Menin dependence of KMT2A::MLLT1: real, repeatedly demonstrated, and always as one lane of a partner panel
also called MLL-ENL menin dependence · menin inhibitor in t(11;19) · MEN1 dependency MLLT1
Every kind of experiment that has been tried says this fusion needs menin, and the cell line carrying it is often the most sensitive one in the dish. But no study has ever set out to ask the question about this partner specifically - it is always one column in a table of several fusions.
The same thing, for a scientist
The evidence that KMT2A::MLLT1 is menin-dependent is consistent across genetic, chemical and in vivo settings, and it is entirely incidental. **Genetic:** the founding menin paper is reported by a later primary paper to have shown that excising the menin-binding motif from the N-terminus of MLL-GAS7 and **MLL-ENL** fusion proteins reproduces the effects of menin loss [E1][E2]. **Chemical, in vitro:** MI-2 gave a GI50 of 7.2 uM in KOPN-8 (MLL-ENL), the lowest of the four MLL-rearranged lines in that panel, and MI-2/MI-3 blocked proliferation of MLL-ENL-transduced mouse marrow at about 5 uM with differentiation and Hoxa9/Meis1 downregulation [E2]. KOPN-8 recurs in the MI-503 and ziftomenib panels [E3]. **In vivo and translational:** KOPN8 and HB11;19 are both in the VTP-50469 (revumenib-lineage) line panel, and **three of the eight patient-derived xenografts in that study are B-ALL t(11;19) MLL-ENL models** [E4]. **DepMap:** this graph's existing entry records a MEN1 CRISPR gene effect of -1.983 for HB11;19 (ACH-001736), the strongest of five KMT2A-rearranged lines checked, against -0.463 for the KMT2A::AFF1 line RS4;11. Against all of that, the paper that examined the partner question directly concluded the opposite of partner specificity: menin-MLL inhibitors block transformation by MLL fusions **in a fusion-partner-independent manner** [E5].
The one thing that would make this partner special, and the evidence is one number
If reciprocal status modulates menin dependence, KMT2A::MLLT1 is the partner in which it would show, because roughly half its patients have no reciprocal allele. The single suggestive data point is that HB11;19 - the one reciprocal-positive KMT2A::MLLT1 line - is the most menin-dependent line in the panel this graph checked. That is n = 1, it is a CRISPR gene effect rather than a drug response, and the direction is the opposite of what the enforceable-remainder hypothesis predicts: if the reciprocal product were a menin-blind remainder that survives menin inhibition, reciprocal-positive cells should be less completely served by menin loss, not more. Either the number means nothing, or it is a point against the hypothesis. It should not be read as a point for it.
The PDX observation is the most actionable thing here
Three of eight PDX models in the pivotal preclinical menin-inhibitor paper are t(11;19) MLL-ENL 4. The paper reports extended survival in 6 of 7 additional MLL-ALL PDXs treated with VTP-50469 and does not name the non-responder in its text. If the single non-responding PDX is one of the three MLL-ENL models, that is a partner-specific preclinical failure hiding in the pivotal dataset; if it is not, it is reassurance. Resolving it needs Figure 6B and Figure S5D of that paper, which are images and were not parsed here. This is a small, concrete, high-value verification task.
What is genuinely absent
No menin-inhibitor experiment anywhere uses a reciprocal-negative KMT2A::MLLT1 system, because none exists: both human lines are reciprocal-positive or reciprocal-unassayed, all mouse transduction models deliver the direct fusion alone (and so are reciprocal-negative, but have no reciprocal-positive counterpart to compare against), and no reciprocal construct has been cloned for this partner. So the preclinical menin package for KMT2A::MLLT1 is unstratified in a known direction, exactly as it is for t(4;11).
Bleximenib has no MLL-ENL data at all
A full-text search of the bleximenib preclinical paper returned zero occurrences of MLLT1, MLL-ENL, ENL, KOPN or 11;19; its KMT2A-rearranged models are MOLM14 and two B-ALL PDX lines 6. That is a bounded absence for one of the four clinical-stage menin inhibitors, and it means partner coverage across the drug class is uneven in a way nobody has tabulated.
| partner | MLLT1 |
|---|---|
| menin dependence | established, never partner-specifically studied |
| kopn8 mi2 gi50 uM | 7.2 |
| hb1119 men1 gene effect | -1.983 |
| pdx total in study | 8 |
| reciprocal stratified menin experiment exists | false |
| bleximenib mll enl data | none |
evidence classes
[ "genetic menin ablation (second-hand attribution)", "tool-compound GI50 in KOPN-8", "clinical-stage compound line panels", "three of eight PDXs in the pivotal in vivo study", "DepMap MEN1 gene effect in HB11;19" ]
pdx mll enl models
[ "PDX MLL-6", "PDX MLL-8", "PDX MLL-14" ]
- Menin inhibitors cannot touch the reciprocal product, which carries no menin-binding motif. In the reciprocal-positive half of this population, menin inhibition is incomplete by construction, and that would present clinically as partner-independent resistance rather than as anything identifiable.
- KOPN-8, the line carrying most of this partner's menin pharmacology, is KRAS G12D mutant. RAS pathway mutations are an established route to menin-inhibitor resistance, so the reference line for this fusion already contains the commonest resistance lesion.
- If the exceptional MEN1 dependency of the one reciprocal-positive line is real and general, then reciprocal-negative patients would be the LESS menin-dependent group - the opposite of the enforceable-remainder prediction, and a direct argument against stratifying menin therapy the way this graph has been contemplating.
- Which of the seven MLL-ALL PDXs failed to show extended survival on VTP-50469, and was it one of the three MLL-ENL models? The answer is in a figure panel of an open-access paper.
- Is HB11;19's exceptional MEN1 dependency reproducible, and is it related to its reciprocal-positive status? One additional reciprocal-negative KMT2A::MLLT1 line would turn a curiosity into a comparison.
- Does bleximenib have unpublished MLL-ENL data? Its published preclinical package has none, and it is in a phase 1/2 trial enrolling KMT2A-altered acute leukaemia.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
Menin is essential for maintenance of MLL-associated but not other oncogene-induced myeloid transformation; acute genetic ablation of menin reverses aberrant Hox gene expression mediated by MLL-menin promoter-associated complexes and specifically abrogates the differentiation arrest and oncogenic properties of MLL-transformed leukemic blasts.
The menin tumor suppressor protein is an essential oncogenic cofactor for MLL-associated leukemogenesis. · Yokoyama A, Somervaille TC, Smith KS, Rozenblatt-Rosen O, et al. · Cell 2005What this evidence cannot show
ABSTRACT ONLY - the abstract does not name MLL-ENL, and Europe PMC returns pmcid null with isOpenAccess N so PMC routes 2-4 are unavailable; a delegated worker also reports cell.com returning HTTP 403. The MLL-ENL specificity of this paper is attested here only through a later paper's citation of it [E2], which is second-hand.
Evidence that argues against this (1)
- Menin is also required for normal haematopoiesis and is a tumour suppressor in endocrine tissue, so essentiality in MLL-transformed cells does not by itself define a therapeutic window; that argument is made elsewhere in this graph. He S et al. 2016
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
MI-2 gave GI50 values of 9.5 uM in MV4;11 (MLL-AF4), 7.2 uM in KOPN-8 (MLL-ENL), 8.7 uM in ML-2 (MLL-AF6) and 18 uM in MonoMac6 (MLL-AF9); MI-2 and MI-3 blocked proliferation of mouse bone marrow cells transduced with MLL-AF9 and MLL-ENL with GI50 values of about 5 uM, induced differentiation in MLL-ENL transformed cells and downregulated Hoxa9 and Meis1 in them; and the authors state that these effects closely recapitulate the effects of excising the menin-binding motif from the N-terminus of MLL-GAS7 and MLL-ENL fusion proteins.
Menin-MLL inhibitors reverse oncogenic activity of MLL fusion proteins in leukemia. · Grembecka J, He S, Shi A, Purohit T, et al. · Nat Chem Biol 2012What this evidence cannot show
Figures relayed from a full-text reading (PMC3401603) by a delegated worker in this session; PMID, title and DOI independently re-fetched from PubMed esummary on 2026-09-19. MI-2/MI-3 are micromolar tool compounds. Four lines is too few to support a between-partner ranking, and the lines differ in lineage and co-lesions as much as in partner.
Evidence that argues against this (1)
- The same laboratory's follow-up study concluded explicitly that the effect is fusion-partner-independent, which is the correct reading of a four-line panel with overlapping GI50 values. He S et al. 2016
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
KOPN-8 (MLL-ENL / KMT2A-ENL) appears in the human cell line panels used to establish menin-KMT2A dependency for MI-503 and for ziftomenib.
Discovery and preclinical activity of the menin-KMT2A inhibitor ziftomenib in acute leukemia models. · Miao H, Wu T, Purohit T, Chen D, et al. · Blood 2026What this evidence cannot show
Relayed from a full-text reading (PMC13507986) by a delegated worker; identifiers independently re-fetched 2026-09-19. Panel GI50s are reported as a range (4.5-28 nM) rather than per line, so no MLLT1-specific potency is available. Presence in a panel is not a partner-specific result.
Evidence that argues against this (1)
- The clinical KMT2A-rearranged arm of the ziftomenib programme was halted over differentiation syndrome, so preclinical panel membership did not carry through to a KMT2A-r registrational dataset for this drug. Wang ES et al. 2024
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In the pivotal VTP-50469 study the MLL-rearranged ALL cell lines exposed to the inhibitor included KOPN8 and HB11;19, three of the eight patient-derived xenografts are recorded in the key resources table as B-ALL t(11;19) MLL-ENL models (PDX MLL-6, MLL-8 and MLL-14), and extended survival was observed in 6 of 7 additional MLL-ALL PDXs treated with 120 mg/kg VTP50469 for 28 days, with the identity of the non-responder not given in the text.
A Menin-MLL Inhibitor Induces Specific Chromatin Changes and Eradicates Disease in Models of MLL-Rearranged Leukemia. · Krivtsov AV, Evans K, Gadrey JY, Eschle BK, et al. · Cancer Cell 2019What this evidence cannot show
Relayed from a full-text reading (PMC7227117) by a delegated worker; identifiers independently re-fetched 2026-09-19. The non-responder's identity is in Figure 6B / S5D, which are images and were not parsed. The PDXs are recorded as t(11;19) without a sub-band in the resource table as quoted, so MLLT1 versus ELL is assigned by the 'MLL-ENL' label rather than by cytogenetic band.
Evidence that argues against this (1)
- Preclinical PDX efficacy did not translate proportionately: in the phase II registration cohort the t(11;19) subgroup was the largest named partner group at 13 of 57 and the overall CR+CRh rate was 22.8%. Issa GC et al. 2025
- E5strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Six MLL fusions representing the major translocation types, including MLL-AF9 and MLL-ENL as the most frequent nuclear-partner fusions plus MLL-CBP, MLL-AF6, MLL-GAS7 and MLL-AF1P, were transformed into murine bone marrow cells, and MI-2-2 strongly inhibited growth of all of them but not of control cells transformed with Hoxa9-Meis1 or E2A-HLF; colony number and size were substantially reduced in all primary MLL patient samples tested, which included one MLL-ENL case.
Menin-MLL inhibitors block oncogenic transformation by MLL-fusion proteins in a fusion partner-independent manner. · He S, Malik B, Borkin D, Miao H, et al. · Leukemia 2016What this evidence cannot show
Relayed from a full-text reading (PMC4830728) by a delegated worker; identifiers independently re-fetched 2026-09-19. No per-partner GI50 values are given in the letter text, so 'partner-independent' is supported qualitatively rather than by an equivalence test. One MLL-ENL primary sample is n = 1.
Evidence that argues against this (1)
- A qualitative 'all partners respond' conclusion from six constructs and six primary samples cannot exclude the two-fold to three-fold differences that would matter clinically, and the trial data show subgroup CR+CRh rates ranging from 0% to 50% across partner buckets. Issa GC et al. 2025
- E6moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
The bleximenib (JNJ-75276617) preclinical paper contains zero occurrences of MLLT1, MLL-ENL, ENL, KOPN or 11;19 in its full text; its named KMT2A-rearranged models are MOLM14 AML and the B-ALL models CBAM-68552 and CBAB-62871.
Preclinical efficacy of the potent, selective menin-KMT2A inhibitor JNJ-75276617 (bleximenib) in KMT2A- and NPM1-altered leukemias. · Kwon MC, Thuring JW, Querolle O, Dai X, et al. · Blood 2024What this evidence cannot show
The zero counts come from a full-text search (PMC11419783) performed by a delegated worker; the PMID, title and DOI were independently re-fetched 2026-09-19 but the text search was not repeated by this entry's author. The named B-ALL models' fusion partners are not stated in the quoted text, so one of them could be a t(11;19) model under a code name.
Evidence that argues against this (1)
- Absence of a partner from one preclinical paper is not evidence the drug is inactive against it; the menin-binding motif is identical across partners and the partner-independence result would predict activity. He S et al. 2016
menin dependence of this partner, demonstrated genetically, chemically and in PDXs
partner-specific menin dossier
three of eight PDXs in the pivotal VTP-50469 study are t(11;19) MLL-ENL
bounded absence: no MLL-ENL model appears anywhere in the bleximenib preclinical package
the one reciprocal-positive line is the MOST menin-dependent, the opposite of the remainder prediction
the same question for the paralogous partner, with the opposite evidence profile