Bounded absence: not one menin-inhibitor study, trial or preclinical paper has ever recorded reciprocal fusion status
also called menin inhibitor reciprocal status absence · unstratified menin trials · no reciprocal eligibility criterion
About half of these children carry a second fusion protein that the menin drug physically cannot bind. No trial, no laboratory study and no eligibility criterion anywhere records which children those are.
The same thing, for a scientist
This is a counted absence with its queries stated. On 2026-09-19, PubMed returns: `(revumenib OR ziftomenib OR "menin inhibitor" OR SNDX-5613 OR KO-539) AND (reciprocal OR AF4-MLL)` = **1 record**, which concerns JUND and AP-1 in menin-inhibitor resistance and does not mention reciprocal fusion status [E1]; `menin inhibitor AND AF4-MLL` = **0**; `AF4-MLL AND menin` = **0**; `(revumenib OR ziftomenib OR bleximenib OR "menin inhibitor") AND "AF4-MLL"` = **0** [E4]. The absence matters because the architecture is not in doubt: the reciprocal product lacks the menin-binding motif by construction [E2], and the one independently characterised menin-blind KMT2A architecture, NUP98::KMT2A, is reported menin-inhibitor resistant [E3]. Meanwhile every standard t(4;11) cell line used in menin-inhibitor pharmacology is reciprocal-positive [E5], so the preclinical package was generated entirely in one arm of a variable nobody measured - and it is unstratified in a known rather than unknown direction.
What would break this absence claim, and what would not
Would break it: any trial protocol, supplementary table or conference abstract recording reciprocal transcript status in menin-inhibitor-treated patients; any preclinical paper comparing a reciprocal-positive with a reciprocal-negative model under menin inhibition. Would not break it: a paper that merely mentions both topics. The absence is bounded by three things and each is a way it could be wrong: PubMed indexing only, so a registry document or a company analysis is out of scope; the alias set used, so a paper writing 'der(4) fusion' or 'A4M' in a way these queries miss would be invisible; and the der(n) parser defect - PubMed returns a clean zero for any query that combines a der(n) token with explicit parentheses, so a search written that way would produce this exact result for a purely mechanical reason.
The eligibility-criterion half
No menin-inhibitor eligibility criterion this beat could find mentions reciprocal status. Eligibility for the pivotal revumenib programme is written on KMT2A rearrangement, not on fusion-product architecture 6. That is the correct clinical choice given the evidence available, and it is also the reason the data that would settle the question is not being generated as a by-product of care.
| kind | bounded absence |
|---|---|
| what would overturn | any protocol, supplement or abstract recording reciprocal transcript status in menin-inhibitor-treated patients, or any preclinical comparison of reciprocal-positive and reciprocal-negative models under menin inhibition |
queries run 2026 09 19 — the numbers behind the chart
{
"(revumenib OR ziftomenib OR \"menin inhibitor\" OR SNDX-5613 OR KO-539) AND (reciprocal OR AF4-MLL)": 1,
"menin inhibitor AND AF4-MLL": 0,
"AF4-MLL AND menin": 0,
"(revumenib OR ziftomenib OR bleximenib OR \"menin inhibitor\") AND \"AF4-MLL\"": 0,
"reciprocal fusion AND menin inhibitor": 1
}aliases searched
[ "AF4-MLL", "AF4.MLL", "AFF1-KMT2A", "AFF1::KMT2A", "reciprocal fusion", "MLL-AF4 reciprocal" ]
scope limits
[ "PubMed index only", "no conference abstracts", "no trial protocol documents", "der(n)-plus-parenthesis parser defect" ]
- If reciprocal-positive patients respond worse to menin inhibition, the effect is currently invisible and would present as unexplained heterogeneity in response rate.
- If they respond better - which the chemotherapy-era prognostic data would naively predict - then the trials have been enriched for responders in a way nobody controlled.
- Either way, an approved therapy is being given without regard to a variable that is architecturally guaranteed to matter, measurable by one PCR, and present in roughly half of the commonest paediatric KMT2A rearrangement.
- Do any menin-inhibitor programmes hold banked diagnostic RNA from t(4;11) patients with consent permitting a single additional RT-PCR?
- Is MV4-11 reciprocal-positive? If every line in the preclinical package is positive, the package is uniform in an unmeasured variable.
- Has any reciprocal-only patient - about 5% of KMT2A-rearranged leukaemia - ever received a menin inhibitor, and what happened?
- Would a prospective protocol collect reciprocal status if asked? The assay is one RT-PCR and it is already validated in a published cohort.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Menin-MLL inhibitors disrupt menin-JUND interactions as well as menin-MLL, and JUND inactivation or JNK inhibition synergises with menin-MLL inhibitors, with increased JUND chromatin occupancy contributing to resistance - a mapped resistance mechanism in which reciprocal fusion status plays no part and is not mentioned.
Menin-MLL inhibitors enhance JUND activity in MLLr leukemic cells contributing to tumorigenesis and therapy resistance. · Ozyerli-Goknar E, Nizamuddin S, Fetsch V, et al. · Blood 2026What this evidence cannot show
Cited as the single record returned by the absence query, to show what the field is studying instead, not as support for any reciprocal claim. The study is in MLLr AML and ALL lines whose reciprocal status is not reported.
Evidence that argues against this (1)
- That resistance mechanisms are being actively and successfully mapped through other axes shows the absence of any reciprocal-status analysis is a framing choice rather than an inability to do the work. Marschalek R 2020
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
The reciprocal product comprises AF4 exons 1-3 fused to MLL exons 12-37 and therefore lacks the KMT2A N-terminal menin-binding motif entirely.
The role of reciprocal fusions in MLL-r acute leukemia: studying the chromosomal translocation t(4;11). · Wilhelm A, Marschalek R · Oncogene 2021What this evidence cannot show
Architectural inference from reported exon composition; no menin binding or competition experiment has been performed on any reciprocal KMT2A product.
Evidence that argues against this (1)
- No published experiment has tested whether menin binds a reciprocal KMT2A fusion protein, so the absence of a handle is inferred from sequence rather than measured. Marschalek R 2020
- E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
NUP98::KMT2A, which lacks the menin-binding exon while retaining the WIN motif and SET domain, is transforming in an inducible transgenic mouse and is reported resistant to menin inhibition with reduced rather than elevated HOXA output.
Transforming activities of the NUP98-KMT2A fusion gene associated with myelodysplasia and acute myeloid leukemia. · Fisher JN, Thanasopoulou A, Juge S, et al. · Haematologica 2020What this evidence cannot show
Murine, long latency, tool-compound resistance; one rearrangement, not a class. It is the nearest available analogue of the reciprocal architecture under menin blockade, not a test of the reciprocal product itself.
Evidence that argues against this (1)
- The reciprocal product differs from NUP98::KMT2A in its N-terminal partner and in carrying only KMT2A exons 12-37 rather than nearly the whole ORF, so transferring a resistance phenotype between them is an architectural analogy, not a measurement. Wilhelm A et al. 2021
- E4moderate confidence 3 of 5: One strong study, or several smaller consistent ones.database
PubMed returns zero records for 'menin inhibitor AND AF4-MLL', zero for 'AF4-MLL AND menin', zero for '(revumenib OR ziftomenib OR bleximenib OR "menin inhibitor") AND "AF4-MLL"', and one record for '(revumenib OR ziftomenib OR "menin inhibitor" OR SNDX-5613 OR KO-539) AND (reciprocal OR AF4-MLL)', run 2026-09-19.
PubMed E-utilities esearch, queries and counts run 2026-09-19 · NCBI · NCBI E-utilities 2026What this evidence cannot show
A counted absence is only as good as its symbol set and its index. 'AF4-MLL' and 'AF4.MLL' each return 50 records and are the dominant spellings; 'AFF1-KMT2A' and 'AFF1::KMT2A' return 2 each. Unindexed material - trial protocols, company analyses, conference abstracts - is entirely out of scope. And the PubMed parser returns a clean zero for any query combining a 'der(n)' token with explicit parentheses, which is a failure mode indistinguishable from a real absence.
Evidence that argues against this (1)
- The same field has produced a clinical stratification variable from an equally cheap assay before: reciprocal transcript status itself was measured by RT-PCR in an Interfant cohort and produced a 50-point survival separation, so the assay is neither expensive nor unprecedented. Agraz-Doblas A et al. 2019
- E5strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
The standard human t(4;11) lines RS4;11 and SEMK2-M1 both express MLL-AF4 and AF4-MLL.
t(4;11) leukemias display addiction to MLL-AF4 but not to AF4-MLL. · Kumar AR, Yao Q, Li Q, et al. · Leuk Res 2011What this evidence cannot show
Stated for two lines only; MV4-11, also used in menin-inhibitor pharmacology, is not covered and its reciprocal status was not found by this beat.
Evidence that argues against this (1)
- Because no reciprocal-negative t(4;11) line is documented, the preclinical package cannot be re-stratified retrospectively even in principle - the comparator arm does not exist as material. Kowarz E et al. 2007
- E6moderate confidence 3 of 5: One strong study, or several smaller consistent ones.trial_registry
AUGMENT-101 (NCT04065399, recruiting, minimum age 30 days) admits participants with acute leukaemia harbouring a KMT2A rearrangement, NUP98 rearrangement or NPM1 mutation, with phase 2 cohorts defined as KMT2A-rearranged ALL/MPAL (2A), KMT2A-rearranged AML (2B), NPM1-mutant AML (2C) and, for cohort 2D, 'acute leukemia with a genetic mutation expected to lead to HOX/MEIS upregulation'; the word 'reciprocal' appears zero times in the eligibility criteria, as does any 'der(' notation.
ClinicalTrials.gov NCT04065399 (AUGMENT-101, revumenib in relapsed/refractory acute leukaemia) · Syndax Pharmaceuticals · ClinicalTrials.gov 2026What this evidence cannot show
Fetched from the raw ClinicalTrials.gov v2 API on 2026-09-19 and read directly; overallStatus RECRUITING, no whyStopped. A registry record states what the protocol permits, not what correlative analyses were run or what was collected and left unreported. Cohort 2D's HOX/MEIS criterion is notable rather than decisive: in the one cohort that measured both, HOXA expression was exclusive to reciprocal-expressing t(4;11) patients, so a HOX-based criterion may be selecting on reciprocal status without naming it.
Evidence that argues against this (1)
- A registry entry is opposable on scope: banked correlative material may exist under this protocol with reciprocal status obtainable retrospectively, in which case the absence is a reporting gap rather than a collection gap. Agraz-Doblas A et al. 2019
extends the same absence from menin trials to all KMT2A::MLLT1 analyses
supplies the reciprocal-positive reference population
Bounded absence: no study of this drug class records reciprocal fusion status.
Same absence across the class; the query covered ziftomenib, bleximenib, SNDX-5613 and KO-539.
The protein whose status has never been recorded in any menin-inhibitor study.
Establishes that the proposed reanalysis has no precedent anywhere in the indexed literature.
Confirms that hypothesis's central claim as a counted absence with stated queries.
Supplies the alias set and the parser caveat that bound that absence claim.