A nine-month-old with AML whose fusion was found only after every standard test failed
also called Ikeda 2019 KMT2A-USP2 infant AML · first KMT2A-USP2 AML case
A nine-month-old boy had acute myeloid leukaemia that tests said was KMT2A-related, but no test could say what KMT2A was joined to. Sequencing all his cells' RNA finally showed it was joined to USP2 - the first time this had been seen in AML.
The same thing, for a scientist
A 9-month-old boy with KMT2A-rearranged AML in whom conventional screening for known translocation partners failed; whole-transcriptome sequencing identified a KMT2A-USP2 fusion, reported as the first AML case with this fusion [E1]. The authors' framing is the point of the case: in paediatric AML, prognosis with a KMT2A rearrangement depends on the partner - unlike infant ALL, where outcome is poor regardless of partner - so a patient whose partner cannot be named cannot be risk-stratified [E1]. They argue that identifying the partner in all KMT2A-rearranged AML is necessary to learn what outcomes specific rearrangements carry, and that better detection methods are needed. **This case is the clinical face of the under-ascertainment problem**: a rare partner stays rare partly because the assays used to look for it are built around the common ones, and the individual cost is a child who cannot be assigned to a risk group.
What the case shows that a cohort cannot1 deeper
The Frankfurt series establishes that KMT2A::USP2 exists at scale and is missed by PCR methods 2. This case shows the same failure happening to one child in real time, in a different country, with a different assay chain, and being solved only by an unbiased method.
It also lands on a specific clinical decision. In paediatric AML, partner identity is prognostic - MLLT3, MLLT10, AFDN and others carry different outcomes - whereas in infant ALL the rearrangement itself dominates 1. So 'KMT2A-rearranged AML, partner unknown' is not a complete diagnosis, and this child's treating team could not place him on the risk axis their protocol uses until the transcriptome came back.
Priority, and why it is a footnote rather than the point
The paper is titled as the first KMT2A-USP2 fusion in infant AML. The Frankfurt letter published weeks earlier already contained one AML case among its 17 2, so 'first' means first dedicated report rather than first observation. Two AML cases among roughly 47 reported KMT2A::USP2 patients makes AML a real but clear minority phenotype for this fusion, which is overwhelmingly B-ALL and MPAL.
What is not reported
No outcome, no treatment response, no follow-up duration is available to this entry from the abstract, and no junction structure. So this case contributes to the diagnostic and stratification argument and contributes nothing to prognosis. It should not be cited for outcome in KMT2A::USP2 AML, because it contains no outcome.
| age at dx | 9 months |
|---|---|
| fusion | fusion-kmt2a-usp2 |
| course | Not reported in the source abstract. |
| outcome | Not reported. This entry must not be cited for prognosis. |
| source type | case report |
| consent note | public source only, no PHI; no identifying detail beyond age, sex and diagnosis as published |
| why it matters | The clinical face of under-ascertainment: in paediatric AML, partner identity carries prognostic weight, so a patient whose partner cannot be named cannot be risk-stratified. |
| diagnostic path | conventional partner screening (failed) -> whole-transcriptome sequencing (successful) |
presentation
[ "acute myeloid leukaemia in infancy", "KMT2A rearrangement suspected but partner not identifiable by conventional screening", "KMT2A-USP2 identified by whole-transcriptome sequencing" ]
- Recommending whole-transcriptome sequencing for every KMT2A-rearranged AML with an unnamed partner is the right diagnostic answer and an unaffordable one in most of the world. The children most affected are infants, who are concentrated in settings least able to run it.
- Naming the partner only helps if the partner's prognosis is known. For USP2 it is not - two AML cases, no outcome data - so the immediate clinical gain from solving this particular diagnosis was smaller than the diagnostic effort suggests.
- What was this child's outcome? The report establishes a diagnosis and stops there, and with only two KMT2A::USP2 AML cases known, one follow-up would double what the field knows about the phenotype's course.
- How many 'KMT2A-rearranged, partner unknown' paediatric AML patients are in cooperative-group registries, and what fraction would resolve to minor-BCR fusions on re-sequencing? This is answerable from banked material.
- Does partner identity carry the same prognostic weight for minor-BCR partners as it does for the major-BCR ones on which that principle was established? Nobody has enough patients to say.
- E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
A 9-month-old boy with KMT2A-rearranged AML required whole-transcriptome sequencing for diagnosis after conventional screening for known translocation partners failed, identifying a KMT2A-USP2 fusion; the authors note that prognosis of paediatric KMT2A-rearranged AML depends on the translocation partner, unlike infant ALL, and argue that partner identification in all KMT2A-rearranged AML is critical.
Whole transcriptome sequencing reveals a KMT2A-USP2 fusion in infant acute myeloid leukemia. · Ikeda J, Shiba N, Tsujimoto SI, Yoshida M, Nakabayashi K, Ogata-Kawata H, Okamura K, Takeuchi M, Osumi T, Tomizawa D, Hata K, Kiyokawa N, Ito S, Kato M · Genes, Chromosomes & Cancer 2019What this evidence cannot show
A single patient. No treatment, response, outcome or follow-up duration is given in the abstract, and no junction structure. The claim that partner identity is prognostic in paediatric KMT2A-r AML is the authors' framing citing prior literature, not a finding of this report. Abstract read in full; body not read.
Evidence that argues against this (1)
- A case report is a biased estimator of how often diagnosis is difficult - cases reach print partly because they were hard. The unbiased version of this observation is the cohort figure: 23% of KMT2A-rearranged patients whose breakpoints two PCR methods could not resolve turned out to be minor-BCR disease. Meyer C et al. 2019
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
The Frankfurt discovery cohort, published weeks earlier, already contained one AML case among 17 KMT2A::USP2 patients, so this report is the first dedicated case description rather than the first observation of the fusion in AML.
Human MLL/KMT2A gene exhibits a second breakpoint cluster region for recurrent MLL-USP2 fusions. · Meyer C, Lopes BA, Caye-Eude A, Cavé H, Arfeuille C, et al., Marschalek R · Leukemia 2019What this evidence cannot show
The Frankfurt letter reports phenotype counts (B-ALL 12, MPAL 4, AML 1) without individual case descriptions, so its AML patient cannot be compared with this one in any detail. Both were published in 2019 within weeks of each other, so neither group could have cited the other. Full text read (PMC6756029).
Evidence that argues against this (1)
- Simultaneous independent publication makes priority ambiguous rather than settled, and nothing turns on it. The substantive point - that AML is a minority phenotype for this fusion - rests on two cases in roughly 47 and is itself a small-number statement. Meyer C et al. 2023
9-month-old with AML; fusion found only by whole-transcriptome sequencing after conventional partner screening failed
the individual-patient face of under-ascertainment: no partner named means no risk stratification in paediatric AML
one of only two reported KMT2A::USP2 AML cases