Oncofusion

start / graph / hyp-the-menin-interface-requirement-is-polarity-dependent-not-kmt2a-intrinsic

Hypothesismoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q92-kmt2a-sarcoma

The menin interface is a requirement of 5'-KMT2A fusions in haematopoiesis, not of KMT2A rearrangement as such - and a solid tumour has been quietly proving it for six years

also called polarity-dependent menin requirement · KMT2A sarcoma as a menin-interface negative control

This project assumes that a KMT2A rearrangement causes cancer by keeping its grip on menin. In a rare sarcoma, the rearrangement throws that grip away and still causes cancer, which suggests the menin requirement belongs to the leukaemia version of the molecule rather than to KMT2A rearrangement in general.

The same thing, for a scientist

Hypothesis: the requirement for an intact KMT2A menin-binding motif is a property of the 5'-KMT2A fusion architecture operating in a haematopoietic cell, not an intrinsic property of KMT2A rearrangement. Evidence in favour: every KMT2A-rearranged sarcoma arrangement with a published breakpoint (YAP1::KMT2A::YAP1, VIM::KMT2A, LDB1::KMT2A) excludes KMT2A exon 1 and therefore the menin-binding motif at aa 6-25, yet these tumours are clonal, recurrent, methylation-distinct and aggressive; and the YAP1 arm shows no HOXA9/MEIS upregulation, i.e. no leukaemic program to require a menin interface for. Consequence if true, and it cuts both ways for this project: it STRENGTHENS the menin-interface hypothesis as a statement about leukaemogenesis (the interface is not dispensable in the disease where it is present, and where it is absent the leukaemic program does not appear), while it REFUTES any broader reading in which menin dependence follows from the words 'KMT2A rearrangement'. The hypothesis is currently untested in the only way that matters: nobody has measured menin binding, menin dependence or menin-inhibitor sensitivity in any KMT2A-rearranged sarcoma or model.

Go deeper
The argument in four steps
  1. In leukaemia KMT2A is always the 5' partner and the break is in exons 8-14, so the menin-binding motif at aa 6-25 (exon 1) is always retained 1.
  2. In the defined sarcoma arms KMT2A is internal or 3', and its contribution begins at exon 2, 4 or 5 - so exon 1 and the motif are absent 23.
  3. These tumours are nonetheless real neoplasms: recurrent partner, reproducible morphology, a distinct DNA methylation cluster, 50% metastasis 4.
  4. And they show no HOXA9/MEIS program for a menin interface to be required by 5.

Therefore the menin requirement travels with the 5' architecture and the haematopoietic context, not with the gene.

Motif content, not polarity, is the discriminating fact

Stated separately because the first version of this entry led with polarity and that is one step removed from the thing that matters. Polarity is the mechanism; motif content is the discriminator. A 3'-KMT2A fusion loses exon 1 because the break is upstream of the retained fragment - but what decides whether a menin inhibitor has a target is only whether exon 1 is in the transcript.

The consequence is a dissociation this graph's model does not contain. The model pairs two things - the CpG address (CXXC, aa 1147-1195) and the menin/LEDGF tether (MBM aa 6-25, IBM1 aa 123-134, IBM2 aa 147-152) - and treats the pair as the leukaemogenic determinant. The sarcoma arrangements separate them and keep the address. In the YAP1 arm the KMT2A contribution is essentially nothing else 2.

So the open question this hypothesis really poses is not 'does polarity matter' but: what can the CXXC domain do on its own, attached to a transcription-factor scaffold, with no menin, no LEDGF, no AT-hooks and no PAF1c-binding repression domain 2? That question has never been asked in any system.

Why this is not simply good news for the hypothesis

The comfortable reading is: the sarcoma lacks the interface, lacks the HOXA program, and is therefore consistent with the interface being the leukaemogenic determinant. That reading is available and this entry endorses it as the most likely one.

The uncomfortable reading is the one worth stating plainly: a KMT2A rearrangement transforms a human cell without menin, without LEDGF, without the AT-hooks and, in the YAP1 arm, without anything of KMT2A except a 49-residue zinc finger. If the project's model is 'KMT2A rearrangement plus a partner-supplied effector, docked by menin', then this entity is a second, independent solution to the same genetic accident that uses none of the docking. That should make the project curious about what the CXXC domain alone can do - the one element retained under both polarities - rather than satisfied that the negative control came out the right way.

What would kill it

Listed in data.falsifiers. The cheapest and most decisive is a menin co-immunoprecipitation: if the YKY protein pulls down menin despite lacking aa 6-25, the coordinate argument is wrong and some other surface is doing the work. The second is a reciprocal-transcript protein blot: if the reciprocal KMT2A::YAP1 species is translated and abundant, the tumour contains a menin-competent KMT2A after all and the entity is not a clean control.

Facts and figures
statementThe discriminating fact is WHICH MOTIFS THE RETAINED KMT2A FRAGMENT CONTAINS, not polarity as such - polarity is the mechanism by which the motifs are lost. Every KMT2A-rearranged sarcoma arrangement with a published breakpoint contributes neither KMT2A exon 1 (menin-binding motif aa 6-25, LEDGF IBM1 aa 123-134) nor, except in the VIM arm, exon 2 (IBM2 aa 147-152), while all of them contribute the CXXC zinc finger (aa 1147-1195). These tumours therefore keep the CpG ADDRESS and discard the MENIN AND LEDGF TETHER, and they transform. The requirement for an intact menin-binding motif is a property of the 5'-KMT2A architecture in a haematopoietic cell, not of KMT2A rearrangement.
originliterature-gap
prior probability0.7
what it does to the projects central claimStrengthens it as a claim about leukaemogenesis and about the drug class's mechanism; refutes any reading in which menin dependence is entailed by the phrase 'KMT2A rearrangement'.
predictions
[
  "A construct that retains KMT2A exon 1 but not the CXXC domain will bind menin and fail to transform; one that retains CXXC but not exon 1 will transform in a mesenchymal context and not bind menin. The two arms of that experiment separate address from tether, which no existing study does.",
  "A YKY protein will not co-precipitate menin, and a VIM::KMT2A protein will not either, whereas a KMT2A::AFF1 control in the same cells will.",
  "A menin inhibitor will show no activity in any KMT2A-rearranged sarcoma model, at concentrations that kill MV4;11.",
  "MEN1 knockout will not reduce viability or fusion-target expression in a YKY-expressing cell beyond its effect on the parental line.",
  "The CXXC domain, not the menin interface, will be the KMT2A element whose deletion abolishes YKY-driven transcriptional change.",
  "If any KMT2A-rearranged solid tumour is found that RETAINS exon 1, it will look transcriptionally more leukaemia-like than the YAP1 arm does."
]
experiments
[
  "exp-menin-pulldown-on-yky-versus-kmt2a-aff1"
]
analyses
[
  "ana-kmt2a-exon-to-domain-map",
  "absence-no-menin-dependence-measurement-in-kmt2a-r-sarcoma"
]
falsifiers
[
  "A menin co-immunoprecipitation in which the YKY protein pulls down menin despite lacking aa 6-25 - which would mean a second, uncharacterised menin surface exists on KMT2A and would undermine the coordinate argument everywhere it is used.",
  "Demonstration that the reciprocal KMT2A::YAP1 transcript is translated into an abundant protein in these tumours, which would mean a menin-competent KMT2A species is present and the entity is not a clean negative control.",
  "A menin inhibitor showing single-agent activity in a KMT2A-rearranged sarcoma model or patient.",
  "Discovery that the YAP1-arm tumours DO run a HOXA program when measured properly (single-cell, or ChIP at the HOXA cluster), which would remove the transcriptional half of the argument.",
  "A published sarcoma KMT2A breakpoint that retains exon 1 in the dominant transcript of a defined arm."
]
How this could bite us back
  • If this hypothesis is right, the addressable population for menin inhibitors is narrower than 'KMT2A-rearranged cancer' and every market or trial estimate that used the broader phrase is wrong in the same direction.
  • If it is wrong - if YKY does bind menin through an unknown surface - then this project's whole coordinate-based reasoning about which KMT2A fragments can engage menin is unreliable, including in leukaemia, and several entries would need re-examination.
  • The hypothesis is attractive because it makes the project's central claim look good, which is exactly the condition under which a hypothesis gets less scrutiny than it deserves. It is recorded at moderate confidence with no functional evidence whatsoever behind it.
Still open
  1. Stated as the sharpest form of this hypothesis: what does the KMT2A CXXC domain do when it is the ONLY part of KMT2A present? The YAP1 arm is that molecule and nobody has characterised it.
  2. What can the CXXC domain do on its own? It is the only KMT2A element retained under both polarities, and in the YAP1 arm it is essentially the whole KMT2A contribution. A CXXC-deleted YKY construct is the obvious experiment and has not been reported.
  3. Is there a lineage in which a 3'-KMT2A fusion causes leukaemia, or a lineage in which a 5'-KMT2A fusion causes a solid tumour? KMT2A::MAML2 thymoma may be the second case and its polarity should be checked against the exon map.
  4. If the menin requirement is contextual, what supplies the equivalent docking function in mesenchyme - or is no docking needed because YAP1 supplies TEAD-mediated targeting directly?
Every source behind this entry (5)
  1. E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    The pathologists defining this entity state that in KMT2A-rearranged leukaemia KMT2A is the 5' partner with breakpoints between exon 9 and intron 11, retaining the N-terminal DNA-binding domains and losing the SET domain, whereas in these sarcomas KMT2A is the 3' partner with breakpoints in exons 4-5 that leave the DNA-binding domain partially disrupted or lost and the SET domain intact.

    PMID 31592798DOI
    What this evidence cannot show

    An interpretive statement in a pathology discussion rather than a protein measurement, and it predates the insertional YAP1::KMT2A::YAP1 model. Read from the PMC full text via efetch db=pmc.

    Evidence that argues against this (1)
    • Seven of the 16 non-recurrent-partner KMT2A sarcomas in the pan-sarcoma screen carried 5'-KMT2A fusions, so 5' polarity is not unique to leukaemia. Those tumours had diverse histologies, median age 57 and 44% TP53 mutation and are judged of unclear significance, but no functional or methylation work has been done on them to confirm they are passengers. Massoth LR et al. 2020
  2. E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    Computed here (2026-09-21): the KMT2A menin-binding motif lies at aa 6-25 within exon 1, so the YAP1 arm (KMT2A exons 4/5-6 only), the VIM arm (KMT2A exon 2 onward) and the LDB1 case (KMT2A exons 4-36) each exclude it, while all three retain the CXXC zinc finger at aa 1147-1195 in exons 5-6.

    What this evidence cannot show

    A coordinate derivation from published exon indices. No sarcoma KMT2A fusion protein has ever been observed by any method, so this states what the reported transcripts can encode and not what protein the tumours make.

    Evidence that argues against this (1)
    • Reciprocal KMT2A::YAP1 transcripts with 5' breakpoints at KMT2A exons 3, 4 and 6 are detected in a substantial minority of YAP1-arm cases and DO begin at exon 1. If any of these is translated, the tumour contains a menin-competent KMT2A species and the hypothesis loses its cleanest evidence. Warmke LM et al. 2025
  3. E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    In a KMT2A-rearranged spindle cell sarcoma, next-generation sequencing identified an LDB1::KMT2A fusion involving exons 1-10 of LDB1 and exons 4-36 of KMT2A, confirmed by FISH and RT-PCR - the most explicit published statement that a sarcoma KMT2A fusion retains nearly the entire KMT2A coding sequence while beginning at exon 4.

    LDB1::KMT2A Fusion in a Spindle-Cell Sarcoma: A Case Report. · Wang Z, Zhang Y, Zhu M, et al. · Genes Chromosomes Cancer 2025
    PMID 40944346DOI
    What this evidence cannot show

    A single case, read at abstract level, with a concurrent pathogenic BCOR frameshift so the fusion is not the only lesion, and 5 months of follow-up. n=1 cannot establish a polarity rule; it is quoted here because the exon range is stated unambiguously.

    Evidence that argues against this (1)
    • This tumour's immunoprofile (CD99, SATB2, cyclin D1, BCL2, TLE1, pan-TRK, NKX2.2, focal BCOR) plus a pathogenic BCOR frameshift is compatible with a BCOR-rearranged or BCOR-altered sarcoma in which the KMT2A fusion is incidental. A single case with a second driver is weak support for any architectural claim. Nakamura H et al. 2023
  4. E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    In 22 human YAP1::KMT2A-rearranged sarcomas, DNA methylation profiling formed a cluster clearly separate from both conventional sclerosing epithelioid fibrosarcoma and low-grade fibromyxoid sarcoma, and among 14 patients with follow-up local recurrence occurred in 21% and metastasis in 50% - establishing that these tumours are a reproducible, aggressive neoplastic entity and not an incidental finding.

    PMID 39641785DOI
    What this evidence cannot show

    Retrospective, referral-biased, median follow-up 25 months with follow-up available in 14 of 22, and the methylation dataset is by-request rather than deposited so the clustering cannot be independently reproduced. A methylation cluster establishes distinctness, not that the fusion is the driver.

    Evidence that argues against this (1)
    • A distinct methylation cluster does not attribute causation to the fusion. Both cases of the two rarest arrangements were disease-free after surgery alone, and the entity is defined by its assay rather than by morphology in at least one published case that could not be classified even after the fusion was found. Fumagalli C et al. 2025
  5. E5moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    In three human YAP1-KMT2A sarcomas, RNA sequencing showed no upregulation of the KMT2A leukaemia target genes HOXA9 and MEIS, and KMT2A expression was slightly lower than control - so no leukaemic transcriptional program is present for a menin interface to be required by.

    PMID 31592798DOI
    What this evidence cannot show

    n=3, reported as 'data not shown' in a discussion, undeposited (elink to GEO returns zero datasets, checked 2026-09-21), two genes only. A negative on two genes in three bulk samples cannot exclude HOXA engagement in a subpopulation or at a subset of loci.

    Evidence that argues against this (1)
    • The same authors record that HOXA upregulation WAS noted in the VIM-KMT2A sarcoma. If so, the arm that retains the AT-hooks, CXXC and SET does run a HOXA program while lacking the menin motif - which would make the hypothesis STRONGER on the polarity point and WEAKER on the claim that a HOXA program needs a menin interface. Yoshida A et al. 2019
Connected to