Oncofusion

start / graph / ana-calm-af10-menin-bounded-absence

Analysisstrong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.Q55-calm-af10

Nobody has ever published a menin experiment in CALM-AF10 - counted, controlled and dated

also called CALM-AF10 menin literature absence

We searched for any study that tested whether CALM-AF10 leukaemia needs menin, and found none. To make sure the search itself was not broken, we ran the same searches with the pieces separated and got thousands of hits, so the zero is real.

The same thing, for a scientist

A rule-67-compliant counted absence, run 2026-09-20. **Every query joining a CALM-AF10 name to a menin name returns zero**, while each operand alone returns a large set, and the only full-text co-occurrences in Europe PMC are reviews and conference abstract books. The 2026 comprehensive disease review does not contain the word 'menin'. This bounds the *published literature*; it does not bound the *experiment*, and the caveat says so.

Go deeper
The counts

PubMed E-utilities esearch, 2026-09-20 (querytranslation read back on every one):

queryrecords
CALM-AF10 AND menin0
CALM-AF10 AND MEN10
CALM AF10 AND menin0
PICALM AND menin0
AF10 AND menin0
"CALM-AF10" AND (revumenib OR ziftomenib OR VTP-50469 OR VTP50469 OR MI-503 OR MI-2-2 OR SNDX-5613)0
MLLT10 AND menin3 - all KMT2A::MLLT10, none PICALM::MLLT10
MLLT10 AND (revumenib OR ziftomenib OR "menin inhibitor" OR "menin inhibitors" OR "menin inhibition")2 - both KMT2A::MLLT10

Operand controls, identical shape (rule 67):

controlrecords
CALM-AF10 alone81
menin alone2,237
CALM-AF10 AND DOT1L (positive control - a join that SHOULD return)6
CALM-AF10 AND HOXA (positive control)17

The last two matter most: the join operation itself works for this term against other biology, so the zero against menin is not a query artefact.

Europe PMC full text, same day: CALM-AF10 AND menin = 34; CALM-AF10 AND MEN1 = 23; (CALM-AF10 OR "CALM/AF10") AND ("menin inhibitor" OR "menin inhibition" OR VTP-50469 OR MI-503) = 23. Every hit inspected was a review, a haematology-congress abstract book, or a paper about a different fusion that mentions CALM-AF10 in passing. No primary study.

Name list, stated because a drug-class absence is bounded by its name list (rule 72)

Target side: menin, MEN1, 'menin inhibitor', 'menin inhibitors', 'menin inhibition'. Compound side: revumenib, SNDX-5613, ziftomenib, VTP-50469, VTP50469 (hyphenated and unhyphenated treated as different strings), MI-503, MI-2-2, bleximenib, JNJ-75276617, BMF-219, enzomenib, DSP-5336. Fusion side: CALM-AF10, 'CALM AF10', CALM/AF10, PICALM, PICALM-MLLT10, PICALM::MLLT10, MLLT10, AF10.

One token to avoid: bare AF10 on ClinicalTrials.gov returns atrial-fibrillation studies (AF = atrial fibrillation), and bare SET in the SET::NUP214 comparison collects SMYD2 ('SET- and MYND-domain') and C. elegans SET-2. Both were title-checked out.

What the absence does and does not bound

Bounded: no primary paper indexed in PubMed or in Europe PMC's full-text corpus reports a menin knockdown, a MEN1 knockout or a menin-inhibitor exposure in a CALM-AF10 system, as of 2026-09-20.

Not bounded: (a) unpublished screening. U-937 is a standard control line in cell-line panels, so a menin-inhibitor programme almost certainly has exposed CALM-AF10 cells to a menin inhibitor and reported it, if at all, as an unremarked negative in a supplementary table. (b) Conference abstracts not indexed by title. (c) The indirect question - whether CALM-AF10 needs wild-type KMT2A1 - which would be published under different vocabulary entirely.

Per the brief's 'write down what you read and did NOT use' rule, three classes of hit were enumerated and rejected: menin-inhibitor review articles that list CALM-AF10 among HOXA-driven leukaemias without any data (e.g. the 2025-2026 ASH-update reviews), haematology congress abstract books where the two strings co-occur across unrelated abstracts, and NUP98/KAT6A papers that cite CALM-AF10 as background.

Facts and figures
scriptrun ad hoc from the beat scratchpad 2026-09-20 using NCBI E-utilities esearch and the Europe PMC REST search endpoint
statusrun
result summaryZero primary publications test menin dependence in CALM-AF10, bounded by operand controls (81 / 2,237) and two same-shape positive controls (DOT1L 6, HOXA 17). All 34 Europe PMC full-text co-occurrences are reviews or abstract books.
date2026-09-20
inputs
[
  "https://eutils.ncbi.nlm.nih.gov/entrez/eutils/esearch.fcgi (db=pubmed, retmode=json)",
  "https://www.ebi.ac.uk/europepmc/webservices/rest/search"
]
name list target
[
  "menin",
  "MEN1",
  "menin inhibitor",
  "menin inhibitors",
  "menin inhibition"
]
name list compound
[
  "revumenib",
  "SNDX-5613",
  "ziftomenib",
  "VTP-50469",
  "VTP50469",
  "MI-503",
  "MI-2-2",
  "bleximenib",
  "JNJ-75276617",
  "BMF-219",
  "enzomenib",
  "DSP-5336"
]
name list fusion
[
  "CALM-AF10",
  "CALM AF10",
  "CALM/AF10",
  "PICALM",
  "PICALM-MLLT10",
  "PICALM::MLLT10",
  "MLLT10",
  "AF10"
]
token traps found
[
  "`AF10` on ClinicalTrials.gov returns atrial-fibrillation trials",
  "`SET` returns SMYD2 (SET- and MYND-domain) and C. elegans SET-2"
]
How this could bite us back
  • A clean, well-bounded zero is exactly the kind of result that gets read as 'it does not work'. It means nobody looked.
Still open
  1. Do any menin-inhibitor cell-line panel supplementary tables include U-937? That is the cheapest place an existing answer could be hiding.
  2. Has any CALM-AF10 patient been enrolled on AUGMENT-101 cohort 2D or an equivalent HOX/MEIS-defined cohort? The registry records eligibility, not enrolled genotypes.
Every source behind this entry (2)
  1. E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database

    As of 2026-09-20, PubMed returns 0 records for every join of a CALM-AF10 name with a menin or menin-inhibitor name, while CALM-AF10 alone returns 81, menin alone returns 2,237, and the positive-control joins CALM-AF10 AND DOT1L and CALM-AF10 AND HOXA return 6 and 17.

    What this evidence cannot show

    Rule-67 compliant: operand counts and two same-shape positive controls were run, and querytranslation was read back on each query to confirm the parser did not rewrite the terms. Bounds publication, not experiment.

    Evidence that argues against this (1)
    • An absence in PubMed is not an absence of data. AUGMENT-101 cohort 2D enrols leukaemia with 'a genetic mutation expected to lead to HOX/MEIS upregulation', which a CALM-AF10 patient satisfies - so the sponsor's own hypothesis predicts sensitivity, and any patient treated under that criterion constitutes an unreported test. ClinicalTrials.gov 2026
  2. E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.review

    A 2026 comprehensive review of PICALM::MLLT10 leukaemia, covering structure, mechanism, prognosis and treatment options, contains no mention of menin.

    PICALM::MLLT10 translocated leukemia. · Cullen JM, et al. · FEBS Lett 2026
    PMID 41532548DOI
    What this evidence cannot show

    REVIEW, cited as a measure of field attention. Full text fetched from Europe PMC fullTextXML and searched (69,891 characters). The same review DOES discuss DOT1L inhibition and venetoclax at length, so the omission is informative about what the field thinks is relevant rather than about what it had space for.

    Evidence that argues against this (1)
    • The same review also contains a wrong identifier, which is a reminder not to inherit citations from it: its reference 78 gives PMID 27389536 for the pinometostat paediatric phase-1 final report, and PMID 27389536 is in fact 'Blood's 70th anniversary: CARs on the Blood highway'. The pinometostat item is an ASH abstract with no separate PMID. Verified by direct efetch 2026-09-20. Dunbar CE 2016
Connected to