Oncofusion

start / graph / cell-state-stem-like-self-renewal

Cell statestrong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.Q4-cells

The stem-like self-renewal state — a limited stem-cell module grafted onto a committed cell

also called leukaemia stem cell state · LSC self-renewal program · self-renewal-associated signature · stemness module

Normal blood cells lose the ability to copy themselves as they mature. KMT2A fusions switch a small part of that self-copying program back on without making the cell a stem cell again, which is what lets an already-committed cell keep the leukaemia going forever.

The same thing, for a scientist

The self-renewal state in KMT2A-rearranged leukaemia is best characterised not as reversion to a haematopoietic stem cell identity but as re-activation of a restricted, self-renewal-associated gene module inside a cell that otherwise retains the global identity of the progenitor it came from. The defining experiment showed that MLL-AF9 leukaemia stem cells arising in committed granulocyte-macrophage progenitors transferred disease from four cells while keeping a GMP-like immunophenotype and global expression profile, re-activating only a limited subset of normal HSC genes. The same directionality is visible in normal human cells: MLL-AF4 expression in cord blood CD34-positive HSPCs measurably enhances repopulating and clonogenic capacity without causing leukaemia, so the self-renewal gain is separable from, and precedes, transformation.

Go deeper
The defining measurement: four cells1 deeper

Leukaemia stem cells isolated from leukaemias initiated by MLL-AF9 in committed granulocyte-macrophage progenitors transferred leukaemia to secondary recipients when only four cells were transferred, while possessing an immunophenotype and global gene expression profile very similar to normal GMPs. A subset of genes highly expressed in normal haematopoietic stem cells was re-activated in the leukaemia stem cells 1.

That sentence contains the whole conceptual shift. Cancer stem cells do not have to look like tissue stem cells. Self-renewal is a module that can be switched on in an inappropriate context, and the paper's own conclusion is the therapeutic one: targeting a self-renewal programme expressed in an abnormal context may be possible 1. The context-dependence is the therapeutic window.

Why this is more hopeful than a stem-cell model

If the leukaemia-propagating cell were a genuine haematopoietic stem cell, anything that killed it would kill normal haematopoiesis. Because it is a committed progenitor running a borrowed module, there is at least a logical target: the module in that context. This is the same logic that makes the menin-KMT2A interface attractive, and it is the reason menin can be not broadly essential in DepMap yet sharply required by KMT2A-rearranged lines.

Self-renewal gain is measurable before leukaemia exists

The cleanest evidence that self-renewal is an early, separable effect of the fusion rather than a late consequence of transformation comes from normal human cells. Enforced MLL-AF4 in human cord blood CD34-positive HSPCs enhanced haematopoietic repopulating cell function and clonogenic potential, promoted proliferation, conveyed a survival advantage with changes in apoptosis, DNA-damage-sensing and DNA-repair gene expression, and improved multilineage engraftment and homing after intra-bone-marrow transplantation, without initiating leukaemia 2.

So the fusion buys self-renewal first. Leukaemia requires something else, and in the infant setting the leading candidate for that something else is the fetal developmental context.

In the murine Mll-AF4 system the propagating compartment is explicit and lineage-dependent: miR-130b-driven mixed B-cell-precursor/myeloid disease was propagated by the LMPP population, while miR-128a-driven pro-B ALL was maintained by an expanded Il7r-positive Kit-positive blast population 3. Two different self-renewing compartments from one fusion.

The essential-function problem sitting underneath

Any therapy that attacks self-renewal in the leukaemia is attacking a process normal blood also needs, and for menin the overlap is documented. Conditional menin deletion in mouse haematopoiesis modestly reduced neutrophil, lymphocyte and platelet counts and decreased B lymphoid progenitors even without stress; multilineage and myelo-erythroid progenitor numbers were preserved at steady state; but competitive transplantation revealed a marked functional defect in long-term haematopoietic stem cells, with the phenotype most apparent during haematopoietic recovery. HoxA9 expression was only modestly reduced in menin-deficient HSCs, implying the HSC role is not simply HOXA-mediated 4.

Two uncomfortable readings follow. First, the toxicity that matters will surface under stress, which is exactly the post-induction state of a relapsed infant. Second, menin's normal HSC function appears not to run through the HOXA axis that the drug is designed to shut down, so the therapeutic index is not as cleanly mechanism-separated as the marketing story implies.

Facts and figures
state definitionRe-activation of a restricted, self-renewal-associated gene module in a cell that otherwise retains the global identity of its parent progenitor; functionally defined by limiting-dilution serial transplantability
normal counterpart dependencymenin is required for long-term HSC function under competitive stress, and its HSC role appears largely HOXA9-independent
functional readouts
[
  {
    "assay": "limiting-dilution secondary transplantation",
    "result": "leukaemia transferred by 4 cells (MLL-AF9 GMP-derived LSC)",
    "pmid": "16862118",
    "title": "Transformation from committed progenitor to leukaemia stem cell initiated by MLL-AF9.",
    "authors": "Krivtsov AV, Twomey D, Feng Z, et al.",
    "venue": "Nature",
    "year": 2006,
    "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."
  },
  {
    "assay": "human repopulating cell and clonogenic assays",
    "result": "enhanced by MLL-AF4 in cord blood CD34+ HSPC without leukaemia",
    "pmid": "21389315",
    "title": "Enforced expression of MLL-AF4 fusion in cord blood CD34+ cells enhances the hematopoietic repopulating cell function and clonogenic potential but is not sufficient to initiate leukemia.",
    "authors": "Montes R, Ayllón V, Gutierrez-Aranda I, et al.",
    "venue": "Blood",
    "year": 2011,
    "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."
  }
]
propagating compartments reported
[
  "GMP-like (MLL-AF9, mouse)",
  "LMPP (Mll-AF4 plus miR-130b, mouse)",
  "Il7r+Kit+ pro-B blast (Mll-AF4 plus miR-128a, mouse)"
]
markers
[
  "retains parent-progenitor immunophenotype",
  "limited HSC gene module re-expressed",
  "NG2 is NOT a marker of this state in MLL-r infant B-ALL"
]
How this could bite us back
  • Self-renewal is a normal HSC function, so attacking it risks impairing haematopoietic recovery exactly when a relapsed infant needs it most.
  • If the propagating compartment differs by co-driver, then a therapy validated in one model may target a compartment that a given patient does not depend on.
  • A self-renewal module grafted onto a committed cell means the reservoir does not need to be a stem cell, so stem-cell-directed strategies such as niche disruption may miss it entirely.
Still open
  1. What is the human KMT2A::AFF1 leukaemia-propagating cell, phenotypically, and is it menin-dependent? Almost all of the self-renewal evidence in this node is mouse or cord blood, not human infant disease.
  2. Menin's normal long-term HSC function appears not to run through HOXA9. If the drug blocks the HOXA-dependent leukaemic function and the HOXA-independent normal function equally, the therapeutic index is narrower than assumed. Which residues or partner interactions separate the two?
  3. Is the self-renewal module the same in the ALL and AML forms of KMT2A::AFF1 disease, so that a single agent covers both, or does lineage switch also switch the self-renewal program?
  4. Can the self-renewal module be scored from a diagnostic single-cell dataset and used as a pharmacodynamic readout of menin inhibition, instead of waiting for marrow blast counts?
Every source behind this entry (4)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    MLL-AF9 leukaemia stem cells generated in committed granulocyte-macrophage progenitors transfer leukaemia from as few as four cells while retaining a GMP-like immunophenotype and global expression profile, with re-activation of a limited subset of normal haematopoietic stem cell genes, showing that leukaemia stem cells can arise from committed progenitors without widespread reprogramming.

    Transformation from committed progenitor to leukaemia stem cell initiated by MLL-AF9 · Krivtsov AV, Twomey D, Feng Z, Stubbs MC, Wang Y, Faber J, Levine JE, Wang J, Hahn WC, Gilliland DG, Golub TR, Armstrong SA · Nature 2006
    PMID 16862118DOI
    What this evidence cannot show

    Mouse retroviral MLL-AF9 model of myeloid leukaemia. Limiting-dilution transplantation measures transplantability in an irradiated recipient, which is not identical to self-renewal in a patient. 2006-era microarray profiling cannot resolve the heterogeneity single-cell methods now reveal within a GMP-like population.

    Evidence that argues against this (1)
    • At endogenous fusion dosage the committed progenitor is NOT the cell that transforms. With Mll-AF9 under endogenous regulatory control, LSK stem cells were efficiently transformed while committed granulocyte-monocyte progenitors were transformation-resistant and produced no leukaemia; GMPs were transformable only at the much higher fusion dosage delivered by retroviral transduction, which is the system the four-cell transfer result used. The 'limited stem-cell module grafted onto a committed cell' framing may therefore be describing the phenotype of an experimentally over-expressed fusion rather than the natural disease. Chen W et al. 2008
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Enforced MLL-AF4 expression in human cord blood CD34-positive HSPCs enhances haematopoietic repopulating cell function and clonogenic potential, promotes proliferation and conveys a survival advantage with altered apoptosis, DNA-damage-sensing and DNA-repair gene expression, while being insufficient to initiate leukaemia, showing that the self-renewal gain is separable from transformation.

    Enforced expression of MLL-AF4 fusion in cord blood CD34+ cells enhances the hematopoietic repopulating cell function and clonogenic potential but is not sufficient to initiate leukemia · Montes R, Ayllon V, Gutierrez-Aranda I, Prat I, Hernandez-Lamas MC, Ponce L, Bresolin S, te Kronnie G, Greaves M, Bueno C, Menendez P · Blood 2011
    PMID 21389315DOI
    What this evidence cannot show

    Enforced lentiviral expression at non-physiological levels, without the reciprocal AF4-MLL product. Repopulating capacity is measured in immunodeficient mice, where human self-renewal is only partly supported.

    Evidence that argues against this (1)
    • The 'insufficient to initiate leukemia' conclusion was overturned by a construct change rather than a biology change. Using MLL fused to murine Af4 - highly conserved with human AF4 but permitting high-titre retrovirus - transduction of human CD34+ cells generated a pro-B ALL reported to fully recapitulate the immunophenotypic and molecular features of t(4;11) disease. That argues the earlier negative results in cord-blood CD34+ cells reflected inadequate fusion expression, not a genuine requirement for a second hit. Lin S et al. 2016
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    In a murine Mll-AF4 model the self-renewing propagating compartment depends on the co-driver: miR-130b-driven mixed B-cell-precursor/myeloid leukaemia was propagated by the LMPP population, whereas miR-128a-driven pro-B ALL was maintained by a highly expanded Il7r-positive Kit-positive blast population.

    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 strong artificial perturbation; the identity of the propagating compartment may be a function of co-driver dose. Murine Mll-AF4 models do not faithfully 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
  4. E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Conditional menin loss in mouse haematopoiesis modestly reduces blood counts and decreases B lymphoid progenitors at steady state while preserving multilineage and myelo-erythroid progenitor numbers, yet competitive transplantation reveals a marked long-term HSC functional defect that is most apparent during haematopoietic recovery; HoxA9 expression is only modestly decreased in menin-deficient HSCs.

    Menin regulates the function of hematopoietic stem cells and lymphoid progenitors · Maillard I, Chen YX, Friedman A, Yang Y, Tubbs AT, Shestova O, Pear WS, Hua X · Blood 2009
    PMID 19228930DOI
    What this evidence cannot show

    Complete conditional deletion in mouse is harsher than reversible pharmacological occupancy, and competitive transplantation is a severe stress test that may overstate the clinical effect. Mouse HSC biology under irradiation stress is an imperfect model of human post-chemotherapy recovery.

    Evidence that argues against this (1)
    • An independent conditional-genetics study reaches the opposite practical conclusion and is the reason the field considered menin-MLL disruption safe enough to drug. Comparing menin-null and Mll1-null haematopoiesis, menin and MLL1 were found to regulate largely DISTINCT gene sets in HSCs and developing B cells, with no genetic interaction detectable in steady-state or regenerative haematopoiesis or in B-cell differentiation despite MLL1 being critical for those processes; only a small shared set (including Hoxa9, Meis1, Mecom, Eya1) was co-regulated. The authors conclude menin is not a requisite cofactor for MLL1 in normal haematopoiesis and that this supports menin-MLL1-disrupting drugs as safe and selective. If both reports are right, the HSC defect on menin loss is real but menin-independent of the MLL1 axis the drug targets. Li BE et al. 2013
Connected to