How a spliced fusion works: the break lands in front of MLLT1, the gene is never cut, and there is nothing left to make a reciprocal product from
also called spliced KMT2A fusion mechanism · read-through fusion transcript · MLLT1 upstream breakpoint
In half of these patients the chromosome does not break inside the partner gene at all - it breaks just in front of it, so the partner gene arrives whole and the KMT2A promoter simply reads straight through into it. Because the partner gene was never cut, the other derivative chromosome has no partner-gene piece on it, and no mirror-image fusion can exist.
The same thing, for a scientist
A spliced fusion is generated when the 5' portion of KMT2A is recombined **upstream of an intact partner gene** rather than within it. There is no transcription terminator downstream of the translocated 5'-KMT2A fragment, so transcripts initiated at the KMT2A promoter run into the partner gene; the most 3' KMT2A exon then splices to the partner's **exon 2**, which is the first partner exon carrying a bona fide splice acceptor site. Only the direct KMT2A::partner transcript is expressed and no reciprocal one is [E1]. Across 3,401 KMT2A-rearranged patients this mechanism accounts for 275 cases in total, and it is rare for every partner except MLLT1, where about 50% of breakpoints (241 of 460) lie upstream of MLLT1 exon 1 [E1]. The founding description of the mechanism is a 2007 Leukemia letter on t(11;19) specifically [E2]. Three consequences follow that the literature does not draw out, and all three bear on whether the reciprocal-negative arm is a clean control: the direct transcript in a spliced case is forced to an exon-2 junction; the MLLT1 coding unit is never interrupted, so partner-gene dosage is not the same as in a balanced case; and the reciprocal derivative carries the 3' half of KMT2A next to non-coding sequence rather than next to a gene.
The mechanism, in the source's own terms
Verbatim from the recombinome full text: spliced fusions "are generated by recombination events where the 5'-portion of the KMT2A gene is located upstream of an intact fusion partner gene. Since there is no transcription termination downstream of the translocated 5'-KMT2A gene fragment, transcripts starting at the KMT2A promoter are transcribing into the downstream located fusion partner gene. In this case, the most 3'-exon of the 5'-KMT2A is usually splicing to exon 2 the this fusion partner gene, as this exon is the first to exhibit a bona fide splice acceptor site." For MLLT1 specifically: "about 50% of the breakpoints (n = 241) localize upstream of MLLT1 exon 1. In these cases, only KMT2A::MLLT1 fusion transcripts but no reciprocal one are expressed" 1.
Why no reciprocal allele results - the part that is geometry, not biology
In a balanced t(11;19) the break falls inside MLLT1, so der(11) receives KMT2A exons 1-n joined to MLLT1 exons m-13 and der(19) receives MLLT1 exons 1-(m-1) joined to KMT2A exons (n+1)-37. Both derivatives carry coding sequence from both genes, and a reciprocal MLLT1::KMT2A transcript is possible. In a spliced fusion the break falls 5' of MLLT1 exon 1, so all of MLLT1's coding sequence travels to der(11) and der(19) receives the 3' half of KMT2A abutting whatever non-coding 19p13.3 sequence lay upstream of MLLT1. There is no MLLT1 coding sequence on der(19) to fuse to, so no reciprocal fusion gene exists - not a reciprocal gene that is silent, and not one that is out of frame. It is absent 1.
The der(19) in a spliced case is not empty, and this is where KMT2A* enters
Absence of a reciprocal fusion is not absence of reciprocal protein. The recombinome reports a gene-internal promoter upstream of KMT2A exon 12 that can drive a 5'-truncated KMT2A protein, KMT2A*, which "has no ability to bind Menin1, LEDGF or MYB, but still carries all enzymatic functions and necessary domains" 1. That promoter travels with the 3'-KMT2A fragment on der(19). So a spliced-fusion patient is unambiguously free of a reciprocal MLLT1::KMT2A chimera and is not thereby shown to be free of a KMT2A C-terminal species. Nobody has looked for KMT2A* in a spliced KMT2A::MLLT1 case.
An inference this graph is making and the literature is not: partner-gene dosage differs between the arms
In a balanced t(11;19) one MLLT1 allele is split, leaving one intact copy. In a spliced fusion MLLT1 is never cut, so the cell carries the normal allele plus an uninterrupted MLLT1 coding unit on der(11). Whether that second copy also produces wild-type ENL depends on whether MLLT1's own promoter sits proximal or distal to the breakpoint, which the published description ('upstream of MLLT1 exon 1') does not resolve. Since ENL is a selective dependency in acute leukaemia in its own right and its paralogue AF9 is not 3, a systematic difference in wild-type ENL dosage between the reciprocal-negative and reciprocal-positive arms would be a confound of exactly the kind this control arm was supposed to be free of. This is our inference from gene geometry, not a measured result, and it is testable directly from the breakpoint sequences the Frankfurt archive already holds.
| mechanism class | read-through transcription plus splicing to the first downstream bona fide splice acceptor |
|---|---|
| partner | MLLT1 |
| spliced cases | 241 |
| total partner cases | 460 |
| spliced cases all partners | 275 |
| cohort n | 3401 |
| direct transcript junction | last KMT2A exon -> MLLT1 exon 2 |
| reciprocal fusion gene | absent by construction, not silent and not out of frame |
| der19 contents | 3'-KMT2A (exons ~12-37) abutting non-coding 19p13.3 sequence upstream of MLLT1, including the KMT2A gene-internal promoter upstream of exon 12 |
| founding reference pmid | 17252016 |
- If the spliced-fusion class carries an extra intact MLLT1 transcription unit, an ENL-directed therapeutic would face more target protein in precisely the patients this graph wants to treat as the clean arm.
- A diagnostic that scores reciprocal status by RT-PCR for the reciprocal transcript will call both spliced cases and out-of-frame balanced cases 'negative', even though only the first is genuinely free of a KMT2A C-terminal product. Any stratified trial built on such an assay would mix two biologically different groups.
- Does the MLLT1 promoter travel with the gene in a spliced fusion, or is it left behind on der(19)? 'Upstream of MLLT1 exon 1' does not resolve it, and the answer determines whether these patients carry one or two functional wild-type ENL transcription units.
- Is the direct KMT2A::MLLT1 transcript in a spliced case quantitatively the same as in a balanced exon-2 case? A read-through transcript initiating at the KMT2A promoter and traversing an entire intergenic region before its first partner exon has a different processing burden from a compact intronic junction, and nobody has compared fusion transcript or protein levels between the two classes.
- Is KMT2A* expressed from der(19) in spliced KMT2A::MLLT1 cases? If it is, the 'reciprocal-negative' arm still contains a menin-blind, SET-bearing KMT2A species and the comparison measures something narrower than it appears to.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Read in full text (PMC10169636): spliced fusions arise when the 5' portion of KMT2A is recombined upstream of an intact partner gene, so transcripts from the KMT2A promoter read through and the most 3' KMT2A exon splices to the partner's exon 2, the first exon with a bona fide splice acceptor; they are rare except for MLLT1, where of n = 460 cases about 50% (n = 241) of breakpoints localise upstream of MLLT1 exon 1 and only KMT2A::MLLT1 transcripts and no reciprocal one are expressed; 275 cases in total used this mechanism, the other partners being EPS15 (12/51), ELL (10/119), PRPF19 (3/3), AFDN (2/138), MLLT3 (2/612), AFF1 (1/1262), DCPS, MYO1F (1/5), SEPTIN5 and CT45A2; and a gene-internal promoter upstream of KMT2A exon 12 can drive a 5'-truncated KMT2A protein, KMT2A*, which has no ability to bind Menin1, LEDGF or MYB but still carries all enzymatic functions.
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 (body element present). The 241/460 figure is a count of breakpoint positions with the transcript consequence inferred from gene structure and attributed to a prior reference, not a per-patient RT-PCR measurement of reciprocal transcript absence. Single-centre referral series, so partner denominators reflect referral patterns rather than population incidence.
Evidence that argues against this (1)
- The functional premise that a missing reciprocal product matters is contested: removing the reciprocal transcript from human t(4;11) lines produced no growth, cell-cycle or apoptotic phenotype. Kumar AR et al. 2011
- E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
The spliced-fusion mechanism was first described for t(11;19)(q23;p13.3) in a 2007 Leukemia letter titled 'Spliced MLL fusions: a novel mechanism to generate functional chimeric MLL-MLLT1 transcripts in t(11;19)(q23;p13.3) leukemia'; its opening section, which is outside the paywall, states that MLLT1 accounts for about 5% of all MLL rearrangements, that in most cases MLLT1 is rearranged with MLL via a balanced reciprocal translocation, that t(11;19)(q23;p13.3) occurs in lymphoid, myeloid and biphenotypic malignancies, that therapy-related MLL-MLLT1 translocations are very rare with only one t-AML case described in the literature at the time, and that two subsets within MLL-MLLT1 seem to have a good prognosis, ALL patients between 1 and 9 years and patients with T-lineage ALL.
Spliced MLL fusions: a novel mechanism to generate functional chimeric MLL-MLLT1 transcripts in t(11;19)(q23;p13.3) leukemia. · Meyer C, Burmeister T, Strehl S, Schneider B, et al. · Leukemia 2007What this evidence cannot show
PARTIALLY UNREADABLE. PubMed carries no abstract for this letter. All four PMC routes fail: Europe PMC core returns pmcid null, isOpenAccess N, inEPMC N, so routes 2, 3 and 4 have no PMCID to use. The publisher page at https://www.nature.com/articles/2404542 returns HTTP 200 and serves the first three paragraphs plus the full reference list before the paywall; everything quoted above comes from that visible portion, read on 2026-09-19. The paper's data - how many cases, how the spliced junction was demonstrated, whether reciprocal transcripts were assayed - lie behind the paywall and were NOT read. No claim here is made about what the unread portion says.
Evidence that argues against this (1)
- The good-prognosis subsets this letter cites come from a 35-patient series, so the claim it is relaying rests on small numbers. Rubnitz JE et al. 1999
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
ENL, but not its paralogue AF9, is required in acute myeloid leukaemia: CRISPR-Cas9 depletion of ENL is anti-leukaemic while AF9 depletion is not, and the ENL YEATS domain binds acetylated histone H3.
ENL links histone acetylation to oncogenic gene expression in acute myeloid leukaemia. · Wan L, Wen H, Li Y, Lyu J, et al. · Nature 2017What this evidence cannot show
Abstract-level reading for this entry; the paralogue asymmetry is established in AML models, not in KMT2A::MLLT1 B-ALL, and the dosage argument made in the section above is an inference from it rather than something this paper tests.
Evidence that argues against this (1)
- gnomAD constraint for MLLT1 is high, which is often read as a safety signal against inhibiting it; constraint measures intolerance to lifelong heterozygous loss from conception in a population sample and is not a statement about partial reversible pharmacological occupancy in an adult. It is not evidence either way about dosage effects in a leukaemic clone. Erb MA et al. 2017
the recombination geometry of the reciprocal-negative half of this fusion
the mechanism that creates the 241-patient reciprocal-negative arm
supplies the partner-matched reciprocal-negative population the hypothesis needs
break lands 5' of MLLT1 exon 1; the gene itself is never interrupted
the arm assignment rests on the spliced-fusion mechanism holding
ACER1 is listed among in-frame reciprocal KMT2A partners, so the intergenic-break class is not reciprocal-free by construction