Oncofusion

start / graph / hyp-ikk-inhibition-is-a-menin-independent-chromatin-eviction-route

Hypothesisspeculative confidence 1 of 5: Our own inference or analogy. Not shown to be true.Q38-nfkb-signalling

IKK inhibition and menin inhibition evict the fusion from chromatin by routes that do not share a node, so they should be additive — and the same reasoning predicts that a menin inhibitor alone partially defeats itself. Prior 0.30

also called two-route eviction hypothesis · IKK plus menin combination hypothesis

Two completely different drugs both end up knocking the leukaemia fusion protein off the genes it controls: one by removing its anchor protein menin, the other by blocking an immune kinase cascade. Because they work through different molecules, using both should knock it off harder — but there is a catch, because removing menin may switch the kinase pathway on.

The same thing, for a scientist

**Statement.** In KMT2A-rearranged leukaemia, IKK/NF-kB signalling and the menin-KMT2A interaction are **two non-overlapping requirements for fusion occupancy at HOXA9 and MEIS1**. Blocking either reduces occupancy; blocking both reduces it more than either alone, and does so with sub-maximal doses of each, which is the only version of an IKK-directed strategy that could survive the toxicity of this target class. **Prior 0.30**, higher than a typical combination guess because the two routes are documented not to touch: **RELA and p50 do not co-immunoprecipitate with MLL fusion proteins in AML cells**, so NF-kB is not a component of the fusion complex and does not tether it [E1]; menin binds the fusion's own N-terminal MBM motifs, which is a direct physical tether. Two requirements, one direct and one indirect, converging on one measurable endpoint. **The complication that makes this more than a combination proposal.** Menin independently **represses** RELA transactivation by recruiting SIRT1 [E2]. So a menin inhibitor is predicted to *raise* canonical NF-kB output — exactly the signal that retains the fusion. **The two hypotheses in this graph are therefore the same claim read forwards and backwards**: `hyp-menin-inhibition-liberates-rela-and-reinforces-the-fusion-programme` says menin inhibition partially defeats itself through this axis, and this entry says an IKK inhibitor is the specific counter to that. **The single strongest objection** is that the menin-NF-kB contact is a linear region (menin 305-381) rather than the MBM pocket [E2], so a pocket-binding drug may not release RELA at all — in which case the liberation half dies and only the plain additivity half survives.

Go deeper
Predictions, in the order they should be tested

P1 (cheapest, no new chemistry). In a KMT2A-rearranged line, an IKK inhibitor reduces fusion occupancy at HOXA9/MEIS1 while total fusion protein is unchanged — reproducing the mouse MLL-AF10 result 1 in a human cell.

P2. A menin inhibitor plus an IKK inhibitor reduces fusion occupancy more than either alone at doses where each alone is sub-maximal.

P3. A clinical menin inhibitor increases RELA-dependent transcription and/or nuclear RELA in a KMT2A-rearranged cell. If P3 fails, the self-defeat half of the hypothesis is dead and the combination is an ordinary additivity claim.

P4. Menin-inhibitor-resistant or persister cells carry an NF-kB signature. This is answerable from published transcriptomes with no bench work.

P5 (the lineage test). All of P1-P3 hold in a KMT2A::AFF1 B-ALL line, not only in a myeloid one. Nothing in the literature supports P5 today (mechanism-nfkb-lineage-split-myeloid-story-lymphoid-silence).

Falsifiers, stated so they can actually fire
  • F1. An IKK inhibitor at a dose that suppresses NF-kB reporter output leaves fusion occupancy at HOXA9/MEIS1 unchanged in human KMT2A-rearranged cells. Kills the whole premise outside the original mouse system.
  • F2. Menin inhibition and IKK inhibition are not additive on occupancy — that is, either alone reaches the same floor. Then they are one requirement measured twice, contrary to the co-IP negative.
  • F3. Menin inhibition reduces rather than raises NF-kB output (plausible: NF-kB drives NFKBIA transcription, so liberating RELA raises its own inhibitor). Kills the self-defeat half.
  • F4. The combination is additive on occupancy but not on the therapeutic index — that is, marrow and immune toxicity rise at least as fast as the anti-leukaemic effect. Kills the clinical proposition while leaving the biology intact. This is the most likely outcome and the one the experiment must be designed to detect.
  • F5. IKK inhibition raises the non-canonical arm's contribution enough that net leukaemia burden falls for reasons unrelated to occupancy. Not a falsifier of the therapy but a falsifier of the mechanism, and only a chromatin read-out distinguishes them.
Why nobody has done this, as far as this beat can tell

A PubMed search for (menin OR MEN1 OR revumenib OR ziftomenib) AND ("IKK"[tiab] OR "NF-kappaB inhibitor"[tiab] OR bortezomib OR "IKK inhibitor"[tiab]) returns 3 records, and title-checking removes all three: a liposomal siRNA/bortezomib SHARP1 study in MLL-AF6 AML, a habenular-nucleus cytokine study, and a menin-deficiency depression study 3. There is no published combination of a menin inhibitor with any IKK or NF-kB-directed agent, in any system. The most likely reason is disciplinary: the 2013 paper mentions menin nowhere except its reference list, despite coming from the laboratory that established the menin-MLL axis.

Facts and figures
statementIKK/NF-kB signalling and the menin-KMT2A interaction are non-overlapping requirements for fusion occupancy at HOXA9 and MEIS1; blocking both is additive, and because menin represses RELA, menin inhibition alone partially reinforces the requirement it is trying to remove.
originliterature-gap
prior probability0.3
predictions
[
  "An IKK inhibitor reduces KMT2A-fusion occupancy at HOXA9/MEIS1 in a human KMT2A-rearranged line with total fusion protein unchanged",
  "Menin inhibitor plus IKK inhibitor reduces occupancy more than either alone at sub-maximal doses of each",
  "A clinical menin inhibitor raises RELA-dependent transcription or nuclear RELA in a KMT2A-rearranged cell",
  "Published menin-inhibitor-resistant or persister transcriptomes carry an NF-kB signature",
  "All of the above hold in a KMT2A::AFF1 B-ALL line, not only in myeloid cells"
]
experiments
[
  "exp-ikk-plus-menin-inhibitor-occupancy-combination"
]
analyses
[]
falsifiers
[
  "An IKK inhibitor at a NF-kB-suppressing dose leaves fusion occupancy unchanged in human KMT2A-rearranged cells",
  "Menin inhibition and IKK inhibition are not additive on occupancy - either alone reaches the same floor",
  "Menin inhibition reduces rather than raises NF-kB output, for example because liberated RELA induces its own inhibitor NFKBIA",
  "The combination is additive on occupancy but the therapeutic index does not improve, because marrow and immune toxicity rise at least as fast",
  "IKK inhibition reduces leukaemia burden without reducing fusion occupancy, which would refute the mechanism while leaving the therapy intact"
]
How this could bite us back
  • Both arms of this combination suppress normal marrow. Revumenib causes differentiation syndrome and cytopenias; IKK inhibition causes neutrophilia and systemic inflammation. Adding them in an infant is adding two haematological and immune toxicities, and the combination has never been dosed in any species.
  • If the eviction is signalling-dependent it is also reversible, so the fusion returns when either drug is held - and the schedule that maintains eviction is the schedule that maximises immunosuppression.
  • A leukaemia with constitutively active downstream NF-kB, or with HOXA9/MEIS1 driven independently of the fusion, escapes both arms. The founding paper already shows Hoxa9/Meis1-transformed progenitors are relatively resistant to IKK inhibition, and HOXA9 upregulation is the documented escape from the non-canonical arm too.
  • If menin inhibition liberates RELA, then the combination is partly the IKK inhibitor cleaning up after the menin inhibitor rather than two independent hits - which predicts a smaller benefit than plain additivity and a worse benefit-to-toxicity ratio.
  • Both drugs remove wild-type KMT2A function as well as fusion function, in a developing haematopoietic system.
Still open
  1. Does a clinical menin inhibitor compete with NF-kB for menin at all? Menin 305-381 is not the MBM pocket, so the answer may be simply no - and one in vitro titration settles it.
  2. Is the co-IP negative that this hypothesis rests on reproducible? It is reported as 'data not shown' in a 2013 paper and has never been repeated.
  3. Could a NEMO-directed agent achieve the eviction without the IKKa-mediated loss of the protective non-canonical arm? That would be the arm-clean version of this combination.
  4. In a combination, does the menin inhibitor's own effect on NF-kB change the dose of IKK inhibitor needed - upwards, because more RELA must be suppressed, or downwards?
Every source behind this entry (4)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    IKK inhibition substantially reduced promoter occupancy of MLL-AF10 and of wild-type MLL at Hoxa9 and Meis1, with reduced H3K79me2 and H3K4me3, despite stable cellular MLL-AF10 protein levels; and physical association of RELA or p50 with MLL fusion proteins was not detected by immunoprecipitation-western in AML cells, leading the authors to conclude NF-kB is not a component of the MLL fusion protein complex and does not directly tether it to chromatin.

    Epigenetic roles of MLL oncoproteins are dependent on NF-κB. · Kuo HP, Wang Z, Lee DF, Iwasaki M, Duque-Afonso J, Wong SHK, Lin CH, Figueroa ME, Su J, Lemischka IR, Cleary ML · Cancer Cell 2013
    PMID 24054986DOI
    What this evidence cannot show

    Mouse MLL-AF10 AML, two loci, pharmacological IKK inhibition with non-selective compounds. The co-IP negative that establishes non-overlap is reported as 'data not shown' - so the load-bearing premise of this hypothesis is the least documented result in the paper, and P2 is partly a test of that premise. Full text read this session (PMC3816582).

    Evidence that argues against this (1)
    • Menin and IKK/NF-kB might not be independent after all: menin binds p50, p52 and p65 directly, so a menin-fusion complex and an NF-kB pool could interact without the fusion itself co-precipitating NF-kB. A failed co-IP against the fusion does not exclude a menin-mediated contact. Heppner C et al. 2001
  2. E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    Menin interacts specifically and directly with the NF-kB proteins p50, p52 and p65 via their N-termini, with menin amino acids 305-381 essential for binding, and represses p65-mediated transcriptional activation at NF-kB sites in a dose-dependent and specific manner, also suppressing PMA-stimulated NF-kB activation.

    PMID 11526476DOI
    What this evidence cannot show

    Abstract fetched and read this session. 2001 methods, in vitro and transfected reporters, non-haematopoietic backgrounds. Crucially, the mapped menin region 305-381 is NOT the MBM pocket that revumenib and ziftomenib occupy, so there is no published reason to expect a clinical menin inhibitor to compete with NF-kB for menin. That is F3's main route to firing. The SIRT1-dependent execution of the repression is established in hepatocellular carcinoma, not in a haematopoietic cell - see mechanism-menin-represses-nfkb-transactivation.

    Evidence that argues against this (1)
    • This is a 2001 experiment on WILD-TYPE menin in an endocrine-tumour context, using transactivation reporters. It establishes a binding surface, not that IKK inhibition would evict anything from chromatin in a KMT2A-rearranged blast - and the hypothesis needs the second claim, not the first. Kuo HP et al. 2013
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database

    Counted absence. (menin OR MEN1 OR revumenib OR ziftomenib) AND ("IKK"[tiab] OR "NF-kappaB inhibitor"[tiab] OR bortezomib OR "IKK inhibitor"[tiab]) returns 3 PubMed records, and title-checking removes all three: a liposomal siRNA/bortezomib SHARP1 knockdown study in MLL-AF6 AML, a study of inflammatory cytokines in the lateral habenular nucleus, and a menin-deficiency depressive-behaviour study. Title-checked count: 0. No published work combines a menin inhibitor with any IKK or NF-kB-directed agent in any system.

    What this evidence cannot show

    querytranslation was read; PubMed expanded 'menin' to include the MEN1 supplementary concept and the multiple endocrine neoplasia MeSH term, which widens rather than narrows recall. Bleximenib, enzomenib and the preclinical MI-series compounds were not included as terms, so a combination study naming only one of those would be missed. This bounds the absence over the named clinical agents.

    Evidence that argues against this (1)
    • A zero at the symbol level is not a zero at the pathway level, and on this beat that is demonstrable rather than hypothetical: PMID 29320732 studies the non-canonical arm directly in MLL-AF9 leukaemia and is unreachable from any of these subunit symbol queries, because it names the kinase and the pathway rather than the subunit. The count is honest about what was searched and should not be read as evidence that the biology is unstudied - only that it is not indexed under this symbol. Xiu Y et al. 2018
  4. E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    IKK inhibitor pretreatment for two days reduced colony-forming cells in the remaining viable population and gave at least a ten-fold reduction in leukaemia stem cell frequency on limit-dilution secondary transplantation, while differential effects on colony formation and cell growth were seen in MLL leukaemia cells compared with normal haematopoietic progenitors.

    Epigenetic roles of MLL oncoproteins are dependent on NF-κB. · Kuo HP, Wang Z, Lee DF, Iwasaki M, Duque-Afonso J, Wong SHK, Lin CH, Figueroa ME, Su J, Lemischka IR, Cleary ML · Cancer Cell 2013
    PMID 24054986DOI
    What this evidence cannot show

    A ten-fold LSC reduction is the strongest functional number in the paper and it is a two-day in vitro pretreatment followed by transplantation, not a treated animal. Full text read this session (PMC3816582).

    Evidence that argues against this (1)
    • A two-day inhibitor pretreatment followed by limit-dilution transplant measures how many cells survive the pretreatment in a transplantable state, which is not the same as stem-cell frequency in an untreated animal. A tenfold reduction is also compatible with general cytotoxicity, since pan-IKK inhibition is broadly toxic to haematopoietic cells. Maharjan CK et al. 2026
Connected to