Early T-cell precursor (ETP) — the thymocyte that has not yet given up being myeloid, and where KMT2A-rearranged T-ALL sits
also called ETP · early T-cell precursor · ETP-ALL · early thymic progenitor · pro-T cell
The early T-cell precursor is the newly arrived cell in the thymus that is starting to become a T cell but has not yet abandoned the option of becoming a myeloid cell instead. Leukaemias arising at this stage look like T-cell leukaemias but behave like myeloid ones, and they respond badly to standard T-cell treatment.
The same thing, for a scientist
The ETP (CL:0000827, pro-T cell) is the most immature intrathymic population, derived from a circulating bone-marrow progenitor and retaining stem-cell and myeloid features. ETP-ALL was defined as a distinct high-risk subtype on gene expression and a characteristic immunophenotype — **CD1a-negative, CD8-negative, CD5-weak, with stem-cell or myeloid markers** — present in 12.6 percent of 239 children with T-ALL, with markedly increased genomic instability and a 10-year risk of remission failure or haematological relapse of 72 percent versus 10 percent for typical T-ALL at the same institution [E1]. Whole-genome sequencing then showed its mutational spectrum is **similar to myeloid tumours** and its global transcriptional profile similar to normal and myeloid-leukaemia haematopoietic stem cells [E2]. KMT2A enters this compartment twice over: KMT2A rearrangements are found in both ETP-ALL and T/myeloid MPAL in childhood series [E3], and across 522 consecutively screened T-ALLs the HOXA trans-deregulated group carrying MLLT10, KMT2A or SET fusions had 5-year event-free survival of **35.7 percent versus 63.7 percent** [E4].
What an ETP is and what it would be doing if it were healthy
T cells are the only blood cells that are not finished in the bone marrow. A progenitor leaves the marrow, travels in the blood, and enters the thymus, a small organ behind the breastbone that is large in infancy and shrinks through life. Inside, it undergoes a long education: rearranging its T-cell receptor genes, being tested against the body's own proteins, and either passing (and leaving as a mature T cell) or dying.
The early T-cell precursor is the very first stage of that process — the cell that has just arrived and started down the T path. Crucially it has not yet fully committed. It still carries stem-cell and myeloid gene programmes, and under the right conditions it can still become a myeloid cell. The marker signature reflects this: it lacks CD1a and CD8, expresses CD5 only weakly, and carries stem-cell or myeloid markers such as CD34, CD117, CD13 or CD33 that a proper thymocyte would have switched off 1.
That incomplete commitment is exactly why leukaemia at this stage behaves the way it does.
ETP-ALL is a T-cell leukaemia with a myeloid genome
Leukaemic cells from 239 children with T-ALL (139 at St Jude, 100 in the Italian AIEOP ALL-2000 study) were assessed by gene-expression profiling, flow cytometry and SNP array. Thirty patients, 12.6 percent, had an ETP-related expression signature or the distinctive immunophenotype. Their cells showed increased genomic instability in both number and size of lesions. Outcomes were dramatically worse: 72 percent risk of remission failure or haematological relapse at 10 years versus 10 percent for typical T-ALL at St Jude, and 57 percent at 2 years versus 14 percent in the AIEOP trial 1.
Whole-genome sequencing of 12 ETP-ALL cases with frequency assessment across 94 T-ALL then characterised the genetics: activating mutations in cytokine-receptor and RAS signalling (67 percent — NRAS, KRAS, FLT3, IL7R, JAK3, JAK1, SH2B3, BRAF), inactivating lesions disrupting haematopoietic development (58 percent — GATA3, ETV6, RUNX1, IKZF1, EP300) and histone-modifying genes (48 percent — EZH2, EED, SUZ12, SETD2, EP300). The conclusion was explicit: the mutational spectrum is similar to myeloid tumours, the global transcriptional profile resembles normal and myeloid-leukaemia HSCs, and myeloid-directed therapy might improve outcomes 2.
For this graph the pattern is familiar. A 67 percent RAS-pathway mutation rate in the most immature, most refractory T-lineage subgroup sits alongside the 100 percent RAS-pathway mutation rate in IC2, the most refractory and least differentiated cluster of infant KMT2A-rearranged ALL. The same pairing — least differentiated, most RAS-mutated, worst outcome — appears independently in both lineages.
Where KMT2A fits
In the ETP compartment itself. Among 857 consecutive childhood leukaemias (266 T-ALL, 591 AML) screened before any treatment decision, 26 ETP-ALL and 10 T/myeloid MPAL cases were identified. The immunophenotypes overlapped heavily — CD34, CD117, CD13/CD33 and CD11b were expressed similarly in both, with CD2 and HLA-DR more frequent in T/M-MPAL — and KMT2A rearrangements were found in both entities, as were NOTCH1, FLT3-ITD and N/KRAS mutations and TLX3 rearrangement. What separated the two entities definitively was cytoplasmic MPO positivity and myelomonocytic differentiation. Five-year overall survival was 56-57 percent in both, and NOTCH1 mutation marked a substantially better outcome (90 percent versus 37 percent) 3.
In T-ALL more broadly. An RT-MLPA assay applied to 522 consecutive adult and paediatric T-ALLs found a fusion transcript in 20 percent, with PICALM-MLLT10 (4 percent), NUP214-ABL1 (3 percent) and SET-NUP214 (3 percent) commonest. In the 235 patients with full clinical annotation, those with HOXA trans-deregulated T-ALL carrying MLLT10, KMT2A or SET fusions (17 percent, 39 of 235) had 5-year EFS of 35.7 percent versus 63.7 percent (HR 1.63, p = 0.04), with a trend to higher cumulative relapse incidence 4.
Note what that grouping does: it puts KMT2A fusions in the same prognostic bin as MLLT10 and SET fusions on the grounds that all three trans-activate HOXA. That is the same axis that makes U-937 an unsuitable control in ana-kmt2a-r-line-panel-monocytic-audit, arrived at independently from patient outcomes.
| ontology | CL:0000827 |
|---|---|
| location | Thymus (cortico-medullary junction on entry); derived from a circulating bone-marrow lymphoid progenitor |
| immediate upstream | cell-type-clp |
markers
[ "CD1a-", "CD8-", "CD5 weak", "cytoplasmic CD3+", "CD34+ and/or CD117+", "CD13 and/or CD33+", "MPO-negative (MPO positivity reclassifies as T/myeloid MPAL)" ]
lineage potential
[ "T cell", "retained myeloid potential", "NK cell" ]
kmt2a relevance
{
"etp_all": "KMT2A rearrangements occur in both ETP-ALL and T/myeloid MPAL in childhood series",
"t_all_prognosis": "HOXA trans-deregulated T-ALL with MLLT10, KMT2A or SET fusions had 5-year EFS 35.7 percent versus 63.7 percent in a 235-patient annotated cohort",
"rarity": "partner-level KMT2A fusions in T-ALL are individually rare; only five KMT2A-ELL T-ALL patients are reported in the entire literature",
"shared_axis": "the prognostic grouping pools KMT2A with MLLT10 and SET on HOXA trans-deregulation, the same axis that disqualifies U-937 as a control in ana-kmt2a-r-line-panel-monocytic-audit"
}cohort numbers
{
"etp_frequency_in_paediatric_t_all": "12.6 percent (30 of 239)",
"fusion_positive_t_all": "20 percent of 522 consecutive cases",
"hoxa_group": "17 percent (39 of 235)"
}atlas refs — the numbers behind the chart
[
{
"label": "Definition of ETP-ALL as a high-risk subtype",
"pmid": "19147408",
"title": "Early T-cell precursor leukaemia: a subtype of very high-risk acute lymphoblastic leukaemia.",
"authors": "Coustan-Smith E, Mullighan CG, Onciu M, et al.",
"venue": "Lancet Oncol",
"year": 2009,
"title_provenance": "Title/authors/venue backfilled from PubMed esummary on 2026-09-20 because this citation stored a bare PMID with no title, which made it unscoreable by verify-citations.py. BACKFILLING IS NOT VERIFICATION: it confirms the PMID resolves to a real paper, not that this is the right paper for the claim. That still requires reading."
},
{
"label": "Genetic basis of ETP-ALL, myeloid-like spectrum",
"pmid": "22237106",
"title": "The genetic basis of early T-cell precursor acute lymphoblastic leukaemia.",
"authors": "Zhang J, Ding L, Holmfeldt L, et al.",
"venue": "Nature",
"year": 2012,
"title_provenance": "Title/authors/venue backfilled from PubMed esummary on 2026-09-20 because this citation stored a bare PMID with no title, which made it unscoreable by verify-citations.py. BACKFILLING IS NOT VERIFICATION: it confirms the PMID resolves to a real paper, not that this is the right paper for the claim. That still requires reading."
},
{
"label": "Fusion landscape and outcomes in 522 T-ALL",
"pmid": "35082269",
"title": "Clinico-biological features of T-cell acute lymphoblastic leukemia with fusion proteins.",
"authors": "Steimlé T, Dourthe ME, Alcantara M, et al.",
"venue": "Blood Cancer J",
"year": 2022,
"title_provenance": "Title/authors/venue backfilled from PubMed esummary on 2026-09-20 because this citation stored a bare PMID with no title, which made it unscoreable by verify-citations.py. BACKFILLING IS NOT VERIFICATION: it confirms the PMID resolves to a real paper, not that this is the right paper for the claim. That still requires reading."
}
]- ETP-ALL retains myeloid potential, so a T-lineage-directed immunotherapy faces the same lineage-escape route that CD19-directed therapy faces in B-ALL, and the escape destination is myeloid in both cases.
- The thymus is a partial sanctuary site with its own vascular barrier, so a leukaemia sitting in this compartment is less exposed to systemic therapy than marrow disease.
- KMT2A fusions in T-ALL are too rare to power a dedicated trial, so this subgroup will be treated on evidence generated in other lineages whether or not it behaves the same way.
- If the prognostic signal belongs to HOXA trans-deregulation rather than to KMT2A, then a menin inhibitor might help MLLT10 and SET fusion patients too, and conversely a KMT2A-restricted eligibility criterion would exclude patients who would benefit.
- Is KMT2A-rearranged T-ALL menin-dependent in the way KMT2A-rearranged B-ALL and AML are? Menin-inhibitor trials have enrolled almost entirely B-lineage and myeloid disease, and the HOXA-driven T-lineage cases would be the natural extension.
- Why does the same pairing — least differentiated, heaviest RAS-pathway mutation burden, worst outcome — appear independently in ETP-ALL (67 percent RAS-pathway mutated) and in the IC2 cluster of infant KMT2A-rearranged ALL (100 percent)? A shared mechanism would be more interesting than a coincidence.
- Can a KMT2A-specific effect size be extracted from T-ALL cohorts, or are the numbers permanently too small? Pooling with MLLT10 and SET on HOXA grounds is defensible biologically but prevents partner-level inference.
- Given that MPO positivity is the only definitive separator between ETP-ALL and T/myeloid MPAL, are the two being distinguished on a genuinely biological boundary or on an assay threshold?
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
Among 239 children with T-ALL studied by gene-expression profiling, flow cytometry and SNP array, 30 (12.6 percent) had leukaemic lymphoblasts with an ETP-related expression signature or its distinctive immunophenotype (CD1a-negative, CD8-negative, CD5-weak, with stem-cell or myeloid markers); ETP-ALL showed increased genomic instability and a high risk of remission failure or haematological relapse — 72 percent at 10 years versus 10 percent for typical T-ALL at St Jude, and 57 percent at 2 years versus 14 percent in the AIEOP trial.
Early T-cell precursor leukaemia: a subtype of very high-risk acute lymphoblastic leukaemia · Coustan-Smith E, Mullighan CG, Onciu M, Behm FG, Raimondi SC, Pei D, et al., Campana D · Lancet Oncology 2009What this evidence cannot show
Thirty ETP cases across two cohorts, with outcomes on therapy protocols of the 1990s and early 2000s; the very poor prognosis reported here has been substantially attenuated by later MRD-directed intensification, so the 72 percent failure figure should not be quoted as current. Read: abstract only.
Evidence that argues against this (1)
- The prognostic penalty attached to ETP status is not fixed and depends on the treatment protocol. In a childhood series applying contemporary ALL protocols, 5-year overall survival for ETP-ALL was 56 percent and indistinguishable from T/myeloid MPAL at 57 percent, and within that group NOTCH1 mutation marked a far better outcome (90 percent versus 37 percent). ETP status is therefore a risk marker whose weight is protocol-dependent and is modified by co-occurring genetics. Read: abstract only. Noronha EP et al. 2019
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
Whole-genome sequencing of 12 ETP-ALL cases with frequency assessment across 94 T-ALL showed activating mutations in cytokine-receptor and RAS signalling in 67 percent (NRAS, KRAS, FLT3, IL7R, JAK3, JAK1, SH2B3, BRAF), inactivating lesions disrupting haematopoietic development in 58 percent (GATA3, ETV6, RUNX1, IKZF1, EP300) and histone-modifying gene lesions in 48 percent (EZH2, EED, SUZ12, SETD2, EP300); the mutational spectrum is similar to myeloid tumours and the global transcriptional profile of ETP-ALL resembles that of normal and myeloid-leukaemia haematopoietic stem cells.
The genetic basis of early T-cell precursor acute lymphoblastic leukaemia · Zhang J, Ding L, Holmfeldt L, Wu G, Heatley SL, Payne-Turner D, et al., Mullighan CG · Nature 2012What this evidence cannot show
Twelve whole genomes is a small discovery set, and the frequency estimates come from targeted assessment in 94 additional T-ALL rather than from genome sequencing of all of them. The suggestion that myeloid-directed therapy might help is a hypothesis the paper raises, not a result. Read: abstract only.
Evidence that argues against this (1)
- A myeloid-like mutational spectrum does not by itself dictate myeloid-directed treatment. In a childhood series where ETP-ALL and T/myeloid MPAL had essentially identical genetics and immunophenotypes, 31 of 36 patients received ALL protocols and 5 received AML therapy, with 5-year overall survival of 56.4 percent on ALL protocols and no difference between the two entities. The genetic resemblance to myeloid disease has not translated into a demonstrated benefit from myeloid therapy. Read: abstract only. Noronha EP et al. 2019
- E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
Among 857 consecutive childhood leukaemias (266 T-ALL, 591 AML) immunophenotyped before treatment decisions, 26 ETP-ALL and 10 T-lymphoid/myeloid MPAL cases were identified with heavily overlapping immunophenotypes (similar CD34, CD117, CD13/CD33, CD11b; CD2 and HLA-DR more frequent in T/M-MPAL), and KMT2A rearrangements were found in both entities alongside NOTCH1, FLT3-ITD, N/KRAS mutations and TLX3 rearrangement; cytoplasmic MPO positivity and myelomonocytic differentiation were definitive in separating the two, 5-year overall survival was 56-57 percent in both on ALL protocols, and NOTCH1 mutation marked better survival (90 percent versus 37 percent, p = 0.017).
T-lymphoid/myeloid mixed phenotype acute leukemia and early T-cell precursor lymphoblastic leukemia similarities with NOTCH1 mutation as a good prognostic factor · Noronha EP, Marques LVC, Andrade FG, Sardou-Cezar I, Dos Santos-Bueno FV, Zampier CDP, et al., Pombo-de-Oliveira MS · Cancer Management and Research 2019What this evidence cannot show
Thirty-six cases total across the two entities, from a single Brazilian registry, with KMT2A status assessed by RT-PCR and FISH but no partner-level breakdown and no statement of how many of the 36 were KMT2A-rearranged. The NOTCH1 survival difference rests on very small numbers. Note the abstract contains a typographical error, 'KTM2A' for KMT2A. Read: abstract only.
Evidence that argues against this (1)
- The claim that these are near-identical entities is contradicted by the same study's own discriminator: cytoplasmic MPO positivity and myelomonocytic differentiation separated them definitively, and MPO is precisely the criterion the WHO uses to assign myeloid lineage in mixed-phenotype acute leukaemia. Overlapping surface markers with a decisive lineage-defining difference is an argument that the two entities are distinct, not that they are the same. Read: abstract only. Khoury JD et al. 2022
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
An RT-MLPA assay applied to 522 consecutive adult and paediatric T-ALLs identified a fusion transcript in 20 percent of cases, with PICALM-MLLT10 (4 percent), NUP214-ABL1 (3 percent) and SET-NUP214 (3 percent) most frequent; among 235 clinically annotated patients, those with HOXA trans-deregulated T-ALL carrying MLLT10, KMT2A or SET fusions (17 percent, 39 of 235) had worse prognosis with 5-year event-free survival of 35.7 percent versus 63.7 percent (HR 1.63, p = 0.04) and a trend to higher 5-year cumulative incidence of relapse (45.7 percent versus 25.2 percent).
Clinico-biological features of T-cell acute lymphoblastic leukemia with fusion proteins · Steimle T, Dourthe ME, Alcantara M, Touzart A, Simonin M, Mondesir J, et al., Asnafi V · Blood Cancer Journal 2022What this evidence cannot show
KMT2A fusions are pooled with MLLT10 and SET fusions into a single HOXA trans-deregulated group of 39 patients, so no KMT2A-specific effect size is reported; the cohort mixes adults from GRAALL2003/05 with children from FRALLE2000 on different protocols, and the relapse-incidence difference did not reach significance (p = 0.11). Read: abstract only.
Evidence that argues against this (1)
- The HOXA grouping may be doing the prognostic work rather than KMT2A rearrangement itself, and KMT2A fusions in T-ALL are individually rare enough that their impact is genuinely uncertain. A case report plus literature search for KMT2A-ELL in T-ALL found only four other reported patients in the entire literature and concluded the fusion is a rare recurrent event in T-ALL with uncertain prognostic implications. An effect estimated on a pooled group of 39 cannot be attributed to a partner-level lesion seen in single digits. Read: abstract only. Panagopoulos I et al. 2021
about half of PICALM::MLLT10 T-ALL is ETP-ALL
the CLP circulates and seeds the thymus as the most immature thymocyte
ETP-ALL and T/myeloid MPAL share immunophenotype and KMT2A rearrangement, separated only by MPO
- related to →HOXA9 — required for survival of established MLL-rearranged leukemia, and dispensable for starting it
KMT2A fusions in T-ALL are grouped prognostically by HOXA trans-deregulation
- arises in →KMT2A::ELL (MLL-ELL), t(11;19)(q23;p13.1)
KMT2A-ELL is a rare recurrent event in T-ALL with only five reported patients
- ← expressed inGATA3 - one of only two transcription-factor baits in a 109-bait BioID screen that recover DOT1L
T-lineage master transcription factor; the cell type where a GATA3-DOT1L relation would matter
- ← expressed inCCNT1 / cyclin T1 — the regulatory half of P-TEFb, and the one part of the elongation core with any dependency variance
the only primary haematopoietic cell type in which cyclin T1 abundance has been measured and manipulated is the resting CD4-positive T lymphocyte, in the HIV-latency literature rather than the leukaemia one
WEAK: acute Taspase1 deletion markedly reduces thymocytes in mice (PMID 16951254) - one abstract clause with no immunophenotype, so the nearest node is used rather than a measured population
REQUIRED IN: Atm-null mice have defects in T lymphocyte maturation and develop malignant thymic lymphoma at 2-4 months - a lineage requirement in an animal with no fusion (murine germline knockout; PMID 8689683)
the murine knockout result covers lymphocyte receptor genes in both lineages, so the T-precursor expression is part of the same claim rather than a separate one (murine; PMID 8356452)
the nearest wrong answer, deliberately not used by Q69: a thymic precursor is not a mature CD4 effector fate, and forcing the RTF1 result onto it would have asserted a developmental stage the source does not
- ← expressed inCCND3 (cyclin D3) — the scan's rank-4 hit, and the only one of the four with a drug class already in children
Ccnd3-null mice fail to expand immature T lymphocytes and resist T-cell malignancy; knockdown blocks human immature-T-derived ALL
CHK1 is overexpressed and required in human T-ALL, with primary leukaemic cells sensitive at doses that spare normal thymocytes
- ← expressed inAFF3 (LAF4) — the AFF-family scaffold of SEC-L3, and a KMT2A fusion partner in its own right
UniProt P51826 tissue specificity: preferentially lymphoid, HIGHEST IN THYMUS - the most lineage-restricted of the four AFF scaffolds, and an awkward competitor to place against AFF1 in a B-lineage disease (Q89 2026-09-21)