Fiduciary conflict of interest — law does not remove a conflicted agent, it separates the mandates
also called conflict of interest analogy · segregation of duties · information barrier
Menin works for two bosses with opposite interests, using the same office for both. When a professional is caught in that position, the law almost never bars them from working; it separates the two jobs. Current drugs do the opposite and shut the whole office down.
The same thing, for a scientist
Menin binds KMT2A and JUND in the same central pocket, with opposite transcriptional consequences [E1]. It is simultaneously a tumour suppressor in endocrine tissue and a required cofactor in KMT2A-rearranged leukemia. That is a textbook fiduciary conflict: one agent, two principals, one office, irreconcilable interests. The legal response to such a conflict is developed and specific — segregation of duties, information barriers, informed consent, or resignation from one mandate only — and it is almost never a blanket disqualification. Current menin inhibitors are blanket disqualification: they occupy the shared pocket and terminate both mandates at once. The 2026 demonstration that menin-MLL inhibitors also disrupt menin-JUND, increase JUND chromatin occupancy and thereby contribute to resistance [E2] shows this is not an abstract concern — the second mandate, terminated collaterally, produces a gain of function that works against the drug.
The conflict, set out precisely1 deeper
| Principal | What menin does for it | Consequence of menin doing it | Consequence of menin stopping |
|---|---|---|---|
| KMT2A / KMT2A::AFF1 | binds MBM1, bridges to LEDGF and chromatin | HOXA9/MEIS1 program, leukemia | leukemic program collapses — the therapeutic goal |
| JUND | binds the same pocket, blocks JNK-mediated JUND phosphorylation | JUND-driven transcription suppressed | JUND released onto chromatin, target genes up, resistance 2 |
| Endocrine tissue (parathyroid, pancreas, pituitary) | tumour-suppressor function; germline loss causes MEN1 syndrome | normal endocrine restraint | unknown over years; raised intact PTH in 34% of patients at 2.5 months median exposure |
The drug is aimed at row 1 and hits rows 2 and 3 as collateral. Rows 2 and 3 are not side effects in the ordinary sense; they are the other mandates of the same office.
Why the JUND arm changes the picture
Until 2026 the JUND interaction was an interesting structural fact. It is now a reported resistance mechanism: menin-MLL inhibitors increase JUND chromatin occupancy, upregulate JUND target genes, and JUND knockout or JNK inhibition synergises with menin inhibition in vitro and reduces tumour burden in vivo 2. That means part of the ceiling on menin-inhibitor efficacy is self-inflicted — created by the drug terminating the second mandate. A ligand that left the menin-JUND contact intact would not create it.
What law does with a conflicted fiduciary, and the biological counterpart
- Segregation of duties. Keep the agent but partition the roles so one cannot contaminate the other. Biological counterpart: a ligand selective for the KMT2A-binding mode over the JUND-binding mode, or for the leukemic complex composition over the endocrine one.
- Information barrier (a Chinese wall). Separate by context rather than by person. Biological counterpart: context-selective engagement — a ligand whose binding depends on a partner present only in the leukemic complex, such as LEDGF, or a lineage-restricted delivery strategy.
- Resignation from one mandate. Biological counterpart: a degrader or targeted-protein-degradation approach restricted to the leukemic cell, which removes the agent entirely but only where it is conflicted.
- Informed consent and monitoring. The unglamorous option and the one actually available today: accept the conflict and monitor the second mandate. Nobody is measuring JUND target-gene output in treated patients, and intact PTH monitoring is not universal.
The strongest objection, which is structural
The original crystallography reports that menin binds short peptides of MLL1 or JUND in the same manner, in the same deep pocket 1. If the two binding modes are genuinely superimposable, partner-selective chemistry may be impossible and the whole segregation-of-duties branch of this analogy collapses onto the last option, monitoring. This is the honest state of the argument: the analogy identifies a strategy that the structural biology may already have ruled out, and the first thing to do is look hard at whether the two binding modes differ anywhere — in flanking contacts, in induced fit, in the LEDGF-facing surface — rather than assume they do.
| domain | law |
|---|---|
| source doctrine | fiduciary duty, conflict of interest, segregation of duties and information barriers |
| strongest objection | The 2012 structures report MLL1 and JUND peptides binding the same pocket in the same manner, which may mean no selective chemistry exists. The analogy identifies a strategy the structural biology may already exclude, and testing that is the first step rather than an afterthought. |
mapping
[
{
"bio": "menin, required by the leukemic program and tumour-suppressive in endocrine tissue",
"analog": "a fiduciary serving two principals with opposing interests"
},
{
"bio": "the single central pocket that binds both KMT2A and JUND",
"analog": "one office used for both mandates, with no partition"
},
{
"bio": "a pan-pocket menin inhibitor",
"analog": "blanket disqualification of the conflicted fiduciary"
},
{
"bio": "JUND released onto chromatin when the drug displaces it, driving resistance",
"analog": "collateral termination of the second mandate producing unintended consequences"
},
{
"bio": "a ligand selective for the KMT2A binding mode that preserves menin-JUND",
"analog": "segregation of duties"
},
{
"bio": "a ligand whose engagement depends on a leukemia-restricted complex partner such as LEDGF",
"analog": "an information barrier that separates by context rather than by person"
},
{
"bio": "leukemia-restricted targeted protein degradation of menin",
"analog": "resignation from one engagement only"
},
{
"bio": "monitoring intact PTH and JUND target-gene output on therapy",
"analog": "disclosure, consent and ongoing supervision — the option available today"
}
]generated hypotheses
[ "hyp-partner-selective-menin-ligand-separates-mandates" ]
- If the two binding modes are truly superimposable, then every menin-pocket drug that will ever exist carries the JUND liberation arm, and the resistance it causes is a permanent property of the target rather than of the compound.
- A partner-selective ligand would necessarily be lower-affinity for the shared pocket than a pan-pocket binder, so selectivity would be bought with potency, in a disease where complete responses are already hard to reach.
- Nobody is monitoring the second mandate. Endocrine consequences of years of menin inhibition in a one-year-old are untested rather than excluded, and JUND output is not measured at all.
- Do the MLL1 and JUND binding modes differ anywhere — flanking contacts, induced fit, the LEDGF-facing surface — enough for a selective ligand to exist?
- Do revumenib, ziftomenib, bleximenib and enzomenib differ in how much menin-JUND they displace? If they do, that difference is a plausible and immediately measurable explanation for their differing resistance and toxicity profiles.
- Is JUND target-gene output measurable in on-treatment patient samples, and does it track resistance?
- Is the endocrine mandate separable by complex composition as well, which would be the route to a drug that spares the parathyroid?
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Crystal structures of free menin and of menin bound to MLL1, to JUND, and to an MLL1-LEDGF heterodimer show a deep central pocket that binds short MLL1 or JUND peptides in the same manner but with opposite transcriptional consequences: menin-JUND blocks JNK-mediated JUND phosphorylation and suppresses JUND-induced transcription, while menin-MLL1 promotes transcription and simultaneously engages LEDGF at a distinct surface formed by both menin and MLL1.
The same pocket in menin binds both MLL and JUND but has opposite effects on transcription · Huang J, Gurung B, Wan B, et al. · Nature 2012What this evidence cannot show
Short peptides in crystals; the full-length proteins may engage additional surfaces that the peptide structures do not capture, which is precisely where any selectivity would have to live.
Evidence that argues against this (1)
- WHERE THE ANALOGY BREAKS. The legal remedies the entry lists — segregation of duties, information barriers, informed consent, resignation from one mandate — all presuppose an agent that can be *instructed*, and a physical separation of the two engagements. Menin offers neither: the crystal structures show MLL1 and JUND peptides occupying the same deep central pocket in the same manner. A pocket-occupying ligand terminates both mandates by one steric act, and there is no structural place to put the information barrier. The analogy's chief output is therefore a prescription with an empty action set for the current chemical series. What it does correctly identify is that the interesting chemistry is *not* in the pocket: the LEDGF-engaging surface is formed jointly by menin and MLL1 and is distinct from the peptide pocket, so an agent directed there is the structurally licensed version of 'segregation of duties'. That is a different and better hypothesis than a partner-selective pocket ligand. Huang J et al. 2012
- E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
Quantitative mass spectrometry in MLL-rearranged leukemic cells showed menin-MLL inhibitors also disrupt menin-JUND interactions and increase JUND chromatin occupancy, upregulating JUND target genes and contributing to therapy resistance; JUND knockout or JNK inhibition synergised with menin-MLL inhibitors, and mice engrafted with JUND-deficient leukemia cells had reduced tumour burden.
Menin-MLL inhibitors enhance JUND activity in MLLr leukemic cells contributing to tumorigenesis and therapy resistance · Ozyerli-Goknar E, Perner F, Armstrong SA, Zeiser R, Timmers HTM, et al. · Blood 2026What this evidence cannot show
Very recent, single group, not independently replicated. The magnitude of the JUND contribution relative to other resistance routes is unquantified.
Evidence that argues against this (1)
- Searched for independent replication of, or contradiction to, the finding that menin-MLL inhibitors disrupt menin-JUND and drive JUND chromatin gain. None found either way: the observation is from a single 2026 report and has not, as far as this search could establish, been reproduced or challenged. That is not a contradiction but it is a single-laboratory dependency for one of this analogy's two load-bearing facts. Recorded so the status is known to have been checked. Q12-analogy-counter (search record) 2026
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Menin and MLL1 are genetically separable in normal haematopoiesis, with no genetic interaction in steady-state or regenerative haematopoiesis or B-cell differentiation, which is the basis for expecting a therapeutic window at all.
Distinct pathways regulated by menin and by MLL1 in hematopoietic stem cells and developing B cells · Li BE, Gan T, Meyerson M, Rabbitts TH, Ernst P · Blood 2013What this evidence cannot show
Mouse genetics; separability of genetic requirement does not guarantee separability of a drug-binding surface.
Evidence that argues against this (1)
- Genetic separability of menin and MLL1 in haematopoiesis does not establish the therapeutic window the entry infers from it, because the second principal in this conflict is not MLL1 and is not haematopoietic. Menin is a tumour suppressor whose loss or inactivation causes multiple endocrine neoplasia type 1, a hereditary syndrome characterised by tumorigenesis in multiple endocrine organs, and a subset of MEN1 missense mutations act specifically by disrupting the menin-JUND interaction that pharmacological pocket occupancy also disrupts. So the on-target long-term liability of sustained menin-pocket blockade is endocrine neoplasia, which a haematopoietic knockout study cannot speak to and which no menin-inhibitor trial has run long enough to observe. This belongs in `bite_back` for the whole menin programme, not only for this analogy. Huang J et al. 2012
- E4moderate confidence 3 of 5: One strong study, or several smaller consistent ones.review
Menin inhibitors differ from one another in safety, QTc, cytopenia and pharmacokinetic profile, and acquired MEN1 mutations reported in 39% of post-revumenib relapses have not been identified after ziftomenib or bleximenib, indicating that the class is not chemically or biologically interchangeable at the pocket.
The promise of menin inhibitors: from approval to triplet regimens · Thakur RK, Wang ES · Hematology, American Society of Hematology Education Program 2025What this evidence cannot show
Review-level. Differences in reported resistance mutations may reflect differing follow-up and sequencing intensity rather than chemistry.
Evidence that argues against this (1)
- The claim that the class is 'not chemically or biologically interchangeable at the pocket', evidenced by MEN1 mutations after revumenib but not after ziftomenib or bleximenib, is currently absence of evidence rather than evidence of difference: menin-binding-site mutations emerged in 13% of 42 patients on a revumenib-containing triplet, so the rate is a function of exposure, regimen and follow-up, and the comparator agents have smaller and shorter-followed exposed populations. Matched-exposure sequencing is needed before compound-level selectivity claims are safe. Issa GC et al. 2026
- E5established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
A fiduciary facing a conflict of interest is subject to duties of loyalty and disclosure; the recognised responses are disclosure and informed consent, segregation of duties or information barriers, and withdrawal from one engagement, rather than automatic disqualification from all.
Conflict of interest (Wex legal encyclopedia) · Legal Information Institute, Cornell Law School · Cornell Law School Legal Information Institute 2024What this evidence cannot show
Source-domain reference. It shapes the space of strategies considered and supports no biological claim.
Evidence that argues against this (1)
- The legal doctrine is also less uniform than the entry's summary implies, which matters because the entry treats the remedy list as settled law that biology should imitate. Even in the closely related area of severing or saving over-broad contractual terms, US jurisdictions split at least four ways — refusing to rescue the agreement at all, deleting offending words only, reforming the language (in some states by legislative mandate), or reforming conditional on good-faith drafting. A prescription drawn from 'the legal response' is therefore drawn from one of several legal responses, and the analogy should say which jurisdiction's rule it is importing and why that one. Beck R 2025
structural constraint on the profiling's premise: MLL1 and JUND occupy the same pocket in the same manner, so expect selectivity ratios near unity and pre-specify what would count as meaningful
menin as a fiduciary serving the leukemic, JUND and endocrine mandates from one office
the shared pocket as an unpartitioned office
- related to →Adaptive, non-genetic resistance — the leukemia keeps the fusion and rewires how it is maintained
JUND liberation is a drug-created, non-genetic resistance arm
generated by segregation of duties
menin as an agent serving two principals with conflicting interests