Oncofusion

start / graph / ana-unstated-system-leptin-lepr-adipocyte-resolved

Analysismoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q18b-unstated-systems

A partial resolution: the leptin-LEPR chemoresistance model is murine MLL-AF9, but the two AML arms were never separated and the proteomics deposition is unreachable

also called PXD066693 provenance · leptin LEPR AML arms · MLL-AF9 versus AML1-ETO9a

An entry on marrow fat cells notes that a paper mixes two different mouse leukaemia models and never separates their results. Reading the full text confirms the leukaemia model is MLL-AF9 and that the arms stay mixed; the study's protein dataset could not be retrieved at all, so the separation the entry needs is still missing.

The same thing, for a scientist

`cell-type-bone-marrow-adipocyte` carries the caveat *"MLL-AF9 is one of two mouse models and the abstract does not separate the KMT2A-rearranged results from the AML1-ETO9a ones, so the KMT2A-specific effect size is not established"* [E1]. This is a **partial resolution and the beat's clearest example of a caveat that a deposition cannot fix**. What resolved: the open-access full text names MLL-AF9 18 times against C57BL/6 3 times and human U937, HL-60 and THP-1 3 times each, confirming a murine MLL-AF9 primary model with human line corroboration [E2]. What did not: the paper's data-availability statement lists **GSE1159, GSE37642 and GSE6891** - three long-standing **public patient expression cohorts** that the authors re-analysed rather than generated - and **PXD066693**, a PRIDE proteomics accession that returns HTTP 404 from both the v2 and v3 PRIDE archive APIs, and NCBI elink returns no GEO, SRA, BioSample or BioProject record for the PMID [E3]. So there is **no deposition of this study's own mouse experiments at all**, and the per-arm separation the entry asks for cannot be recovered from any repository. The distinction is worth naming precisely because it is easy to mistake a re-analysis accession for a deposition. GSE1159, GSE37642 and GSE6891 are other people's AML cohorts. Listing them satisfies a journal's data-availability requirement and tells a provenance reader nothing about the paper's own systems.

Go deeper
Three kinds of accession in one data-availability statement
AccessionWhat it actually isUse for provenance
GSE1159, GSE37642, GSE6891pre-existing public human AML expression cohorts, re-analysednone - not this study's samples
PXD066693PRIDE proteomics, presumably this study's ownunreachable - HTTP 404 from PRIDE v2 and v3 APIs on 2026-09-19
(mouse MLL-AF9 and AML1-ETO9a experiments)the arms the entry needs separatednot deposited

This is a failure mode the project should expect again. A data-availability statement is a list of accessions, not a list of the study's depositions, and re-analysis accessions look identical to primary ones in a regular-expression sweep. Any automated provenance pipeline that harvests accessions from full text will pick up all three human cohorts here and conclude, wrongly, that this is a large human study.

What the entry should say now

cell-type-bone-marrow-adipocyte should keep its caveat and sharpen it: the KMT2A-rearranged model is murine MLL-AF9, the comparator model is AML1-ETO9a, the two are not separated in the abstract, and no deposited data exists from which a reader could separate them - the study's only primary accession is a PRIDE record that does not resolve. That converts an open question into a bounded one, which is the most this route can do here.

Facts and figures
scriptnone - Europe PMC full text, PRIDE REST API, NCBI elink
statusrun
result summaryPartial. Leukaemia model resolved to murine MLL-AF9 with human U937/HL-60/THP-1 corroboration. The two AML arms remain unseparated and cannot be separated from any repository: the listed GEO accessions are re-analysed public patient cohorts, the PRIDE accession returns 404, and no primary deposition of the study's own experiments exists.
resolution classpartial
counts
accessions_retrievable0.000accessions_retrievable: 0kmt2a_aff1_systems0.000kmt2a_aff1_systems: 0accessions_that_are_this_…1.000accessions_that_are_this_study: 1accessions_listed4.000accessions_listed: 4mll_af9_mentions18.0mll_af9_mentions: 18fulltext_chars63,475fulltext_chars: 63475
inputs
[
  "PMID 41730833",
  "PMC12966444",
  "PXD066693",
  "GSE1159",
  "GSE37642",
  "GSE6891"
]
counts — the numbers behind the chart
{
  "mll_af9_mentions": 18,
  "fulltext_chars": 63475,
  "accessions_listed": 4,
  "accessions_that_are_this_study": 1,
  "accessions_retrievable": 0,
  "kmt2a_aff1_systems": 0
}
entries affected
[
  "cell-type-bone-marrow-adipocyte"
]
import axes
[
  "fusion: KMT2A::MLLT3",
  "species: mouse",
  "lineage: myeloid"
]
How this could bite us back
  • A metabolic intervention against leptin-LEPR signalling would act on marrow adipocytes and on normal haematopoietic metabolism, and a pooled two-model mouse experiment with no deposited data gives no basis for estimating that cost.
  • If the chemoresistance effect is driven mainly by the AML1-ETO9a arm, then an entry importing it into KMT2A biology has imported a result that may not exist in the index genotype at all - and nothing public can currently rule that out.
  • Marrow adiposity rises sharply with age and is low in infants, so a niche mechanism established in adult mice may be close to absent in the population this project treats.
Still open
  1. What is the leptin-LEPR effect size in the MLL-AF9 arm alone? The study's own data are not deposited and the arms are pooled in the text, so this is answerable only by the authors.
  2. Is PXD066693 embargoed rather than absent? A retry in a later cycle would distinguish the two and is the cheapest open item in this beat.
Every source behind this entry (3)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    The marrow leptin-LEPR paper reports that leptin-LEPR signalling rewires mitochondrial oxidative metabolism to confer chemoresistance in acute myeloid leukaemia, and its abstract does not separate the MLL-AF9 results from the AML1-ETO9a ones.

    Marrow leptin-LEPR signaling rewires mitochondrial oxidative metabolism to confer chemoresistance in acute myeloid leukemia. · Liao X, Dai W, Xu X, Cai D, Tan M, Wang Z, Huang Y, Hou D, Liu J, Wang L, Wang J, Wang X, Zhang S, Lin X, Huang H · Cell Death Dis 2026
    PMID 41730833DOI
    What this evidence cannot show

    Identifiers taken from the NCBI esummary record fetched on 2026-09-19. Full text read from Europe PMC PMC12966444 (open access) on the same date.

    Evidence that argues against this (1)
    • A metabolic chemoresistance mechanism operating through a marrow-adipocyte-derived ligand is plausibly independent of the driving fusion, in which case pooling two AML models is a strength rather than a confound - the objection is that the paper does not let a reader check. Viñado AC et al. 2022
  2. E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    The paper's full text names MLL-AF9 18 times against 3 mentions each of C57BL/6, U937, HL-60 and THP-1 and 2 of HEK293T, indicating a murine MLL-AF9 primary model with human myeloid line corroboration.

    Marrow leptin-LEPR signaling rewires mitochondrial oxidative metabolism to confer chemoresistance in acute myeloid leukemia. · Liao X, Dai W, Xu X, Cai D, Tan M, Wang Z, Huang Y, Hou D, Liu J, Wang L, Wang J, Wang X, Zhang S, Lin X, Huang H · Cell Death Dis 2026
    PMID 41730833DOI
    What this evidence cannot show

    Counted by regular-expression tally over 63,475 characters of flattened Europe PMC full text on 2026-09-19. A mention count is not an experiment inventory and is graded moderate accordingly. Notably the tally returned no occurrences of an AML1-ETO or RUNX1-RUNX1T1 pattern under the expression used, so the second model's prominence in the text was not measured here and no claim is made about it.

    Evidence that argues against this (1)
    • None of the three human lines named is KMT2A::AFF1 or B-lymphoid, so even the human corroboration arm imports away from the index disease on both fusion and lineage. Dietsche L et al. 2024
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database

    The paper deposited no retrievable primary dataset: its data-availability statement lists GSE1159, GSE37642 and GSE6891, which are pre-existing public human AML expression cohorts, plus PRIDE accession PXD066693, which returns HTTP 404 from both the v2 and v3 PRIDE archive REST endpoints; NCBI elink returns no GEO, SRA, BioSample or BioProject record for this PMID.

    What this evidence cannot show

    Queried on 2026-09-19 from a single network location. An HTTP 404 from PRIDE may mean the accession is embargoed pending publication rather than absent, which for a 2026 paper is a live possibility; it is therefore an unreachable rather than a non-existent deposition, and should be retried. The identification of GSE1159, GSE37642 and GSE6891 as pre-existing cohorts rests on their accession ranges and on their appearing in a re-analysis context in the text, not on separately fetching each series.

    Evidence that argues against this (1)
    • A 404 on a very recent accession is the same signature as an embargo, and this project has twice been wrong about a clean negative, so this bounded failure is one that should be re-run in a later cycle rather than treated as settled. Rogic S et al. 2026
Connected to