Oncofusion

start / graph / interface-aff4-btk-declined-an-osteosarcoma-row-on-the-one-partner-with-an-approved-drug

Interfaceweak confidence 2 of 5: Preliminary, small, indirect, or not yet peer reviewed.q160-the-prey-only-interactome

AFF4-BTK - DECLINED. One osteosarcoma AP-MS row, and it is the only partner in the uncovered prey-only tail with an approved drug against it, which is exactly why the decline is written down

also called AFF4-BTK · Bruton tyrosine kinase · AFF4-Bruton

Bruton's tyrosine kinase, the target of ibrutinib, turned up with the elongation-complex scaffold AFF4 in one bone-cancer-cell experiment. BTK belongs in B cells, this experiment was not in one, and one row is not enough.

The same thing, for a scientist

**Verdict: DECLINED.** IntAct holds one AFF4-BTK (Q9UHB7-Q06187) record: anti-tag co-IP, *association*, **U2OS osteosarcoma**, AFF4 as prey, intact-miscore 0.35, from **PMID 40205054** [E1] - the Schaffer et al. *Nature* 2025 U2OS multimodal cell map, a proteome-scale resource integrating biophysical interaction measurement with immunofluorescence over more than 5,100 proteins and resolving 275 molecular assemblies [E2]. **The decline is straightforward - one row, one line, an atlas, a tagged bait, no reciprocal - and it is written up rather than dropped because of what BTK is.** BTK (Q06187, 659 aa) is a **B-cell receptor-pathway kinase** and the target of **ibrutinib, acalabrutinib and zanubrutinib**, all approved. KMT2A::AFF1 is the defining lesion of **infant B-ALL**. **A row connecting the fusion's partner scaffold to a druggable B-lineage kinase is the single most promotable claim in this entire census, and it rests on one osteosarcoma pull-down.**

Go deeper
Why this row is worse than it looks, not better

The instinct on seeing AFF4-BTK in a B-cell leukaemia project is that the lineage fits. It does not fit; that is the problem. The experiment was done in U2OS, an osteosarcoma line, which does not express BTK at meaningful levels and has no B-cell receptor. So the pull-down is a tagged kinase introduced into a cell with none of its physiological context - precisely the pattern that killed interface-menin-cd226-declined-proteome-scale-ap-ms, where a lymphocyte surface receptor was over-expressed in kidney, colon and bone lines.

The lineage argument is doing no work here at all. BTK being a B-cell kinase and KMT2A::AFF1 being a B-ALL lesion are two true facts about different systems. Joining them across a U2OS row is the shape this corpus records as two true facts joined is not a verified fact.

What would make it real: an AFF4 or AFF1 co-IP from a KMT2A-rearranged B-ALL line blotted for BTK, or the reciprocal. Neither exists. The experiment is cheap, the reagents are standard, and the reason to run it is the drug rather than the row.

The clinical question the row points at, which is separable from the row

Independently of whether AFF4 binds BTK, BTK inhibitors have essentially never been tested in KMT2A-rearranged infant ALL, and the reason is developmental rather than mechanistic: KMT2A-rearranged infant ALL is characteristically pro-B / CD10-negative, arrested before the stage at which the B-cell receptor - and therefore BTK signalling - is the cell's dominant survival input. A BTK inhibitor works where BCR signalling is load-bearing.

So the honest statement is a conjunction of two absences: nobody has shown AFF4 contacts BTK in a relevant cell, and nobody has shown that KMT2A-rearranged blasts depend on BTK. Either would be interesting; this row is evidence for neither.

And there is a bite-back that does not depend on the row being true at all. Ibrutinib is given to large numbers of patients, and its off-target profile is broad. If the super elongation complex has any BTK-dependent regulation in any lineage, that is an unexamined transcriptional effect of an approved drug.

Facts and figures
partner agene-aff4
partner bBTK
partner b uniprotQ06187
partner b length AA659
partner a uniprotQ9UHB7
partner a length AA1163
verdictDECLINED
adjudicated byq160, cycle 146, 2026-09-24
druggabilityBTK is the only partner in the uncovered prey-only tail with approved covalent inhibitors against it (ibrutinib, acalabrutinib, zanubrutinib). That is the reason this decline is written down rather than left implicit.
n methods1
n platforms1
n publications1
expansion methodnot flagged on this row; the same publication's rows are spoke-expanded elsewhere in this corpus
endogenous sideno
intact miscore0.35
cell stateunstimulated; no B-cell receptor engagement, and U2OS has no B-cell receptor
epigeneticsNone.
absence classnever-measured
absence class noteTwo separate absences: no AFF4 or AFF1 co-IP in a KMT2A-rearranged B-ALL cell has been blotted for BTK, and no BTK-dependence experiment exists in KMT2A-rearranged leukaemia at all.
interface residues a
[]
structures
[]
ligands
[]
records
[
  "AFF4-BTK, anti tag coip, association, U2OS, AFF4 prey, miscore 0.35, PMID 40205054"
]
cell systems
[
  "U2OS osteosarcoma"
]
expression mismatch
{
  "partner_lineage": "B-lymphoid (BCR pathway kinase)",
  "host_line_used": "U2OS osteosarcoma",
  "line_expresses_partner_natively": false
}
localization
{
  "compartment": "cytoplasm and plasma membrane for BTK; nucleus for AFF4",
  "subnuclear": [],
  "shown_for_fusion": "NO - nobody has looked in any fusion, and no BTK-dependence experiment exists in KMT2A-rearranged leukaemia."
}
measurements
[
  {
    "quantity": "count",
    "subject": "uncovered prey-only partners with an FDA-approved targeted inhibitor",
    "value": 1,
    "units": "partners",
    "n": 135,
    "universe": "the 135 uncovered human prey-only partners, 2026-09-24; judged by hand against approved-drug target knowledge, not against a drug database",
    "null": null,
    "note": "BTK. Hand-judged - EGFR and BRAF also have approved inhibitors but are already covered by q128, so they are not in the 135."
  }
]
How this could bite us back
  • Ibrutinib's off-target profile is wide and includes several TEC- and SRC-family kinases. Any elongation phenotype attributed to BTK in this setting would need a second, more selective inhibitor before it meant anything.
  • A B-lineage kinase in a B-lineage leukaemia is the most seductive kind of wrong answer, because the lineage argument feels like evidence. The row is an osteosarcoma pull-down and the lineage contributes nothing to it.
Still open
  1. Do KMT2A-rearranged B-ALL blasts depend on BTK at all? The pro-B, CD10-negative arrest predicts no, but the prediction has not been tested and DepMap has no KMT2A-rearranged infant-ALL line to test it in.
  2. Does an AFF4 co-IP from a KMT2A::AFF1 line recover BTK? One blot, standard reagents, and currently nobody has done it in any B-lineage cell.
  3. Do BTK inhibitors have transcription-elongation effects in any lineage? If the SEC is BTK-regulated anywhere, that is an unexamined pharmacology of three approved drugs.
Every source behind this entry (2)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    IntAct REST returns one AFF4-BTK (Q9UHB7-Q06187) record: detection method anti-tag co-immunoprecipitation, interaction type 'association', host Homo sapiens U2OS osteosarcoma cell, AFF4 as PREY, intact-miscore 0.35, from PMID 40205054. UniProtKB Q06187 (BTK_HUMAN) is 659 aa.

    What this evidence cannot show

    One record, one cell line, one method. The Schaffer map is spoke-expanded for its AP-MS arm, as this corpus already records for the CD226 row from the same publication, so the binary AFF4-BTK pair was probably never measured as such.

    Evidence that argues against this (1)
    • The Schaffer map's distinguishing feature is that it integrates AP-MS with size-exclusion and imaging, so a row there is in principle better supported than a bare AP-MS row. That integration is not visible at row level in IntAct, and this entry does not credit it because it cannot check it. Schaffer LV et al. 2025
  2. E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    PMID 40205054 (Schaffer et al., Nature 2025) is a U2OS multimodal cell map integrating biophysical interaction measurement with immunofluorescence across more than 5,100 proteins and resolving 275 molecular assemblies. It is a proteome-scale resource; neither AFF4 nor BTK is its subject. This corpus already cites it in five entries.

    Multimodal cell maps as a foundation for structural and functional genomics. · Schaffer LV, Hu M, Qian G, Moon KM, Pal A, Soni N, Latham AP, Pontano Vaites L, et al. (34 authors), Ideker T · Nature 2025
    PMID 40205054DOI
    What this evidence cannot show

    Abstract read in full; the interaction tables and Extended Data were not opened. Rule 82 check run: this PMID is cited five times already in this graph, including in interface-menin-cd226-declined-proteome-scale-ap-ms, which reaches the same verdict about a different partner from the same map - so the reasoning here is not new, only its application.

    Evidence that argues against this (1)
    • A large map is not the same as an unfiltered list; Schaffer et al. apply confidence thresholds and integrate orthogonal modalities. Declining every row from it would discard a genuine resource, and the specific objection here is the cell line rather than the study design. Schaffer LV et al. 2025
Connected to