The clearest clinical rationale for a LEDGF agent - second line after menin-inhibitor failure - has never been tested, and one 2025 measurement now argues against it
also called LEDGF in menin-resistant disease
The obvious use for a LEDGF drug is in patients whose leukaemia has stopped responding to the menin drug. Nobody has ever tested a LEDGF-directed agent in a menin-resistant model, and a new measurement suggests the LEDGF grip may already be loosened by the menin drug, which would weaken the idea.
The same thing, for a scientist
The graph records as unanswered whether a LEDGF-directed agent retains activity in menin-inhibitor-resistant cells. Checked this cycle: it is still unanswered, and the absence is now counted. `(PSIP1 OR LEDGF OR DFS70)` intersected with the named menin inhibitors returns **0 PubMed records**; intersected with resistance terms it returns 2, neither of which tests a LEDGF agent in a resistant model. The rationale is structurally sound - MEN1 gatekeeper mutations attenuate inhibitor binding **without disturbing the menin-KMT2A interaction**, so the clamp and therefore the LEDGF arm should be intact in resistant disease - and that soundness is exactly why the absence is worth recording rather than shrugging at. But the same structure supplies the counter-argument. If the menin-KMT2A interaction is intact in a resistant cell, the tenfold cooperative enhancement is intact too, so the LEDGF grip in resistant disease is at its **tight** state (Kd 1.4 uM) rather than its loose one (14.7 uM), and a competitive IBD ligand faces a harder problem there than in a naive cell, not an easier one.
The counted absence
Run 2026-09-20, translations checked:
| Query | Count | Content |
|---|---|---|
| LEDGF alias set AND (revumenib OR ziftomenib OR bleximenib OR enzomenib OR SNDX-5613 OR KO-539 OR JNJ-75276617 OR MI-503 OR MI-3454 OR VTP-50469) | 0 | - |
| LEDGF alias set AND ("menin inhibitor resistance" OR "resistant to menin" OR "MEN1 mutation") | 2 | PMID 40102405 (IMPDH2 blockade sensitising to menin inhibition) and PMID 34685704 (IBD interactome in docetaxel-resistant prostate cancer) |
| LEDGF alias set AND (ChIP OR CUT&RUN OR occupancy) AND menin | 2 | PMID 41279447 (preprint) and PMID 29084774 |
Neither of the two resistance hits is a test of a LEDGF-directed agent in a menin-inhibitor-resistant leukaemia. The "resistant to menin" phrase was reported in quotedphrasesnotfound, so that leg contributed nothing and the count rests on the other two legs.
No clinical trial exists either: there is no LEDGF-directed agent to put in one.
The case for, stated at full strength
Resistance to menin inhibitors is driven substantially by somatic MEN1 mutations at the drug-menin interface, which attenuate inhibitor binding while leaving the menin-KMT2A interaction functional 1. A resistant cell therefore still has an assembled menin-KMT2A-LEDGF module; only the drug has been excluded from it. An agent acting on a different contact within the same module - the LEDGF IBD-IBM interface - is not affected by a mutation in the menin pocket. On its face this is the cleanest orthogonal-mechanism argument in the whole dossier, and it is the reason the graph has kept the question open across several cycles.
It is strengthened by the observation that the resistance route the graph most recently added - JunD liberation on menin displacement - is also not addressable from the menin pocket.
The case against, which is new this cycle
Two arguments, both from measurements made since the question was first posed.
1. Cooperativity cuts the wrong way. The LEDGF IBD binds MLL 1-160 at 14.7 uM alone and the preformed menin-MLL complex at 1.4 uM 2. In a MEN1 gatekeeper-mutant resistant cell the menin-KMT2A interaction is intact, so the LEDGF contact sits in its tenfold-enhanced state. A competitive ligand must therefore beat a 1.4 uM interaction in precisely the population where it would be deployed. The naive intuition - resistant cells have lost menin function, so the LEDGF arm is exposed - is backwards for the mutational resistance class.
2. In drug-sensitive cells, LEDGF occupancy already falls. A 2025 preprint reports that revumenib decreases LEDGF chromatin binding 3. If that holds, the LEDGF axis is partly downstream of the existing drug in responders, which narrows a LEDGF agent's value proposition to non-responders - and argument 1 says that is where it is hardest.
The combination does not kill the idea. It reframes it: the LEDGF arm is most attractive against non-genetic, adaptive resistance, where menin engagement is presumably reduced and the clamp weak, and least attractive against MEN1 gatekeeper mutations, where it is tight. The graph currently treats menin-inhibitor resistance as one thing; on this axis the two classes have opposite predictions.
What to do about it
The discriminating measurement is arm 3 of exp-ledgf-occupancy-menin-inhibitor-monotherapy-panel: LEDGF CUT&RUN in a MEN1 gatekeeper-mutant line under revumenib. The clamp model predicts LEDGF occupancy is preserved there while it falls in the parental line. That would confirm the module stays assembled in resistant disease, which is the premise the whole second-line rationale rests on, and it would do so before anyone spends a chemistry programme on it.
Until that is done, the second-line rationale should be graded as an inference from structure, not as a finding.
| question | Does a LEDGF-directed agent retain activity in menin-inhibitor-resistant disease? |
|---|---|
| status | untested; no LEDGF-directed agent exists to test |
| pubmed count ledgf x named menin inhibitors | 0 |
| pubmed count ledgf x resistance terms | 2 |
| structural prediction for MEN1 mutant resistance | clamp intact, LEDGF contact in its tight 1.4 uM state - harder to break, not easier |
| structural prediction for adaptive resistance | clamp state unknown; the more favourable setting of the two |
| discriminating experiment | exp-ledgf-occupancy-menin-inhibitor-monotherapy-panel arm 3 |
| date run | 2026-09-20 |
| absence class | never-measured |
| absence class basis | The experiment requires a LEDGF agent and a resistant model; neither has been combined. |
| absence class assigned | 2026-09-21 (cycle 47, king) |
- The cooperativity that makes the clamp interesting makes the LEDGF contact tightest in exactly the resistant population a LEDGF agent would target.
- If revumenib already reduces LEDGF occupancy in responders, a LEDGF agent adds nothing where the drug works and faces the tight clamp where it does not - a narrow window in both directions.
- The second-line rationale has been carried in this graph for several cycles as an open question, which reads as a promising lead; it is an untested structural inference and should be graded as one.
- Is LEDGF occupancy preserved in MEN1 gatekeeper-mutant cells under drug? The prediction is yes and the experiment is one CUT&RUN.
- Do the two resistance classes - genetic MEN1 mutation and adaptive non-genetic state - differ in clamp status? The graph currently treats menin-inhibitor resistance as a single category and on this axis it should not.
- Is there any leukaemia sample, anywhere, with both a MEN1 resistance mutation and a LEDGF ChIP or CUT&RUN profile? If a published resistant-sample dataset exists, this could be answered from an archive rather than at the bench.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Somatic MEN1 mutations at the drug-menin interface mediate clinical resistance to menin inhibition and attenuate inhibitor binding without disturbing the menin-KMT2A interaction.
MEN1 mutations mediate clinical resistance to menin inhibition. · Perner F, Stein EM, et al. · Nature 2023What this evidence cannot show
The paper characterises drug binding and the menin-KMT2A interaction; LEDGF was not examined, so every statement here about the LEDGF arm in resistant cells is an inference from the structural argument.
Evidence that argues against this (1)
- Genetic MEN1 mutation is only one resistance class; adaptive non-genetic states and lineage switch are documented in this graph and would not preserve the clamp in the same way, so the inference does not generalise across resistance mechanisms. Perner F et al. 2023
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
The LEDGF IBD binds MLL 1-160 at Kd 14.7 +/- 0.7 uM alone and the preformed menin-MLL complex at Kd 1.4 uM, so an intact menin-KMT2A interaction holds the LEDGF contact in a tenfold-tighter state.
The same site on the integrase-binding domain of lens epithelium-derived growth factor is a therapeutic target for MLL leukemia and HIV. · Murai MJ, Pollock J, et al. · Blood 2014What this evidence cannot show
In vitro affinities on isolated fragments by two different methods. Whether the cooperative term survives in a chromatinised, phosphorylated, full-length context is untested.
Evidence that argues against this (1)
- If PWWP-nucleosome avidity dominates occupancy in cells, the IBD affinity difference may be irrelevant to how hard the complex is to break pharmacologically, in which case the argument in this section does not apply. Koutná E et al. 2023
- E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.preprint
A 2025 preprint reports by CUT&RUN that the menin inhibitor SNDX-5613 decreases LEDGF chromatin binding at MLL target loci in MOLM-13 and MV4-11, which would place the LEDGF axis partly downstream of the existing drug in responding cells.
Co-targeting menin and LSD1 dismantles oncogenic programs and restores differentiation in MLL-rearranged AML. · Tayari MM, Dos Santos HG, et al. · bioRxiv 2025What this evidence cannot show
Preprint, combination design, qualitative reporting, myeloid lines only, and the LSD1 inhibitor alone produced a comparable reduction.
Evidence that argues against this (1)
- If the fall in LEDGF occupancy is caused by loss of H3K36me3 or by loss of MLL occupancy rather than by loss of the cooperative clamp, it carries no implication for what a LEDGF-directed agent would add. Zhu L et al. 2016
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
The one intervention that lowers LEDGF protein and is combined with a menin inhibitor in vivo is indirect: IMPDH2 blockade reduces guanine nucleotides and rRNA transcription, lowers LEDGF and menin protein, impairs MLL-fusion complex formation and chromatin binding, and suppresses MLLr AML when combined with a menin inhibitor.
Guanine nucleotide biosynthesis blockade impairs MLL complex formation and sensitizes leukemias to menin inhibition. · Shi X, Li M, et al. · Nat Commun 2025What this evidence cannot show
The mechanism runs through global nucleotide and translation effects, so this is not a LEDGF-directed agent and cannot answer whether targeting LEDGF specifically helps in resistant disease. Open access in Europe PMC (PMC11920272). The models are MLLr AML, not lymphoid.
Evidence that argues against this (1)
- An agent that lowers two complex members plus rRNA transcription cannot attribute its benefit to LEDGF, and IMPDH inhibitors have their own established haematological toxicity. Shi X et al. 2025
- related to →MEN1 drug-interface mutations — the first proof that a chromatin drug can drive Darwinian escape in patients
the resistance class the second-line rationale rests on
the new measurement that weakens the rationale
the drug whose failure defines the proposed indication
arm 3 is the discriminating measurement
independently predicts activity against MEN1-mutant but not non-genetic resistance
SURVIVES: recall went 0 -> 116 at full text and the 50 read are all reviews naming both ideas without testing them together
SURVIVES: 115 LEDGIN/ALLINI records, every one HIV-1 integrase pharmacology
the twelve-name LEDGF-agent list executed as nine names; the published list is not the list that ran