t(11;19) is the largest named partner group in the revumenib registration trial - and the trial never resolves whether those patients have MLLT1 or ELL
also called AUGMENT-101 t(11;19) · revumenib partner distribution · 19p13.1 versus 19p13.3 in menin trials
In the trial that got the first menin inhibitor approved, the single biggest identified group of patients was the one with a break between chromosomes 11 and 19 - the fusion this beat is about. But the trial's records never say which of two different genes on chromosome 19 was involved, and in children with myeloid leukaemia the other one is actually more common.
The same thing, for a scientist
AUGMENT-101's phase II registration cohort reports KMT2A translocations in Table 1 as: t(9;11)(p22;q23) 11 (19.3%), **t(11;19)(q23;p13) 13 (22.8%)**, t(10;11)(p12;q23) 7 (12.3%), t(6;11)(q27;q23) 7 (12.3%), t(4;11)(q21;q23) 2 (3.5%), t(1;11) 2 (3.5%), fusion partner unknown 13 (22.8%) and other translocations 2 (3.5%), in an efficacy population of n = 57; in the 94-patient safety population t(11;19) is 17 (18.1%) [E1]. t(11;19) is therefore tied with 'partner unknown' as the largest single entry and is the largest *named* one. The trial reports it as **t(11;19)(q23;p13)** with no sub-band, so KMT2A::MLLT1 at 19p13.3 and KMT2A::ELL at 19p13.1 are pooled [E1]. That matters quantitatively: in 1,256 children with KMT2A-rearranged AML, 19p13.1/KMT2A::ELL is *more* common than 19p13.3/KMT2A::MLLT1 (75 versus 56), and the two have opposite risk assignments - MLLT1 independently adverse (EFS HR 1.6, P = .01; CIR HR 1.8, P < .01), ELL not significant on any endpoint [E2]. The efficacy population is 86.0% AML [E1]. **So there is no published KMT2A::MLLT1-specific menin-inhibitor response rate, and the number that is sometimes read as one is a mixture of two partners in unknown proportion.**
The partner-response analysis exists and is not publicly readable
The paper states: "Exploratory analysis with limited sample sizes demonstrated efficacy for KMT2Ar across various translocation partners (Data Supplement, Table S3)." and, in the Discussion, "Although sample sizes were small, efficacy was observed across subgroups, including age, lineage, prior venetoclax, and KMT2A translocation partner." 1 Table S3 is not in PMC - the PMC record carries the property pmc-prop-has-supplement no, verified on reading the full text - and the publisher's supplementary file was not retrievable from this environment. So the one partner-stratified response table that is known to exist for any menin inhibitor is referenced in an open-access paper and is not itself open. This is a bounded unreadability, not an absence: the analysis was done and the paper says so.
Why the sub-band problem is worse than it looks
KMT2A::ELL and KMT2A::MLLT1 are not near-neighbours biologically. ELL is a direct elongation factor and a core SEC component; MLLT1 is a YEATS acyl-lysine reader whose fusion retains a chromatin-reading module in 84% of patients. Their disease distributions differ (ELL in this graph's recombinome row is 119 cases with zero infant ALL; MLLT1 is 460 cases of which 234 are infant ALL), and their paediatric AML risk assignments are opposite 23. Pooling them produces a group whose biology is the average of two things that share only a chromosome arm. The graph already holds concept-cytogenetic-band-versus-molecular-partner-misassignment on this; the new fact here is that the pooling reaches into the registration dataset of an approved drug.
What would fix it, and it is cheap
Central molecular confirmation of KMT2A rearrangement was already performed for the efficacy population 1. Whether that confirmation resolved the 3' partner at the gene level is not stated in the paper. If it did, the sub-band split for the 13 t(11;19) patients already exists in the trial database and is one query. If it did not, archived material would need re-testing. Either way the number that the field has been quoting as a t(11;19) response rate cannot be attributed to KMT2A::MLLT1 until this is done.
| metric | fusion partner distribution in a menin-inhibitor registration trial |
|---|---|
| value | 13 |
| unit | patients with t(11;19) in the efficacy population |
| population | relapsed/refractory KMT2A-rearranged acute leukaemia, AUGMENT-101 phase II |
| year | 2025 |
| source | PMID 39121437 Table 1 |
| nct | NCT04065399 |
| efficacy population n | 57 |
| safety population n | 94 |
| band resolution | t(11;19)(q23;p13) - sub-band NOT resolved anywhere in the paper; MLLT1 (19p13.3) and ELL (19p13.1) are pooled |
| overall cr crh PCT | 22.8 |
| overall orr PCT | 63.2 |
| partner stratified table | Data Supplement Table S3 - exists, referenced by the paper, not in the PMC deposit (pmc-prop-has-supplement no), not retrievable from this environment |
partner rows efficacy — the numbers behind the chart
{
"t(9;11)": 11,
"t(11;19)": 13,
"t(10;11)": 7,
"t(6;11)": 7,
"t(4;11)": 2,
"t(1;11)": 2,
"unknown": 13,
"other": 2
}partner rows safety — the numbers behind the chart
{
"t(9;11)": 21,
"t(11;19)": 17,
"t(10;11)": 12,
"t(6;11)": 8,
"t(4;11)": 7,
"t(1;11)": 2,
"unknown": 24,
"other": 3
}lineage mix efficacy — the numbers behind the chart
{
"AML": 49,
"ALL": 7,
"ambiguous": 1
}- A clinician reading 't(11;19)' in the revumenib label context may be treating a KMT2A::ELL patient on the strength of data mostly generated in KMT2A::MLLT1 models, or the reverse. The two fusions retain different partner domains and only one of them contributes a ligandable reader pocket.
- If a partner-specific response difference exists and the t(11;19) bucket is a 50:50 mixture, the mixture halves the apparent effect size and would make a real difference look like noise in a 13-patient subgroup.
- Did AUGMENT-101's central molecular confirmation resolve the 3' partner gene? If it did, the MLLT1-versus-ELL split of those 13 patients exists in the trial database and no new testing is needed.
- What is in Data Supplement Table S3? It is the only partner-stratified menin-inhibitor response table known to exist, it is referenced from an open-access paper, and it is not open.
- Does the ELL:MLLT1 ratio in relapsed/refractory adult AML resemble the paediatric 75:56, or the recombinome's 119:460? The two give opposite readings of how contaminated a t(11;19) trial bucket is, and no series reports the ratio in the relevant population.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Read in full text (PMC11687943): in AUGMENT-101's phase II registration cohort, Table 1 gives KMT2A translocations for the efficacy population (n = 57) and safety population (N = 94) as t(9;11)(p22;q23) 11 (19.3) / 21 (22.3), t(11;19)(q23;p13) 13 (22.8) / 17 (18.1), t(10;11)(p12;q23) 7 (12.3) / 12 (12.8), t(6;11)(q27;q23) 7 (12.3) / 8 (8.5), t(4;11)(q21;q23) 2 (3.5) / 7 (7.4), t(1;11) 2 (3.5) / 2 (2.1), fusion partner unknown 13 (22.8) / 24 (25.5) and other translocations 2 (3.5) / 3 (3.2); acute leukaemia type in the efficacy population was AML 49 (86.0%), ALL 7 (12.3%) and ambiguous lineage 1 (1.8%); the CR + CRh rate was 22.8% (95% CI 12.7 to 35.8) and the overall response rate 63.2% (95% CI 49.3 to 75.6); and the paper states that exploratory analysis with limited sample sizes demonstrated efficacy for KMT2Ar across various translocation partners, referring the reader to Data Supplement Table S3.
Menin Inhibition With Revumenib for KMT2A-Rearranged Relapsed or Refractory Acute Leukemia (AUGMENT-101). · Issa GC, Aldoss I, Thirman MJ, DiPersio J, et al. · J Clin Oncol 2025What this evidence cannot show
Read in full text via Europe PMC fullTextXML (body element present, 64,201 characters of flattened text) on 2026-09-19. Data Supplement Table S3, which holds the partner-stratified response rates, is NOT in the PMC deposit - the record carries pmc-prop-has-supplement no - and was not retrievable from this environment, so no partner-specific response number is taken from this source. The t(11;19) rows give (q23;p13) without a sub-band, so MLLT1 and ELL are not separated anywhere in the paper. The efficacy population is 86% AML, which is the lineage in which KMT2A::MLLT1 is a 10% minority of its own partner group.
Evidence that argues against this (1)
- The phase I report states plainly that in evaluable patients with baseline targeted sequencing, no specific KMT2A translocation or comutation was clearly associated with response in a limited analysis - so the trial programme's own position is that partner does not predict response, which if true makes the sub-band ambiguity less consequential for prescribing, though not for mechanism. Issa GC et al. 2023
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In 1,256 children with KMT2A-rearranged AML, 19p13.1/KMT2A::ELL (n = 75) was more frequent than 19p13.3/KMT2A::MLLT1 (n = 56), with a further 23 cases recorded as t(11;19)(q23;p13) without an ascertained sub-band; the MLLT1 group was independently associated with inferior event-free survival (HR 1.6, P = .01) and higher cumulative incidence of relapse (HR 1.8, P < .01) and was assigned adverse risk, while the ELL group was non-significant on EFS, CIR and OS and was assigned intermediate risk; median age at diagnosis in the MLLT1 group was 6.8 years against under 4 years for most KMT2A-r groups.
Optimized cytogenetic risk-group stratification of KMT2A-rearranged pediatric acute myeloid leukemia. · van Weelderen RE, Harrison CJ, Klein K, Jiang Y, et al. · Blood Adv 2024What this evidence cannot show
Numbers relayed from a full-text reading (PMC11225675) performed by a delegated worker in this session; the PMID, title and DOI were independently re-fetched from PubMed esummary on 2026-09-19 by the author of this entry, and the abstract's own statement that 19p13.3/KMT2A::MLLT1 is independently associated with adverse outcomes was read directly. This is a paediatric AML cohort; the ELL:MLLT1 ratio in a mostly adult, relapsed/refractory trial population such as AUGMENT-101 is not known and may differ.
Evidence that argues against this (1)
- The ELL:MLLT1 ratio reverses outside paediatric AML. In the 3,401-patient recombinome, MLLT1 has 460 cases against ELL's 119, and ELL has zero infant ALL cases against MLLT1's 234 - so in a lymphoid or mixed-age population most t(11;19) will be MLLT1, and the contamination argument is lineage- and age-dependent. Meyer C et al. 2023
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Read in full text (PMC10169636), Table 1: the ELL row is infant ALL 0 / AML 30 / other 1, paediatric ALL 0 / AML 32 / other 1, adult ALL 1 / AML 51 / other 3, total 119, against the MLLT1 row of infant ALL 234 / AML 2 / other 6, paediatric ALL 100 / AML 28 / other 1, adult ALL 66 / AML 16 / other 2 and 4/1/0 unclassified, total 460.
The KMT2A recombinome of acute leukemias in 2023. · Meyer C, Larghero P, Almeida Lopes B, et al. · Leukemia 2023What this evidence cannot show
Read in full text via Europe PMC fullTextXML; the twelve cell counts per row were read from the flattened Table 1. Single-centre referral series, so the ELL:MLLT1 ratio reflects referral patterns.
Evidence that argues against this (1)
- A childhood t(11;19) RT-PCR series found that all 17 ALL cases had MLL/ENL while of 9 AML cases 6 had MLL/ENL, 2 had MLL/ELL and 1 had no detectable fusion - so ELL is a minority even within childhood t(11;19) AML in that series, which sits awkwardly with the 75-versus-56 count in the larger modern cohort. Rubnitz JE et al. 1996
t(11;19) is 13/57 of the efficacy population, the largest named partner group
trial representation of this fusion, pooled with KMT2A::ELL
- co-occurs with →KMT2A::ELL (MLL-ELL), t(11;19)(q23;p13.1)
19p13.1 versus 19p13.3 are pooled as t(11;19)(q23;p13) throughout the trial record
Table S3 exists and is referenced but is not in the PMC deposit
bounds how contaminated an unresolved t(11;19) bucket is, by lineage
the same MLLT1-versus-ELL band ambiguity recurs here, but the pinometostat trial resolved it by central NGS