KMT2A::SEPTIN11 (MLL-SEPT11) — the fusion that cytogenetically impersonates t(4;11)
also called MLL-SEPT11 · KMT2A-SEPT11 · MLL::SEPTIN11 · t(4;11)(q21;q23) mimic
SEPTIN11 sits on chromosome 4 only a short distance from AFF1, so a KMT2A break that joins it produces a picture under the microscope that is indistinguishable from the commonest and best-known KMT2A leukaemia. The only way to tell them apart is a molecular test.
The same thing, for a scientist
KMT2A::SEPTIN11 is the rarest of the recurrent KMT2A-septin fusions (n=3 of 3401: 1 paediatric ALL, 1 paediatric AML, 1 adult ALL) [E2] and the only one with more lymphoid than myeloid cases, which matters because it is also the only one that can masquerade as KMT2A::AFF1. In a 55-year-old man with aggressive ALL arising eight years after chemotherapy for nephrotic syndrome, the sole cytogenetic abnormality was t(4;11)(q21;q23); RT-PCR for MLL-AFF1 was negative, FISH narrowed the 4q21 breakpoint to a 250 kb region proximal to AFF1 containing four genes, and RT-PCR showed an in-frame fusion of MLL exon 11 to SEPT11 exon 2 [E1]. The authors' own proposed mechanism is the one this beat is testing: 'forced MLL dimerization by the septin coiled coil and GTP-binding domains, which could convert MLL to an oncogene' [E1]. SEPTIN11 (UniProt Q9NVA2, 429 aa) carries the longest coiled coil of any KMT2A-partner septin, aa 320-415 [E3].
The impersonation, and why it matters clinically
t(4;11)(q21;q23) is the signature of KMT2A::AFF1, the commonest KMT2A fusion in infant and adult ALL, and it carries specific prognostic and treatment implications. SEPTIN11 lies within 250 kb proximal to AFF1 on 4q21 1. A KMT2A break joining SEPTIN11 therefore produces an identical karyotype. The index patient was correctly reclassified only because MLL-AFF1 RT-PCR came back negative and somebody followed it up rather than assuming assay failure 1.
Two of the three SEPTIN11 patients in the systematic series are ALL 2, so this is not an isolated coincidence: the septin class as a whole is essentially myeloid, and SEPTIN11 is where its lymphoid cases sit.
The mechanism the authors proposed, and its status
Stevens and colleagues explicitly hypothesised that the oncogenic role of the septin partners relates to forced MLL dimerisation through the septin coiled coil and GTP-binding domains 1. That is the same hypothesis Ono and colleagues had already tested experimentally on MLL-SEPT6 and largely confirmed — both elements are required — while also showing that a synthetic dimerisation module does not substitute 4. Nothing of the kind has been done for SEPTIN11. Given that SEPTIN11 has the longest annotated coiled coil of the partner septins (96 residues versus SEPTIN5's 32) 3, it is the natural test case for whether coiled-coil length predicts transforming strength.
| gene5 | KMT2A |
|---|---|
| gene3 | SEPTIN11 |
| cytoband | t(4;11)(q21;q23) — cytogenetically indistinguishable from KMT2A::AFF1 |
| uniprot5 | Q03164 |
| uniprot3 | Q9NVA2 |
| uniprot3 verified | Q9NVA2 = SEP11_HUMAN, PRIMARY gene name SEPTIN11, 429 aa, Homo sapiens (UniProt REST, checked 2026-09-19). |
| prognosis | No estimate. The index patient's disease was described as aggressive. |
| PDB note | No SEPTIN11 structure was identified and none of the fusion. |
| mechanism class note | Proposed by the reporting authors, untested for this partner. |
| druggability class | unexplored |
| stage of hematopoiesis | lymphoid precursor in two of three reported patients, myeloid progenitor in the third |
| age peak | paediatric and adult; no infant case reported |
diseases
[ "adult ALL", "paediatric ALL", "paediatric AML" ]
frequency — the numbers behind the chart
{
"dcal_recombinome_n": 3,
"dcal_recombinome_total": 3401,
"paediatric_all_n": 1,
"paediatric_aml_n": 1,
"adult_all_n": 1
}breakpoints
[
{
"id": "bp-kmt2a-exon-11-septin11",
"gene5_exon_last": 11,
"gene3_exon_first": 2,
"population": "the index adult ALL case; the only KMT2A::SEPTIN11 junction sequenced in the literature",
"domains_retained_5": [
"MBM1",
"MBM2",
"AT-hook 1",
"AT-hook 2",
"AT-hook 3",
"SNL1/2",
"CXXC"
],
"domains_lost_5": [
"PHD1",
"PHD2",
"PHD3",
"Bromodomain",
"PHD4",
"FYRN",
"TAD",
"FYRC",
"WIN",
"SET"
],
"domains_retained_3": [
"Septin-type G domain (aa 38-304)",
"G1/P-loop motif (aa 48-55)",
"G3 motif (aa 100-103)",
"G4 motif (aa 183-186)",
"coiled coil (aa 320-415)"
],
"domains_lost_3": [
"SEPTIN11 exon 1"
],
"phenotype_notes": "An exon-11 KMT2A junction lies at the distal end of the major breakpoint cluster region and retains CXXC."
}
]PDB IDS
[]
mechanism class
[ "forced-oligomerisation" ]
drugs
[]
key interactors
[ "gene-men1" ]
trials
[]
- The clinical hazard here is misassignment, not mechanism: a patient carrying KMT2A::SEPTIN11 will be reported, counted, risk-stratified and possibly enrolled as KMT2A::AFF1 unless the transcript is confirmed. Every downstream inference about that patient, including response to a partner-agnostic drug, is then attached to the wrong fusion.
- This also contaminates the denominator in the other direction: cohorts of 't(4;11) ALL' assembled on karyotype include an unknown number of non-AFF1 fusions, which weakens any claim about AFF1-specific biology built on them.
- How many reported t(4;11) cases have never had the MLL-AFF1 transcript confirmed, and would systematic transcript typing of karyotype-only t(4;11) cases turn up more SEPTIN11 fusions?
- SEPTIN11 has the longest coiled coil of the KMT2A-partner septins. Does coiled-coil length predict transforming strength, as the dimerisation model implies it should?
- Why is SEPTIN11 the septin with lymphoid cases while SEPTIN6 and SEPTIN9 have none? Partner-driven lineage bias is real for the elongation partners and nobody has asked whether it holds here.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In a 55-year-old man with aggressive ALL developing eight years after chemotherapy for nephrotic syndrome, the sole cytogenetic abnormality was t(4;11)(q21;q23) but RT-PCR for MLL-AFF1 was negative; FISH narrowed the 4q21 breakpoint to a 250 kb region proximal to AFF1 comprising four genes, and RT-PCR revealed an in-frame fusion of MLL exon 11 to SEPT11 exon 2, a previously undescribed translocation cytogenetically indistinguishable from the recurrent t(4;11); the authors hypothesise the septins' oncogenic role relates to forced MLL dimerisation by the septin coiled coil and GTP-binding domains.
A translocation in acute lymphoblastic leukemia that cytogenetically mimics the recurrent MLL-AFF1 translocation and fuses SEPT11 to MLL. · Stevens SJ, Meers LE, Albrechts JC, et al. · Cancer Genet Cytogenet 2010What this evidence cannot show
A single case. The dimerisation mechanism is the authors' hypothesis in the discussion, explicitly speculative, and was not tested in this paper.
Evidence that argues against this (1)
- The proposed mechanism is only partly supported where it has been tested: in MLL-SEPT6 both the coiled coil and the GTP-binding domain are indeed required, but substituting a synthetic dimerisation module for the coiled coil failed to restore transformation, so 'forced dimerisation converts MLL to an oncogene' is not a sufficient account for a septin partner. Ono R et al. 2005
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
In 3401 genomically characterised KMT2A-rearranged patients, SEPTIN11 appears three times: one paediatric ALL, one paediatric AML and one adult ALL, ranking 31st of the partner list.
The KMT2A recombinome of acute leukemias in 2023. · Meyer C, Larghero P, Almeida Lopes B, et al. · Leukemia 2023What this evidence cannot show
Three patients. Any statement about lineage distribution from n=3 is descriptive only.
Evidence that argues against this (1)
- A fusion that produces an unremarkable t(4;11) karyotype is the one most likely to be mis-assigned to AFF1 and never molecularly typed, so 3 of 3401 is a floor and the true lymphoid share of the septin class could be higher than this beat's headline suggests. Stevens SJ et al. 2010
- E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
Human septin-11 is UniProt Q9NVA2 (SEP11_HUMAN), primary gene name SEPTIN11, 429 aa, with a septin-type G domain at aa 38-304 (G1/P-loop 48-55, G3 100-103, G4 183-186), a coiled coil at aa 320-415 and a disordered C-terminus at 398-429.
UniProtKB Q9NVA2 (SEP11_HUMAN) Septin-11, Homo sapiens, 429 aa · UniProt Consortium · UniProtKB 2026What this evidence cannot show
Canonical isoform; coiled-coil boundaries are prediction-derived.
Evidence that argues against this (1)
- Sept11 is functionally dispensable in the one genetic test available: depleting Sept11 in Sept6-null cells produced no phenotype, which the authors read as septin-system redundancy. Ono R et al. 2005
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
MLL-SEPT6 myeloid immortalisation requires both the septin GTP-binding domain and the coiled coil, and a synthetic 4-OHT-inducible dimerisation module substituted for the coiled coil does not rescue it.
Dimerization of MLL fusion proteins and FLT3 activation synergize to induce multiple-lineage leukemogenesis. · Ono R, Nakajima H, Ozaki K, et al. · J Clin Invest 2005What this evidence cannot show
Demonstrated for SEPTIN6; extension to SEPTIN11 is an inference and is flagged as an open question, not a finding.
Evidence that argues against this (1)
- For other cytoplasmic KMT2A partners the synthetic module does work — the AF1p/EPS15 and GAS7 coiled coils are necessary and sufficient and synthetic dimerisation modules mimic the bona fide fusions — so the septin result may be construct-specific rather than a class property. So CW et al. 2003
- E5weak confidence 2 of 5: Preliminary, small, indirect, or not yet peer reviewed.dataset
In a genome-wide DepMap CRISPR (Chronos) scan, SEPTIN11 had a mean gene effect of -0.149 across 1090 lines and ranked 15,347 of 17,728 for KMT2A selectivity (86.6th percentile), i.e. no selective requirement in KMT2A-rearranged leukaemia.
DepMap CRISPR (Chronos) genome-wide KMT2A-selectivity scan of 17,728 protein-coding genes, 11 Cellosaurus-annotated KMT2A-rearranged leukaemia lines vs other leukaemia lines (analyses/depmap-kmt2a-selective-scan-full/results.json, all_results; septin/ABI1/MLLT11 rows looked up for this beat) · Broad Institute DepMap; Cellosaurus; Open Targets Platform; scan run by this project · Database and derived analysis 2026What this evidence cannot show
No KMT2A-SEPTIN11 line exists in DepMap; this is wild-type SEPTIN11 in leukaemias driven by other partners.
Evidence that argues against this (1)
- A single-gene CRISPR score cannot resolve a requirement buffered by paralogues, and SEPTIN11 is the closest homologue of SEPTIN6, so the two would be expected to buffer each other. Ono R et al. 2005
SEPTIN-FAMILY PARTNERS, joining the septin2/6/9 group already linked here. SEPTIN11 carries the same architecture the family analogy rests on - a septin-type G domain and a coiled coil retained 3' of the breakpoint - so the oligomerisation account either covers all of them or none. SEPTIN11 is the rarest of the group at 3 of 3401 patients, which makes it the member where a family-level claim is doing the most work.
- part of →KMT2A / MLL1 — the gene that breaks
retained KMT2A N-terminus including CXXC
- co-occurs with →KMT2A::AFF1 (MLL-AF4), t(4;11)(q21;q23)
cytogenetically indistinguishable; SEPTIN11 lies within 250 kb proximal to AFF1
2 of 3 reported patients had ALL
G domain plus the longest partner-septin coiled coil