The OXPHOS-dependent state — established in KMT2A-rearranged AML, and inverted in KMT2A-rearranged B-ALL where OXPHOS inhibition makes leukaemia-initiating cells appear
also called oxidative phosphorylation dependence · OXPHOS-high state · mitochondrial respiration dependence · metabolic state of KMT2A-rearranged blasts
Cells make energy two ways: burning fuel with oxygen in the mitochondria, or a faster, cruder route that does not need oxygen. Myeloid KMT2A-rearranged leukaemia leans hard on the mitochondrial route and can be attacked there, but in the lymphoid form of the same disease blocking that route backfires — it shrinks the bulk tumour while increasing the cells that can restart it.
The same thing, for a scientist
This entry states what is **published** about the metabolic state of KMT2A-rearranged blasts, and the honest headline is that the published position splits by lineage rather than by fusion. In AML, MLL/AF9-genotype cell lines showed upregulated mitochondrial activity and depended mainly on OXPHOS for energy production within a metabolically heterogeneous AML panel, and metformin repressed their proliferation by inhibiting mitochondrial respiration [E1]. In MLL-rearranged **B-ALL**, the opposite operational conclusion was reached by single-cell transcriptomics plus quantitative xenotransplantation: *"while inhibiting oxidative phosphorylation blunts blast proliferation, this intervention promotes LIC emergence"*, whereas inhibiting hypoxia and glycolysis impaired leukaemia-initiating cells and gave therapeutic benefit in xenografts [E2]. A third strand is neither AML nor ALL-specific and is the closest published analogue of this project's own heme finding: **de novo heme biosynthesis is a selective dependency in AML**, underpinned by a propensity of AML cells and especially LSCs to downregulate heme biosynthesis enzymes, with inhibition collapsing mitochondrial Complex IV and inducing cuproptosis, and with **glycolysis synthetic-lethal with heme biosynthesis** [E3]. Read together these say that 'the metabolic state of KMT2A-rearranged leukaemia' is not one state, that the bulk-versus-initiating-cell axis matters more than the bulk measurement, and that an OXPHOS-directed agent given to infant ALL could enrich for exactly the compartment that causes relapse.
What is actually established, split by lineage
Myeloid. Across an AML cell-line panel with demonstrable metabolic heterogeneity, the MLL/AF9-bearing lines had upregulated mitochondrial activity and depended primarily on OXPHOS for ATP; metformin, a Complex I inhibitor, repressed their proliferation through inhibition of mitochondrial respiration 1. This is a cell-line study with a genotype comparison, not a patient-sample study.
Lymphoid. In MLL-rearranged B-ALL, LICs were characterised by single-cell transcriptomics combined with quantitative (limiting-dilution) xenotransplantation. Two results matter here. First, engraftable LICs in MLL-r B-ALL are abundant compared with reported frequencies in AML, and while multipotent self-renewing LICs are enriched among phenotypically undifferentiated blasts, LICs able to replenish the full leukaemic diversity can also emerge from more mature fractions 2. Second, and decisively for this entry, OXPHOS inhibition blunted blast proliferation but promoted LIC emergence, while hypoxia and glycolysis inhibition impaired LICs and gave therapeutic benefit 2.
Those two findings do not contradict each other as biology — a myeloid blast and a lymphoid LIC are different cells — but they do contradict each other as a therapeutic instruction, and the disease shares a founding lesion across both.
The heme rheostat, and why this project should read it before trusting its own scan
This project's internal DepMap work found the whole haem biosynthesis pathway selectively required in its KMT2A-rearranged arm (median rank 115 of 17,728) with the monocytic-lineage confound unresolved. That is an internal result and is not cited here as evidence.
What the published literature now says is close and instructive. Integrated analysis of mouse models, human cell lines and primary patient samples identifies de novo heme biosynthesis as a selective dependency in AML — not in KMT2A-rearranged AML specifically, in AML generally 3. The mechanism runs the opposite way to the naive reading: AML cells, and leukaemic stem cells especially, downregulate heme biosynthesis enzymes, and that downregulation promotes self-renewal. Inhibiting the enzymes collapses mitochondrial Complex IV, dysregulates the copper-chaperone system and induces cuproptosis. Glycolysis is among the synthetic-lethal partners 3.
Two consequences for this graph. (1) A pathway-level haem dependency in a KMT2A-rearranged arm is consistent with a pan-AML property, so the selectivity claim needs a myeloid comparator, not a pan-cancer one — which is the same correction this project applied to CEBPA and SPI1. (2) The published mechanism is Complex IV collapse and copper, which is a different prediction from 'the cells need more haem', and an experiment designed on the wrong one of those will measure nothing.
Separately, SETD1A degradation in AML lines immediately downregulates heme biosynthesis and DNA repair transcripts and metabolomics shows an essential SETD1A role in maintaining mitochondrial respiration 4 — a chromatin-to-heme-to-respiration link in a COMPASS-family methyltransferase, which is the family KMT2A belongs to.
The bulk-versus-initiating-cell trap, stated once because it recurs
Every metabolic measurement cited here was made on one of two things: a bulk population, or a functionally defined initiating compartment. The Morris result is the only one that measured both, and it found them to move in opposite directions under the same drug 2.
So a metabolic claim in this disease should always be read with the question 'measured in what compartment, and scored by what endpoint?'. Proliferation of bulk blasts and limiting-dilution engraftment are different experiments that can both be done correctly and still disagree, and the second is the one that corresponds to relapse.
This is the same shape as the quiescence problem recorded in cell-state-quiescence: the compartment that survives therapy is not the compartment the assay is measuring.
| state definition | A metabolic configuration in which ATP production and survival depend primarily on mitochondrial oxidative phosphorylation rather than on glycolysis; operationally scored by oxygen consumption rate, Complex I-IV activity, and sensitivity to respiratory-chain inhibitors |
|---|---|
| internal scan not cited as evidence | this project's DepMap haem-pathway result (median rank 115 of 17,728) and its four respiratory-complex result are internal analyses, held in ana-depmap-haem-lineage-control and related entries, and are deliberately not used as citations in this node |
| compartment caveat | bulk proliferation and limiting-dilution engraftment gave opposite answers to the same drug class in the one study that measured both (PMID 35476984) |
lineage split
{
"myeloid": "MLL/AF9 AML lines OXPHOS-dependent; metformin represses proliferation (PMID 35158754)",
"lymphoid": "in MLL-r B-ALL, OXPHOS inhibition blunts bulk proliferation but PROMOTES leukaemia-initiating cell emergence; hypoxia/glycolysis inhibition impairs LICs (PMID 35476984)"
}published metabolic dependencies in kmt2a r — the numbers behind the chart
[
{
"node": "OXPHOS / Complex I",
"lineage": "AML",
"tool": "metformin",
"pmid": "35158754",
"title": "High Metabolic Dependence on Oxidative Phosphorylation Drives Sensitivity to Metformin Treatment in MLL/AF9 Acute Myeloid Leukemia.",
"authors": "Liu L, Patnana PK, Xie X, et al.",
"venue": "Cancers (Basel)",
"year": 2022,
"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."
},
{
"node": "hypoxia + glycolysis",
"lineage": "B-ALL LIC",
"tool": "not named in abstract",
"pmid": "35476984",
"title": "Hypoxic, glycolytic metabolism is a vulnerability of B-acute lymphoblastic leukemia-initiating cells.",
"authors": "Morris V, Wang D, Li Z, et al.",
"venue": "Cell Rep",
"year": 2022,
"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."
},
{
"node": "de novo heme biosynthesis",
"lineage": "AML (class-wide, not KMT2A-stratified)",
"tool": "HBE inhibition; death by cuproptosis",
"pmid": "41265435",
"title": "Inhibition of heme biosynthesis triggers cuproptosis in acute myeloid leukemia.",
"authors": "Lewis AC, Gruber E, Franich R, et al.",
"venue": "Cell",
"year": 2026,
"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."
},
{
"node": "ACSL4 / polyunsaturated fatty acid utilisation",
"lineage": "KMT2Ar AML specifically",
"tool": "CRISPRi and shRNA",
"pmid": "41712378",
"title": "ACSL4-associated lipid metabolism is a distinct therapeutic vulnerability in KMT2A-rearranged acute myeloid leukemia.",
"authors": "Schäfer S, Rahimian E, Schmitz-Hübsch L, et al.",
"venue": "Cell Rep",
"year": 2026,
"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."
}
]atlas refs — the numbers behind the chart
[
{
"label": "MLL/AF9 AML OXPHOS dependence and metformin sensitivity",
"pmid": "35158754",
"title": "High Metabolic Dependence on Oxidative Phosphorylation Drives Sensitivity to Metformin Treatment in MLL/AF9 Acute Myeloid Leukemia.",
"authors": "Liu L, Patnana PK, Xie X, et al.",
"venue": "Cancers (Basel)",
"year": 2022,
"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."
},
{
"label": "OXPHOS inhibition promotes LIC emergence in MLL-r B-ALL",
"pmid": "35476984",
"title": "Hypoxic, glycolytic metabolism is a vulnerability of B-acute lymphoblastic leukemia-initiating cells.",
"authors": "Morris V, Wang D, Li Z, et al.",
"venue": "Cell Rep",
"year": 2022,
"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."
},
{
"label": "Heme biosynthesis as a selective AML dependency via cuproptosis",
"pmid": "41265435",
"title": "Inhibition of heme biosynthesis triggers cuproptosis in acute myeloid leukemia.",
"authors": "Lewis AC, Gruber E, Franich R, et al.",
"venue": "Cell",
"year": 2026,
"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."
}
]- An OXPHOS inhibitor given to KMT2A-rearranged B-ALL is predicted by the only quantitative xenotransplantation study in this disease to shrink the blast burden while INCREASING leukaemia-initiating cells — a response that would look like a good response and be a bad one.
- Metformin and other Complex I inhibitors are attractive because they are old and cheap; the flip side is that their in vitro concentrations are far above achievable plasma levels, so a positive cell-line result is weak grounds for a trial.
- Heme biosynthesis is required by every erythroid precursor in the marrow. A heme-pathway inhibitor in an infant with a cytopenic marrow after induction carries an obvious on-target toxicity that no cited study measures.
- Cuproptosis depends on copper handling, and copper status is nutritionally and genetically variable in infants; a death mechanism that depends on a trace metal is a death mechanism with a pharmacodynamic confounder.
- If the metabolic state differs between AML and ALL forms of the same lesion, then a metabolic combination partner for a menin inhibitor cannot be chosen once for 'KMT2A-rearranged acute leukaemia' — the class the trials actually enrol.
- Is the OXPHOS-versus-glycolysis split a lineage property or a compartment property? The two cited studies differ in BOTH lineage and in whether they measured bulk or initiating cells, so the experiment that separates them — limiting-dilution engraftment after OXPHOS inhibition in a KMT2A-rearranged AML model — has not been done.
- Does the pan-AML heme dependency hold specifically in KMT2A-rearranged AML, and does it hold in KMT2A-rearranged ALL at all? The published work stratifies by disease, not by fusion, and this project's own interest in the pathway came from a KMT2A-rearranged arm with an unresolved monocytic confound.
- Is the heme signal a metabolic requirement or a chromatin phenotype? SETD1A degradation collapses heme biosynthesis transcription in AML; if the same is true for KMT2A-rearranged cells the dependency would be a transcriptional readout of COMPASS activity rather than a metabolic vulnerability, and the two predict different combination partners.
- Does menin inhibition change the metabolic state? Nobody has measured oxygen consumption, extracellular acidification or heme flux before and after revumenib in a KMT2A-rearranged cell, which is a one-plate Seahorse experiment on a drug already in patients.
- E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
Across an AML cell-line panel with high metabolic heterogeneity, AML cells with an MLL/AF9 genotype had upregulated mitochondrial activity and depended mainly on oxidative phosphorylation for energy production, and metformin repressed the proliferation of MLL/AF9 AML cells by inhibiting mitochondrial respiration.
High Metabolic Dependence on Oxidative Phosphorylation Drives Sensitivity to Metformin Treatment in MLL/AF9 Acute Myeloid Leukemia · Liu L, Patnana PK, Xie X, Frank D, Nimmagadda SC, Rosemann A, et al., Khandanpour C · Cancers (Basel) 2022What this evidence cannot show
Established cell lines, not primary blasts or xenografts; the MLL/AF9 'genotype' arm is a small number of lines within a heterogeneous panel and lineage is not controlled, so monocytic identity is an unexcluded alternative explanation exactly as it is in this project's own dependency scan. Metformin at the concentrations used in vitro is a blunt Complex I inhibitor with well-documented off-target and pharmacokinetic limitations, and no in vivo efficacy is claimed. Read: abstract only.
Evidence that argues against this (1)
- In MLL-rearranged B-ALL the same intervention class points the other way at the level that matters. Single-cell transcriptomics with quantitative xenotransplantation found that 'while inhibiting oxidative phosphorylation blunts blast proliferation, this intervention promotes LIC emergence', whereas inhibiting hypoxia and glycolysis impaired MLL-r B-ALL leukaemia-initiating cells and provided therapeutic benefit in xenotransplantation systems. An OXPHOS-directed agent could therefore shrink the measurable disease and expand the compartment that regenerates it. Read: abstract only. Morris V et al. 2022
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In MLL-rearranged B-ALL characterised by single-cell transcriptomics and quantitative xenotransplantation, engraftable leukaemia-initiating cells are abundant relative to reported frequencies in AML; multipotent self-renewing LICs are enriched among phenotypically undifferentiated blasts but LICs capable of replenishing the leukaemic cellular diversity can emerge from more mature fractions; and while inhibiting oxidative phosphorylation blunts blast proliferation it promotes LIC emergence, whereas inhibiting hypoxia and glycolysis impairs MLL-r B-ALL LICs and provides therapeutic benefit in xenotransplantation systems.
Hypoxic, glycolytic metabolism is a vulnerability of B-acute lymphoblastic leukemia-initiating cells · Morris V, Wang D, Li Z, Marion W, Hughes T, Sousa P, et al., Rowe RG · Cell Rep 2022What this evidence cannot show
Xenotransplantation into immunodeficient mice, so the niche imposing the hypoxic state is murine and the human marrow oxygen gradient is only approximated. 'Inhibiting hypoxia and glycolysis' covers a class of pharmacological tools whose selectivity is imperfect, and the abstract does not name the agents or separate the HIF and glycolysis arms. The infant subset of MLL-r B-ALL is not distinguished from older KMT2A-rearranged B-ALL, and partner identity is not reported. Read: abstract only.
Evidence that argues against this (1)
- Genetic deletion of the hypoxia response in a KMT2A-rearranged myeloid model does the opposite of what a hypoxia-targeting strategy predicts. Conditional Hif-1a deletion in a mouse MLL-AF9 AML model produced FASTER disease progression after chemotherapy, and the authors explicitly warn against combining HIF inhibition with chemotherapy in AML. Independently, deletion of Hif-2a, or of Hif-1a and Hif-2a together, ACCELERATED Mll-AF9-initiated leukaemogenesis while established LSCs lacking either or both propagated AML with wild-type latency, and pharmacological HIF inhibition or HIF-2a CRISPR knockout in human MLL-AF9 leukaemic cells had no effect on their function. The hypoxia-directed arm of E2 is therefore lineage-restricted at best and is contradicted by genetics in the myeloid form of the same disease. Read: abstracts only. Vukovic M et al. 2015
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Integrated analysis of mouse models, human cell lines and primary patient samples identifies de novo heme biosynthesis as a selective dependency in acute myeloid leukaemia, underpinned by a propensity of AML cells and especially leukaemic stem cells to downregulate heme biosynthesis enzymes in a way that promotes self-renewal; inhibition of those enzymes collapses mitochondrial Complex IV, dysregulates the copper-chaperone system and induces cuproptosis, and glycolysis is among the pathways synthetic-lethal with heme biosynthesis.
Inhibition of heme biosynthesis triggers cuproptosis in acute myeloid leukemia · Lewis AC, Gruber E, Franich R, Armstrong J, Kelly MJ, Opazo CM, et al., Kats LM · Cell 2026What this evidence cannot show
The dependency is claimed for AML as a class, not for KMT2A-rearranged AML, and the abstract reports no genotype breakdown; whether the KMT2A-rearranged arm is more or less dependent than other AML is unstated. Nothing here addresses lymphoid disease, which is where infant KMT2A-rearranged leukaemia mostly lives. Cuproptosis is a recently defined death modality whose in vivo relevance is still being established. Read: abstract only.
Evidence that argues against this (1)
- The heme pathway's link to leukaemic survival has been reported running through apoptosis rather than through respiration, in B-lineage disease, and in the opposite mechanistic direction. A genome-wide CRISPR screen found that loss of heme synthesis genes renders B-ALL cells RESISTANT to dihydroartemisinin-induced death, because the drug works by disrupting heme binding to the eIF2-alpha kinase HRI and triggering the integrated stress response, which translationally represses MCL-1 and synergises with BH3-mimetics. On that account heme is a signal that sets apoptotic priming, not a metabolite whose synthesis is required; and cells with less heme synthesis are harder, not easier, to kill by that route. Two mechanisms cannot both be the reason a heme-pathway perturbation changes survival, and which applies in KMT2A-rearranged disease has not been tested. Read: abstract only. Smith KH et al. 2021
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Rapid degrader-induced loss of SETD1A in AML cell lines immediately downregulates transcripts of the DNA repair and heme biosynthesis pathways, and CRISPR functional analysis with metabolomics reveals an essential role for SETD1A in maintaining mitochondrial respiration in AML cells, through a non-enzymatic SETD1A domain-dependent cyclin K function controlling RNA polymerase II promoter-proximal pause release.
SETD1A regulates transcriptional pause release of heme biosynthesis genes in leukemia · Hoshii T, Perlee S, Kikuchi S, Rahmutulla B, Fukuyo M, Masuda T, et al., Kaneda A · Cell Rep 2022What this evidence cannot show
SETD1A, not KMT2A, and AML cell lines with no KMT2A-rearrangement stratification reported in the abstract. The link to this beat is that a COMPASS-family methyltransferase controls heme biosynthesis transcription and mitochondrial respiration, which makes a heme-pathway dependency in a KMT2A-rearranged arm potentially a chromatin phenotype rather than a metabolic one; that is an interpretation, not a result of this paper. Read: abstract only.
Evidence that argues against this (1)
- If heme biosynthesis genes are SETD1A-dependent transcriptional targets in AML generally, then a haem-pathway signal in a KMT2A-rearranged dependency arm need not be fusion-driven at all, and the comparator that would settle it — KMT2A-rearranged versus other AML rather than versus all cancer lines — is the comparator this project has repeatedly had to add after the fact. The same paper's finding that the effect runs through a non-enzymatic domain also means that inhibiting the catalytic activity of a COMPASS methyltransferase would not reproduce it, which weakens any read-across from this result to KMT2A SET-domain biology. Read: abstract only. Lewis AC et al. 2026
MLL/AF9-genotype AML lines show upregulated mitochondrial activity and depend mainly on OXPHOS; metformin represses their proliferation
- related to →The stem-like self-renewal state — a limited stem-cell module grafted onto a committed cell
OXPHOS inhibition blunts bulk proliferation but PROMOTES leukaemia-initiating cell emergence in MLL-r B-ALL — bulk and initiating compartments move in opposite directions
the published pan-AML heme-biosynthesis dependency is the external comparator this project's internal haem-pathway scan needs; mechanism is Complex IV collapse and cuproptosis, not increased haem demand
a class-wide AML heme dependency is an alternative explanation to monocytic identity for the same signal, and neither has been excluded
the two halves of one metabolic choice; the same study found opposite LIC effects from inhibiting each
third instance in this beat of a metabolic property splitting by lineage rather than by fusion
metabolic axis: OXPHOS dependence in myeloid, LIC-promoting OXPHOS inhibition in lymphoid
MCL-1 dependence is created by complex I inhibition, linking it to the OXPHOS state
Hnf4a loss downregulates glycolysis AND oxidative phosphorylation gene sets together
the mechanism requires an OXPHOS-dependent starting state
knockdown raises mitochondrial respiration and membrane potential in KMT2A::AFF1 lines
the lineage split that cuts across all three directions
complex I blockade attacks the OXPHOS-dependent state
proposes ontogeny rather than lineage as the variable behind the OXPHOS split
rapid SETD1A degradation downregulates DNA-repair and heme-biosynthesis transcripts and CRISPR plus metabolomics show SETD1A maintains mitochondrial respiration in AML cells; lines and fusion status unnamed in the source, and this graph records OXPHOS dependence as inverted in KMT2A-rearranged B-ALL
the guanine-nucleotide axis is the documented ESCAPE from venetoclax-imposed OXPHOS inhibition, so it sits on the far side of this state rather than inside it
sibling metabolic state; that one splits by lineage and reverses direction between AML and B-ALL, this one is evidenced in KMT2A-rearranged ALL
the compartment that would pay for a haem-pathway intervention: that entry raises haem-biosynthesis inhibition as a bite-back and had no cell-type node to attach it to. Erythroid progenitors run the pathway at maximum flux while AML cells reportedly downregulate it
Fbxl10-transgenic HSCs enrich mitochondrial OXPHOS genes with raised ATP, and KDM2B ChIP sits on those regulatory regions
SIRT3, an NAD+-dependent deacetylase, antagonises the p300 acetylation of SKP2's NLS that drives cytoplasmic retention (PMID 22770219) - a metabolic input to a nuclear E3 receptor's address, untested in leukaemia (Q91 2026-09-22)