The marrow oxygen gradient — low everywhere, lowest at the sinusoids, and in this leukaemia low oxygen helps the drug rather than the cell
also called bone marrow hypoxia · hypoxic niche · marrow pO2
Bone marrow runs on very little oxygen even though it is full of blood vessels, and the least oxygenated part is deep around the drainage vessels rather than at the bone surface, which is the opposite of what textbooks usually say. In this particular leukaemia, putting cells in low oxygen made a promising drug combination work better, not worse — so the usual story that low oxygen shelters cancer cells does not hold here.
The same thing, for a scientist
Direct in vivo two-photon phosphorescence lifetime microscopy in live mice gives absolute marrow pO2 **below 32 mmHg throughout despite very high vascular density**, with the lowest values **about 9.9 mmHg (1.3 percent) in deep peri-sinusoidal regions** and the **endosteal region less hypoxic** because it is perfused by small, often Nestin-positive arteries; pO2 changes markedly after radiation and chemotherapy [E1]. That inverts the conventional hypoxic-endosteal-niche picture, which rested on indirect evidence - HIF-1alpha expression, proteomics and pimonidazole staining. The leukaemia literature has not absorbed it: MLL-AF9 metabolic imaging places glycolytic, LIC-enriched cells in what it calls a hypoxic endosteal niche [E2]. Most consequentially for this project, the one experiment that manipulated oxygen in KMT2A-rearranged AML found that **hypoxia induction SENSITISED cells to venetoclax plus the menin inhibitor MI-503**, with the combination acting partly by repressing HIF-1A and its target HDAC9, and HDAC9 indispensable for KMT2A-rearranged AML proliferation [E3]. This graph's resistance beat already rejected hypoxic niche protection on that basis. A marrow-niche beat is exactly where that rejection would quietly be reversed by the field's default framing, so it is restated here.
What was measured, and why the measurement matters
The hypoxic stem-cell niche was inferred, not measured, for about a decade. The inference came from HIF-1alpha expression, from proteomic signatures and from pimonidazole adducts — all of which report a cell's response to low oxygen rather than the oxygen itself, and all of which can be driven by metabolic state independently of pO2.
Spencer and colleagues measured it, using a phosphorescent probe whose excited-state lifetime depends on local oxygen, read by two-photon microscopy in a live animal. Three results 1:
- The whole marrow is hypoxic by tissue standards — under 32 mmHg — despite dense vasculature.
- The gradient runs the other way from the textbook. Deep peri-sinusoidal regions are lowest at about 9.9 mmHg (1.3 percent oxygen); the endosteum is better oxygenated because arteries perfuse it.
- Treatment changes it. pO2 shifts markedly after radiation and chemotherapy, so the oxygen map of a treated marrow is not the oxygen map of a normal one.
Point 3 is the one most likely to be useful and least likely to be remembered: the compartment a residual leukaemic cell sits in after induction may have a different oxygen tension from the compartment it sat in at diagnosis.
In KMT2A-rearranged AML, hypoxia is on the drug's side
The default oncology framing is that hypoxia protects: it slows the cell cycle, it selects for glycolysis, it drives HIF-dependent survival programmes, and it limits the delivery of oxygen-dependent cytotoxics. In KMT2A-rearranged AML the one direct experiment says otherwise.
Ling and colleagues found that hypoxia induces HDAC9 in KMT2A-rearranged AML, that HDAC9 is indispensable for its proliferation, that venetoclax plus a menin inhibitor represses HIF-1A and thereby HDAC9, and that hypoxia induction sensitised KMT2A-rearranged AML to venetoclax plus MI-503 for both proliferation inhibition and apoptosis 3.
The logic is coherent: if the combination's mechanism includes shutting down a hypoxia-driven dependency, then a cell already leaning on that dependency is more vulnerable, not less. And it yields a clean prediction that combines with 1: the deep peri-sinusoidal compartment, the most hypoxic part of the marrow, should be where venetoclax plus a menin inhibitor clears disease best. Untested; carried in hyp-niche-remodelling-precedes-relapse-kmt2a.
The caution is real. 3 is a Letter, uses cell lines and a MOLM-13 xenograft, induces hypoxia in culture rather than measuring it in a marrow compartment, and uses MI-503 rather than a clinical menin inhibitor. It is one experiment. But it is the only one done in this genotype, and writing a niche entry that assumed the opposite would be writing against the only available data — which is the specific failure this project has recorded before.
| treatment effect | pO2 changes markedly after radiation and chemotherapy, so the oxygen map of treated marrow differs from that of normal marrow |
|---|---|
| direction in kmt2a | hypoxia SENSITISES KMT2A-rearranged AML to venetoclax plus MI-503; HIF-1A and its target HDAC9 are repressed by the combination and HDAC9 is indispensable for proliferation |
| graph status | hypoxic niche protection was REJECTED as a menin-inhibitor resistance candidate by the resistance beat on exactly this evidence; this entry restates the rejection so a niche beat does not reintroduce it by default framing |
| unresolved | the MLL-AF9 imaging literature calls the endosteal niche hypoxic; direct pO2 measurement calls it the better-oxygenated compartment |
| prediction | the deep peri-sinusoidal compartment should be where venetoclax plus a menin inhibitor clears KMT2A-rearranged disease best, which is the opposite of a sanctuary prediction |
measurements
[
{
"region": "whole marrow",
"value": 32,
"unit": "mmHg (upper bound)",
"method": "two-photon phosphorescence lifetime microscopy in vivo",
"source_evidence": "E1"
},
{
"region": "deep peri-sinusoidal",
"value": 9.9,
"unit": "mmHg (approx 1.3 percent O2)",
"method": "two-photon phosphorescence lifetime microscopy in vivo",
"source_evidence": "E1"
},
{
"region": "endosteal",
"value": null,
"unit": "not quantified; explicitly LESS hypoxic than peri-sinusoidal, arterially perfused",
"method": "two-photon phosphorescence lifetime microscopy in vivo",
"source_evidence": "E1"
}
]atlas refs
[
{
"label": "Direct in vivo marrow pO2 measurement",
"pmid": "24590072",
"title": "Direct measurement of local oxygen concentration in the bone marrow of live animals.",
"authors": "Spencer JA, Ferraro F, Roussakis E, et al.",
"venue": "Nature",
"year": 2014,
"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": "Hypoxia sensitises KMT2A-r AML to venetoclax plus menin inhibitor",
"pmid": "38053201",
"title": "Repressing HIF-1α-induced HDAC9 contributes to the synergistic effect of venetoclax and MENIN inhibitor in KMT2Ar AML.",
"authors": "Ling Q, Zhou Y, Qian Y, et al.",
"venue": "Biomark Res",
"year": 2023,
"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."
}
]- The sensitisation result rests on a single Letter with cell lines, a tool compound and culture-induced hypoxia; building a therapeutic argument on it without replication would be the same kind of error as building a niche-protection argument on a review heading.
- If hypoxia sensitises, then any supportive intervention that raises marrow oxygenation - transfusion to a high haemoglobin target, for instance - could in principle reduce the efficacy of this combination, which is an interaction nobody has looked for and which cuts against routine practice.
- Manipulating tissue oxygen therapeutically is not achievable with the precision this mechanism would require, so the practical use of the finding is predictive and mechanistic rather than interventional.
- Where is the hypoxic compartment in LEUKAEMIC marrow? Both primary measurements are in normal or surrogate-marked tissue, and the direct-measurement paper itself shows pO2 shifts after treatment.
- Does venetoclax plus a menin inhibitor clear the deep peri-sinusoidal compartment preferentially? A spatially resolved response readout in a xenograft would test the entry's central prediction.
- HDAC9 is reported indispensable for KMT2A-rearranged AML proliferation. Is it a target in its own right, and does its dependency hold in KMT2A::AFF1 B-lymphoid disease?
- Does the sensitisation hold for clinical menin inhibitors (revumenib, ziftomenib, bleximenib) or only for the tool compound MI-503?
- If three niche compartments protect by buffering oxidative stress while hypoxia sensitises to this combination, what happens when both manipulations are applied together? Nobody has run stromal co-culture and hypoxia in the same experiment.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Direct in vivo two-photon phosphorescence lifetime measurement of local oxygen tension in live mouse bone marrow gives absolute pO2 below 32 mmHg despite very high vascular density, with heterogeneity in local pO2 and the lowest values (about 9.9 mmHg, or 1.3 percent) in deeper peri-sinusoidal regions, while the endosteal region is LESS hypoxic because it is perfused with small arteries that are often positive for the marker nestin; these pO2 values change markedly after radiation and chemotherapy.
Direct measurement of local oxygen concentration in the bone marrow of live animals · Spencer JA, Ferraro F, Roussakis E, Klein A, Wu J, Runnels JM, Zaher W, Mortensen LJ, et al. · Nature 2014What this evidence cannot show
Mouse calvarial marrow, which is thin, accessible and may not represent long-bone or vertebral marrow where most human haematopoiesis occurs. The measurement is extracellular local pO2 and is not equivalent to intracellular hypoxic signalling; a cell can mount a HIF response at an oxygen tension this method scores as unremarkable. Normal, not leukaemic, marrow. Read: abstract only.
Evidence that argues against this (1)
- The leukaemia literature makes the opposite spatial assignment. Metabolic imaging with the SoNar sensor in MLL-AF9 murine AML reports that the glycolytic, leukaemia-initiating-cell-enriched fraction homes to and locates in what its authors describe as the hypoxic ENDOSTEAL niche, maintaining itself there by symmetric division. Both are primary mouse experiments and they cannot both be right about where the low-oxygen compartment is; possible reconciliations include different bones, surrogate versus direct measurement, or a genuinely different oxygen distribution in leukaemic marrow. Read: abstract only. Hao X et al. 2019
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In an MLL-AF9-induced murine AML model, metabolically defined SoNar-high cells are more glycolytic, enriched for leukaemia-initiating cell frequency and develop leukaemia much faster than SoNar-low cells; they mainly home to and locate in the hypoxic endosteal niche and maintain their activities through efficient symmetric division, with PDK2 fine-tuning glycolysis, homing and symmetric division.
Metabolic Imaging Reveals a Unique Preference of Symmetric Cell Division and Homing of Leukemia-Initiating Cells in an Endosteal Niche · Hao X, Gu H, Chen C, Huang D, Zhao Y, Xie L, Zou Y, Shu HS, et al. · Cell Metabolism 2019What this evidence cannot show
Mouse MLL-AF9; SoNar reports cytosolic NADH/NAD+ ratio, so the 'hypoxic' characterisation of the niche is inferred from the cells' redox and glycolytic state and from surrogate markers rather than from oxygen measurement. LIC enrichment is relative, not absolute. Read: abstract only.
Evidence that argues against this (1)
- Direct in vivo pO2 measurement reports the endosteum as the better-oxygenated compartment and deep peri-sinusoidal marrow as the hypoxic one, specifically testing and contradicting the surrogate-marker-based assignment this claim relies on. Read: abstract only. Spencer JA et al. 2014
- E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
In KMT2A-rearranged AML, hypoxia induces HDAC9 expression; HDAC9 is indispensable for KMT2A-rearranged AML proliferation and its repression contributes to the proliferation inhibition produced by venetoclax plus a menin inhibitor; the combination inhibits the hypoxia pathway, especially HIF-1A, and its target HDAC9, with HDAC9 downregulation partially dependent on HIF-1A repression; and hypoxia induction SENSITISED KMT2A-rearranged AML to venetoclax plus MI-503-mediated proliferation inhibition and apoptosis induction.
Repressing HIF-1alpha-induced HDAC9 contributes to the synergistic effect of venetoclax and MENIN inhibitor in KMT2Ar AML · Ling Q, Zhou Y, Qian Y, Qian J, Zhang Y, Wang J, Zhu Y, Zhou Y, et al. · Biomarker Research 2023What this evidence cannot show
PubMed classifies the record as a Letter, so methods are abbreviated. Cell lines, primary ex vivo cells and a MOLM-13 xenograft, with hypoxia induced in culture rather than measured in a marrow compartment; the link to the peri-sinusoidal region is this entry's inference. MI-503 is a tool compound. Adult-derived myeloid lines, not infant ALL. One experiment, unreplicated. Read: abstract only.
Evidence that argues against this (1)
- Three independent niche compartments protect KMT2A-rearranged or MLL-AF9 leukaemia by buffering oxidative stress - Prx1+ mesenchymal CXCL12 controlling LSC redox balance, nestin+ BMSCs transferring OXPHOS capacity and glutathione, and adipocyte leptin driving complex I with a compensatory antioxidant programme. Those results predict that an oxidative-stress-adapted, hypoxia-tolerant state should be protective, which is the opposite of the sensitisation reported here. Neither literature has run the other's drug, so the contradiction is untested rather than resolved. Read: abstract only. Forte D et al. 2020
the oxygen dimension of the niche; lowest pO2 is peri-sinusoidal, not endosteal
deep peri-sinusoidal marrow is the measured minimum, about 9.9 mmHg
hypoxia SENSITISES KMT2A-r AML to venetoclax plus a menin inhibitor, via HIF-1A and HDAC9 repression
contradicts a hypoxic-sanctuary account of niche protection in this genotype
direct pO2 measurement contradicts the MLL-AF9 imaging literature's hypoxic-endosteal assignment