MECOM (EVI1) - a cell-of-origin memory that MLL-AF9 maintains but cannot install, present in about 43% of MLL-rearranged AML and marking the chemoresistant half

This is a prediction, not a measurement. No one has seen this protein in this shape. The colours say how much the software trusts itself, and that is worth reading before anything else — the orange stretches are places the model is effectively saying this part has no fixed shape.

• EVI1 is a zinc-finger transcription factor with no ligandable pocket and a decade of failed attempts at PRDM-family chemistry. The only routes are protein-protein interfaces or degraders, and the degrader route is complicated by eight isoforms.
• EVI1 is essential for normal HSC regulation, so an EVI1-directed agent has a marrow-failure toxicity built into its mechanism - the same shape as the essential-function problem this graph records for menin.
• The therapeutic logic that works is INVERSE: knocking EVI1 down sensitises to chemotherapy [E1]. That means EVI1 status is a stratification biomarker before it is a target, and the immediate clinical action is to measure it, not to drug it.
• If EVI1 is a cell-of-origin memory the fusion cannot install, then an agent blocking the fusion does not remove EVI1 - and the chemoresistant, hypermethylated programme survives menin inhibition by construction. Nobody has tested this and it is the most consequential prediction on this entry.
• 43% is a cohort statistic from one series under one threshold. A stratification biomarker whose prevalence has been measured once is not ready to select patients.

1. Does the EVI1-to-ERG axis operate in the EVI1-positive 43% of MLL-rearranged AML? [E4] establishes it in 3q26-driven AML with an epistasis result; the transfer experiment has never been run, and if it transfers then ERG becomes a drug target for a defined subset of this disease rather than a bystander.
2. What is EVI1 status in KMT2A::AFF1 infant B-ALL? Every EVI1 datum in this disease is myeloid. If the cell of origin for infant t(4;11) is a fetal-liver HSC or LMPP rather than a committed pro-B cell - which this graph argues elsewhere - then the cell-of-origin logic predicts EVI1-positivity, and nobody has looked.
3. Is DepMap's KMT2A-rearranged arm stratifiable by EVI1 expression? The data exist publicly. A CRISPR effect computed within the EVI1-high half is the one number that would test E1 in human cells, and this beat did not have the budget to run it.
4. Does menin inhibition move EVI1? If EVI1 is a cell-of-origin memory rather than a fusion output, it should be MENIN-INDEPENDENT - and a menin inhibitor that collapses HOXA9/MEIS1 while leaving EVI1 standing would leave the chemoresistant programme intact, which is a testable resistance prediction.
5. Which isoform is measured in the MLL-rearranged cohorts - EVI1 or MDS1-EVI1? The papers say EVI1, and the PR/SET domain difference is functionally decisive, but no source read this cycle distinguishes them at the protein level in MLL-rearranged material.

1. E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

   In a large AML cohort carrying 11q23 translocations, about 43% of MLL-rearranged leukaemias are EVI1-positive, with high EVI1 occurring in AMLs expressing MLL-AF6, -AF9, -AF10, -ENL or -ELL; EVI1-positive MLL-rearranged AMLs differ molecularly, morphologically and immunophenotypically from EVI1-negative ones; in murine bone marrow, MLL-AF9 maintains Evi1 expression on transformation of Evi1-positive HSCs but does not activate Evi1 in MLL-AF9-transformed GMPs that were initially Evi1-negative; and shRNA knockdown of Evi1 in an Evi1-positive MLL-AF9 mouse model inhibits leukaemia growth in vitro and in vivo and increases sensitivity to chemotherapeutic agents.

   EVI1 is critical for the pathogenesis of a subset of MLL-AF9-rearranged AMLs. · Bindels EM, Havermans M, Lugthart S, Erpelinck C, Wocjtowicz E, Krivtsov AV, Rombouts E, Armstrong SA, et al. · Blood 119:5838-5849 2012
   PMID 22553314DOI
   ▸What this evidence cannot show

   Abstract fetched from PubMed efetch XML on 2026-09-20 and read in full; the body (PMC3382941) was not retrieved this cycle, so the cohort size behind the 43% figure, the EVI1-positivity threshold used and the effect sizes are NOT asserted here. RULE 73: the mechanistic arms are MURINE, MYELOID, MLL-AF9/HSC-versus-GMP; the human arm is an AML cohort with 11q23 translocations and no B-lineage or infant material. The 43% is a property of that cohort and of the EVI1-positivity cut-point the authors chose, both of which this entry has not read. Note the fusion list does NOT include MLL-AF4/KMT2A::AFF1.

   ▸Evidence that argues against this (1)

   • DepMap sees no MECOM dependency in any leukaemia arm, including the 11 KMT2A-rearranged lines (z +0.08 myeloid, -0.57 B-lineage, genome-scan rank 4,664 of 17,728). That is not a contradiction of this paper so much as a demonstration of the paper's own point: a dependency confined to the EVI1-positive subset is invisible to a screen that pools EVI1-positive and EVI1-negative lines, and no public stratification of the DepMap KMT2A-rearranged arm by EVI1 status exists. Broad Institute DepMap; Open Targets Platform; query run by this project 2026-09-20 2026
2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

   Evi-1 is frequently up-regulated in murine bone-marrow cells transformed by the MLL chimeric genes MLL-ENL or MLL-AF9; analysis of the Evi-1 promoter region showed that MLL-ENL activates transcription of Evi-1; the up-regulation occurs exclusively in the undifferentiated haematopoietic population, in which Evi-1 particularly contributes to the propagation of MLL-ENL-immortalised cells; and human AML cases with MLL rearrangement and high Evi-1 show a stem-cell-like gene-expression signature.

   Evi-1 is a transcriptional target of mixed-lineage leukemia oncoproteins in hematopoietic stem cells · Arai S, Yoshimi A, Shimabe M, Ichikawa M, Nakagawa M, Imai Y, Goyama S, Kurokawa M · Blood 117:6304-6314 2011
   PMID 21190993DOI
   ▸What this evidence cannot show

   Abstract fetched from PubMed efetch XML on 2026-09-20 and read in full; body not retrieved, so the promoter region tested and whether MLL-ENL occupancy was shown by ChIP (as opposed to reporter activation) are NOT asserted - the abstract says 'analysis of the Evi-1 promoter region revealed that MLL-ENL activates transcription', which is compatible with a reporter assay. MURINE, MYELOID, MLL-ENL and MLL-AF9. THIS SITS IN TENSION WITH E1, which finds MLL-AF9 unable to activate Evi1 in an Evi1-negative GMP; the reconciliation offered by both papers is that the activation is restricted to the undifferentiated compartment, but the two results are not identical and the entry does not pretend they are.

   ▸Evidence that argues against this (1)

   • E1 shows MLL-AF9 does NOT activate Evi1 in initially Evi1-negative GMPs, which is hard to square with a straightforward 'MLL fusion activates the Evi-1 promoter' reading of this paper. The compartment restriction both papers report is the likeliest reconciliation, but a promoter-reporter activation in a permissive cell and a failure to activate an endogenous silent locus in a committed progenitor are different experiments, and the graph should not merge them. Bindels EM et al. 2012
3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

   In murine single-cell transduction and transplantation, MLL-AF9-transduced HSC-derived clones were more likely than GMP-derived clones to initiate AML; leukaemia stem cells from both origins shared an immunophenotype but inherited gene-expression programmes from the cell of origin, including high-level Evi-1 expression only in HSC-derived leukaemia stem cells; the HSC-derived signature was enriched in poor-prognosis human MLL-rearranged AML in three independent data sets; global 5-methylcytosine was elevated in HSC-derived leukaemias, mirroring the difference between EVI1-positive and EVI1-negative human MLL-rearranged leukaemias; and HSC-derived leukaemias were more resistant to chemotherapy.

   Cell of origin determines clinically relevant subtypes of MLL-rearranged AML · Krivtsov AV, Figueroa ME, Sinha AU, Stubbs MC, Feng Z, Valk PJ, Delwel R, Döhner K, et al. · Leukemia 27:852-860 2013
   PMID 23235717DOI
   ▸What this evidence cannot show

   Abstract fetched from PubMed efetch XML on 2026-09-20 and read in full; body (PMC4693300) not retrieved, so the chemotherapy agent, the 5mC assay and the three validation data sets are NOT named here. MURINE MLL-AF9 MYELOID for the causal arms; the human arms are gene-expression and methylation comparisons in existing MLL-rearranged AML cohorts. This paper does not itself test whether Evi-1 is REQUIRED - it establishes that Evi-1 status is inherited from the cell of origin and that it tracks methylation and drug response. The requirement statement is E1's.

   ▸Evidence that argues against this (1)

   • An inherited gene-expression programme that correlates with chemoresistance need not be causal for it: HSC-derived and GMP-derived leukaemias differ in many genes and in global 5mC, and attributing the drug-response difference to Evi-1 specifically requires the knockdown, which is done in E1 in a different model and shows chemosensitisation. The two together make the case; neither does alone. Bindels EM et al. 2012
4. E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

   In experimentally tractable human and murine EVI1-driven AML models with genome-wide CRISPR screening, the ETS transcription factor ERG is a direct transcriptional target of EVI1 that is aberrantly expressed and selectively required in both human and murine EVI1-driven AML; EVI1 controls ERG expression and occupies a conserved intragenic enhancer region in AML cell lines and primary patient samples; suppression of ERG induces terminal differentiation of EVI1-driven AML cells, and ectopic expression of ERG abrogates their dependence on EVI1.

   EVI1 drives leukemogenesis through aberrant ERG activation · Schmoellerl J, Barbosa IAM, Minnich M, Andersch F, Smeenk L, Havermans M, Eder T, Neumann T, et al. · Blood 141:453-466 2023
   PMID 36095844DOI
   ▸What this evidence cannot show

   Abstract fetched from PubMed efetch XML on 2026-09-20 and read in full; body not retrieved, so the models' genotypes, the enhancer coordinates and the screen's hit list are NOT asserted. THE SYSTEM IS 3q26/EVI1-REARRANGED AML, NOT KMT2A-REARRANGED AML. Whether this axis operates in the EVI1-positive subset of MLL-rearranged AML is untested and is written here as an open question rather than as an inference.

   ▸Evidence that argues against this (1)

   • The epistasis result - ectopic ERG abrogating EVI1 dependence - is a strong claim about EVI1-driven AML and says nothing about a leukaemia driven by a KMT2A fusion where EVI1 is a retained cell-of-origin marker rather than the driver. In KMT2A-rearranged lines DepMap finds ERG is if anything LESS required than in other leukaemia lines (z +0.48 myeloid), so the axis does not visibly transfer. Broad Institute DepMap; Open Targets Platform; query run by this project 2026-09-20 2026
5. E5established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

   Fetched live on 2026-09-20, DepMap Chronos gene effect for MECOM has mean +0.0199 across 1,257 cell lines with 2.07% of lines at or below -0.5; among leukaemia lines the KMT2A-rearranged myeloid half sits at z +0.08 and the B-lineage half at z -0.57 relative to 44 other leukaemia lines, and this project's genome-wide scan ranks MECOM 4,664 of 17,728 on KMT2A-selectivity with a difference of -0.0365. MEN1 through the identical query shape returned z -2.41 / -2.43.

   DepMap CRISPR (Chronos) gene effect for MECOM (ENSG00000085276) and MEN1 (ENSG00000133895), retrieved via the Open Targets Platform GraphQL API, with the lineage split of analyses/lineage-confound-myeloid-tf-dependencies re-run for this gene · Broad Institute DepMap; Open Targets Platform; query run by this project 2026-09-20 · Database query 2026
   ▸What this evidence cannot show

   A bounded negative with a date, a route and a positive control through the identical query shape (rule 67). THE BOUND IS THE WHOLE POINT: the published dependency is confined to the EVI1-POSITIVE subset, which is about 43% of MLL-rearranged AML [E1], and none of the 11 KMT2A-rearranged DepMap lines is annotated for EVI1 status in this query. A screen that pools an EVI1-positive minority with an EVI1-negative majority returns the null it returned here, and that null is uninterpretable as evidence against E1. The only informative row is B-lineage z -0.57, driven largely by RS4;11 at -0.433 on four lines - one line, not a result. Ensembl id from HGNC primary (HGNC:3498); approvedSymbol read back as MECOM.

   ▸Evidence that argues against this (1)

   • MECOM is the clearest case on this beat where a clean DepMap null costs the reader nothing and tells them nothing. The measurement that would be informative - CRISPR effect stratified by EVI1 expression within the KMT2A-rearranged arm - is computable from public data and has not been run by anyone, including this project. Bindels EM et al. 2012

• ← regulatesKMT2A::MLLT3 (MLL-AF9), t(9;11)(p21.3;q23.3)

  MLL-AF9 MAINTAINS Evi1 in an Evi1-positive HSC and cannot activate it in an Evi1-negative GMP

• expressed in →Bone marrow haematopoietic stem cell — the postnatal comparator, and why the same fusion behaves differently here

  Evi-1 expression in MLL-AF9 leukaemia stem cells is inherited from an HSC cell of origin, not from a GMP

• regulates →ERG - a heptad stem-cell transcription factor whose 15 mentions in this graph are mostly about OTHER diseases, and whose best-established role in B-ALL is loss of function, not cooperation

  EVI1 occupies a conserved intragenic enhancer and drives ERG; ectopic ERG abrogates the EVI1 dependence

• co-occurs with →KMT2A-rearranged acute myeloid leukaemia in children and infants

  about 43% of MLL-rearranged AMLs are EVI1-positive, across MLL-AF6/AF9/AF10/ENL/ELL

• supports →A KMT2A fusion is necessary but not sufficient: the cell it lands in decides whether there is a disease and which one

  the fusion keeps Evi1 on where the founding cell had it on and cannot switch it on otherwise

• related to →The KMT2A-r DNA methylation signature — focal hypermethylation, no global hypomethylation, and a hypomethylating agent that did not help

  EVI1-positive MLL-rearranged leukaemias carry globally elevated 5mC, mirroring HSC-derived versus GMP-derived murine leukaemias

• regulates →KMT2A::MLLT1 (MLL-ENL), t(11;19)(q23;p13.3)

  MLL-ENL activates transcription from the Evi-1 promoter region, exclusively in the undifferentiated haematopoietic population

• ← regulatesGATA2 - the least KMT2A-selective gene on this beat, ranking 17,591 of 17,728, and a factor whose leukaemia biology is about stemness and chemoresistance rather than about the fusion

  inv(3)/t(3;3) repositions a distal GATA2 enhancer onto EVI1, activating EVI1 and leaving GATA2 haploinsufficient