KMT2A exon 11 / intron 11 breakpoints (BCR subregion C): the infant, HOXA-low, topoisomerase-II-associated break
also called MLL intron 11 break · 3' BCR break · KMT2A e11 junction · BCR subregion C · therapy-related MLL break
The second hot spot for KMT2A breaks sits in intron 11, a piece of DNA that is unusually easy to cut. Babies under six months and patients whose leukaemia was caused by earlier chemotherapy break here most often, and this is the break linked to worse survival.
The same thing, for a scientist
Breaks in KMT2A exon 11 through exon 13 (5026 bp; BCR1 subregion C) account for 39.5% of all 3401 genomically characterised KMT2A rearrangements [E1]. Intron 11 is the most feature-dense fragile element in the gene: it carries a DNase I hypersensitive site, an apoptotic cleavage site, an RNA polymerase II binding site and topoisomerase II binding/cleavage sites [E1]. It is the break of infancy and of therapy: ALL in infants under ~6 months preferentially breaks in intron 11 and this preference reverses thereafter; therapy-induced acute leukaemias shift significantly to intron 11; KMT2A::MLLT1 and KMT2A::SEPTIN6 prefer intron 11 constitutively; and the intron-11 shift seen in adult KMT2A::ELL is specifically attributable to therapy-induced cases [E1]. The intron 10 / exon 11 boundary is the reported prognostic cut point: infant acute leukaemias with poor outcome preferentially broke within intron 11 [E2]. The proposed mechanism is not a change in the direct fusion's domain content but a change in the reciprocal protein: breaks within or downstream of exon 11 disrupt the folding of PHD1, impairing CYP33/PPIE binding and PHD1 homodimerisation, whereas upstream breaks leave PHD1 intact on the reciprocal product [E1].
Why intron 11 is the fragile site
Intron 11 is where three different kinds of DNA vulnerability overlap 1:
- Topoisomerase II cleavage sites. This is the mechanistic link to epipodophyllotoxins (etoposide) and anthracyclines, which trap topo-II-DNA cleavage complexes. It explains why therapy-induced leukaemias shift significantly towards intron 11.
- A DNase I hypersensitive site and an RNA Pol II binding site. Open, transcribed chromatin is more exposed to nucleases and to transcription-associated breakage.
- An apoptotic cleavage site. A cell that starts but survives apoptosis can end up with a cut here.
The practical consequence: the same genomic element is implicated in transplacental/in-utero exposure-driven infant leukaemia and in iatrogenic therapy-related leukaemia. That is a strong hint that intron 11 breaks are chemically induced rather than stochastic, and links this node to the fusion-genesis beat.
The prognostic claim, stated honestly1 deeper
What is claimed: breakpoints within KMT2A intron 11 are associated with poorer outcome, with the cut point exactly at the intron 10 / exon 11 boundary 12.
What the evidence actually is: Emerenciano et al. studied 545 children <=24 months (385 ALL, 160 AML) but determined the genomic breakpoint location in only 30 KMT2A-rearranged cases 2. In that subset, patients with poor outcome preferentially displayed breakpoints within intron 11. Age <=6 months (P<0.001), high WBC (P=0.001) and KMT2A-r itself (P=0.002) were the significant prognostic factors in the main analysis; the breakpoint finding is presented as 'first evidence', not as a validated stratifier, and is confounded with age because infants <6 months are precisely the group that breaks in intron 11 12.
What the largest dataset says: Meyer et al., with 3401 patients, write that whether the partner- and age-specific breakpoint shifts have an impact on clinical outcome 'is yet unclear' 1.
Verdict for the lab: breakpoint-position prognostics are a hypothesis supported by one small analysis, not an established fact. It is a genuinely open and testable question, and it is the highest-value piece of missing evidence in this beat.
Proposed mechanism: the reciprocal PHD1
Meyer et al. offer a structural rationale 1. PHD1-PHD2-ePHD3 is encoded by KMT2A exons 11-16, with the bromodomain and ePHD4 encoded by exons 17-21. A break upstream of exon 11 transfers the entire, correctly folded PHD1 to the reciprocal partner::KMT2A protein; a break within or downstream of exon 11 truncates PHD1 mid-domain, which is reported to change its folding, impair binding of the peptidyl-prolyl isomerase CYP33/PPIE, and destroy PHD1 homodimerisation. Because CYP33 binding to PHD3 is the switch that recruits BMI1/HDAC/CBX4/KDM5B repressors to the CXXC domain, an intron-11 break removes that regulatory brake from the reciprocal protein.
Note carefully: this mechanism operates on the reciprocal fusion protein, not on the direct KMT2A::partner oncoprotein, whose domain content is identical whether the break is in intron 9 or intron 11. So if breakpoint position really changes prognosis, the most parsimonious explanation is that it changes the reciprocal product - which makes the reciprocal protein a therapeutic and biomarker target in its own right (see mechanism-reciprocal-kmt2a-fusion-chromatin-opening).
Who breaks here
From the DCAL cohort (n=3401) 1:
- Mean share of all breaks in subregion C (exons 11-13): 39.5%.
- Infant ALL (<~6 months): clear preference for intron 11; the preference switches to intron 9 at about 6 months of age. Meyer et al. interpret this as evidence that infant ALL under 6 months is a distinct entity, most likely arising from rapidly proliferating pro-B fetal liver cells (CD10-, CD19+, CD34+) rather than bone-marrow HSPCs.
- Infant KMT2A::AFF1: predominantly intron 11 (adults shift to introns 9-10).
- KMT2A::MLLT1 and KMT2A::SEPTIN6: preferentially intron 11 across the cohort.
- KMT2A::ELL: paediatric and adult groups prefer intron 11; in adults this shift is specifically linked to therapy-induced leukaemia.
- Therapy-induced leukaemia (n=142): significant shift to intron 11. This shift was not seen for AFF1, MLLT3 or MLLT10 fusions.
- AML: breaks nearly exclusively in BCR1 and preferentially in intron 9, i.e. AML is the low-intron-11 lineage.
| gene | KMT2A |
|---|---|
| cytoband | 11q23.3 |
| genome build | hg38 |
| region | KMT2A exon 11 + intron 11 + exon 12 + intron 12 + exon 13 (BCR1 subregion C) |
| length bp | 5026 |
| frequency PCT | 39.5 |
| population | all KMT2A-rearranged acute leukaemia, n=3401 (DCAL 2003-2022) |
| exon last kmt2a mrna | 11 |
| mechanism class | topoisomerase-II-associated double-strand breakage followed by illegitimate end joining; see mechanism-kmt2a-breakpoint-repair-signature |
| transcriptional signature | HOXA-low |
| reported outcome | worse than upstream (intron 9-10) breaks in infant AL; cut point at the intron 10 / exon 11 boundary; evidence base n=30 typed patients |
| reciprocal product note | Breaks within or downstream of exon 11 truncate PHD1 mid-domain on the reciprocal partner::KMT2A protein, reported to impair CYP33/PPIE binding and PHD1 homodimerisation |
| numbering convention | Meyer/Marschalek (DCAL) KMT2A exon numbering |
repair features
[ "topoisomerase II binding/cleavage sites", "DNase I hypersensitive site", "RNA polymerase II binding site", "apoptotic cleavage site" ]
enriched in
[ "infant ALL under ~6 months", "infant KMT2A::AFF1", "KMT2A::MLLT1", "KMT2A::SEPTIN6", "therapy-induced acute leukaemia (n=142)", "adult therapy-induced KMT2A::ELL" ]
depleted in
[ "AML", "adult KMT2A::AFF1", "KMT2A::AFDN" ]
domains retained 5 prime
[ "AT-hook 1-3", "SNL/repression region", "MBM1", "MBM2", "LEDGF/PSIP1-binding region", "CXXC" ]
domains lost 5 prime
[ "PHD1", "PHD2", "ePHD3", "bromodomain", "ePHD4", "FYRN", "FYRC", "TAD", "WIN motif", "SET" ]
- If the prognostic effect of an intron-11 break runs through the reciprocal protein, then therapies aimed only at the direct fusion (including menin inhibitors) address the wrong molecule in exactly the worst-prognosis group.
- Intron 11 is a topoisomerase-II cleavage hotspot. Any regimen that keeps or intensifies etoposide/anthracycline exposure in KMT2A-r patients risks creating a second KMT2A break in a surviving normal cell - the therapy-related leukaemia signature is literally the same intron.
- The whole prognostic story could collapse to age: infants <6 months break in intron 11 and infants <6 months do badly. Building a biomarker on breakpoint position without adjusting for age would be a false lead.
- Read the first bite-back above with its counter-evidence attached. It assumes the intron-11 effect runs through a truncated PHD1 on the reciprocal protein, but CYP33/PPIE binds PHD3, not PHD1 (PMID 20677832), and inserting PHD1+PHD2 into an MLL fusion has no effect on immortalisation while PHD3 alone abolishes it (PMID 18676843). The 'menin inhibitors address the wrong molecule' worry therefore has no working mechanism behind it at present, and the direct-fusion-only evidence points the other way: MLL::Af4 alone in human CD34+ cells reproduces t(4;11) pro-B ALL (PMID 27846391).
- The one KMT2A fusion in this entry's intron-11 list whose transcriptional output has been measured directly, KMT2A::SEPTIN6, SUPPRESSES HOXA7 and HOXA9 (PMID 42505395). Menin inhibitors act by collapsing the HOXA9/MEIS1 programme, so a HOXA-low KMT2A fusion is exactly the case in which that pharmacology has nothing to collapse. SEPTIN6 is rare, but it is a worked example of a KMT2A-rearranged leukaemia that a menin inhibitor may not touch.
- Is the intron-11/poor-outcome association real after adjusting for age at diagnosis and partner gene? This needs breakpoint typing on a prospective trial cohort (Interfant-06 or COG AALL15P1 scale), which is the single highest-value analysis this beat identifies.
- Does an intron-11 break change the direct fusion protein at all, or only the reciprocal one? If only the reciprocal, then the prognostic effect is evidence that the reciprocal protein matters clinically - a strong, testable prediction.
- Infant ALL <6 months breaks in intron 11 and adult ALL breaks in intron 9-10 for the same partner (AFF1). Is intron 11 fragility a property of the fetal-liver pro-B chromatin state, and if so, is that state reproducible in vitro?
- Do intron-11 KMT2A::AFF1 patients (HOXA-low) respond differently to menin inhibitors than intron-9 (HOXA-high) patients? No trial has reported this.
- Does the intron-11 prognostic signal survive once age, partner gene and the PHD3 boundary are all entered together? The reciprocal-PHD1 mechanism that motivated this entry is not supported by the primary PHD literature, so if the association is real it needs a different explanation, and nobody has proposed one.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Exons 11-13 (5026 bp) carry 39.5% of KMT2A breakpoints; intron 11 contains a DNase I hypersensitive site, an apoptotic cleavage site, an RNA Pol II binding site and topoisomerase II binding sites; infant ALL and therapy-induced leukaemia shift significantly to intron 11; breaks within/downstream of exon 11 alter PHD1 folding, CYP33 binding and PHD1 homodimerisation on the reciprocal protein.
The KMT2A recombinome of acute leukemias in 2023. · Meyer C, Larghero P, Almeida Lopes B, et al. · Leukemia 2023What this evidence cannot show
Single-centre referral cohort; subgroup breakpoint shifts are chi-square comparisons against the cohort mean without reported multiple-testing correction. The PHD1-folding mechanism is cited by Meyer et al. from other work and is not measured in this paper; we have not independently verified the primary structural study, so the PHD1 claim is carried here at the confidence of the review statement, not of a structure.
Evidence that argues against this (2)
- The topoisomerase-II-consensus explanation for intron-11 fragility is not what the mapping experiment found. When the same laboratory actually located etoposide-induced double-strand breaks in the MLL gene, they colocalised with the RNA polymerase II binding site and the transcription initiation region of a newly described gene-internal promoter, and explicitly NOT with a nearby topoisomerase II consensus sequence; the authors conclude the genomic instability of MLL is presumably linked to transcriptional processes. Listing a topo II consensus site inside intron 11 is therefore a sequence annotation, not a demonstration that topo II cuts there, and a transcription-coupled mechanism fits the measured break positions better. Scharf S et al. 2007
- The mechanism this item attributes to a truncated PHD1 is assigned by the primary structural literature to a different finger. CYP33/PPIE binds MLL through its RRM domain to PHD3, not PHD1, and PHD3 is the switch: H3K4me3 binding and CYP33 RRM binding to PHD3 are mutually inhibitory, and CYP33 downregulates HOXC8, HOXA9, CDKN1B and C-MYC in a proline-isomerase-dependent way. PHD3 is encoded well 3' of exon 11, so a break at the intron 10 / exon 11 boundary does not obviously change CYP33 binding on the direct fusion at all - PHD3 is absent from it whether the break is in intron 9 or intron 11. Park S et al. 2010
- E2weak confidence 2 of 5: Preliminary, small, indirect, or not yet peer reviewed.paper
Among infant/young-child acute leukaemias, patients with poor outcome preferentially displayed chromosomal breakpoints within MLL intron 11.
The distribution of MLL breakpoints correlates with outcome in infant acute leukaemia · Emerenciano M, Meyer C, Mansur MB, Marschalek R, Pombo-de-Oliveira MS · British Journal of Haematology 2013What this evidence cannot show
Breakpoint location was determined in only 30 of 545 cases; the breakpoint-outcome comparison is therefore based on tens of patients, was secondary, and is confounded with age <=6 months, which was independently and strongly prognostic (P<0.001). Single-country (Brazil) cohort with 60.5% overall survival, i.e. a different treatment backbone from current Interfant/COG protocols. Not replicated prospectively.
Evidence that argues against this (1)
- The 3401-patient recombinome analysis states that whether these breakpoint shifts affect clinical outcome is yet unclear, declining to confirm the association in the largest available dataset. Meyer C et al. 2023
- E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
In paediatric t(4;11) ALL, MLL breakpoints in introns 11/12 mark the infant, HOXA-low subgroup, distinct from the non-infant HOXA-high intron 9/10 subgroup.
Two independent gene signatures in pediatric t(4;11) acute lymphoblastic leukemia patients · Trentin L, Giordan M, Dingermann T, Basso G, Te Kronnie G, Marschalek R · European Journal of Haematology 2009What this evidence cannot show
Small microarray cohort with n=3 normal bone-marrow references; breakpoint and age are confounded. Same senior author as the recombinome series.
Evidence that argues against this (2)
- HOXA-low tracks the partner gene, not only the breakpoint. KMT2A::SEPTIN6 - one of the two partners this entry lists as constitutively preferring intron 11 - actively INHIBITS HOXA7 and HOXA9 expression in the only cell-line model of that fusion, activating BMP signalling instead. Because partner identity and breakpoint preference are correlated in the recombinome, a HOXA-low signature observed in intron-11 cases can be produced by the partners that happen to break there, and a breakpoint-versus-HOXA association cannot be read as causal without holding the partner constant. Nagel S et al. 2026
- A high HOXA signature is not a KMT2A-breakpoint readout at all. CALM-AF10 (PICALM::MLLT10) T-ALL, which contains no KMT2A break of any kind, overexpresses HOXA5, HOXA9, HOXA10 and MEIS1 to levels comparable with MLL-translocated acute leukaemia. Whatever sets HOXA output high or low is therefore a property of the effector complex the partner assembles, and mapping HOXA status onto intron 9 versus intron 11 is at best an indirect proxy. Dik WA et al. 2005
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Retaining PHD fingers in an MLL fusion protein blocks transformation and reduces Hoxa9 locus occupancy, establishing that the PHD cluster encoded by exons 11-16 is functionally decisive and not inert spacer.
The PHD fingers of MLL block MLL fusion protein-mediated transformation · Muntean AG, Giannola D, Udager AM, Hess JL · Blood 2008What this evidence cannot show
Tests PHD fingers on the direct fusion, not the partial PHD1 fragment created on a reciprocal protein by an intron-11 break; the relevance to the reciprocal-PHD1 mechanism is by analogy.
Evidence that argues against this (1)
- Only the THIRD PHD finger blocks transformation; the ones a break near exon 11 truncates do not. Inserting the first and second PHD fingers of MLL into MLL-ENL had no effect on haematopoietic stem cell immortalisation in serial replating assays, whereas inserting the PHD cassette or PHD3 alone blocked it, and that block runs through CYP33 and HDAC1 recruitment. So the PHD cluster is functionally decisive at PHD3, encoded by exons 14-16, not at the intron 10 / exon 11 cut point on which this entry's prognostic mechanism rests. Chen J et al. 2008
Counter-evidence to this entry's own PHD1 mechanism: CYP33/PPIE binds MLL through its RRM to PHD3, not PHD1 (Park 2010, PMID 20677832), and inserting PHD1+PHD2 into MLL-ENL has no effect on immortalisation while PHD3 alone abolishes it (Chen 2008, PMID 18676843). An intron-11 break cannot plausibly act by impairing CYP33 binding via a truncated PHD1.
- related to →PPIE (CyP33) — the off-switch the fusion throws away, and the reason PHD3 reinsertion kills transformation
The CYP33/PPIE binding site on KMT2A is PHD3 (exons 14-16), not PHD1 (exon 11). This matters for the breakpoint atlas because the reciprocal-PHD1 prognostic mechanism assumes otherwise.
- refutes →HOXA-high versus HOXA-low KMT2A-rearranged leukemia — the split that breaks the one-mechanism story
The HOXA-low/HOXA-high split cannot be read off breakpoint position: KMT2A::SEPTIN6, an intron-11-preferring partner, actively suppresses HOXA7/HOXA9 (Nagel 2026, PMID 42505395), and CALM-AF10, which has no KMT2A break at all, is HOXA-high (Dik 2005, PMID 16107895). Partner identity is confounded with breakpoint preference throughout the recombinome.
Direct evidence for the gap this analysis names: KMT2A::SEPTIN6 in KOPM-88 INHIBITS HOXA7 and HOXA9 and activates BMP signalling instead (Nagel 2026, PMID 42505395, DOI 10.3390/cells15141286). A menin inhibitor works by collapsing the HOXA9/MEIS1 programme, so a HOXA-low KMT2A fusion is the case where that pharmacology has nothing to collapse.
Bearing on the menin-interface hypothesis, in its favour: the reciprocal-protein objection to targeting only the direct fusion lost its mechanism this cycle. CYP33 binds PHD3 not PHD1 (PMID 20677832), PHD1+PHD2 add-back does nothing (PMID 18676843), and MLL::Af4 alone in human CD34+ cells reproduces t(4;11) pro-B ALL with no reciprocal present (PMID 27846391).
- ← related toEach partner prefers a different KMT2A exon: partner-specific break position in 3,401 patients
Counter-evidence (Broeker 1996, PMID 8634439): the intron-9-versus-intron-11 axis was an aetiology axis 27 years before it was a partner axis -- 74% of de novo breaks centromeric versus 25% of therapy-related AML breaks, P=0.02, with the telomeric half carrying a high-affinity SAR and six topoisomerase II consensus sites. Meyer 2023 concedes the adult MLLT3 intron-11 shift is 'clearly linked to therapy-induced leukemia'.
- ← related toEach partner prefers a different KMT2A exon: partner-specific break position in 3,401 patients
infant KMT2A::AFF1, MLLT1 and adult EPS15 all prefer intron 11; adult MLLT3 does so only in therapy-induced disease
subregion C, 39.5% of breaks
infant and therapy-related break class
topoisomerase-II-associated cleavage hot spot
- ← related toRolf Marschalek & Claus Meyer (DCAL, Goethe-University Frankfurt) — the world's KMT2A breakpoint dataset
source of the age-dependent breakpoint shift within the KMT2A breakpoint cluster region
intron 11 breaks differ in retained domain content, repeat context and reciprocal product from intron 9 breaks
breakpoint position determines both the assay design and the retained PHD1-3 architecture
transcription-coupled TOP2 breakage as the shared causal shape with the KMT2A breakpoint cluster region