tyrosine Kinase Fusions, Chronic Myelomonocytic Leukemia and Other MDS/MPN.
  • Former names of BPDCN. Obsolete names for BPDCN (blastic NK-cell lymphoma, CD4+/CD56+ hematodermic neoplasm) reflect CD56 expression and resemblance to lymphoblasts (Sapienza 2019).

Genetics and treatment. Deletions of chromosomes 9, 13 and 12 are frequent and cause loss of CDKN1B, CDKN2A, CDKN2B, RB1 and IKZF1 (Sapienza 2019). In 86 patients, 51% of cytogenetic abnormalities involved chromosomes 13, 12, 9 and 15 (Garnache-Ottou 2019). Most patients die within 1 year with conventional chemotherapy (Sapienza 2019); median overall survival was 8.7 months in 43 patients with leukemic presentation (Pagano 2013). Tagraxofusp, a CD123-targeted therapy, was evaluated in the largest prospective BPDCN trial in treatment-naive and relapsed or refractory disease (Pemmaraju 2022). Allogeneic transplantation in first complete remission is pursued when feasible (Sapienza 2019; Garnache-Ottou 2019).

Outcomes of tagraxofusp and transplantation. Tagraxofusp is interleukin-3 fused to a truncated diphtheria toxin and is approved by the US Food and Drug Administration for BPDCN in patients aged ≥2 years. In 65 treatment-naive patients, complete response plus clinical complete response was 57%, and median overall survival was 15.8 months (Pemmaraju 2022). Median survival was 49 months for patients in persistent complete remission after allogeneic transplantation, versus 8 months for other patients (Garnache-Ottou 2019), and 22.7 versus 7.1 months for transplanted versus non-transplanted patients with leukemic presentation (Pagano 2013).

Acute leukemias of ambiguous lineage and mixed-phenotype acute leukemia

Definition. Acute leukemias of ambiguous lineage (ALAL) have ≥20% blasts in blood or marrow and either show differentiation along more than one lineage (mixed-phenotype acute leukemia, MPAL) or no lineage differentiation (acute undifferentiated leukemia, AUL). MPAL may be a single blast population expressing lineage-defining antigens of 2 or more lineages, distinct populations of different lineages, or both. In bilineal cases, the 2 abnormal populations are added to reach 20% (CAP 2026; Fuda 2025). MPAL accounts for 1%–3% of adult acute leukemias (George 2022), and acute leukemias of ambiguous lineage account for <4% of all acute leukemias (Fuda 2025).

Exclusions. A single-lineage leukemia with aberrant antigens that are not lineage-defining is acute leukemia with aberrant antigen expression. Leukemias that meet MPAL criteria but belong to a defined entity are assigned to that entity, for example AML with RUNX1::RUNX1T1, myeloid/lymphoid neoplasms with eosinophilia and tyrosine kinase fusions, and blast phase of chronic myeloid leukemia (CAP 2026; Fuda 2025).

Therapy-related AML and AML-MR that meet MPAL lineage criteria are also classified in their specific category, and the report notes the mixed-lineage immunophenotype. A leukemia meeting immunophenotypic criteria for acute undifferentiated leukemia but carrying MR cytogenetic abnormalities or mutations is classified as AML (Fuda 2025).

Table 7. MPAL lineage assignment criteria by WHO-HAEM5 (Khoury 2022; CAP 2026; Fuda 2025), ICC 2023 (Weinberg 2023a, as summarized in Fuda 2025) and WHO 2017 (George 2022).

Lineage WHO-HAEM5 ICC 2023 WHO 2017 (previous rule)
Myeloid MPO with intensity in part exceeding 50% of the mature neutrophil level; or monocytic differentiation with ≥2 of NSE, CD11c, CD14, CD64 and lysozyme (CAP lists CD36 as well) MPO, with no specified intensity threshold; or monocytic differentiation with ≥2 of NSE, CD11c, CD14, CD64 and lysozyme MPO by flow cytometry, immunohistochemistry or cytochemistry; or monocytic differentiation with ≥2 of NSE, CD11c, CD14, CD64 and lysozyme
T Cytoplasmic or surface CD3 (CD3 epsilon antibody), intensity in part exceeding 50% of mature T cells by flow cytometry; cytoplasmic CD3 by immunohistochemistry with a non-zeta antibody Cytoplasmic CD3 with at least part of the blasts at the intensity of background mature T cells (surface CD3 also listed); no immunohistochemistry guidance Strong cytoplasmic CD3 (equal to or brighter than normal T cells in the sample) or surface CD3
B CD19 strong (in part >50% of normal B-cell progenitor intensity) with ≥1 of CD10, CD22 or CD79a strongly expressed; or CD19 weak with ≥2 strongly expressed; B lineage may be assigned without CD19 when PAX5, CD79a and CD22 are present Strong CD19 (at least the intensity of stage 1 B-cell progenitors, or of mature B cells when progenitors are absent) with ≥1 of CD10, CD22 or CD79a; weak CD19 with ≥2 of these strongly expressed; PAX5, OCT2 and BOB1 listed for immunohistochemistry without rules Strong CD19 with ≥1 strongly expressed of CD79a, cytoplasmic CD22 or CD10; or weak CD19 with ≥2 of these
CD79a with T lineage Not used for B lineage when T lineage criteria are met No exclusion stated No exclusion stated
Bilineal MPAL The 2 abnormal populations are added to reach 20% blasts; no minimum size for either population The smaller abnormal population must be ≥5% of cells; below that, the leukemia is classified by the main population with a descriptive modifier Not stated

Lineage assignment principles (intensity and pattern resembling the normal counterpart) are described in Classification of Myeloid Neoplasms (WHO 5th Edition and ICC 2022), Flow Cytometry of Myeloid Neoplasms, Clonal Hematopoiesis, VEXAS Syndrome and Germline Predisposition.

ICC. Arber 2022 deferred acute leukemias of ambiguous lineage to a working group, which published the ICC criteria in 2023 (Weinberg 2023a, as summarized in Fuda 2025). The ICC lists MPAL with BCR::ABL1, MPAL with t(v;11q23.3)/KMT2A rearrangement, MPAL with BCL11B activation and MPAL with ZNF384 rearrangement, then B/myeloid, T/myeloid, B/T/myeloid and B/T MPAL, ALAL, NOS and acute undifferentiated leukemia. It does not recognize T/megakaryocytic MPAL (Fuda 2025).

Table 8. WHO-HAEM5 types of acute leukemia of ambiguous lineage (Khoury 2022; CAP 2026; Fuda 2025; Alexander 2018 for ZNF384; Montefiori 2021 for BCL11B).

Family Type Key criteria
Defining genetic abnormalities MPAL with BCR::ABL1 fusion MPAL immunophenotype; BCR::ABL1 at diagnosis; no prior or subsequent CML; no prior cytotoxic therapy; transcript type and baseline level desirable
Defining genetic abnormalities MPAL with KMT2A rearrangement MPAL immunophenotype; KMT2A rearrangement; no prior cytotoxic therapy
Other defined genetic alterations MPAL with ZNF384 rearrangement Usually B/myeloid; about 50% of pediatric B/myeloid MPAL; partners TCF3, EP300, TAF15, CREBBP; in T/myeloid MPAL, biallelic WT1 alterations are common
Other defined genetic alterations ALAL with BCL11B rearrangement Found in AUL and 20%–30% (Khoury 2022) to about one third (Fuda 2025) of T/myeloid MPAL; also in AML with minimal differentiation and 20%–30% of early T-precursor lymphoblastic leukemia; enhancer hijacking deregulates BCL11B in leukemias expressing myeloid, T-lymphoid and stem cell markers
Immunophenotypically defined MPAL, B/myeloid B and myeloid lineage antigens; no defining genetic alteration; no prior cytotoxic therapy
Immunophenotypically defined MPAL, T/myeloid T and myeloid lineage antigens; no defining genetic alteration; no prior cytotoxic therapy
Immunophenotypically defined MPAL, rare types B/T, B/T/myeloid or T/megakaryocytic; no prior myeloid neoplasm
Immunophenotypically defined ALAL, NOS Lineage markers that do not permit definitive assignment
Immunophenotypically defined Acute undifferentiated leukemia No lineage-defining antigens

Previous rule (WHO 2017). Acute leukemias of ambiguous lineage comprised AUL, MPAL with t(9;22)(q34.1;q11.2)/BCR-ABL1, MPAL with t(v;11q23.3)/KMT2A rearranged, MPAL B/myeloid NOS and MPAL T/myeloid NOS (Döhner 2017).

Distribution, genetics and treatment. Immunophenotypic MPAL is B/myeloid in 59%, T/myeloid in 35%, T/B in 4% and T/B/myeloid in 2% (Matutes 2011; George 2022). PHF6 mutations and PICALM::MLLT10 fusions are enriched in MPAL (Khoury 2022). About 42% of B-lymphoblastic leukemias that express MPO carry t(9;22) (George 2022). ALL-type induction followed by allogeneic transplantation in first remission is the usual approach, and a tyrosine kinase inhibitor is added for BCR::ABL1-positive MPAL (George 2022).

MPAL series of 100 patients (WHO 2008 criteria). t(9;22) was present in 20% and 11q23/KMT2A rearrangement in 8%, and morphology resembled ALL in 43% and AML in 42%. ALL-type therapy produced responses in 85% and AML-type therapy in 41%. Age, Ph positivity and AML-type therapy predicted poor outcome, and median overall survival was 18 months (Matutes 2011).

T/megakaryocytic MPAL. About 20 cases are reported. Features overlap with pediatric non-Down syndrome AMKL: infant onset, male predominance, CBFA2T3::GLIS2 (rarely CBFA2T3::GLIS3) and an extremely poor prognosis. Cytoplasmic CD3 separates it from AMKL (Fuda 2025).

Early T-precursor lymphoblastic leukemia. Early T-precursor lymphoblastic leukemia/lymphoma (ETP-ALL) shares BCL11B rearrangements and a biologic continuum with T/myeloid MPAL (Khoury 2022; George 2022; Alexander 2018). ETP-ALL is classified with the precursor lymphoid neoplasms.

CAP reporting elements for AML-MR, post-cytotoxic therapy neoplasms, myeloid sarcoma, BPDCN and ambiguous-lineage leukemias

The CAP case summary (version 1.1.0.2, September 2026) lists AML-MR and the AML types defined by differentiation, myeloid sarcoma with the associated neoplasm specified, myeloid neoplasm post cytotoxic therapy with the neoplasm and therapy specified, TAM and ML-DS, the acute leukemias of mixed or ambiguous lineage and BPDCN. Dysplasia is reported as not applicable for myeloid sarcoma and BPDCN. Sites include bone marrow, blood, lymph node, cutaneous and extranodal or mucosal sites, with percent blasts or blast equivalents (CAP 2026).

References

  1. Ahn JS, Park JH, Lee C, et al. Serial next-generation sequencing for detecting germline predisposition in acute myeloid leukemia. Haematologica. 2026;111(2):718–723. doi:10.3324/haematol.2025.287794. PMID 40605710
  2. Alexander TB, Gu Z, Iacobucci I, et al. The genetic basis and cell of origin of mixed phenotype acute leukaemia. Nature. 2018;562(7727):373–379. doi:10.1038/s41586-018-0436-0. PMID 30209392
  3. Alibrahim MN, Elsayed MA, Alqahtani AN, et al. Defining and managing high-risk acute myeloid leukemia (AML) in 2026. Ann Hematol. 2026;105(10):416. doi:10.1007/s00277-026-07142-6. PMID 42734831
  4. Aperna F, Abdelmagid MG, Kumar M, et al. Systemic mastocytosis in 910 patients: prognostic contribution of the International Consensus Classification in the context of the Mayo Alliance Prognostic System. Am J Hematol. 2025;100(9):1566–1576. doi:10.1002/ajh.27764. PMID 40607735
  5. Apperley JF, Milojkovic D, Cross NCP, et al. 2025 European LeukemiaNet recommendations for the management of chronic myeloid leukemia. Leukemia. 2025;39(8):1797–1813. doi:10.1038/s41375-025-02664-w. PMID 40646132
  6. Araújo HV, Correia RP, Bento LC, et al. Myelodysplastic syndrome: validation of flow cytometry multilineage score system. Einstein (Sao Paulo). 2020;18:eAO4966. doi:10.31744/einstein_journal/2020AO4966. PMID 31994605 (Araújo 2020)
  7. Arber DA, Borowitz MJ, Cessna M, et al. Initial diagnostic workup of acute leukemia: guideline from the College of American Pathologists and the American Society of Hematology. Arch Pathol Lab Med. 2017;141(10):1342–1393. doi:10.5858/arpa.2016-0504-CP. PMID 28225303 (abstract)
  8. Arber DA, Orazi A, Hasserjian RP, et al. International Consensus Classification of Myeloid Neoplasms and Acute Leukemias: integrating morphologic, clinical, and genomic data. Blood. 2022;140(11):1200–1228. doi:10.1182/blood.2022015850. PMID 35767897
  9. Arber DA, Orazi A. Classification of myelodysplastic, myeloproliferative, and myelodysplastic/myeloproliferative neoplasms: the past, present, and future. Am J Hematol. 2025;100 Suppl 4:5–15. doi:10.1002/ajh.27656. PMID 40056065 (Arber and Orazi 2025)
  10. Baliakas P, Tesi B, Cammenga J, et al. How to manage patients with germline DDX41 variants: recommendations from the Nordic working group on germline predisposition for myeloid neoplasms. Hemasphere. 2024;8(8):e145. doi:10.1002/hem3.145. PMID 39139355
  11. Barbui T, Thiele J, Alvarez-Larran A, et al. From essential thrombocythemia to polycythemia vera (ET-2-PV): rationale and design of a multicenter ELN study. Ann Hematol. 2026;105(4):158. doi:10.1007/s00277-026-06898-1. PMID 41776088
  12. Barbui T, Thiele J, Gisslinger H, et al. Masked polycythemia vera (mPV): results of an international study. Am J Hematol. 2014;89(1):52–54. doi:10.1002/ajh.23585. PMID 23996471
  13. Barbui T, Thiele J, Gisslinger H, et al. The 2016 WHO classification and diagnostic criteria for myeloproliferative neoplasms: document summary and in-depth discussion. Blood Cancer J. 2018;8(2):15. doi:10.1038/s41408-018-0054-y. PMID 29426921
  14. Beck DB, Ferrada MA, Sikora KA, et al. Somatic mutations in UBA1 and severe adult-onset autoinflammatory disease. N Engl J Med. 2020;383(27):2628–2638. doi:10.1056/NEJMoa2026834. PMID 33108101
  15. Bennett JM, Orazi A. Diagnostic criteria to distinguish hypocellular acute myeloid leukemia from hypocellular myelodysplastic syndromes and aplastic anemia: recommendations for a standardized approach. Haematologica. 2009;94(2):264–268. doi:10.3324/haematol.13755. PMID 19144661
  16. Bernard E, Nannya Y, Hasserjian RP, et al. Implications of TP53 allelic state for genome stability, clinical presentation and outcomes in myelodysplastic syndromes. Nat Med. 2020;26(10):1549–1556. doi:10.1038/s41591-020-1008-z. PMID 32747829 (abstract)
  17. Bernard E, Tuechler H, Greenberg PL, et al. Molecular International Prognostic Scoring System for Myelodysplastic Syndromes. NEJM Evid. 2022;1(7):EVIDoa2200008. doi:10.1056/EVIDoa2200008. PMID 38319256 (abstract)
  18. Bhatnagar N, Nizery L, Tunstall O, et al. Transient abnormal myelopoiesis and AML in Down syndrome: an update. Curr Hematol Malig Rep. 2016;11(5):333–341. doi:10.1007/s11899-016-0338-x. PMID 27510823
  19. Boggs NA, Sun X, Lyons JJ, et al. Challenges in applying diagnostic criteria for systemic mastocytosis. Blood Adv. 2023;7(13):3150–3154. doi:10.1182/bloodadvances.2023009826. PMID 36848636
  20. Bolouri H, Farrar JE, Triche T Jr, et al. The molecular landscape of pediatric acute myeloid leukemia reveals recurrent structural alterations and age-specific mutational interactions. Nat Med. 2018;24(1):103–112. doi:10.1038/nm.4439. PMID 29227476
  21. Brouwer N, Matarraz S, Nierkens S, et al. Immunophenotypic analysis of acute megakaryoblastic leukemia: a EuroFlow study. Cancers (Basel). 2022;14(6):1583. doi:10.3390/cancers14061583. PMID 35326734
  22. Brunner AM, Leitch HA, van de Loosdrecht AA, et al. Management of patients with lower-risk myelodysplastic syndromes. Blood Cancer J. 2022;12(12):166. doi:10.1038/s41408-022-00765-8. PMID 36517487
  23. Calabretto G, Attardi E, Teramo A, et al. Hypocellular myelodysplastic syndromes (h-MDS): from clinical description to immunological characterization in the Italian multi-center experience. Leukemia. 2022;36(7):1947–1950. doi:10.1038/s41375-022-01592-3. PMID 35597805
  24. Cha CH, Park CJ, Chi HS, et al. CD34 and p53 immunohistochemical stains differentiate hypocellular myelodysplastic syndrome (hMDS) from aplastic anemia and a CD34 immunohistochemical stain provides useful survival information for hMDS. Ann Lab Med. 2014;34(6):426–432. doi:10.3343/alm.2014.34.6.426. PMID 25368817
  25. Cloos J, Valk PJM, Thiede C, et al. 2025 update on MRD in acute myeloid leukemia: a consensus document from the ELN-DAVID MRD Working Party. Blood. 2026;147(11):1147–1167. doi:10.1182/blood.2025031480. PMID 41397238
  26. College of American Pathologists. Protocol for the Examination of Myeloid and Mixed / Ambiguous Lineage Neoplasms, version 1.1.0.2. September 2026. https://www.cap.org/protocols-and-guidelines/cancer-reporting-tools/cancer-protocol-templates (accessed 2026-11-17) (CAP 2026)
  27. College of American Pathologists. Protocol for the examination of specimens from patients with myeloid and mixed/ambiguous lineage neoplasms (Heme.Myeloid_Mixed_Amb.Bx.Res, version 1.1.0.2). September 2026. https://www.cap.org/protocols-and-guidelines/cancer-reporting-tools/cancer-protocol-templates (accessed 2026-11-17) (CAP 2026)
  28. Cross NCP, Ernst T, Branford S, et al. European LeukemiaNet laboratory recommendations for the diagnosis and management of chronic myeloid leukemia. Leukemia. 2023;37(11):2150–2167. doi:10.1038/s41375-023-02048-y. PMID 37794101
  29. D'Angelo G. Copper deficiency mimicking myelodysplastic syndrome. Blood Res. 2016;51(4):217–219. doi:10.5045/br.2016.51.4.217. PMID 28090480
  30. de Figueiredo-Pontes LL, Catto LFB, Chauffaille MLLF, et al. Diagnosis and management of acute promyelocytic leukemia: Brazilian consensus guidelines 2024 on behalf of the Brazilian Association of Hematology, Hemotherapy and Cellular Therapy. Hematol Transfus Cell Ther. 2024;46(4):553–569. doi:10.1016/j.htct.2024.05.002. PMID 38890097
  31. de Rooij JD, Branstetter C, Ma J, et al. Pediatric non-Down syndrome acute megakaryoblastic leukemia is characterized by distinct genomic subsets with varying outcomes. Nat Genet. 2017;49(3):451–456. doi:10.1038/ng.3772. PMID 28112737
  32. Della Porta MG, Travaglino E, Boveri E, et al. Minimal morphological criteria for defining bone marrow dysplasia: a basis for clinical implementation of WHO classification of myelodysplastic syndromes. Leukemia. 2015;29(1):66–75. doi:10.1038/leu.2014.161. PMID 24935723 (abstract)
  33. Döhner H, DiNardo CD, Appelbaum FR, et al. Genetic risk classification for adults with AML receiving less-intensive therapies: the 2024 ELN recommendations. Blood. 2024;144(21):2169–2173. doi:10.1182/blood.2024025409. PMID 39133932 (abstract; table as reproduced in Shimony 2025) (Döhner 2024a)
  34. Döhner H, Estey E, Grimwade D, et al. Diagnosis and management of AML in adults: 2017 ELN recommendations from an international expert panel. Blood. 2017;129(4):424–447. doi:10.1182/blood-2016-08-733196. PMID 27895058
  35. Döhner H, Pratz KW, DiNardo CD, et al. Genetic risk stratification and outcomes among treatment-naive patients with AML treated with venetoclax and azacitidine. Blood. 2024;144(21):2211–2222. doi:10.1182/blood.2024024944. PMID 39133921 (Döhner 2024b)
  36. Döhner H, Wei AH, Appelbaum FR, et al. Diagnosis and management of AML in adults: 2022 recommendations from an international expert panel on behalf of the ELN. Blood. 2022;140(12):1345–1377. doi:10.1182/blood.2022016867. PMID 35797463 (abstract; risk table as reproduced in Shimony 2025)
  37. Drexler B, Schwarz-Furlan S, Baumann I, et al. Long-term outcomes of patients with refractory cytopenia of childhood under observation only. Blood Adv. 2025;9(16):4279–4285. doi:10.1182/bloodadvances.2025016136. PMID 40554414
  38. Duncavage EJ, Bagg A, Hasserjian RP, et al. Genomic profiling for clinical decision making in myeloid neoplasms and acute leukemia. Blood. 2022;140(21):2228–2247. doi:10.1182/blood.2022015853. PMID 36130297
  39. El Achkar H, Bazarbachi A, Chakhachiro Z, et al. High-risk cytogenetic abnormalities impact the cytological and clinical behavior of core binding factor acute myeloid leukemia: a case-control study. BMC Cancer. 2026;26(1):979. doi:10.1186/s12885-026-16281-y. PMID 42252425
  40. El Chaer F, Hourigan CS, Zeidan AM. How I treat AML incorporating the updated classifications and guidelines. Blood. 2023;141(23):2813–2823. doi:10.1182/blood.2022017808. PMID 36758209
  41. Elbaz Younes I, Nguyen L, Zhang L. Moving beyond somatic alterations: uncovering the germline basis of myeloid malignancies. Cancers (Basel). 2026;18(2):240. doi:10.3390/cancers18020240. PMID 41595160
  42. Elena C, Gallì A, Such E, et al. Integrating clinical features and genetic lesions in the risk assessment of patients with chronic myelomonocytic leukemia. Blood. 2016;128(10):1408–1417. doi:10.1182/blood-2016-05-714030. PMID 27385790
  43. Eletrebi MM, Sanchez-Ramirez J, Abdalla A, et al. Hypereosinophilia of persistent/unknown cause: a bone marrow-based study integrating WHO-HAEM5 criteria and myeloid next-generation sequencing. Am J Clin Pathol. 2026;165(5). doi:10.1093/ajcp/aqag031. PMID 42106910
  44. Erdogan Yucel E, Bulbul Dilmen H, Ozer Kaya O, et al. Is unidentified cytopenia truly unidentified? Or are we overlooking clonality? Medicina (Kaunas). 2026;62(5):868. doi:10.3390/medicina62050868. PMID 42195121
  45. Falini B, Brunetti L, Sportoletti P, Martelli MP. NPM1-mutated acute myeloid leukemia: from bench to bedside. Blood. 2020;136(15):1707–1721. doi:10.1182/blood.2019004226. PMID 32609823 (abstract)
  46. Fattizzo B, Serpenti F, Barcellini W, et al. Hypoplastic myelodysplastic syndromes: just an overlap syndrome? Cancers (Basel). 2021;13(1):132. doi:10.3390/cancers13010132. PMID 33401595
  47. Fianchi L, Criscuolo M, Fabiani E, et al. Therapy-related myeloid neoplasms: clinical perspectives. Onco Targets Ther. 2018;11:5909–5915. doi:10.2147/OTT.S101333. PMID 30271175
  48. Forghieri F, Nasillo V, Paolini A, et al. NPM1-mutated myeloid neoplasms with <20% blasts: a really distinct clinico-pathologic entity? Int J Mol Sci. 2020;21(23):8975. doi:10.3390/ijms21238975. PMID 33255988
  49. Fuda F, Chen W. Acute leukemia of ambiguous lineage: diagnosis and evaluation by flow cytometry. Cancers (Basel). 2025;17(5):871. doi:10.3390/cancers17050871. PMID 40075717
  50. Gangat N, Bellam N, Reichard K, et al. Emergence of TP53 mutation during lenalidomide therapy of myelodysplastic syndrome with del(5q) and its subsequent disappearance following salvage therapy with decitabine. Haematologica. 2024;109(4):1306–1309. doi:10.3324/haematol.2023.284547. PMID 38031760 (case report)
  51. Garnache-Ottou F, Vidal C, Biichlé S, et al. How should we diagnose and treat blastic plasmacytoid dendritic cell neoplasm patients? Blood Adv. 2019;3(24):4238–4251. doi:10.1182/bloodadvances.2019000647. PMID 31869411
  52. George BS, Yohannan B, Gonzalez A, et al. Mixed-phenotype acute leukemia: clinical diagnosis and therapeutic strategies. Biomedicines. 2022;10(8):1974. doi:10.3390/biomedicines10081974. PMID 36009521
  53. Ghosh K, Shome DK, Kulkarni B, et al. Fibrosis and bone marrow: understanding causation and pathobiology. J Transl Med. 2023;21(1):703. doi:10.1186/s12967-023-04393-z. PMID 37814319
  54. Gianelli U, Thiele J, Orazi A, et al. International Consensus Classification of myeloid and lymphoid neoplasms: myeloproliferative neoplasms. Virchows Arch. 2023;482(1):53–68. doi:10.1007/s00428-022-03480-8. PMID 36580136
  55. Giannella A, Vianello F, Zoletto S, et al. From variant detection to interpretation in idiopathic erythrocytosis: a structured approach applied to a clinical cohort. Br J Haematol. 2026;209(2):837–841. doi:10.1111/bjh.70597. PMID 42285891
  56. Greenberg PL, Tuechler H, Schanz J, et al. Revised international prognostic scoring system for myelodysplastic syndromes. Blood. 2012;120(12):2454–2465. doi:10.1182/blood-2012-03-420489. PMID 22740453
  57. Grimm J, Heckl D, Klusmann JH. Molecular mechanisms of the genetic predisposition to acute megakaryoblastic leukemia in infants with Down syndrome. Front Oncol. 2021;11:636633. doi:10.3389/fonc.2021.636633. PMID 33777792
  58. Gröschel S, Sanders MA, Hoogenboezem R, et al. A single oncogenic enhancer rearrangement causes concomitant EVI1 and GATA2 deregulation in leukemia. Cell. 2014;157(2):369–381. doi:10.1016/j.cell.2014.02.019. PMID 24703711 (abstract)
  59. Gruber TA, Larson Gedman A, Zhang J, et al. An Inv(16)(p13.3q24.3)-encoded CBFA2T3-GLIS2 fusion protein defines an aggressive subtype of pediatric acute megakaryoblastic leukemia. Cancer Cell. 2012;22(5):683–697. doi:10.1016/j.ccr.2012.10.007. PMID 23153540
  60. Guglielmelli P, Lasho TL, Rotunno G, et al. MIPSS70: Mutation-Enhanced International Prognostic Score System for transplantation-age patients with primary myelofibrosis. J Clin Oncol. 2018;36(4):310–318. doi:10.1200/JCO.2017.76.4886. PMID 29226763
  61. Gupta M, Jafari K, Rajab A, et al. Radar plots facilitate differential diagnosis of acute promyelocytic leukemia and NPM1+ acute myeloid leukemia by flow cytometry. Cytometry B Clin Cytom. 2021;100(4):409–420. doi:10.1002/cyto.b.21979. PMID 33301193
  62. Haferlach T, Nagata Y, Grossmann V, et al. Landscape of genetic lesions in 944 patients with myelodysplastic syndromes. Leukemia. 2014;28(2):241–247. doi:10.1038/leu.2013.336. PMID 24220272
  63. Hasserjian RP, Orazi A, Orfao A, et al. The International Consensus Classification of myelodysplastic syndromes and related entities. Virchows Arch. 2023;482(1):39–51. doi:10.1007/s00428-022-03417-1. PMID 36287260
  64. Hazarika B, Bain BJ. MDS/MPN with SF3B1 mutation and thrombocytosis but without ring sideroblasts. Am J Hematol. 2026;101(1):129–130. doi:10.1002/ajh.70040. PMID 40844204
  65. Heuser M, Freeman SD, Ossenkoppele GJ, et al. 2021 update on MRD in acute myeloid leukemia: a consensus document from the European LeukemiaNet MRD Working Party. Blood. 2021;138(26):2753–2767. doi:10.1182/blood.2021013626. PMID 34724563
  66. Hochhaus A, Baccarani M, Silver RT, et al. European LeukemiaNet 2020 recommendations for treating chronic myeloid leukemia. Leukemia. 2020;34(4):966–984. doi:10.1038/s41375-020-0776-2. PMID 32127639
  67. Holden R, Jeelal Y, McLean-Tooke A, et al. VEXAS: a review of current understandings and emerging treatment strategies. Front Immunol. 2025;16:1644404. doi:10.3389/fimmu.2025.1644404. PMID 40791602
  68. Hollink IH, van den Heuvel-Eibrink MM, Arentsen-Peters ST, et al. NUP98/NSD1 characterizes a novel poor prognostic group in acute myeloid leukemia with a distinct HOX gene expression pattern. Blood. 2011;118(13):3645–3656. doi:10.1182/blood-2011-04-346643. PMID 21813447 (abstract)
  69. Hourigan CS, Beugelink L, Othman J, et al. ELN-DAVID recommendations for NGS-based FLT3-ITD MRD testing for patients with acute myeloid leukemia. Blood. 2026;148(7):910–913. doi:10.1182/blood.2026033569. PMID 42237660
  70. Hwang SM. Genomic testing for germline predisposition to hematologic malignancies. Blood Res. 2024;59(1):12. doi:10.1007/s44313-024-00012-y. PMID 38485837
  71. Invernizzi R, Quaglia F, Porta MG. Importance of classical morphology in the diagnosis of myelodysplastic syndrome. Mediterr J Hematol Infect Dis. 2015;7(1):e2015035. doi:10.4084/MJHID.2015.035. PMID 25960863
  72. Iurlo A, Cattaneo D, Gianelli U, et al. Atypical chronic myeloid leukemia: from diagnosis to molecular features and therapeutic options. Hemasphere. 2025;9(12):e70270. doi:10.1002/hem3.70270. PMID 41378161
  73. Jain AG, Zhang L, Bennett JM, et al. Myelodysplastic syndromes with bone marrow fibrosis: an update. Ann Lab Med. 2022;42(3):299–305. doi:10.3343/alm.2022.42.3.299. PMID 34907099
  74. Jakovic L, Djordjevic V, Kraguljac Kurtovic N, et al. Early prediction and streamline of nucleophosmin mutation status in acute myeloid leukemia using cup-like nuclear morphology. Medicina (Kaunas). 2024;60(9):1443. doi:10.3390/medicina60091443. PMID 39336484
  75. Jung J, Kwag D, Kim Y, et al. Perspectives on acute myeloid leukemia diagnosis: a comparative analysis of the latest World Health Organization and the International Consensus Classifications. Leukemia. 2023;37(10):2125–2128. doi:10.1038/s41375-023-01996-9. PMID 37580414
  76. Khan A, Anwar M, Azam A, et al. Hypoplastic myelodysplastic syndrome: symptom of methotrexate toxicity in rheumatoid arthritis. Cureus. 2023;15(6):e40580. doi:10.7759/cureus.40580. PMID 37469807 (case report)
  77. Khoury JD, Solary E, Abla O, et al. The 5th edition of the World Health Organization Classification of Haematolymphoid Tumours: myeloid and histiocytic/dendritic neoplasms. Leukemia. 2022;36(7):1703–1719. doi:10.1038/s41375-022-01613-1. PMID 35732831
  78. Klampfl T, Gisslinger H, Harutyunyan AS, et al. Somatic mutations of calreticulin in myeloproliferative neoplasms. N Engl J Med. 2013;369(25):2379–2390. doi:10.1056/NEJMoa1311347. PMID 24325356
  79. Konstantinov KN, Konstantinov NK, Kumar V. VEXAS syndrome beyond UBA1: genetic architecture and the role of co-occurring somatic mutations: a focused review. Genes (Basel). 2026;17(6):711. doi:10.3390/genes17060711. PMID 42353870
  80. Kosmidou A, Tragiannidis A, Gavriilaki E. Myeloid leukemia of Down syndrome. Cancers (Basel). 2023;15(13):3265. doi:10.3390/cancers15133265. PMID 37444375
  81. Lee WH, Lin CC, Tsai XC, et al. Comparison of the 2022 World Health Organization classification and International Consensus Classification in myelodysplastic syndromes/neoplasms. Blood Cancer J. 2024;14(1):57. doi:10.1038/s41408-024-01031-9. PMID 38594285
  82. Leguit RJ, Wang SA, George TI, et al. The international consensus classification of mastocytosis and related entities. Virchows Arch. 2023;482(1):99–112. doi:10.1007/s00428-022-03423-3. PMID 36214901
  83. Li JY, Ryder CB, Zhang H, et al. Review and updates on systemic mastocytosis and related entities. Cancers (Basel). 2023;15(23):5626. doi:10.3390/cancers15235626. PMID 38067330
  84. Lindsley RC, Mar BG, Mazzola E, et al. Acute myeloid leukemia ontogeny is defined by distinct somatic mutations. Blood. 2015;125(9):1367–1376. doi:10.1182/blood-2014-11-610543. PMID 25550361
  85. Loghavi S, Hasserjian RP. Chronic myelomonocytic leukemia: history, pathobiology, diagnostic controversies, and evolving classification. Curr Hematol Malig Rep. 2026;21(1). doi:10.1007/s11899-026-00781-6. PMID 42545651
  86. Loscocco GG, Vannucchi AM. Myeloid sarcoma: more and less than a distinct entity. Ann Hematol. 2023;102(8):1973–1984. doi:10.1007/s00277-023-05288-1. PMID 37286874
  87. Lübke J, Metzgeroth G, Reiter A, et al. Approach to the patient with eosinophilia in the era of tyrosine kinase inhibitors and biologicals. Curr Hematol Malig Rep. 2024;19(5):208–222. doi:10.1007/s11899-024-00738-7. PMID 39037514
  88. Malcovati L, Karimi M, Papaemmanuil E, et al. SF3B1 mutation identifies a distinct subset of myelodysplastic syndrome with ring sideroblasts. Blood. 2015;126(2):233–241. doi:10.1182/blood-2015-03-633537. PMID 25957392
  89. Malcovati L, Stevenson K, Papaemmanuil E, et al. SF3B1-mutant MDS as a distinct disease subtype: a proposal from the International Working Group for the Prognosis of MDS. Blood. 2020;136(2):157–170. doi:10.1182/blood.2020004850. PMID 32347921
  90. Masetti R, Bertuccio SN, Pession A, et al. CBFA2T3-GLIS2-positive acute myeloid leukaemia. A peculiar paediatric entity. Br J Haematol. 2019;184(3):337–347. doi:10.1111/bjh.15725. PMID 30592296
  91. Matutes E, Pickl WF, Van't Veer M, et al. Mixed-phenotype acute leukemia: clinical and laboratory features and outcome in 100 patients defined according to the WHO 2008 classification. Blood. 2011;117(11):3163–3171. doi:10.1182/blood-2010-10-314682. PMID 21228332
  92. Maxson JE, Gotlib J, Pollyea DA, et al. Oncogenic CSF3R mutations in chronic neutrophilic leukemia and atypical CML. N Engl J Med. 2013;368(19):1781–1790. doi:10.1056/NEJMoa1214514. PMID 23656643
  93. McMullin MF. Genetic background of congenital erythrocytosis. Genes (Basel). 2021;12(8):1151. doi:10.3390/genes12081151. PMID 34440325
  94. Mehra S, Taylor J. Blastic plasmacytoid dendritic cell neoplasm: a comprehensive review of the disease, central nervous system presentations, and treatment strategies. Cells. 2024;13(3):243. doi:10.3390/cells13030243. PMID 38334635
  95. Meyer C, Larghero P, Almeida Lopes B, et al. The KMT2A recombinome of acute leukemias in 2023. Leukemia. 2023;37(5):988–1005. doi:10.1038/s41375-023-01877-1. PMID 37019990
  96. Monika F, Abu Mehsen S, Zhang L. Molecular advances in juvenile myelomonocytic leukemia and associated RASopathy. Cancers (Basel). 2026;18(10):1655. doi:10.3390/cancers18101655. PMID 42193013
  97. Montefiori LE, Bendig S, Gu Z, et al. Enhancer hijacking drives oncogenic BCL11B expression in lineage-ambiguous stem cell leukemia. Cancer Discov. 2021;11(11):2846–2867. doi:10.1158/2159-8290.CD-21-0145. PMID 34103329
  98. Montesinos P, Sanz MA. The differentiation syndrome in patients with acute promyelocytic leukemia: experience of the PETHEMA group and review of the literature. Mediterr J Hematol Infect Dis. 2011;3(1):e2011059. doi:10.4084/MJHID.2011.059. PMID 22220256
  99. Morales-Camacho RM, Caballero-Velázquez T, Borrero JJ, et al. Hematological neoplasms with eosinophilia. Cancers (Basel). 2024;16(2):337. doi:10.3390/cancers16020337. PMID 38254826
  100. Mufti GJ, Bennett JM, Goasguen J, et al. Diagnosis and classification of myelodysplastic syndrome: International Working Group on Morphology of myelodysplastic syndrome (IWGM-MDS) consensus proposals for the definition and enumeration of myeloblasts and ring sideroblasts. Haematologica. 2008;93(11):1712–1717. doi:10.3324/haematol.13405. PMID 18838480 (abstract)
  101. Nachmias B, Haran A, Yisraeli Salman M, et al. Measurable residual disease-guided therapy in acute myeloid leukaemia: practical insights. Br J Haematol. 2025;207(4):1192–1212. doi:10.1111/bjh.70038. PMID 40887043
  102. Nangalia J, Massie CE, Baxter EJ, et al. Somatic CALR mutations in myeloproliferative neoplasms with nonmutated JAK2. N Engl J Med. 2013;369(25):2391–2405. doi:10.1056/NEJMoa1312542. PMID 24325359
  103. Pagano L, Valentini CG, Pulsoni A, et al. Blastic plasmacytoid dendritic cell neoplasm with leukemic presentation: an Italian multicenter study. Haematologica. 2013;98(2):239–246. doi:10.3324/haematol.2012.072645. PMID 23065521
  104. Papadakis S, Liapis I, Papadhimitriou SI, et al. Approach to acute myeloid leukemia with increased eosinophils and basophils. J Clin Med. 2024;13(3):876. doi:10.3390/jcm13030876. PMID 38337573
  105. Papaemmanuil E, Cazzola M, Boultwood J, et al. Somatic SF3B1 mutation in myelodysplasia with ring sideroblasts. N Engl J Med. 2011;365(15):1384–1395. doi:10.1056/NEJMoa1103283. PMID 21995386
  106. Papaemmanuil E, Gerstung M, Malcovati L, et al. Clinical and biological implications of driver mutations in myelodysplastic syndromes. Blood. 2013;122(22):3616–3627. doi:10.1182/blood-2013-08-518886. PMID 24030381
  107. Pardanani A, Lasho TL, Laborde RR, et al. CSF3R T618I is a highly prevalent and specific mutation in chronic neutrophilic leukemia. Leukemia. 2013;27(9):1870–1873. doi:10.1038/leu.2013.122. PMID 23604229
  108. Passamonti F, Cervantes F, Vannucchi AM, et al. A dynamic prognostic model to predict survival in primary myelofibrosis: a study by the IWG-MRT (International Working Group for Myeloproliferative Neoplasms Research and Treatment). Blood. 2010;115(9):1703–1708. doi:10.1182/blood-2009-09-245837. PMID 20008785
  109. Patel BA, Calvo KR, Reichard KK, et al. Myelodysplasia in VEXAS syndrome. Blood Adv. 2026;10(17):5951–5961. doi:10.1182/bloodadvances.2025018827. PMID 42284697
  110. Patnaik MM, Tefferi A. Chronic myelomonocytic leukemia: 2024 update on diagnosis, risk stratification and management. Am J Hematol. 2024;99(6):1142–1165. doi:10.1002/ajh.27271. PMID 38450850
  111. Pemmaraju N, Sweet KL, Stein AS, et al. Long-term benefits of tagraxofusp for patients with blastic plasmacytoid dendritic cell neoplasm. J Clin Oncol. 2022;40(26):3032–3036. doi:10.1200/JCO.22.00034. PMID 35820082
  112. Pfirrmann M, Baccarani M, Saussele S, et al. Prognosis of long-term survival considering disease-specific death in patients with chronic myeloid leukemia. Leukemia. 2016;30(1):48–56. doi:10.1038/leu.2015.261. PMID 26416462
  113. Pfirrmann M, Clark RE, Prejzner W, et al. The EUTOS long-term survival (ELTS) score is superior to the Sokal score for predicting survival in chronic myeloid leukemia. Leukemia. 2020;34(8):2138–2149. doi:10.1038/s41375-020-0931-9. PMID 32601376
  114. Pileri SA, Ascani S, Cox MC, et al. Myeloid sarcoma: clinico-pathologic, phenotypic and cytogenetic analysis of 92 adult patients. Leukemia. 2007;21(2):340–350. doi:10.1038/sj.leu.2404491. PMID 17170724
  115. Plander M, Kányási M, Szendrei T, et al. Flow cytometry in the differential diagnosis of myelodysplastic neoplasm with low blasts and cytopenia of other causes. Pathol Oncol Res. 2024;30:1611811. doi:10.3389/pore.2024.1611811. PMID 39040799
  116. Platzbecker U, Larson RA, Gurbuxani S. Diagnosis and treatment of AML in the context of WHO and ICC 2022 classifications: divergent nomenclature converges on common therapies. Hemasphere. 2025;9(2):e70083. doi:10.1002/hem3.70083. PMID 39897085
  117. Porwit A, Béné MC, Duetz C, et al. Multiparameter flow cytometry in the evaluation of myelodysplasia: analytical issues: recommendations from the European LeukemiaNet/International Myelodysplastic Syndrome Flow Cytometry Working Group. Cytometry B Clin Cytom. 2023;104(1):27–50. doi:10.1002/cyto.b.22108. PMID 36537621
  118. Quessada J, Cuccuini W, Saultier P, et al. Cytogenetics of pediatric acute myeloid leukemia: a review of the current knowledge. Genes (Basel). 2021;12(6):924. doi:10.3390/genes12060924. PMID 34204358
  119. Rausch C, Rothenberg-Thurley M, Dufour A, et al. Validation and refinement of the 2022 European LeukemiaNet genetic risk stratification of acute myeloid leukemia. Leukemia. 2023;37(6):1234–1244. doi:10.1038/s41375-023-01884-2. PMID 37041198
  120. Reichard KK, Tefferi A, Abdelmagid M, et al. Pure (acute) erythroid leukemia: morphology, immunophenotype, cytogenetics, mutations, treatment details, and survival data among 41 Mayo Clinic cases. Blood Cancer J. 2022;12(11):147. doi:10.1038/s41408-022-00746-x. PMID 36323674
  121. Roberts I, Alford K, Hall G, et al. GATA1-mutant clones are frequent and often unsuspected in babies with Down syndrome: identification of a population at risk of leukemia. Blood. 2013;122(24):3908–3917. doi:10.1182/blood-2013-07-515148. PMID 24021668
  122. Rudelius M, Weinberg OK, Niemeyer CM, et al. The International Consensus Classification (ICC) of hematologic neoplasms with germline predisposition, pediatric myelodysplastic syndrome, and juvenile myelomonocytic leukemia. Virchows Arch. 2023;482(1):113–130. doi:10.1007/s00428-022-03447-9. PMID 36445482
  123. Sandahl JD, Coenen EA, Forestier E, et al. t(6;9)(p22;q34)/DEK-NUP214-rearranged pediatric myeloid leukemia: an international study of 62 patients. Haematologica. 2014;99(5):865–872. doi:10.3324/haematol.2013.098517. PMID 24441146 (abstract)
  124. Sanz MA, Fenaux P, Tallman MS, et al. Management of acute promyelocytic leukemia: updated recommendations from an expert panel of the European LeukemiaNet. Blood. 2019;133(15):1630–1643. doi:10.1182/blood-2019-01-894980. PMID 30803991
  125. Sapienza MR, Pileri A, Derenzini E, et al. Blastic plasmacytoid dendritic cell neoplasm: state of the art and prospects. Cancers (Basel). 2019;11(5):595. doi:10.3390/cancers11050595. PMID 31035408
  126. Sauta E, Robin M, Bersanelli M, et al. Real-world validation of Molecular International Prognostic Scoring System for myelodysplastic syndromes. J Clin Oncol. 2023;41(15):2827–2842. doi:10.1200/JCO.22.01784. PMID 36930857
  127. Scott LM, Tong W, Levine RL, et al. JAK2 exon 12 mutations in polycythemia vera and idiopathic erythrocytosis. N Engl J Med. 2007;356(5):459–468. doi:10.1056/NEJMoa065202. PMID 17267906
  128. Selimoglu-Buet D, Wagner-Ballon O, Saada V, et al. Characteristic repartition of monocyte subsets as a diagnostic signature of chronic myelomonocytic leukemia. Blood. 2015;125(23):3618–3626. doi:10.1182/blood-2015-01-620781. PMID 25852055
  129. Shah MV, Arber DA, Hiwase DK. TP53-mutated myeloid neoplasms: 2024 update on diagnosis, risk-stratification, and management. Am J Hematol. 2025;100 Suppl 4:88–115. doi:10.1002/ajh.27655. PMID 40066944
  130. Shao R, Ryder C, Wang L, et al. A review of the pathological and molecular diagnosis of primary myelofibrosis. Cancers (Basel). 2025;18(1):50. doi:10.3390/cancers18010050. PMID 41514563
  131. Shimony S, Stahl M, Stone RM. Acute myeloid leukemia: 2025 update on diagnosis, risk-stratification, and management. Am J Hematol. 2025;100(5):860–891. doi:10.1002/ajh.27625. PMID 39936576
  132. Sukswai N, Aung PP, Yin CC, et al. Dual expression of TCF4 and CD123 is highly sensitive and specific for blastic plasmacytoid dendritic cell neoplasm. Am J Surg Pathol. 2019;43(10):1429–1437. doi:10.1097/PAS.0000000000001316. PMID 31261288
  133. Szuber N, Orazi A, Tefferi A. Chronic neutrophilic leukemia and atypical chronic myeloid leukemia: 2024 update on diagnosis, genetics, risk stratification, and management. Am J Hematol. 2024;99(7):1360–1387. doi:10.1002/ajh.27321. PMID 38644693
  134. Tarlock K, Lamble AJ, Wang YC, et al. CEBPA-bZip mutations are associated with favorable prognosis in de novo AML: a report from the Children's Oncology Group. Blood. 2021;138(13):1137–1147. doi:10.1182/blood.2020009652. PMID 33951732
  135. Tashakori M, Wang W, Kadia TM, et al. Differential characteristics of TP53 alterations in pure erythroid leukemia arising after exposure to cytotoxic therapy. Leuk Res. 2022;118:106860. doi:10.1016/j.leukres.2022.106860. PMID 35636055
  136. Taube F, Georgi JA, Kramer M, et al. CEBPA mutations in 4708 patients with acute myeloid leukemia: differential impact of bZIP and TAD mutations on outcome. Blood. 2022;139(1):87–103. doi:10.1182/blood.2020009680. PMID 34320176 (abstract)
  137. Tefferi A, Barbui T. Polycythemia vera: 2024 update on diagnosis, risk-stratification, and management. Am J Hematol. 2023;98(9):1465–1487. doi:10.1002/ajh.27002. PMID 37357958 (Tefferi and Barbui 2023)
  138. Tefferi A, Vannucchi AM, Barbui T. Essential thrombocythemia: 2024 update on diagnosis, risk stratification, and management. Am J Hematol. 2024;99(4):697–718. doi:10.1002/ajh.27216. PMID 38269572
  139. Tefferi A. Primary myelofibrosis: 2023 update on diagnosis, risk-stratification, and management. Am J Hematol. 2023;98(5):801–821. doi:10.1002/ajh.26857. PMID 36680511
  140. Tesi B, Robelius A, Baskin B, et al. Validation of guidelines for genetic investigation of myeloid neoplasms with germline predisposition: results from a prospective cohort study. Clin Cancer Res. 2025;31(14):3062–3071. doi:10.1158/1078-0432.CCR-24-4251. PMID 40388595
  141. Thiele J, Kvasnicka HM, Facchetti F, et al. European consensus on grading bone marrow fibrosis and assessment of cellularity. Haematologica. 2005;90(8):1128–1132. PMID 16079113
  142. Thiele J, Kvasnicka HM, Gianelli U, et al. Evolution of WHO diagnostic criteria in "classical myeloproliferative neoplasms" compared with the International Consensus Classification. Blood Cancer J. 2025;15(1):31. doi:10.1038/s41408-025-01235-7. PMID 40038244
  143. Thiele J, Kvasnicka HM, Orazi A, et al. The international consensus classification of myeloid neoplasms and acute leukemias: myeloproliferative neoplasms. Am J Hematol. 2023;98(1):166–179. doi:10.1002/ajh.26751. PMID 36200127
  144. Thomopoulos TP, Symeonidis A, Kourakli A, et al. Chronic neutrophilic leukemia: a comprehensive review of clinical characteristics, genetic landscape and management. Front Oncol. 2022;12:891961. doi:10.3389/fonc.2022.891961. PMID 35494007
  145. Tzankov A, Reichard KK, Hasserjian RP, et al. Updates on eosinophilic disorders. Virchows Arch. 2023;482(1):85–97. doi:10.1007/s00428-022-03402-8. PMID 36068374
  146. Umeda M, Liu YC, Karol SE, Klco JM. Novel classification system and high-risk categories of pediatric acute myeloid leukemia. Haematologica. 2025;110(9):1962–1973. doi:10.3324/haematol.2024.285644. PMID 39781610
  147. Umeda M, Ma J, Westover T, et al. A new genomic framework to categorize pediatric acute myeloid leukemia. Nat Genet. 2024;56(2):281–293. doi:10.1038/s41588-023-01640-3. PMID 38212634
  148. Valent P, Akin C, Hartmann K, et al. Updated diagnostic criteria and classification of mast cell disorders: a consensus proposal. Hemasphere. 2021;5(11):e646. doi:10.1097/HS9.0000000000000646. PMID 34901755
  149. Van Galen J, Willis SD, Bains A, et al. Measurable residual disease testing in acute myeloid leukemia: current state, foundational models, and tools for future development. Cancer Metastasis Rev. 2026;45(2). doi:10.1007/s10555-026-10322-5. PMID 41894068
  150. Vanner RJ, Zeng AGX, Kim RH, et al. Clinical decisions in clonal hematopoiesis: a contemporary review for clinicians. Haematologica. 2026;111(5):1584–1609. doi:10.3324/haematol.2025.287422. PMID 41641651
  151. Veenstra CR, Alhan C, Westers TM, et al. Validation of the diagnostic potential of 17 immunophenotypic core markers defined by the ELN-iMDS-Flow Working Group in an independent cohort of patients with MDS. Hemasphere. 2025;9(3):e70075. doi:10.1002/hem3.70075. PMID 40060116
  152. Wagner-Ballon O, Saft L, Johansson U, et al. Should flow cytometry be required for CMML diagnosis? Position of the ELN iMDS Flow Working Group. Hemasphere. 2026;10(7):e70410. doi:10.1002/hem3.70410. PMID 42396131
  153. Wahab A, Mushtaq K, Borak SG, Bellam N. Zinc-induced copper deficiency, sideroblastic anemia, and neutropenia: a perplexing facet of zinc excess. Clin Case Rep. 2020;8(9):1666–1671. doi:10.1002/ccr3.2987. PMID 32983473
  154. Wechsler J, Greene M, McDevitt MA, et al. Acquired mutations in GATA1 in the megakaryoblastic leukemia of Down syndrome. Nat Genet. 2002;32(1):148–152. doi:10.1038/ng955. PMID 12172547
  155. Weinberg OK, Arber DA, Döhner H, et al. The International Consensus Classification of acute leukemias of ambiguous lineage. Blood. 2023;141(18):2275–2277. doi:10.1182/blood.2022019493. PMID 36877915 (Weinberg 2023a; content as summarized in Fuda 2025)
  156. Weinberg OK, Porwit A, Orazi A, et al. The International Consensus Classification of acute myeloid leukemia. Virchows Arch. 2023;482(1):27–37. doi:10.1007/s00428-022-03430-4. PMID 36264379
  157. Wong TN, Ramsingh G, Young AL, et al. Role of TP53 mutations in the origin and evolution of therapy-related acute myeloid leukaemia. Nature. 2015;518(7540):552–555. doi:10.1038/nature13968. PMID 25487151
  158. Xiao W, Nardi V, Stein E, et al. A practical approach on the classifications of myeloid neoplasms and acute leukemia: WHO and ICC. J Hematol Oncol. 2024;17(1):56. doi:10.1186/s13045-024-01571-4. PMID 39075565
  159. Zhang X, Sun J, Yu W, Jin J. Current views on the genetic landscape and management of variant acute promyelocytic leukemia. Biomark Res. 2021;9(1):33. doi:10.1186/s40364-021-00284-x. PMID 33957999