KMT2A::TET1 (MLL-LCX), t(10;11)(q22;q23) — the fusion that throws away a second CXXC domain
also called MLL-TET1 · MLL-LCX · KMT2A-TET1 · t(10;11)(q22;q23)
A rare leukaemia rearrangement joins the KMT2A gene to TET1. What makes it interesting is what gets discarded: TET1 normally carries the same kind of DNA-reading module that KMT2A does, and the join removes it. Two proteins that both read the same chemical mark on DNA are fused, and only one of the two readers survives.
The same thing, for a scientist
t(10;11)(q22;q23) fuses KMT2A to TET1, cloned as LCX in a de novo adult AML-M2 with trilineage dysplasia [E1] and independently reported as TET1 the following year [E2]. The point of interest for this project is structural: TET1 is one of the twelve human CXXC-domain proteins, and the fusion DISCARDS TET1's CXXC while retaining its alpha-helical coiled coil. A fusion of two CXXC readers keeps exactly one CXXC — KMT2A's. Reported across AML, adult CD10-negative BCP-ALL and one T-lymphoblastic lymphoma with lineage switch; three of the four clinical reports are Letters whose full text is not retrievable, and their contents are NOT claimed here [E4].
Why a rare fusion earns an entry: it is the CXXC family's only self-referential case
| karyotype | t(10;11)(q22;q23) |
|---|---|
| cxxc status | TET1's own CXXC domain is NOT retained in the fusion; the alpha-helical coiled coil is. The fusion therefore carries exactly one CXXC domain, KMT2A's. |
| shown for fusion | NOT APPLICABLE as a localisation question. As a partner-biology question: the CXXC-discard is established from the founding cloning paper's own domain description, in this fusion, not imported from another family member. |
| gene5 | KMT2A |
| gene3 | TET1 |
| uniprot5 | Q03164 |
| uniprot3 | Q8NFU7 |
| cytoband | t(10;11)(q22;q23) |
| partner note 5prime | KMT2A (11q23) |
| partner note 3prime | TET1 (10q22), cloned as LCX |
diseases reported
[ "de novo adult AML-M2 with trilineage dysplasia (index case)", "adult CD10-negative B-cell precursor ALL (recombinome series)", "T-cell lymphoblastic lymphoma with subsequent lineage switch to acute myelomonocytic leukaemia" ]
full text not retrievable
[ "23100278 (Haematologica 2013, the dedicated breakpoint paper)", "24323992 (Haematologica 2013, the T-LBL lineage-switch case)", "28798069 (Haematologica 2017, the immortalisation study)" ]
- This entry describes a rare fusion and supports no clinical action. Nothing here bears on whether a TET-directed or menin-directed agent would help a patient carrying t(10;11)(q22;q23), and the three reports that describe clinical course have not been read.
- TET1's presence in a fusion name should not be read as implying that TET inhibitors or TET-modulating agents are relevant. Whether the fusion retains TET1 catalytic function is explicitly unresolved above.
- Does discarding TET1's CXXC matter? The fusion joins two of the twelve human CXXC proteins and keeps one CXXC. If the KMT2A CXXC address is the leukaemogenic determinant, a fusion that had retained BOTH readers would be the natural comparison — and no such fusion is known, so the question cannot currently be asked experimentally.
- TET1's catalytic domain converts 5-methylcytosine to 5-hydroxymethylcytosine. Is it retained in the fusion, and if so does the chimera carry a demethylase activity next to a CXXC that requires unmethylated CpG? The founding paper's domain description was not read in full and this is not resolved here.
- The three unread Letters. PMID 23100278 in particular is titled for exactly the breakpoint and clinical detail this entry lacks; an interlibrary copy would close it in an afternoon.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
The founding report. MLL on 11q23 was fused to LCX on 10q22 in a de novo adult AML-M2 with trilineage dysplasia carrying t(10;11)(q22;q23). LCX was predicted to encode a 2,136-amino-acid protein of ~235 kDa carrying a zinc-binding CXXC domain of the kind MLL also contains, three nuclear localisation signals, and an alpha-helical coiled-coil region. The paper's own name for the gene — leukemia-associated protein with a CXXC domain — records that the CXXC was the salient feature at discovery.
LCX, leukemia-associated protein with a CXXC domain, is fused to MLL in acute myeloid leukemia with trilineage dysplasia having t(10;11)(q22;q23). · Ono R, Taki T, Taketani T, Taniwaki M, Kobayashi H, Hayashi Y · Cancer Res 2002What this evidence cannot show
Abstract fetched and read 2026-09-21; full text not read, and PubMed carries no DOI for this record. A single index patient. The 2,136-residue length is the paper's prediction from its own cloning and matches UniProt's current TET1 length, which is a consistency check rather than independent confirmation.
Evidence that argues against this (1)
- One patient is not a disease entity, and the same translocation was reported independently the following year under the name TET1 rather than LCX — so the founding paper is one of two simultaneous namings, not a sole description. Which name the literature uses determines what a query retrieves, and LCX is additionally a severe collision token: it is the standard abbreviation for the left circumflex coronary artery, and a PubMed search for LCX returns mass-spectrometry and cardiology papers well ahead of leukaemia. Lorsbach RB et al. 2003
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Independent contemporaneous description under the name TET1, establishing the same t(10;11)(q22;q23) MLL fusion and placing the partner in a then-novel protein family. This is why the gene carries two names in the literature and why searches on either alone are incomplete.
TET1, a member of a novel protein family, is fused to MLL in acute myeloid leukemia containing the t(10;11)(q22;q23). · Lorsbach RB, Moore J, Mathew S, Raimondi SC, Mukatira ST, Downing JR · Leukemia 2003What this evidence cannot show
PubMed carries no abstract for this record, so only the title's assertions are used here: the fusion, the karyotype and the novel-family placement. Nothing about breakpoint position, domain retention or clinical course is claimed from it (rule 80).
Evidence that argues against this (1)
- Being a title-only source, this cannot corroborate the structural claim that matters here — that the fusion discards TET1's CXXC. That rests on the founding paper's domain description alone, and a second independent structural confirmation does not exist in a form this project has read. Ono R et al. 2002
- E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
The fusion is not confined to AML. It appears in the MLL recombinome of adult CD10-negative B-cell precursor ALL, a systematic series that determined fusion-partner frequencies in that immunophenotype rather than reporting single cases — which is the evidence class that establishes a partner as recurrent rather than anecdotal.
The MLL recombinome of adult CD10-negative B-cell precursor acute lymphoblastic leukemia: results from the GMALL study group. · Burmeister T, Meyer C, Schwartz S, Hofmann J, Molkentin M, Kowarz E · Blood 2009What this evidence cannot show
Abstract read; full text not read, so the exact patient count carrying KMT2A::TET1 in that series is not stated here. Adult CD10-negative BCP-ALL is a different disease context from infant KMT2A::AFF1, and the series is adult.
Evidence that argues against this (1)
- Appearing in a recombinome series establishes recurrence but says nothing about frequency — a partner can be recurrent and still vanishingly rare. The 3,401-patient recombinome census this graph uses elsewhere is the better denominator, and TET1 does not appear among its high-frequency partners, so this fusion belongs in the long tail whatever the CD10-negative series contributes. Burmeister T et al. 2009
- E4weak confidence 2 of 5: Preliminary, small, indirect, or not yet peer reviewed.paper
BOUNDED ABSENCE OF READING, stated rather than worked around. Three further reports exist and none could be read: PMID 23100278, the dedicated genomic-breakpoint and clinical-features paper; PMID 24323992, the first t(10;11)/MLL-TET1 in T-lymphoblastic lymphoma with lineage switch to acute myelomonocytic leukaemia; and PMID 28798069, reporting that the fusion protein immortalises myeloid progenitors. All three are Haematologica Letters with NO PubMed abstract. Both PMC records that exist (PMC3660007, PMC5664403) return front matter only — no <body> and no <abstract> element via efetch, checked 2026-09-21. Their TITLES are used here and nothing else.
Genomic breakpoints and clinical features of MLL-TET1 rearrangement in acute leukemias. · Lee SG, Cho SY, Kim MJ, Oh SH, Cho EH, Lee S · Haematologica 2013What this evidence cannot show
This item asserts only that these papers exist with these titles. Titles are informative here — a paper titled for genomic breakpoints and clinical features almost certainly contains the breakpoint detail this entry lacks — but an inference from a title is not evidence, and the entry does not make one. Retrieving these three is the single cheapest improvement available to this entry.
Evidence that argues against this (1)
- A reader could reasonably object that citing three papers for their titles inflates the apparent evidence base. The objection is fair, which is why they are gathered into one explicitly bounded item rather than spread across three that would each look like a supported claim. The honest position is that this fusion has four clinical reports and this project has read one of them. Ittel A et al. 2013
- analogous to →KMT2A::MLLT10 (MLL-AF10), t(10;11)(p12;q23)
SAME BREAKPOINT REGION, DIFFERENT PARTNER CLASS, AND THE PAIR ISOLATES WHAT THE COILED COIL DOES. Both arise from 10q22-q23 rearrangements with KMT2A and both retain their partner's alpha-helical coiled coil while discarding its other modules - MLLT10's OM-LZ delivers DOT1L, whereas TET1's retained region is a coiled coil with its CXXC and catalytic domains left behind. If the two behave alike, the coiled coil is doing the work and the discarded domains are irrelevant to transformation; if they differ, the difference locates in what MLLT10 brings and TET1 does not.
The 3' partner. TET1 is one of the twelve human CXXC-domain proteins, and the fusion DISCARDS its CXXC while keeping the alpha-helical coiled coil - so a join between two CpG readers yields a chimera with exactly one CpG reader, KMT2A's. TET1's own entry additionally records it as a disputed fusion dependency: indispensable in conventional-knockout MLL-AF9 work and not required in purpose-built conditional-knockout MLL-ENL mice.
THE ONLY SELF-REFERENTIAL CASE THE CXXC ADDRESS HAS. Of twelve human CXXC proteins, exactly two are recurrent fusion partners with one another, and this is that fusion. It is therefore the single natural experiment in which the address-reading family is on both sides of the breakpoint - and the breakpoint keeps one reader and cuts the other. Whether that is selection or an accident of gene structure cannot be settled from one rare translocation.
A CAUTION FOR THE SELECTIVITY GATE. The dossier now lists the CXXC family as closed at twelve and treats the other eleven as off-targets for a KMT2A-CXXC ligand. This fusion shows one of those eleven is also a KMT2A fusion PARTNER - so in a t(10;11) patient, TET1 is simultaneously an off-target of the ligand and half of the oncoprotein, and the fusion has already removed the domain the ligand would bind.