Oncofusion

start / graph / cell-state-primed-bivalent-chromatin

Cell statemoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q4-cells

The primed chromatin state — what is actually established, and why the word 'bivalent' should be used carefully here

also called primed chromatin · poised chromatin state · bivalent domain state · epigenetic priming in KMT2A-r

The leukaemia cells keep two sets of instructions accessible at once: the B-cell program they are running and the myeloid program they are not. That dual readiness is what lets them change lineage later, and it is written in the chemical marks on their DNA and histones.

The same thing, for a scientist

There is strong, primary evidence for a distinctive chromatin state in KMT2A::AFF1 leukaemia: fusion and menin binding spread into gene bodies at unmethylated CpG regions with elevated H3K79me2/3 and disrupted H3K36me3, and t(4;11) is one of the extensively promoter-hypermethylated infant ALL subtypes while t(9;11) is not. There is also strong functional evidence for lineage priming, in that the cells retain and can express a myeloid program and share a core regulatory network with the AML form of the disease. What this cycle could NOT verify is the specific claim that these loci carry canonical H3K4me3/H3K27me3 bivalent domains in KMT2A-rearranged leukaemia. That claim is treated here as an open question rather than a finding, and the node is deliberately named so that the gap is visible rather than buried.

Go deeper
What is established: the chromatin address of the fusion1 deeper

At a subset of target genes, MLL-AF4 binding spreads into the gene body. The spreading is accompanied by 1:

featuredirection
menin bindingspreads with the fusion
transcriptionincreased
H3K79me2/3increased
H3K36me3normal patterning disrupted
CpG methylation in the gene bodyunmethylated regions

And it carries a therapeutic prediction that was tested in the same paper: compared with other H3K79me2/3-marked genes, spreading-gene expression is preferentially downregulated by DOT1L inhibitors, which then synergise with additional targeted agents 1.

The unmethylated-CpG association is mechanistically important because every KMT2A fusion retains the CXXC domain, which binds unmethylated CpG DNA. The fusion is not choosing targets arbitrarily; it is reading a DNA methylation state.

DNA methylation: the second layer, and it is fusion-partner-specific

Promoter CpG island methylation separates genetic subtypes of MLL-rearranged infant ALL. t(4;11) and t(11;19) are extensively hypermethylated; t(9;11)-positive infant ALL and MLL-wild-type infant ALL epigenetically resemble normal bone marrow. Among t(4;11)/t(11;19) patients, the degree of promoter hypermethylation influenced relapse-free survival, with accentuated methylation marking high relapse risk, and the demethylating agent zebularine reversed the aberrant methylation and induced apoptosis in MLL-rearranged ALL cells 2.

This is one of the most actionable and least exploited findings in the dossier. It says that t(4;11) specifically, not KMT2A-rearrangement generally, carries a pharmacologically reversible silencing layer on top of the fusion's activating layer. A hypomethylating agent plus a menin inhibitor is a combination the biology recommends and, as far as this cycle could establish, one that has not been prioritised. It also predicts that testing a hypomethylating combination across unselected KMT2A-r disease would dilute a real t(4;11) effect with inactive t(9;11) cases.

What is established: functional priming for two lineages

Priming is not only a chromatin-mark claim; it has direct functional readouts.

  • MLL-Af4-transformed human cells can assume a myeloid state under environmental pressure while retaining lymphoid-lineage potential, matched by t(4;11) patients whose leukaemia evaded CD19-directed therapy by myeloid lineage switch 3.
  • In a murine Mll-AF4 system, one co-driver produces a mixed B-cell-precursor/myeloid leukaemia propagated by LMPPs and another produces a pro-B ALL, from the same fusion 4.
  • The KMT2A-AFF1 gene regulatory network contains a core set of nodes present in both the ALL and the AML forms of the disease 5.

Taken together: the regulatory apparatus is pre-wired for both outcomes, and which one manifests is set by context. That is what 'primed' should mean operationally, and it is supported without needing to invoke histone bivalency at all.

What is NOT established: canonical bivalency in this disease1 deeper

Bivalent domains, carrying H3K4me3 and H3K27me3 together, are a well-established feature of developmentally poised promoters in embryonic stem cells and progenitors. It is a natural and widely repeated inference that the lineage-primed state of KMT2A-rearranged leukaemia is written as bivalency at myeloid and B-lineage promoters.

This cycle could not verify a primary paper demonstrating H3K4me3/H3K27me3 bivalent domains in KMT2A-rearranged leukaemia specifically. Repeated targeted literature searches returned no such primary study. Rather than cite a review as if it were data, or state the claim with a confidence it has not earned, the claim is recorded here as an explicit gap.

The distinction matters practically. If priming is written as bivalency, then EZH2/PRC2 inhibition should release the silent arm and is a rational combination partner. If priming is instead written as DNA hypermethylation plus retained CXXC-readable unmethylated CpG islands plus H3K79me2/3 spreading — which is what the verified evidence describes 12 — then the rational partner is a hypomethylating agent or a DOT1L inhibitor, and PRC2 inhibition would be the wrong drug. These lead to different trials.

The experiment that would settle it

Sequential or single-cell CUT&Tag for H3K4me3, H3K27me3, H3K79me2 and H3K36me3 on primary infant KMT2A::AFF1 blasts, with matched fetal PreProB and fetal LMPP references, before and after menin inhibition. The readout is whether myeloid-lineage promoters in the leukaemic cell carry both active and repressive marks in the same cell, and whether menin inhibition resolves them toward the active or the repressive arm. This is written up as exp-bivalency-cuttag-kmt2a-aff1.

Facts and figures
state definitionSimultaneous accessibility of the executed B-lymphoid program and the unexecuted myeloid program, evidenced functionally by inducible lineage switch and genomically by a shared ALL/AML regulatory core, and mechanistically by fusion and menin spreading with H3K79me2/3 at unmethylated CpG regions on a background of subtype-specific promoter hypermethylation
therapeutic discriminantIf priming is bivalency, PRC2/EZH2 inhibition is the rational partner. If priming is DNA hypermethylation plus CXXC-readable unmethylated CpG plus H3K79me2/3 spreading, the rational partners are a hypomethylating agent or a DOT1L inhibitor. The two hypotheses imply different trials.
verified chromatin features
[
  "MLL-AF4 and menin spreading into gene bodies",
  "elevated H3K79me2/3 over spreading targets",
  "disrupted H3K36me3 patterning",
  "unmethylated CpG regions in gene bodies of spreading targets, readable by the retained CXXC domain",
  "extensive promoter CpG island hypermethylation in t(4;11) and t(11;19); absent in t(9;11)"
]
unverified claims
[
  "Canonical H3K4me3/H3K27me3 bivalent domains at lineage promoters in KMT2A-rearranged leukaemia. No primary demonstration was found in this cycle; treated as an open question, not a finding."
]
pharmacological probes
[
  {
    "agent": "DOT1L inhibitor",
    "effect": "preferential downregulation of MLL-AF4 spreading genes",
    "pmid": "28076791",
    "title": "MLL-AF4 Spreading Identifies Binding Sites that Are Distinct from Super-Enhancers and that Govern Sensitivity to DOT1L Inhibition in Leukemia.",
    "authors": "Kerry J, Godfrey L, Repapi E, et al.",
    "venue": "Cell Rep",
    "year": 2017,
    "title_provenance": "Title/authors/venue backfilled from PubMed esummary on 2026-09-20 because this citation stored a bare PMID with no title, which made it unscoreable by verify-citations.py. BACKFILLING IS NOT VERIFICATION: it confirms the PMID resolves to a real paper, not that this is the right paper for the claim. That still requires reading."
  },
  {
    "agent": "zebularine (demethylating)",
    "effect": "reversal of aberrant methylation and apoptosis in vitro",
    "pmid": "19855078",
    "title": "Specific promoter methylation identifies different subgroups of MLL-rearranged infant acute lymphoblastic leukemia, influences clinical outcome, and provides therapeutic options.",
    "authors": "Stumpel DJ, Schneider P, van Roon EH, et al.",
    "venue": "Blood",
    "year": 2009,
    "title_provenance": "Title/authors/venue backfilled from PubMed esummary on 2026-09-20 because this citation stored a bare PMID with no title, which made it unscoreable by verify-citations.py. BACKFILLING IS NOT VERIFICATION: it confirms the PMID resolves to a real paper, not that this is the right paper for the claim. That still requires reading."
  }
]
How this could bite us back
  • If the primed myeloid program is opened rather than closed by releasing the differentiation block, menin inhibition could actively facilitate lineage switch.
  • Promoter hypermethylation gives t(4;11) a second, menin-independent silencing layer, so differentiation genes may stay off after the fusion is neutralised.
  • The DOT1L prediction from spreading biology was clean in vitro and did not deliver as single-agent therapy, which is a warning about how far chromatin-state reasoning transfers to the clinic.
  • Asserting bivalency without primary evidence would send a combination programme toward PRC2 inhibitors when the verified data point at hypomethylating agents instead.
Still open
  1. Do lineage promoters in primary infant KMT2A::AFF1 blasts actually carry H3K4me3 and H3K27me3 in the same cell? No primary study establishing this was found, and the answer determines whether PRC2 inhibition is a rational combination partner.
  2. Given that t(4;11) is hypermethylated and t(9;11) is not, is a hypomethylating agent plus a menin inhibitor a t(4;11)-specific combination that would appear inactive if tested across unselected KMT2A-rearranged disease?
  3. Does menin inhibition change the chromatin state at myeloid promoters, and in which direction? If it opens them further, the drug may be priming the lineage switch it is meant to prevent.
  4. Is the CXXC-readable unmethylated CpG landscape different in fetal versus postnatal progenitors, which would give a chromatin-level explanation for the fetal requirement?
Every source behind this entry (6)
  1. E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    MLL-AF4 binding spreads into the bodies of a subset of target genes together with spreading menin binding, increased transcription, increased H3K79me2/3, disruption of normal H3K36me3 patterns and unmethylated CpG regions in the gene body; spreading-gene expression is preferentially downregulated by DOT1L inhibition, which synergises with other targeted drugs.

    MLL-AF4 Spreading Identifies Binding Sites that Are Distinct from Super-Enhancers and that Govern Sensitivity to DOT1L Inhibition in Leukemia · Kerry J, Godfrey L, Repapi E, Tapia M, Blackledge NP, Ma H, Ballabio E, O'Byrne S, Ponthan F, Heidenreich O, Roy A, Roberts I, Konopleva M, Klose RJ, Geng H, Milne TA · Cell Reports 2017
    PMID 28076791DOI
    What this evidence cannot show

    Cell-line-dominated ChIP-seq with limited primary material; 'spreading' is defined by a computational threshold, so the gene set is threshold-dependent. This paper does not report H3K4me3/H3K27me3 bivalency and should not be cited for it. In vitro DOT1L sensitivity did not translate into single-agent clinical benefit for pinometostat.

    Evidence that argues against this (1)
    • Spreading-associated H3K79me2/3 is not load-bearing for the primed state once the cell has adapted. Paediatric KMT2A::AFF1 ALL cells driven to pinometostat resistance became largely independent of DOT1L-mediated H3K79 methylation while still requiring the physical presence of DOT1L, HOXA9 and the fusion; they selectively lost some fusion target genes and upregulated myeloid genes including CD33 and LILRB4. Priming therefore survives the loss of the mark that is supposed to encode it, which argues H3K79me2/3 is a correlate of the state rather than its substrate - and it is the myeloid half of the primed program that is retained. Schneider P et al. 2023
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Promoter CpG island methylation patterns separate MLL-rearranged infant ALL subtypes, with t(4;11) and t(11;19) extensively hypermethylated and t(9;11) or MLL-wild-type infant ALL resembling normal bone marrow; the degree of hypermethylation influenced relapse-free survival among t(4;11)/t(11;19) patients, and zebularine reversed the aberrant methylation and induced apoptosis.

    Specific promoter methylation identifies different subgroups of MLL-rearranged infant acute lymphoblastic leukemia, influences clinical outcome, and provides therapeutic options · Stumpel DJPM, Schneider P, van Roon EHJ, Boer JM, de Lorenzo P, Valsecchi MG, de Menezes RX, Pieters R, Stam RW · Blood 2009
    PMID 19855078DOI
    What this evidence cannot show

    Differential methylation hybridisation is a lower-resolution platform than current bisulfite sequencing, so calls are subtype-level rather than locus-precise. Zebularine data are in vitro with no clinical translation in infant ALL. Survival analysis is retrospective and subgroup-based.

    Evidence that argues against this (1)
    • The 'therapeutic options' half of this 2009 title has been tested and did not deliver. COG AALL15P1 primed infants with newly diagnosed KMT2A-rearranged ALL with azacitidine before each of four post-induction chemotherapy courses; whole-genome bisulfite sequencing confirmed decreased methylation in 87% of samples tested, the regimen was safe, and event-free survival was similar to prior studies with no impact on survival. Promoter hypermethylation can be subgroup-defining, prognostic, and pharmacologically reversible without being the thing that has to be reversed. Guest EM et al. 2024
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    MLL-Af4-transformed cells can assume a myeloid state under environmental pressure while retaining lymphoid-lineage potential, matching t(4;11) patients whose leukaemia evaded CD19-directed therapy by myeloid lineage switch.

    Instructive Role of MLL-Fusion Proteins Revealed by a Model of t(4;11) Pro-B Acute Lymphoblastic Leukemia · Lin S, Luo RT, Ptasinska A, Kerry J, Assi SA, Wunderlich M, Imamura T, Kaberlein JJ, Rayes A, Althoff MJ, Anastasi J, O'Brien MM, Meetei AR, Milne TA, Bonifer C, Mulloy JC, Thirman MJ · Cancer Cell 2016
    PMID 27846391DOI
    What this evidence cannot show

    'Environmental pressure' means culture and xenograft conditions rather than a defined molecular signal; the patient observation establishes existence rather than frequency.

    Evidence that argues against this (1)
    • This is cited in two entries of this beat for two different conclusions. What it shows is that a murine model and human patients both exhibit lineage plasticity; it does not show that bivalent chromatin is the substrate, which is what this entry needs. eLife 2021
  4. E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    The same murine Mll-AF4 fusion produces a mixed B-cell-precursor/myeloid leukaemia propagated by LMPPs with one microRNA co-driver and a pro-B ALL with another, demonstrating that lineage output is context-determined rather than fusion-determined.

    miR-130b and miR-128a are essential lineage-specific codrivers of t(4;11) MLL-AF4 acute leukemia · Malouf C, Antunes ETB, O'Dwyer M, Jakobczyk H, Sahm F, Landua SL, Anderson RA, Soufi A, Halsey C, Ottersbach K · Blood 2021
    PMID 34111240DOI
    What this evidence cannot show

    MicroRNA overexpression is a non-physiological perturbation; murine Mll-AF4 models do not fully reproduce human t(4;11) infant ALL.

    Evidence that argues against this (1)
    • This is murine Mll-Af4 with an enforced microRNA co-driver, and the co-driver is doing the work the conclusion is attributed to. Showing that miR-130b gives an LMPP-propagated mixed leukaemia while miR-128a gives a pro-B ALL demonstrates that TWO DIFFERENT ONCOGENIC COMBINATIONS give two diseases, which is weaker than "lineage output is context-determined": no patient carries either microRNA at those levels. Oncofusion knowledge graph (internal analysis) 2026
  5. E5moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    The KMT2A-AFF1 gene regulatory network, built from ChIP-seq, patient RNA-seq and CRISPR essentiality screens, contains a core set of nodes present in both the ALL and the AML forms of the disease, with RUNX1-centred feed-forward loops and cascades.

    A KMT2A-AFF1 gene regulatory network highlights the role of core transcription factors and reveals the regulatory logic of key downstream target genes · Harman JR, Thorne R, Jamilly M, Tapia M, Crump NT, Rice S, Beveridge R, Morrissey E, de Bruijn MFTR, Roberts I, Roy A, Fulga TA, Milne TA · Genome Research 2021
    PMID 34088716DOI
    What this evidence cannot show

    Inferred network with only a subset of edges experimentally validated; input data are cell-line-heavy.

    Evidence that argues against this (1)
    • The gene regulatory network is a reconstruction, not a measurement. It is assembled from ChIP-seq occupancy, bulk patient RNA and CRISPR essentiality in lines, and edges inferred by combining three assays inherit the weakest one: occupancy does not imply regulation, bulk RNA averages over the clusters this beat says the disease splits into, and essentiality comes from cell lines whose lineage composition this project has shown to be confounded. "A core set of nodes shared between the ALL and AML forms" is also the expected result of reconstructing a network from genes both forms express. Oncofusion knowledge graph (internal analysis) 2026
  6. E6strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Infant KMT2A-rearranged leukaemias can be separated into five integrative clusters by combined RNA, methylation, exome and targeted sequencing, with differentiation-stage assignments spanning B-lymphopoietic and early hemato-endothelial development, confirming that DNA methylation carries subgroup-defining information in this disease.

    Multi-omics analysis defines highly refractory RAS burdened immature subgroup of infant acute lymphoblastic leukemia · Isobe T, Takagi M, Sato-Otsubo A, Nishimura A, Nagae G, Yamagishi C, Tamura M, Tanaka Y, Asada S, Takeda R, Tsuchiya A, Wang X, Yoshida K, Nannya Y, Ueno H, Akazawa R, Kato I, Mikami T, Watanabe K, Sekiguchi M, Seki M, Kimura S, Hiwatari M, Kato M, Fukuda S, Tatsuno K, Tsutsumi S, Kanai A, Inaba T, Shiozawa Y, Shiraishi Y, Chiba K, Tanaka H, Kotecha RS, Cruickshank MN, Ishikawa F, Morio T, Eguchi M, Deguchi T, Kiyokawa N, Arakawa Y, Koh K, Aoki Y, Ishihara T, Tomizawa D, Miyamura T, Ishii E, Mizutani S, Wilson NK, Gottgens B, Miyano S, Kitamura T, Goyama S, Yokoyama A, Aburatani H, Ogawa S, Takita J · Nature Communications 2022
    PMID 36042201DOI
    What this evidence cannot show

    Methylation arrays interrogate a fixed probe set and cannot report histone marks; cluster assignment is a multi-omics consensus, not a chromatin-state measurement. Predominantly Japanese cohort.

    Evidence that argues against this (1)
    • Five integrative clusters were derived by clustering 84 cases on combined RNA, methylation, exome and targeted sequencing, and the number five is a choice made during the analysis rather than a count of things in the world. Differentiation-stage assignment is by similarity to normal reference profiles, so a leukaemia that expresses a fetal programme will be assigned a fetal stage whether or not it is arrested there. Five clusters spanning B-lymphopoietic and early hemato-endothelial stages is also compatible with the fusion arising at different points in one continuum, which is a different claim from chromatin being primed. Oncofusion knowledge graph (internal analysis) 2026
Connected to