Re-bound at full text: the LEDGF-after-menin-failure intersection goes from 0 records at abstract level to 116 at full text and 90 with resistance terms, and the 50 read are review articles - no LEDGF-directed agent has been tested in a menin-inhibitor-resistant model
also called Q115 LEDGF resistance re-bound
The most obvious clinical use for a LEDGF-blocking drug is after the menin drug stops working, and nobody has tested it. Searching whole papers turns a zero into a hundred hits, all of them reviews that mention both ideas without testing either together.
The same thing, for a scientist
The entry records `(PSIP1 OR LEDGF OR DFS70)` crossed with named menin inhibitors as **0 PubMed records**. At full text the same crossing returns **116**, and with resistance terms added **90** [E1]. That is the largest recall change in this tranche and it is entirely explained by what full text indexes: **the first 50 of the 90 are review articles** - *Menin inhibitors for AML: challenges and opportunities*, *Targeting the Menin-KMT2A interaction in leukemia*, *Menin: from molecular insights to clinical impact*, *Synergistic Strategies for KMT2A-Rearranged Leukemias: Beyond Menin Inhibitor*, and two dozen more of the same shape - which name LEDGF once in a mechanism paragraph and a menin inhibitor elsewhere [E1]. **No primary study among the 50 tests a LEDGF-directed agent in a menin-inhibitor-resistant model. The absence survives.** One record in the 50 is the closest thing in print: **PMID 33477970**, *Unlike Its Paralog LEDGF/p75, HRP-2 Is Dispensable for MLL-R Leukemogenesis but Important for Leukemic Cell Survival* (Cells 2021) - LEDGF-family genetics in MLL-rearranged leukaemia, with no menin-inhibitor arm [E1].
What a 0-to-116 jump means for the rest of this corpus
This is the clearest instance in the tranche of what full text adds and what it does not. It adds the review literature, which is where two ideas get named in the same document without being tested together. It does not add a primary study, because there is none.
The practical consequence for a beat writing a bounded absence about a combination or a sequence - drug A after drug B, protein X in a Y-resistant model - is that the full-text hit list will be dominated by reviews and the count is close to meaningless. The screen has to be for study type before it is for topic. A PUB_TYPE facet exclusion would help here more than anywhere else in this tranche and was not attempted.
| script | none - Europe PMC BODY: queries, 2026-09-22 |
|---|---|
| status | run |
| verdict | SURVIVES at full text |
| entry re bounded | absence-ledgf-agent-never-tested-in-menin-inhibitor-resistant-disease |
| instrument before | abstract-level, bounded by a drug name list; 0 records |
| instrument after | full text, 20-name drug list; 116 records, 50 read |
| recall change | 0 -> 116, the largest in this tranche, and the verdict did not change |
| generalisable finding | For a bounded absence about a COMBINATION or a SEQUENCE, the full-text hit list is dominated by reviews that name both ideas without testing them together, and relevance ranking puts those reviews first. Screen for study type before topic. |
| absence class | never-measured |
| absence class basis | UNCHANGED. No LEDGF-directed agent has been tested in a resistant model, so nothing was measured; there is no archive in which such an experiment would be deposited. |
queries — the numbers behind the chart
[
{
"q": "LEDGF group AND 20-name menin-inhibitor list",
"hits": 116,
"read": 100,
"proceedings": 0
},
{
"q": "the same AND (resistance OR resistant)",
"hits": 90,
"read": 50,
"on_topic": 0
},
{
"q": "20-name menin-inhibitor list alone",
"hits": 1218,
"role": "operand control"
}
]nearest record
{
"pmid": "33477970",
"title": "Unlike Its Paralog LEDGF/p75, HRP-2 Is Dispensable for MLL-R Leukemogenesis but Important for Leukemic Cell Survival.",
"venue": "Cells",
"year": 2021,
"status": "TITLE-CHECKED, not opened",
"why": "LEDGF-family genetics in MLL-rearranged leukaemia with no menin-inhibitor arm"
}- If a LEDGF agent were developed for second line after menin-inhibitor failure, the entry's own structural argument cuts both ways: a MEN1 gatekeeper mutation that leaves the menin-KMT2A contact intact leaves the LEDGF grip intact too, so the resistant cell may be no more dependent on LEDGF than the naive one. The absence of a test is not neutral - it is the absence of the experiment that would decide whether the rationale is real.
- 40 of the 90 resistance-scoped records were not read, and relevance ranking puts primary studies last in a query dominated by reviews. Reading the tail is the unfinished half.
- Would a Europe PMC PUB_TYPE exclusion of reviews make combination-and-sequence absences tractable at full text? This is the query where it would matter most.
- Does PMID 33477970 report anything about menin-inhibitor sensitivity in its HRP-2 or LEDGF/p75 arms? It is the only primary LEDGF-family paper in the hit list.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database
The LEDGF alias group crossed with a 20-name menin-inhibitor list returns 116 full-text records and 90 with resistance terms, against 1,218 for the drug list alone; the 50 read are review articles with no primary test of a LEDGF-directed agent in a menin-inhibitor-resistant model, the nearest being PMID 33477970 on HRP-2 versus LEDGF/p75 in MLL-rearranged leukaemia.
Europe PMC RESTful search, BODY: LEDGF aliases x menin-inhibitor name list x resistance, 2026-09-22 · Europe PMC 2026What this evidence cannot show
Europe PMC BODY: reaches only the 12,362,145 full texts the index holds, so a paywalled paper remains invisible. A congress abstract book is one document holding thousands of unrelated abstracts and matches almost any conjunction; across this beat's 18 fully-read intersections 83 of 321 records (25.9%, range 0-50%) are proceedings. Screening was at TITLE level unless a paper is listed as opened. 50 of 90 were retrieved and screened; 40 were not. Zero of the 100 records read across the 116-record intersection are congress abstract books, which is the lowest proceedings rate in this beat and is consistent with the finding that specific compound names exclude them. Judging 50 records as reviews was done from titles and journal names.
Evidence that argues against this (1)
- 40 of the 90 were not read, and relevance ordering puts reviews first because they contain the most term hits - so a primary study would rank LOW in exactly this kind of query. The unread tail is therefore the part most likely to contain the refutation, which inverts the usual reassurance that the top of a hit list is the important part. Europe PMC 2026
SURVIVES: recall went 0 -> 116 at full text and the 50 read are all reviews naming both ideas without testing them together
the second instance in this beat of a drug-plus-disease intersection whose full-text hit list is made of reviews, at four times the scale