Agency law — actual authority, apparent authority, and what revocation cannot undo
also called power of attorney analogy · agency coupled with an interest
Menin acts on behalf of the fusion protein, the way an agent acts on behalf of a principal. Law has spent centuries working out exactly what happens when you revoke an agent's authority, and almost none of it is instantaneous — which is a strange and useful thing to notice about a drug we dose every twelve hours.
The same thing, for a scientist
Menin is an adaptor: it holds no independent function in the leukemic program and exists in this story only as the thing that lets KMT2A::AFF1 act on chromatin. That is exactly the legal relation of agent to principal. Agency law then supplies three structures that biology has not systematically applied here. (1) Revocation operates prospectively; acts already completed under authority continue to bind, which predicts hysteresis between drug occupancy and program output. (2) Apparent authority survives revocation until third parties have notice, which predicts a residual set of fusion binding sites held by menin-independent tethering. (3) An agency coupled with an interest is irrevocable, which is a precise description of what a MEN1 interface mutation does. The first two generate testable hypotheses that are not in the graph; the third re-describes something already known and is included only to show the mapping is not arbitrary.
Why menin is an agent and not a partner1 deeper
Menin contributes no catalytic activity to the leukemic complex. Its role is to bind the extreme N-terminus of KMT2A at MBM1 and bridge it, via LEDGF/PSIP1, onto chromatin. It is a conduit for someone else's authority, which is the defining feature of agency rather than partnership. The same protein simultaneously acts for other principals, most notably JUND, using the same pocket and producing the opposite transcriptional effect 2 — in agency terms, a single agent serving conflicting principals, which is the subject of analogy-law-fiduciary-conflicted-menin.
Revocation is prospective, and that is the interesting part
A principal may revoke an agent's authority at will, but the revocation does not unwind acts already performed. The biological counterpart is that removing menin stops new recruitment; it does not directly erase H3K79me2 already deposited, loops already formed, or a transcriptional state already established. Degron removal of MLL::AF9 changes elongation within 15 minutes, with loss of the active chromatin landscape occurring subsequently 1 — elongation first, chromatin after. The ordering is precisely what prospective revocation predicts, and it means occupancy and output are separated in time in both directions.
The three imported structures and what each predicts
| Agency doctrine | Biological counterpart | Prediction |
|---|---|---|
| Revocation is prospective; completed acts bind | Deposited marks and established loops persist after menin is displaced | Hysteresis: program collapse and program recovery have different time constants (hyp-agency-revocation-hysteresis-pulsed-menin) |
| Apparent authority survives until third parties have notice | The fusion retains CXXC-DNA binding and LEDGF tethering independent of menin | A residual occupancy set under saturating drug (hyp-apparent-authority-residual-fusion-occupancy) |
| Agency coupled with an interest is irrevocable | MEN1 interface mutation: the drug can no longer revoke, while the natural grant to MLL1 is untouched | Already observed in patients 3; the mapping is validated rather than predictive here |
| Ratification: a principal may adopt an unauthorised act, making it its own | A cell that has adopted the fusion program as its identity no longer needs the fusion to maintain it | The adaptive, non-genetic resistance state already recorded in the graph |
Where the analogy fails
Agency has an intending principal who chooses to revoke. Nothing in this system intends anything, and the direction of the metaphor can invert without warning: it is equally coherent to call the fusion the agent and chromatin the principal. The mapping is only useful because the three doctrines above happen to be statements about timing and residue — what survives a withdrawal of authority and for how long — and timing is exactly what has not been measured here.
The analogy also offers nothing about dose-response, cooperativity or pharmacokinetics, which is where most of the real difficulty in an intermittent-dosing proposal lives.
| domain | law |
|---|---|
| source doctrine | agency: actual and apparent authority, prospective revocation, ratification, agency coupled with an interest |
| already validated mapping | agency coupled with an interest maps onto MEN1 interface mutation, which is observed; this is a check on the mapping rather than a new prediction |
mapping
[
{
"bio": "KMT2A::AFF1 fusion protein",
"analog": "the principal"
},
{
"bio": "menin, a non-catalytic adaptor bridging the fusion to chromatin via LEDGF",
"analog": "the agent holding a power of attorney"
},
{
"bio": "chromatin and the target gene loci",
"analog": "third parties who deal with the agent"
},
{
"bio": "a menin inhibitor occupying the MBM1 pocket",
"analog": "revocation of the power of attorney"
},
{
"bio": "H3K79me2, established enhancer-promoter loops and an established transcriptional state persisting after menin displacement",
"analog": "acts completed under authority, which revocation does not unwind"
},
{
"bio": "CXXC-DNA binding and LEDGF/PSIP1 tethering, which do not require menin",
"analog": "apparent authority, which survives revocation until third parties have notice"
},
{
"bio": "MEN1 interface mutation, which blocks drug binding while preserving MLL1 binding",
"analog": "an agency coupled with an interest, which the principal cannot revoke"
},
{
"bio": "a cell state that maintains the leukemic program without the fusion circuit",
"analog": "ratification — the principal adopts the act and no longer needs the agent"
},
{
"bio": "menin also binding JUND in the same pocket with opposite transcriptional effect",
"analog": "an agent acting for two principals with conflicting interests"
}
]generated hypotheses
[ "hyp-agency-revocation-hysteresis-pulsed-menin", "hyp-apparent-authority-residual-fusion-occupancy" ]
- If completed acts really do persist, then a drug holiday is not a clean off-state and intermittent dosing could allow the program to be maintained on deposited marks alone while relieving selection pressure on nothing.
- Apparent authority cuts against the whole interface strategy: any residual menin-independent occupancy is, by construction, a ceiling on what a menin inhibitor can achieve no matter how good the chemistry is.
- How fast does the KMT2A::AFF1 program re-establish after a menin inhibitor is withdrawn? The collapse rate is known within minutes and the recovery rate has never been reported for any KMT2A fusion.
- Is there a measurable fraction of fusion chromatin occupancy that survives saturating menin inhibition, and is it the same fraction across patients?
- Does displacing JUND from the same pocket contribute to any of the observed toxicity, and would a partner-selective ligand avoid it?
menin mapped to an agent holding a revocable power of attorney
- analogous to →Reversibility and kinetics — does the program collapse when the fusion is removed, and how fast?
prospective revocation predicts hysteresis between occupancy and output
- analogous to →MEN1 drug-interface mutations — the first proof that a chromatin drug can drive Darwinian escape in patients
an agency coupled with an interest is irrevocable, which is what the interface mutation creates
generated by prospective revocation
generated by apparent authority surviving revocation