KMT2A-rearranged acute leukemia across all ages: infants are 4% of it
also called KMT2A-r incidence all ages · how big is the KMT2A-rearranged population
Around 29,000 Americans of all ages are diagnosed with acute leukemia each year, and roughly one in ten of those leukemias has a broken KMT2A gene, so about 2,900 people a year. Only about 120 of them are babies, which means infants are about four in every hundred KMT2A patients.
The same thing, for a scientist
SEER projects 22,720 new AML and 6,250 new ALL diagnoses in the United States in 2026, about 28,970 acute leukemias. Taking the widely quoted figure that KMT2A rearrangements occur in up to 10% of acute leukemias gives on the order of 2,900 US KMT2A-rearranged acute leukemias per year across all ages. Against roughly 120 KMT2A-rearranged infant cases per year (70-80% of ~90 infant ALL plus ~50% of ~70 infant AML), infants are approximately 4% of the US KMT2A-rearranged population. That single ratio explains most of the commercial behaviour documented elsewhere in this graph.
The denominator, stated first2 deeper
Denominator: all new acute leukemia diagnoses in the United States in one year, all ages 12. This is not a per-million rate and not a paediatric figure.
| Quantity | Value | Source |
|---|---|---|
| US new AML cases, 2026 estimate | 22,720 | SEER Cancer Stat Facts 1 |
| US new ALL cases, 2026 estimate | 6,250 | SEER Cancer Stat Facts 2 |
| US new acute leukemias, all ages | 28,970 | sum |
| KMT2A-rearranged share of acute leukemias | up to ~10% | Issa et al. 2023 3 |
| Implied US KMT2A-r acute leukemias per year | ~2,900 | arithmetic |
| US KMT2A-r infant cases per year | ~120 | Brown 2013 4 |
| Infant share of the US KMT2A-r population | ~4% | arithmetic |
The 10% figure is a framing statement in a clinical paper's introduction, not a registry measurement, so the ~2,900 carries that uncertainty. Even at 5% the infant share only rises to ~8%.
Age is not the only asymmetry: median AML age is 70
SEER puts the median age at AML diagnosis at 70, with most cases in the 65-74 band; median age at ALL diagnosis is 18 12. KMT2A rearrangement is bimodal across that landscape: it dominates infant ALL, is uncommon in childhood ALL outside infancy, and reappears in adult ALL and in therapy-related AML.
The practical consequence is that a sponsor recruiting 'KMT2A-rearranged acute leukemia' will fill a trial overwhelmingly with adults without doing anything wrong. AUGMENT-101, the trial supporting the first menin-inhibitor approval, had a median patient age of 37 with a range of 1.3 to 75 years despite an indication that formally covers one-year-olds (see trial-augment-101). That is not negligence, it is arithmetic.
How many distinct KMT2A fusions there are, and why that fragments the population further
The Frankfurt diagnostic reference laboratory analysed 3,401 acute leukemia patients between 2003 and 2022 and, together with the published literature, catalogued 107 in-frame KMT2A gene fusions, of which the seven commonest partners plus partial tandem duplications account for more than 90% of recombinations; 37 partners are recurrent and 63 have been seen only once 5.
That matters for the platform argument in two opposite directions.
- Against fragmentation: if the drug targets menin, which every KMT2A fusion protein needs, the 107 partners collapse into one addressable population. This is the central case for interface-directed therapy.
- For fragmentation: if biology or prognosis differs by partner (it demonstrably does in AML, where subgroup 5-year EFS spans 11% to 92%, see
disease-kmt2a-r-aml), then a single trial arm averaging over partners can hide both a large benefit in one subgroup and harm in another.
| metric | derived annual case count |
|---|---|
| value | 2900 |
| unit | KMT2A-rearranged acute leukemia cases per year |
| population | all ages |
| region | United States |
| year | 2026 |
| source | derived: SEER 2026 projections x ~10% KMT2A-r share (Issa 2023) |
| denominator | all new US acute leukemia diagnoses in one year, all ages (28,970) |
| us new aml 2026 | 22720 |
| us new all 2026 | 6250 |
| kmt2a r share PCT | 10 |
| us kmt2a r infant cases per year | 120 |
| infant share of kmt2a r PCT | 4 |
| median age aml dx | 70 |
| median age all dx | 18 |
| distinct in frame kmt2a fusions | 107 |
- The 4% infant share is the economic root of every paediatric gap in this field. A sponsor can hit every enrolment target, satisfy every regulator and still dose almost no infants, because infants are 4% of the eligible population and the hardest 4% to dose.
- If the partner gene turns out to modify menin-inhibitor response, an all-comers KMT2A-r trial dominated by adult partners (MLLT3, PTD) could return a positive result that does not hold for infant AFF1 disease, and the label would still cover infants.
- What is the true population-level frequency of KMT2A rearrangement in unselected adult acute leukemia, as opposed to the 10% framing figure that everyone cites?
- How many of the ~2,900 US KMT2A-rearranged patients per year are therapy-related, and does that subgroup respond to menin inhibition like de novo disease?
- If the addressable US KMT2A-r population is really 2,900 a year and infants are 120 of them, what enrolment mechanism could ever give infants a proportionate share of trial slots?
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
SEER estimates 22,720 new cases of acute myeloid leukemia and 11,500 deaths in the United States in 2026, with an incidence of 4.4 per 100,000 per year, 5-year relative survival of 33.4%, and median age at diagnosis of 70.
Cancer Stat Facts: Leukemia - Acute Myeloid Leukemia (AML) · National Cancer Institute Surveillance, Epidemiology, and End Results Program · SEER 2026What this evidence cannot show
The 2026 case count is a projection, not an observed count; incidence and survival are based on 2019-2023 and 2016-2022 data respectively. SEER does not report KMT2A rearrangement status, so it cannot supply the molecular numerator.
Evidence that argues against this (1)
- SEER figures for 2026 are PROJECTIONS from historical trend, not counts, and a median age at diagnosis of 70 makes adult AML the wrong denominator for this graph's patients. SEER also does not record KMT2A rearrangement status, so no KMT2A-rearranged figure can be derived from it without importing a fraction from elsewhere. National Cancer Institute 2026
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
SEER estimates 6,250 new cases of acute lymphocytic leukemia and 1,600 deaths in the United States in 2026, with an incidence of 1.9 per 100,000 per year, 5-year relative survival of 73.2%, and median age at diagnosis of 18.
Cancer Stat Facts: Leukemia - Acute Lymphocytic Leukemia (ALL) · National Cancer Institute Surveillance, Epidemiology, and End Results Program · SEER 2026What this evidence cannot show
Projection for 2026. SEER's ALL category mixes B- and T-lineage and does not report fusion status.
Evidence that argues against this (1)
- The same projection caveat, and the same missing variable: ALL's 73.2% five-year survival is dominated by children with favourable cytogenetics, while infant KMT2A-rearranged disease sits far below it. Quoting the aggregate next to this graph's population invites exactly the elision the clinical beat had to correct elsewhere. Leukemia 2021
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
KMT2A rearrangements occur in up to 10% of acute leukemias and NPM1 mutations in up to 30% of AML, and revumenib is active across both.
The menin inhibitor revumenib in KMT2A-rearranged or NPM1-mutant leukaemia · Issa GC, Aldoss I, DiPersio J, et al. · Nature 2023What this evidence cannot show
The 10% and 30% shares are introductory framing statements rather than the paper's own measurements; they are the weakest link in the ~2,900 estimate and should be replaced with a registry-derived figure if one becomes available.
Evidence that argues against this (1)
- Series that genotype unselected adult acute leukemia often report KMT2A rearrangement in the low single-digit percent rather than 10%, which would halve or better the implied all-ages population and correspondingly raise the infant share. source 2023
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.review
US infant acute leukemia comprises roughly 160 cases per year (~90 ALL, ~70 AML), with KMT2A rearrangement in 70-80% of infant ALL and about 50% of infant AML, giving roughly 120 KMT2A-rearranged infants per year.
Treatment of infant leukemias: challenge and promise · Brown P · Hematology Am Soc Hematol Educ Program 2013What this evidence cannot show
2013 review; the ~120 figure is arithmetic on its component percentages, not a counted cohort.
Evidence that argues against this (2)
- The 70-80% KMT2A-rearrangement figure in infant leukaemia comes from series assembled where molecular characterisation is routine, and the detection method sets the number: cytogenetics alone misses cryptic rearrangements that RNA-level analysis finds. This graph's own atlas records that KMT2A::SEPTIN11 is cytogenetically indistinguishable from KMT2A::AFF1, so the true proportion is a function of assay. SOURCING CORRECTION 2026-09-20 (Q42 citation-integrity repair). The KMT2A::SEPTIN11 sentence is a fact from another paper carried under this review's PMID. Its source is the 2023 recombinome, attached as the note below, and the fact is slightly more specific than stated: both partners are reported as t(4;11) at conventional banding, and the recombinome assigns SEPTIN11 to 4q21.1 against AFF1 at 4q21.3-q22.1 - sub-band resolution that routine karyotyping does not deliver. Hematology Am Soc Hematol Educ Program 2013
- Primary source for the cytogenetic-collision point above, read from PMC10169636 on 2026-09-20. The recombinome's partner table lists KMT2A::SEPTIN11 as t(4;11)(q21.1;q23.3) and KMT2A::AFF1 as t(4;11)(q21.3-q22.1;q23.3); SEPTIN11 is seen 3 times in 3,401 patients. At the resolution a clinical karyotype reports, both read t(4;11)(q21;q23), so a cohort characterised by banding alone assigns some SEPTIN11 cases to AFF1 and the 70-80% infant KMT2A-r figure inherits that. Meyer C et al. 2023
- E5established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
Analysis of 3,401 acute leukemia patients between 2003 and 2022, combined with the literature, identified 107 in-frame KMT2A gene fusions, with the seven most frequent partner genes plus partial tandem duplications accounting for more than 90% of recombinations; 37 partners are recurrent and 63 have been seen only once.
The KMT2A recombinome of acute leukemias in 2023. · Meyer C, Larghero P, Almeida Lopes B, et al. · Leukemia 2023What this evidence cannot show
A diagnostic reference-laboratory series, not a population sample: referral bias enriches for difficult or research-interesting cases, so partner frequencies are more reliable than any incidence inference drawn from the cohort.
Evidence that argues against this (1)
- A census counts the fusions the contributing laboratories could characterise, so 107 in-frame fusions is a lower bound and the concentration in seven partners is an upper bound on itself. This graph's own audit also finds this paper supporting 75 items across 69 entries in 14 beats, graded at four different confidence levels. Oncofusion knowledge graph (internal analysis) 2026
>580-fold in-vitro resistance in infant ALL; a further >7.4-fold in t(4;11)
11q23-rearranged ALL and t(9;11) AML both more cytarabine-sensitive than non-rearranged
topoisomerase II poison implicated in therapy-related KMT2A rearrangement; cleaves the BCR
12-fold in-vitro resistance in infant ALL; same relative resistance in proB ALL at any age
12-fold in-vitro resistance measured with the native enzyme; transfer to the pegylated product is an inference
the KMT2A::SEPTIN11 cytogenetic-collision fact in that entry's counter-note is from the 2023 recombinome, not from the 2013 review whose PMID it carried; the correct source is now attached beside it
- related to →How many children this affects
infants are ~4% of the US KMT2A-rearranged population
- related to →KMT2A::AFF1 (MLL-AF4), t(4;11)(q21;q23)
AFF1 is one of 107 in-frame KMT2A fusion partners catalogued
partner-dependent prognosis fragments the apparently single population
explains why a KMT2A-r trial fills with adults at median age 37
price x eligible population sets the commercial case
population arithmetic behind the cross-indication case
- ← depends onHypothesis: the paediatric gap is set by the population ratio, not by paediatric biology
the population ratio the hypothesis is built on
population size determines eligibility for designation