MechanismmoderateQ54-immune

MHC class I is a bivalent gene set, and a menin inhibitor derepresses it - shown in K-562, SCLC and neuroblastoma, and measured on HLA-A in primary AML, but never once in a KMT2A-rearranged cell

also called MHC-I bivalency · bivalent MHC class I genes · menin inhibition and MHC class I · HLA-A induction by menin inhibitor · antigen presentation under menin inhibition

The genes that put a leukaemia cell's identity card on its surface are held half-on and half-off, and a menin inhibitor tips them towards on. That has been shown directly in three kinds of cancer cell and in leukaemia cells from patients, but never in a cell carrying the KMT2A rearrangement this whole project is about.

The same thing, for a scientist

Two independent papers place MHC class I inside the menin-inhibitor mechanism, and they are easy to over-read. **Sparbier et al.** built genome-scale CRISPR screens around **a cell-surface readout of bivalent MHC class I gene expression** and showed that VTP50469 and the chemically distinct MI-503 activate MHC-I dose-dependently and potentiate IFN-gamma-induced surface MHC-I [E1]. That is a real pharmacological result, but the cell is **K-562, a BCR::ABL1 CML line**, and the other systems are small-cell lung cancer, neuroblastoma, DLBCL and pluripotent stem cells. The string 'KMT2A-rearranged' occurs **once** in the whole paper, in the sentence explaining where the inhibitors came from [E1]. **Hogeling et al.** is the leukaemia arm: bleximenib (JNJ-75276617) raised H3K4me3 and H3K27ac at the **HLA-A** locus, raised HLA-A mRNA, and raised surface MHC class I in **9 of 25** primary AML samples, with both class I and class II significantly up in **5 of 25** [E2]. Functionally, allogeneic T-cell killing rose in **4 of 8** samples and **all four carried NPM1c and DNMT3A mutations**, not a KMT2A rearrangement [E2]. So the honest statement for this graph is: *a menin inhibitor raises MHC class I* is **shown in AML and in non-leukaemic cancers**, is **heterogeneous between patients**, is **measured on HLA-A only**, and is **not established in any KMT2A-rearranged cell, in any species, in either lineage**.

Go deeper

The distinction the beat exists to draw: readout versus result

The tempting reading of Sparbier is 'the menin-inhibitor screen used MHC class I, therefore menin inhibitors raise MHC class I on KMT2A-rearranged blasts.' Two separate steps are being collapsed.

Step one is a result. VTP50469 activated MHC-I gene expression dose-dependently and enhanced IFN-gamma-induced surface MHC-I, and MI-503 reproduced it, so the effect is not a property of one chemical series 1. In an SCLC-OVA model, pre-treating tumour cells with VTP50469 before co-culture with OT-I T cells increased tumour killing and T-cell IFN-gamma production 1. That is a genuine pharmacology-to-immunology chain.

Step two is an extrapolation. The chain was measured in K-562, murine and human SCLC, neuroblastoma and hESC/iPSC. Whether MHC-I promoters are bivalent in a KMT2A-rearranged blast at all has, as far as this beat could establish, never been measured. Bivalency was confirmed by H3K27me3/H3K4me3 ChIP-reChIP in K-562 specifically 1.

And the extrapolation has a specific reason to fail. Sparbier's mechanism requires KMT2A/KMT2B methyltransferase activity: combined KMT2A/KMT2B knockout impaired IFN-gamma-induced MHC-I, and KMT2A/B were essential to drive the increased MHC-I expression induced by menin inhibition 1. In a KMT2A-rearranged cell one KMT2A allele produces a fusion that has lost the SET domain (see subtraction-kmt2a-set-domain). The menin inhibitor evicts the fusion and the wild-type protein alike, but only the wild-type allele and KMT2B can deposit H3K4me3 at a liberated bivalent promoter. The dose of competent enzyme available to do the job is therefore lower in the index disease than in any cell the experiment was done in. That is a prediction, not a finding, and it is the first thing exp-mhc-class-i-timecourse-under-menin-inhibition-kmt2a-r is designed to test.

The mechanism, as far as it is established

Menin does not bind MHC class I promoters. Menin ChIP-seq found ubiquitous menin at active gene promoters and no significant menin binding at bivalent MHC-I promoters, which rules out a direct repressive role 1. The effect is a redistribution:

Menin loss or menin inhibition globally reduces menin chromatin occupancy and displaces KMT2A from 9,258 of 11,583 bound sites - high-H3K4me3, low-SUZ12, active genes 1.
A much smaller set of loci, 142 genes with increased KMT2A occupancy at the TSS, of which 63% were bivalent beforehand, gain KMT2A. Bivalent MHC-I genes are in this set 1.
At those loci H3K4me3 rises, H3K27me3 and SUZ12 fall, and expression rises 1.
Total cellular H3K4me3, H3K27me3, H2AK119Ub, SUZ12 and EZH2 are unchanged - this is a redistribution, not a global shift 1.

So the correct verb is liberation: the drug takes KMT2A off the genes menin was holding it on, and some of it lands where polycomb was winning. mechanism-menin-inhibition-shifts-the-kmt2a-polycomb-balance states the general form; this entry states what the class I arm of it is.

Which gene, exactly - rule 69 applied to 'MHC class I'

'MHC class I' is not one gene. It is three classical heavy-chain loci (HLA-A, HLA-B, HLA-C) plus the invariant light chain B2M, and a W6/32 flow antibody sees the assembled heterodimer on the surface without telling you which heavy chain carried it. This beat therefore records, per source, what was actually measured:

source	readout	gene named
Sparbier screens	FACS with W6/32 (pan HLA-A,B,C) and REA143 (HLA class I Bw6) in parallel	HLA-B for the genome-browser and CUT&Tag panels; hits required in both screens for validation 1
Sparbier, IFN-gamma validation	flow	'this concentration range provided more reliable induction of HLA-B' 1
Hogeling, ChIP and qPCR	H3K27ac/H3K4me3 tracks, mRNA	HLA-A 2
Hogeling, protein	flow, MHC class I	heavy chain not resolved 2
Fetsch	class II only	CIITA and MHC-II 3

B2M is measured by nobody. No source in this beat reports beta-2 microglobulin under a menin inhibitor, which matters because a surface MHC-I signal requires B2M and a B2M-null cell cannot present class I however much heavy-chain mRNA it makes. See gene-b2m.

What would have to be true for this to matter clinically

Four conditions, none of which is established in this disease:

MHC-I promoters are bivalent in a KMT2A-rearranged blast. Untested.
Enough competent KMT2A/KMT2B remains after the fusion has taken one allele. Untested; and Sparbier shows the requirement is real 1.
An IFN-gamma signal is present. Almost everything in both papers is measured as enhancement of IFN-gamma-induced MHC-I. Without cytokine the menin-inhibitor effect on surface protein was small 1. A heavily pre-treated lymphopenic infant may not supply it.
There is an MHC-restricted effector to see the result. CAR T cells and BiTEs do not read MHC. See the note in ana-immune-evasion-routes-documented-in-kmt2a-r-versus-aml-generally.

Facts and figures

claim status

{
  "shown": [
    "A menin inhibitor raises cell-surface MHC class I in K-562 (BCR::ABL1 CML), in human SCLC and neuroblastoma lines, and in murine SCLC - dose-dependently and, mostly, as enhancement of IFN-gamma-induced expression (PMID 36635503).",
    "Two chemically distinct menin inhibitors (VTP50469, MI-503) do it, so it is a target effect, not a series effect (PMID 36635503).",
    "Menin inhibitor pre-treatment increases antigen-specific T-cell killing of a tumour cell in vitro (murine SCLC-OVA / OT-I, PMID 36635503).",
    "A clinical-stage menin inhibitor raises H3K4me3/H3K27ac at HLA-A and raises HLA-A mRNA and surface MHC-I in primary human AML - in 9 of 25 samples (PMID 39704147).",
    "Menin inhibitor pre-treatment increases allogeneic and autologous T-cell killing of primary AML blasts - in 4 of 8 samples, all NPM1c/DNMT3A-mutant (PMID 39704147)."
  ],
  "inferred_not_shown": [
    "That MHC class I promoters are bivalent in a KMT2A-rearranged blast.",
    "That a menin inhibitor raises MHC class I on a KMT2A-rearranged blast, in any species or lineage.",
    "That the effect survives the loss of one wild-type KMT2A allele to the fusion, when KMT2A/B activity is required for it.",
    "Anything at all about MHC class I in KMT2A-rearranged ALL, which is where infants are."
  ],
  "explicitly_excluded": [
    "Viral mimicry: STAT1 depletion and MDA-5/IFIH1 depletion did not impair class I activation after menin targeting (PMID 36635503).",
    "NF-kB: RELA was not required, and TNF-alpha-induced p65 phosphorylation did not rise in menin- or LEDGF-deficient cells (PMID 36635503).",
    "A direct repressive role for menin at class I promoters: no significant menin binding at bivalent MHC-I promoters (PMID 36635503)."
  ]
}

genes actually measured

{
  "HLA-B": "PMID 36635503 (screen antibody REA143, CUT&Tag and browser tracks)",
  "HLA-A": "PMID 39704147 (ChIP tracks, qPCR)",
  "HLA-C": "not measured by any source in this beat",
  "B2M": "not measured by any source in this beat"
}

cell systems

[
  "K-562 (BCR::ABL1 CML)",
  "human SCLC lines",
  "KELLY neuroblastoma",
  "murine SCLC-OVA (Trp53/Rb1, +/- Myc)",
  "EZH2-mutant DLBCL lines",
  "H9 hESC and PBMC-derived iPSC",
  "primary human AML (N up to 37; MHC arm N=25)",
  "MV4-11 and MOLM13 (KMT2A-rearranged, used for the CLASS II CIITA knockout and the PBMC killing assay only)"
]

drugs

[
  "VTP50469",
  "MI-503",
  "JNJ-75276617 (bleximenib)",
  "EPZ-011989 (EZH2i, comparator)"
]

How this could bite us back

Raising MHC class I is not unambiguously good. MHC class I is the ligand for inhibitory KIR and Ly49 receptors, and in the MLL-AF9 model the class I heavy chain H2-K1 is an in vivo leukaemia dependency that works by suppressing NK cells - so the same intervention that makes a blast visible to T cells makes it less visible to NK cells (see `risk-raising-mhc-class-i-trades-t-cell-visibility-for-nk-inhibition`).
The effect is largely an amplification of an IFN-gamma signal. In a lymphopenic, heavily pre-treated infant there may be no such signal to amplify, and the monotherapy effect on surface protein was small.
None of the three immunotherapy failures this graph documents - CD19 lineage switch, CLL-1 antigen-negative escape, the limits of blinatumomab - is MHC-mediated. CAR T cells and BiTEs engage their target independently of MHC, so raising MHC class I cannot fix any of them.
If the mechanism requires KMT2A/KMT2B catalysis, then combining a menin inhibitor with anything that removes KMT2A function - including a second KMT2A-axis agent - would be expected to abolish the immune benefit while preserving the cytoreduction, and the combination would look fine on a viability readout.
Bivalent gene derepression is not selective for MHC. The same drug action derepresses whatever else is bivalent in that cell, in normal tissue as well, and in pluripotent cells it drove spontaneous differentiation.

Still open

Are MHC class I promoters bivalent in a KMT2A-rearranged blast? A single H3K4me3/H3K27me3 ChIP-reChIP or CUT&Tag at HLA-A, HLA-B and HLA-C in SEM, RS4;11, MV4-11 and a primary infant sample would settle the premise the whole inference rests on, and nobody has done it.
Does the requirement for KMT2A/KMT2B methyltransferase activity make the index disease a poor host for this mechanism? The fusion allele has lost the SET domain; the wild-type allele and KMT2B have not. An allele-resolved experiment could measure whether the surviving enzyme dose is sufficient.
Is the effect on class I separable from differentiation? A menin inhibitor drives myeloid differentiation, and differentiated myeloid cells raise MHC on their own. No source in this beat reports a differentiation-matched control.
Does the heterogeneity (9 of 25) track genotype, baseline bivalency, or baseline MHC-I level? If it tracks baseline bivalency it is a predictive biomarker; if it tracks differentiation it is an epiphenomenon.

Every source behind this entry (4)

E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Whole-genome CRISPR-Cas9 screens in K-562 built on a cell-surface readout of bivalent MHC class I gene expression identified MTF2-PRC2.1, PCGF1-PRC1.1 and Menin-KMT2A/B as maintainers of bivalency; the Menin-KMT2A/B inhibitor VTP50469 activated MHC-I gene expression dose-dependently and enhanced IFN-gamma-induced surface MHC-I protein, the chemically distinct inhibitor MI-503 reproduced this, no further induction occurred in Menin- or LEDGF-deficient cells, menin was not detectably bound at bivalent MHC-I promoters, MEN1 knockout displaced KMT2A from 9,258 of 11,583 bound sites while 142 genes (63% bivalent) gained KMT2A with rising H3K4me3 and falling H3K27me3/SUZ12, combined KMT2A/KMT2B knockout impaired IFN-gamma-induced MHC-I and abolished the menin-inhibitor-induced increase, and pre-treating murine SCLC-OVA cells with VTP50469 before OT-I T-cell co-culture increased tumour killing and T-cell IFN-gamma production.
Targeting Menin disrupts the KMT2A/B and polycomb balance to paradoxically activate bivalent genes. · Sparbier CE, Gillespie A, Gomez J, Kumari N, Motazedian A, Chan KL, et al., Dawson MA · Nat Cell Biol 2023
PMID 36635503 DOI
What this evidence cannot show
The index cell line is K-562, a BCR::ABL1-positive CML line, and the other systems are human and murine SCLC, KELLY neuroblastoma, EZH2-mutant DLBCL and H9/iPSC pluripotent stem cells. No KMT2A-rearranged leukaemia cell appears anywhere in the paper; the string 'KMT2A-rearranged' occurs exactly once, in the sentence explaining the provenance of the inhibitors. Bivalency of MHC-I promoters was demonstrated by ChIP-reChIP in K-562 and is not shown for any leukaemic blast. The functional T-cell killing is murine OT-I against a model antigen (OVA) on H-2Kb, not a human leukaemia neoantigen. Almost every surface-protein effect is an enhancement of IFN-gamma-induced expression rather than an effect of the drug alone. Read: full text via Europe PMC fullTextXML (PMC7614190), 113,400 characters, 2026-09-20.
Evidence that argues against this (1)
- The paper's own KMT2A/KMT2B knockout arm is the strongest reason to doubt the extrapolation to KMT2A-rearranged disease: KMT2A/B are REQUIRED for the menin-inhibitor-induced MHC-I increase, so a cell in which one KMT2A allele has been converted into a SET-domain-less fusion is not obviously a cell in which the mechanism works. The paper also shows the requirement can be bypassed - once H3K27me3 is removed, IFN-gamma induces MHC-I in KMT2A/KMT2B double-knockout cells entirely through the NLRC5-RFX5 enhanceosome with SETD1A/B supplying H3K4me3 - so the direction of the net effect in a KMT2A-rearranged blast is genuinely unpredictable from this data. Read: full text (same paper). Sparbier CE et al. 2023
E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In primary AML patient cells, the menin-KMT2A inhibitor JNJ-75276617 (bleximenib) increased H3K4me3 and H3K27ac at the HLA-A locus and at HLA-DPA1/B1 and HLA-DRA, raised HLA-A and HLA-DR mRNA and surface protein, with HLA-DR up in the majority of cases, HLA-A up in 9 of 25 cases and both class I and class II consistently up in 5 of 25; the class II arm was MEIS1-independent and CIITA-dependent, with menin binding confirmed at CIITA and HLA-DR loci by ChIP-qPCR and HLA-DR induction lost in CIITA-null MV4-11 cells; and 4 of 8 primary AML samples became more sensitive to allogeneic T cells after treatment, all four of which carried NPM1c and DNMT3A mutations.
Bleximenib, the novel menin-KMT2A inhibitor JNJ-75276617, impairs long-term proliferation and immune evasion in acute myeloid leukemia. · Hogeling SM, Le DM, La Rose N, Kwon MC, Wierenga ATJ, Van den Heuvel FAJ, et al., Schuringa JJ · Haematologica 2025
PMID 39704147 DOI
What this evidence cannot show
AML only, adults, no lymphoid arm and no infant material. The class I effect is the minority outcome: HLA-A rose in 9 of 25 samples against HLA-DR in most, and the two rose together in 5 of 25. The functional immune benefit was demonstrated in NPM1c/DNMT3A-mutant disease - all four responders in the allogeneic assay and the whole autologous cohort (N=9) were NPM1c - so the immune arm of this paper is not a KMT2A-rearranged result even though the drug is a menin-KMT2A inhibitor. The authors state explicitly that the strength of the T-cell response was not correlated with the degree of MHC upregulation. Class I was measured as HLA-A transcript plus pan-MHC-I flow; HLA-B, HLA-C and B2M were not measured. Two of eleven authors are Janssen employees and the compound is Janssen's. Read: full text via the PMC HTML route (PMC12130779), 64,140 characters after tag-stripping, 2026-09-20.
Evidence that argues against this (1)
- The paper contains its own negative case: in sample AML54, bleximenib produced no HLA-A upregulation but strong HLA-DR upregulation, and T-cell cytotoxicity was nevertheless NOT enhanced - the AML cells expanded slightly in the presence of allogeneic T cells. Raised class II without raised class I did not buy T-cell control in that patient, and the authors attribute the strength of the response to unmeasured tumour-intrinsic immune-evasion mechanisms rather than to MHC level. Read: full text (same paper). Hogeling SM et al. 2025
E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In KMT2A-rearranged and NPM1-mutated AML, menin inhibition induced class II transactivator and MHC class II expression in vitro and in vivo through induction of multiple human endogenous retroviruses and consequent interferon-stimulated gene upregulation, sensitising AML cells to T-cell-mediated elimination after allogeneic transplant and enhancing graft-versus-leukaemia.
Menin inhibition enhances graft-versus-leukemia effects by T-cell activation and endogenous retrovirus induction in AML. · Fetsch V, Schwobel L, Ozyerli-Goknar E, Stell AV, Punta M, Plenge T, et al., Zeiser R · Blood 2026
PMID 41144759 DOI
What this evidence cannot show
This is the class II route and is cited here for contrast, not as class I evidence: the paper's mechanism runs through HERV derepression and interferon-stimulated genes, which Sparbier explicitly excludes for the class I effect (see `mechanism-class-i-derepression-under-menin-inhibition-is-not-viral-mimicry`). AML only, murine and xenograft transplant models. This graph's `cell-state-interferon-herv-mhc2` is the entry for this paper; it is not duplicated here. Read: abstract only.
Evidence that argues against this (1)
- Fetsch and Sparbier disagree about the route, not about the destination. Sparbier tested viral mimicry directly in the class I system and excluded it: neither STAT1 depletion, which is required for interferon-induced MHC-I upregulation, nor depletion of the cytoplasmic dsRNA sensor MDA-5 (IFIH1) impaired MHC-I gene activation following menin targeting, and RELA was not required either. An entry that folds class I into the HERV-interferon story would therefore be citing the wrong mechanism for it. Read: full text. Sparbier CE et al. 2023
E4moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset
As of 2026-09-20 the whole published intersection of menin and MHC class I is two papers: PubMed `menin[tiab] AND ("MHC class I"[tiab] OR "HLA class I"[tiab])` returns 2 records, PMID 39704147 and PMID 36635503, and the drug-name form `(menin inhibitor OR menin inhibitors OR menin inhibition OR revumenib OR ziftomenib OR bleximenib OR VTP-50469 OR VTP50469 OR MI-503 OR SNDX-5613 OR JNJ-75276617)[tiab] AND ("MHC class I"[tiab] OR "HLA class I"[tiab] OR HLA-A[tiab] OR B2M[tiab])` returns 3, the third being an unrelated liver-transplant graft-versus-host-disease report (PMID 29619911) excluded on title check.
Q54 antigen-presentation beat: bounded counted-absence sweep, PubMed E-utilities esearch with querytranslation read back -- menin x MHC class I, drug-name list and bare-operand control · Q54-immune, Oncofusion Knowledge Graph · Oncofusion Knowledge Graph 2026
What this evidence cannot show
Bounded absence. Rule 72 applied: the drug-name list carried singular, plural and nominalisation (inhibitor / inhibitors / inhibition), both hyphenations of VTP-50469, and four compound codes, and the bare operand `menin[tiab]` was run as the control. Rule 67 control: `B2M[tiab]` alone returns 2,102 and `HLA-A[tiab] AND leukemia` returns 741, so the instrument returns ones through this shape. The count is a floor: it cannot reach a paper that measures MHC-I without naming it in the title or abstract, which is exactly what PMID 39704147 does for HLA-A - the gene name is in the figures, not the abstract. That is rule 71 operating on this beat's own evidence.
Evidence that argues against this (1)
- The count of 2 understates the literature in the direction that matters. PMID 39704147's abstract says 'MHC class I' and never says HLA-A, so a symbol-keyed query returns it only by accident; `(KMT2A OR MLL) AND leukemia AND (HLA-A[tiab] OR HLA-B[tiab] OR HLA-C[tiab])` returns exactly one record, which title-checks to an unrelated alternative-donor transplant study (PMID 25402415). Any future absence claim about class I heavy chains in this disease has to be phrased at the level of the phrase 'MHC class I', not the gene symbols. Q54-immune et al. 2026

Connected to

part of →Trithorax against Polycomb at HOXA: the textbook antagonism, and the two ways this disease breaks it
the MHC class I arm of the polycomb-balance shift; same paper, PMID 36635503
regulates →H3K27me3 - the repressive mark this graph argued about for seventy-seven cycles without ever writing an entry for it
H3K27me3 falls at bivalent MHC-I promoters after MEN1 knockout, without a change in total H3K27me3
regulates →H3K4me3 - the mark KMT2A's own SET domain writes, that the fusion cannot write, that an interface in this graph already binds, and that switches PRC2 off
H3K4me3 rises at the 142 loci that gain KMT2A after menin loss; KMT2A/B are required for it
related to →The primed chromatin state — what is actually established, and why the word 'bivalent' should be used carefully here
MHC class I genes are the bivalent gene set this screen read out
related to →The interferon state is not baseline in this disease — it is something a menin inhibitor creates, through endogenous retrovirus induction and MHC class II
the class II/HERV route; same drug class, different and explicitly separable mechanism
← related toRevumenib (SNDX-5613, Revuforj) — first approved menin–KMT2A interface inhibitor
class effect inferred; the published class I experiments used VTP50469, MI-503 and bleximenib, not revumenib
← regulatesMEN1 / menin — the adaptor the fusion cannot work without
indirectly - menin is not bound at bivalent MHC-I promoters; the effect is KMT2A redistribution
← depends onKMT2A / MLL1 — the gene that breaks
combined KMT2A/KMT2B knockout abolishes the menin-inhibitor-induced MHC-I increase
← depends onKMT2B (MLL2/MLL4) - the other protein menin binds, and the one this graph never wrote
KMT2B knockout alone modestly impairs MHC-I; dominant at bivalent promoters
← related toLost from the KMT2A half: the SET domain (aa 3829-3945) - the lost module whose loss demonstrably does not matter
the fusion allele cannot supply the H3K4me3 this mechanism requires - stated as a prediction, not a finding
← regulatesPSIP1 / LEDGF p75 — the tether with the best therapeutic index in the whole interactome
LEDGF loss phenocopies menin loss for MHC-I; no further induction by VTP50469 in LEDGF-deficient cells
← regulatesPCGF1 — the PRC1.1 scaffold whose LOSS causes menin-inhibitor resistance, so it is a restraint on the leukaemia rather than a dependency of it
PCGF1-PRC1.1 restricts bivalent MHC-I activation; menin inhibition adds to PCGF1 knockout
← regulatesEED - the PRC2 subunit this disease actually needs, with drugs in the clinic and zero leukaemia studies
EED knockout derepresses MHC-I partially; menin inhibition augments it
← related toB2M (beta-2 microglobulin) - the invariant light chain every MHC class I molecule needs, the classic class I escape route, and a gene this disease's literature has never once named
the light chain every class I readout depends on and that no source in this beat measured
← part ofHLA-A - the class I heavy chain a menin inhibitor was actually measured on, and the human orthologue of the MHC class I molecule that protects MLL-AF9 leukaemia stem cells from NK cells
the only class I gene with a named, measured response to a clinical-stage menin inhibitor
← part ofHLA-B - the class I heavy chain the bivalency screen was actually read out on, and the locus that presents the one KMT2A epitope this graph has a structure for
the locus the bivalency screen was read out on, with the Bw6 clone REA143
← part ofNLRC5 - the MHC class I transactivator, and the route by which interferon activates class I when KMT2A and KMT2B are both gone
the transcription-factor route that bypasses the KMT2A/B requirement once H3K27me3 is removed
← supportsIFNGR1 - the receptor the whole antigen-presentation mechanism runs through, and the one where low-dose interferon-gamma makes MLL-fusion leukaemia worse
the menin-inhibitor class I effect is largely an enhancement of IFN-gamma-induced expression
← related toJAK1 - recruited and stabilised by AF10 fusions, and the kinase whose downstream STAT1 the class I effect of menin inhibition explicitly does not need
STAT1 downstream of JAK1 is NOT required for the menin-inhibitor class I effect
← part ofThe class I and class II arms of menin inhibition are different mechanisms - the class I effect was tested against viral mimicry and NF-kB and excluded both
the exclusion arm of the class I mechanism
← confers resistance toRaising MHC class I is not a free good: the same molecule is the inhibitory ligand that holds NK cells off, and in this fusion family it is an in vivo leukaemia dependency
the cost side of the intervention the mechanism entry describes
← analogous toAnother fusion family did the experiment this beat wants: in FUS::ERG and CBFA2T3::GLIS2 AML, MHC class I and II are absent at diagnosis, EZH2 drives it, and an EZH2 inhibitor restores presentation
polycomb-silenced MHC restored by an epigenetic inhibitor, measured in patient cohorts of another fusion family
← depends onHypothesis: a menin inhibitor must come BEFORE an antigen-directed immunotherapy, not with it or after it - and every published experiment has already assumed this without anyone testing it
the ordering rule only matters if the drug really creates the antigen
← testsThe missing measurement: is MHC class I bivalent, does a menin inhibitor raise it, and does raising it cost NK control - in a KMT2A-rearranged cell, for the first time
tests bivalency and drug response in a KMT2A-rearranged cell for the first time
related to →Graft-versus-leukaemia in KMT2A-rearranged infant leukaemia - the mechanism the whole transplant rests on, demonstrated in ALL generally, assumed in this disease, and with a published mechanistic reason to doubt it
the class I arm of the antigen-presentation question; the GVL entry owns the transplant half
← supportsThe MHC was never invisible - the resolver was wrong - and once the classical chains are joined the class I machinery is bivalent, not unmarked
AN ORTHOGONAL LINE TO THE SAME PLACE: a published occupancy table, read independently, puts the class I locus in the bivalent group.