Blast Percentage Interpreter
Blast Percentage Interpreter
There is no longer one answer to this. The WHO 5th edition and the International Consensus Classification, both published in 2022 and both in use, set different blast thresholds for the same diseases — and both define entities in which a genetic abnormality makes the percentage irrelevant. This page prints what each classification says about your case rather than picking a side, because a report that names one system and a reader who assumes the other is how a diagnosis gets misread.
How many blasts make it acute leukaemia?
Blast % + genetics → both classificationsA marrow aspirate from a 68-year-old with a macrocytic anaemia shows 6 per cent blasts on a 500-cell differential. Molecular studies return a mutated NPM1. There is no BCR::ABL1 and no CEBPA mutation, and the blasts are myeloid.
The two classifications, entity by entity
ICC 2022 sets a floor of 10% for AML with recurrent genetic abnormalities — including mutated NPM1 and in-frame bZIP CEBPA — with one exception, BCR::ABL1, which requires 20%. For TP53, myelodysplasia-related and not-otherwise-specified disease, 10 to 19% is the new category MDS/AML and 20% or more is AML.
Both take the percentage from blood or marrow, whichever reaches the threshold, and both count blast equivalents with the blasts.
Worked example
A marrow aspirate from a 68-year-old with a macrocytic anaemia shows 6 per cent blasts on a 500-cell differential. Molecular studies return a mutated NPM1. There is no BCR::ABL1 and no CEBPA mutation, and the blasts are myeloid.
Myeloid lineage, so the lymphoid route does not apply, and this is a marrow rather than a blood percentage
Mutated NPM1 is an AML-defining abnormality in both classifications and is neither BCR::ABL1 nor CEBPA
6% is below 10%, so the ICC's floor for AML with recurrent genetic abnormalities is not met → the ICC classifies this as a myelodysplastic syndrome carrying the mutation
The WHO 5th edition removed the blast threshold for this group entirely → on that reading it is acute myeloid leukaemia
Two published classifications, the same marrow, opposite diagnoses. Neither is a transcription error and neither is out of date
What to do: refer, and make the report state which classification it followed. Raise the blast count to 12% and both systems say AML; raise it to 25% and both say AML; it is only in this band that the answer depends on the system
Blast threshold by entity, in both classifications
| Entity | WHO 5th edition | ICC 2022 |
|---|---|---|
| AML with a defining genetic abnormality (PML::RARA, RUNX1::RUNX1T1, CBFB::MYH11, NPM1, KMT2A-r, DEK::NUP214, MECOM) | No blast threshold | 10% or more |
| AML with BCR::ABL1 | 20% or more | 20% or more |
| AML with in-frame bZIP CEBPA mutation | 20% or more | 10% or more |
| Mutated TP53 | Not a defining entity; classified as AML-MR or by differentiation, 20% or more | 10 to 19% = MDS/AML with mutated TP53; 20% or more = AML with mutated TP53 |
| Myelodysplasia-related | 20% or more | 10 to 19% = MDS/AML; 20% or more = AML |
| No defining abnormality | 20% or more = AML defined by differentiation; 10 to 19% = MDS-IB2 | 20% or more = AML NOS; 10 to 19% = MDS/AML |
| B or T lymphoblastic | No blast threshold | No blast threshold |
Counting the blasts properly, before any threshold is applied
| Requirement | Why it matters |
|---|---|
| Manual differential, not an analyser flag | Analysers flag immature cells; they do not count blasts, and the flag has neither the specificity nor the numerator to be used as a percentage |
| At least 200 cells on a blood film, 500 nucleated cells on a marrow | At 200 cells the 95% interval around an observed 20% is roughly 15 to 26 — wide enough to cross the threshold in either direction |
| Count blast equivalents with the blasts | Abnormal promyelocytes in acute promyelocytic leukaemia, promonocytes in monocytic disease and megakaryoblasts all count. Excluding promonocytes is a recognised way of under-diagnosing acute monocytic leukaemia |
| Use the higher of blood and marrow | Both classifications accept either specimen, so a blood percentage that reaches the threshold is sufficient even if the marrow is lower |
| Correct for nucleated red cells | An uncorrected white cell count distorts every percentage derived from it |
| Use the trephine when the aspirate is dry or diluted | CD34 immunohistochemistry on the trephine carries the estimate; a percentage from a haemodilute aspirate is not interpretable |
Why there are two classifications, and what to do about it
For twenty years the diagnosis of acute myeloid leukaemia rested on a single number: 20 per cent blasts in the blood or marrow, inherited from an earlier 30 per cent threshold and applied almost universally. It was never a biological boundary. It was a convention, chosen because a line had to be drawn somewhere, and everyone knew it was arbitrary at the edges. What changed is that the genetics became good enough to make the arbitrariness intolerable — a case with PML::RARA and 15 per cent blasts is the same disease as one with PML::RARA and 40 per cent, and treating the first as a myelodysplastic syndrome was indefensible.
Two groups set out to fix this in 2022 and reached different answers. The WHO 5th edition removed the blast requirement from genetically defined acute myeloid leukaemia altogether, keeping 20 per cent only where a threshold does real work — BCR::ABL1, where it separates de novo disease from chronic myeloid leukaemia in blast phase, and CEBPA — and for the entities that have no defining genetics. The International Consensus Classification kept a floor but lowered it to 10 per cent for the genetically defined group, and created a new category, MDS/AML, for the 10 to 19 per cent band in everything else, explicitly to acknowledge a continuum and to let those patients into both MDS and AML clinical trials.
Both are defensible and both are in use. That is the situation a reader has to work in, and pretending otherwise is the failure mode this page exists to prevent. The consequence is concrete: the same marrow can be reported as acute myeloid leukaemia by one laboratory and as a myelodysplastic syndrome by another, without either being wrong, and a clinician who assumes one system while reading a report written under the other will misjudge the urgency. There is also a second-order effect that is easy to miss — published series, trial eligibility criteria and prognostic scores were derived under particular definitions, so a case reclassified from MDS to AML by a change of system does not thereby acquire the outcomes of the AML series it has joined.
The practical response is short. State the classification in the report, by name and edition. Where the two disagree about a case, say so in the report rather than choosing silently. Make sure the blast percentage itself was properly derived, because the arguments above all presume a 500-cell marrow differential that counted blast equivalents. And treat the genetics as the diagnosis rather than as a supplement to the count: in both systems, for most of the entities that matter, the abnormality now carries more weight than the percentage does.
Frequently asked questions
Is 20 per cent still the threshold for acute myeloid leukaemia?
Only for some entities, and which ones depends on the classification. Both systems keep 20 per cent for AML with BCR::ABL1 and for disease with no defining genetic abnormality. The WHO 5th edition also keeps it for CEBPA and for myelodysplasia-related AML, and removes it entirely for the other genetically defined entities. The ICC lowers it to 10 per cent for AML with recurrent genetic abnormalities, including NPM1 and in-frame bZIP CEBPA, and puts the 10 to 19 per cent band elsewhere into a new category called MDS/AML.
Which classification should my laboratory use?
That is a local decision and both are legitimate, so the answer this page can give is narrower and more useful: whichever you use, name it and its edition in the report, and where a case would be classified differently under the other system, say so. The cost of not doing that falls on the clinician reading the report, who otherwise cannot tell whether a diagnosis of myelodysplastic syndrome at 15 per cent blasts with mutated NPM1 reflects the case or the system.
What is MDS/AML?
A category created by the ICC in 2022 for cases with 10 to 19 per cent blasts that do not carry an AML-defining genetic abnormality. It exists to acknowledge that the biology is continuous across that band rather than changing at 20 per cent, and to make those patients eligible for both myelodysplastic syndrome and acute myeloid leukaemia trials. The WHO 5th edition has no equivalent category and classifies the same cases as myelodysplastic syndrome with increased blasts, grade 2, while noting they may be treated as AML where clinically indicated.
Does a normal blast percentage exclude acute leukaemia?
Not under the WHO 5th edition, where a defining genetic abnormality establishes acute myeloid leukaemia at any blast percentage. In practice such cases are very unusual, and a genuinely low blast count alongside a defining abnormality should prompt review of both results rather than immediate acceptance. It also does not exclude a clonal myeloid disorder more generally: myelodysplastic syndrome with an isolated deletion of 5q, or with SF3B1 mutation and ring sideroblasts, presents with a normal blast count.
Does the 20 per cent rule apply to lymphoblastic leukaemia?
No. Neither classification uses a blast threshold to define B or T lymphoblastic leukaemia or lymphoma; the diagnosis rests on an unequivocal lymphoblast population with a compatible immunophenotype. The figure of about a quarter of marrow nucleated cells that is sometimes quoted separates lymphoblastic leukaemia from lymphoblastic lymphoma with marrow involvement — a question about the distribution of the disease rather than about whether it is acute — and treatment is largely the same either way.
Related calculators
References
- 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.
- 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.
- Huber S, Baer C, Hutter S, et al. AML classification in the year 2023: how to avoid a Babylonian confusion of languages. Leukemia. 2023;37(7):1413-1420.
- Aqil B. What’s new in hematopathology 2025: myeloid neoplasms in the WHO 5th edition and ICC. J Pathol Transl Med. 2025;59(6):472-475.
- College of American Pathologists. What’s New in AML Classification: WHO 2022 vs International Consensus Classification. CAP member resources.
Medical Disclaimer: The tools and content provided here are for educational and reference purposes only. They are not intended to substitute for professional medical advice, diagnosis, or treatment. Clinical decisions should always be based on the comprehensive assessment of a qualified healthcare professional.