Oncofusion

start / graph / gene-epha7

Genemoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q152-aff1-only-recurrent-core

EPHA7 - ephrin type-A receptor 7, a published direct promoter target of the ALL1/AF4 (KMT2A::AFF1) fusion protein

also called EphA7 · EHK-3 · EK11 · HEK11 · Receptor, EphA7

EPHA7 makes a receptor that sits on the cell surface and reads signals from neighbouring cells. In 2007 it was shown that the MLL-AF4 fusion protein binds the EPHA7 gene's switch directly and turns it up, and that turning the fusion off turns EPHA7 down.

The same thing, for a scientist

EPHA7 (Q15375, 998 aa) is a receptor tyrosine kinase of the EPH family. Nakanishi et al. 2007 identified it as a direct target of KMT2A (ALL1/MLL) fusion proteins: ChIP showed tagged ALL1 fusion proteins occupying the EphA7 promoter, K562 cells expressing recombinant ALL1/AF4 or ALL1/AF9 up-regulated EphA7, and siRNA knockdown of ALL1/AF4 in SEMK2 t(4;11) cells down-regulated it. Up-regulation is accompanied by ERK phosphorylation and an ERK-phosphorylation blocker kills t(4;11) cells selectively. This project's q117 independently places Epha7 in the top twenty fusion-occupied bins in murine MLL-AF9 with a 0.178 VTP50469/DMSO share ratio, and q152 finds the human EPHA7 promoter occupied in 3 of 8 primary KMT2A::AFF1 samples and 0 of 19 non-AFF1 samples of GSE252378.

Go deeper
The founding experiment and its limits

The 2007 PNAS paper is the founding observation and remains, as far as this beat could find, the only dedicated promoter-occupancy experiment on EPHA7 in a KMT2A fusion context 1. Its strongest arm is the SEMK2 knockdown, because SEMK2 carries an endogenous t(4;11); its ChIP arm uses tagged fusion protein and its expression arm uses K562, a CML line with ectopic fusion expression. No entry in this graph cited this paper before 2026-09-24.

Where the partner scope disagrees

The 2007 work reports EphA7 up-regulation under ALL1/AF9 as well as ALL1/AF4, and q117 confirms occupancy in murine MLL-AF9. In GSE252378 the three primary human KMT2A-AF9 samples are negative at promoter scale 2. See finding-q152-epha7-is-the-one-core-member-with-a-published-kmt2a-aff1-promoter-chip for the three readings that could explain it; none is excluded.

EPHA7 outside leukaemia, which matters for any therapeutic reading

EPHA7 is silenced by promoter hypermethylation in colorectal cancer, expressed at substantial levels in most lung cancers, and has a secreted splice variant first described in malignant lymphoma and then in lung cancer 3. A surface-receptor-directed strategy therefore meets a target whose presence, form and promoter state are all tissue- and tumour-dependent.

Facts and figures
uniprotQ15375
length AA998
gene symbolEPHA7
hgnc symbolEPHA7
chromosome6q16.1
promoter hg38chr6:93,418,559-93,420,559 (TSS 93,419,559, minus strand, NM_001288629)
shown for fusionKMT2A::AFF1 (ALL1/AF4) and KMT2A::MLLT3 (ALL1/AF9) by ChIP on tagged fusion protein in K562, plus endogenous knockdown in SEMK2 t(4;11); murine MLL-AF9 by CUT&RUN in this project's q117; NOT reproduced in primary human KMT2A-AF9 blasts at promoter scale in GSE252378
epigenetic readoutpromoter hypermethylation silences EPHA7 in colorectal cancer; no histone-mark measurement at the EPHA7 promoter in any KMT2A-rearranged system was found by this beat
relationship to existing entriesNamed by beat Q152 because its promoter sits in the 17-gene AFF1-only recurrent core of GSE252378. Q152's headline result is that that core is NOT more recurrence than a label permutation gives (p = 0.11), so this entry does NOT claim the gene is a KMT2A::AFF1 target on the strength of that occupancy. The entry exists because the gene is real biology in this disease and had no entry. EPHA7 is the exception: it has independent published direct-occupancy evidence and is written as a target on that basis, not on this beat's occupancy call.
measurements
[
  {
    "quantity": "n_aff1_samples_occupied",
    "subject": "EPHA7 2 kb promoter in the KMT2A::AFF1 arm of GSE252378",
    "value": 3,
    "units": "samples of 8",
    "n": 27,
    "universe": "27 GSE252378 KMT2A-fusion samples, two-antibody 2.0x within-sample share plus per-library width-matched floor",
    "null": 2.73,
    "note": "p247, p279, p454; occupied calls 2.02-2.23x; 0 of 19 non-AFF1, highest non-AFF1 1.09x (p186), separation margin 1.84. Null 2.73 is q151's per-promoter expectation across 27 samples. Q152 shows a 3-of-8 count at this pass rate is within the chance range."
  }
]
How this could bite us back
  • The only published therapeutic handle is downstream ERK inhibition, which inherits every RAS-pathway resistance route this graph already records for KMT2A-rearranged disease, including RAS and FLT3 co-mutation.
  • EPH receptors are widely expressed in normal neural and epithelial tissue, so a receptor-directed agent has an on-target toxicity problem independent of leukaemia biology.
  • A secreted EPHA7 isoform could neutralise an antibody or ligand-trap approach before it reaches the cell surface.
Still open
  1. Nothing appears to have followed the 2007 ERK result in a KMT2A::AFF1 system. Is EPHA7-driven ERK activation reproducible in a modern patient-derived xenograft?
  2. Is the EPHA7 promoter CpG-island methylated in KMT2A-rearranged blasts that do NOT show occupancy, which would make the partner-scope disagreement a methylation story rather than a fusion-binding one?
  3. Does the secreted EPHA7 isoform exist in leukaemia, and would it act as a decoy against any receptor-directed agent?
Every source behind this entry (4)
  1. E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    ChIP analysis demonstrates occupancy of tagged ALL1 fusion proteins on the EphA7 promoter; EphA7 is up-regulated in K562 cells producing recombinant ALL1/AF4 or ALL1/AF9; siRNA suppression of ALL1/AF4 in SEMK2 t(4;11) cells down-regulates EphA7; up-regulation is accompanied by ERK phosphorylation and an ERK-phosphorylation blocker causes apoptosis specific to t(4;11) cells.

    ALL1 fusion proteins induce deregulation of EphA7 and ERK phosphorylation in human acute leukemias · Nakanishi H, Nakamura T, Canaani E, Croce CM · Proc Natl Acad Sci U S A 104:14442-7 2007
    PMID 17726105DOI
    What this evidence cannot show

    Abstract read in full, figures not opened. Tagged fusion protein, cell lines only, 2007; no patient-derived occupancy and no replication found in this beat's searches.

    Evidence that argues against this (1)
    • EPHA7 is silenced by promoter hypermethylation in colorectal cancer and expressed at substantial levels in most lung cancers, so its promoter state is strongly tissue-dependent; a 2007 result in K562 and SEMK2 may not describe a primary infant blast. Tsuboi M et al. 2010
  2. E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset

    In GSE252378 the EPHA7 2 kb promoter (hg38 chr6:93,418,559-93,420,559, minus-strand refGene TSS 93,419,559, NM_001288629) is occupied in 3 of 8 KMT2A::AFF1 samples (p247, p279, p454) at 2.02-2.23x and 0 of 19 non-AFF1 samples including all three KMT2A-AF9 samples; the highest non-AFF1 reading is 1.09x and no non-AFF1 sample exceeds 1.5x.

    Q152 per-core-gene occupancy and separation margins over GSE252378 · Iqbal Lab oncofusion knowledge graph, beat Q152 · this project 2026
    What this evidence cannot show

    Three positives in three distinct patients, but Q152's own permutation null shows a 3-of-8 pattern is within the chance range at this rule's pass rate. The promoter window contains no other refGene TSS within 2 kb, so the signal is attributable to EPHA7.

    Evidence that argues against this (1)
    • The three primary KMT2A-AF9 samples in this deposit are negative, which contradicts the published AF9 arm of the founding experiment. Either the promoter-scale rule is underpowered in those libraries or the K562 ectopic-expression result does not transfer, and this beat cannot tell which. Nakanishi H et al. 2007
  3. E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    EPHA7 carries a secreted splice variant, first reported in malignant lymphoma and then detected in human lung cancer cell lines and primary lung cancer, and is down-regulated by promoter hypermethylation in colorectal cancer.

    Secreted form of EphA7 in lung cancer · Tsuboi M, Mori H, Bunai T, Kageyama S, Suzuki M, Okudela K, Takamochi K, Ogawa H, Niwa H, Shinmura K, Sugimura H · Int J Oncol 36:635-40 2010
    PMID 20126984
    What this evidence cannot show

    Lung and colorectal cancer, not leukaemia. Cited here only for the target-biology facts (secreted isoform, methylation-dependent silencing) that bear on druggability, not for any leukaemia claim.

    Evidence that argues against this (1)
    • The secreted-isoform and methylation findings are from lung and colorectal cancer. Whether either applies in leukaemia is unmeasured, so citing them as druggability caveats imports a risk that may not exist in this disease. Nakanishi H et al. 2007
  4. E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    EPHA7 is a reviewed human UniProt entry, accession Q15375, 998 residues, ephrin type-A receptor 7, EC 2.7.10.1.

    UniProtKB Q15375 - Ephrin type-A receptor 7 (Homo sapiens), reviewed · UniProt Consortium · UniProtKB/Swiss-Prot 2026
    What this evidence cannot show

    A single reviewed human record is returned by gene_exact:EPHA7, so no accession collision applies; data/symbol-collisions.json has no row for EPHA7 (checked, 211 rows).

    Evidence that argues against this (1)
    • A single canonical length does not capture the secreted EPHA7 splice variant, which is the isoform most relevant to any extracellular therapeutic strategy and is not the reviewed canonical sequence. Tsuboi M et al. 2010
Connected to