Oncofusion

start / graph / mechanism-nfkb-lineage-split-myeloid-story-lymphoid-silence

Mechanismstrong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.Q38-nfkb-signalling

The NF-kB story in KMT2A-rearranged leukaemia is a myeloid story, and the lymphoid half is not a weaker version of it — it is a different subject

also called NF-kB lineage split · myeloid-only NF-kB evidence · the B-ALL silence

Almost everything known about NF-kB in MLL leukaemia was done in myeloid cells and mice. The infant disease this project exists for is a B-cell leukaemia, and the handful of B-cell papers are about something else entirely — how the cells avoid being killed by an inflammatory signal, not how they depend on one.

The same thing, for a scientist

This graph has a standing finding that the differentiation narrative in KMT2A-rearranged leukaemia is an AML story that should not be inherited by infant ALL entries. **The NF-kB story has exactly the same shape and it is worth stating explicitly before any entry in this folder is read across.** **The myeloid column.** Every functional NF-kB experiment in KMT2A-rearranged leukaemia is myeloid: the 2013 screen used MLL-oncogene-transformed *myeloid* progenitors; the transformation assays used MLL-AF6 and MLL-AF9; the Rela-/- fetal liver assay used *myeloid* progenitors; the in vivo work used MLL-AF10 and MLL-AF9 AML; the human lines were MV4;11 and ML-2; the occupancy experiment was mouse MLL-AF10 AML [E1]. The NIK work is mouse MLL-AF9 AML throughout [E2], and the TIFAB work is MLL-AF9 and KMT2A::MLLT3 AML. **The lymphoid column, and what is in it.** Three things, none of them about pathway dependence. (i) **NF-kB activation mediates acquired resistance to interferon-beta in a murine t(4;11) ALL model**, and NF-kB inhibition resensitises [E3]. (ii) **KMT2A::AFF1 B-ALL blasts are resistant to TNF-alpha via S100A6**, a graft-versus-leukaemia evasion mechanism (`gene-tnf`). (iii) In the founding myeloid paper, the one comparator that a *lymphoid* oncogene provided was **E2A-PBX1, which was insensitive** to IKK inhibition and to Ikk knockdown [E1] — which is evidence that the requirement is fusion-specific, and is not evidence that it holds in a KMT2A-rearranged lymphoid cell. **The counted absence.** `(KMT2A OR MLL) AND infant[tiab] AND (NF-kappaB OR NF-kB OR NFKB)[tiab]` returns **0 records** [E4]. **So the honest position is: in infant KMT2A::AFF1 B-ALL, the IKK/NF-kB retention mechanism is untested — neither supported nor refuted.**

Go deeper
Why the myeloid result cannot simply be read across

Three specific reasons, all of them from the same papers.

  1. The named upstream is myeloid-flavoured. The pathway implicated by the screen is TLR/IRAK3/p105, and the transcript enrichment was significant in MLL-AF1p and MLL-AF10 and failed to reach significance in MLL-AF9 even within the myeloid set 1. A pathway whose evidence is unstable across myeloid fusions is not a safe bet across lineages.

  2. The phenotype is myeloid. The differentiation read-out was Mac-1 and Gr-1 with morphological myeloid maturation 1; there is no lymphoid equivalent in the paper and no reason to expect differentiation reversal to be the phenotype in a pro-B blast.

  3. The arms may invert. Non-canonical NF-kB is tumour-suppressive in mouse MLL-AF9 AML 2 and NIK is oncogenic in mature B-cell malignancies. If the sign of the non-canonical arm flips with lineage, an entry that carried the AML conclusion into infant ALL would have the therapeutic direction backwards.
The one disease-matched lymphoid NF-kB result, and its exact scope

In a murine model of t(4;11) ALL treated with continuous AAV-delivered interferon beta, complete remissions were achieved, resistance then emerged, NF-kB activation was identified as the resistance mechanism, and NF-kB inhibition resensitised the resistant cells 3. That is a genuine t(4;11) result and it is about acquired resistance to a specific therapy, not about baseline dependence of the fusion programme. Cited for what it is, it supports the idea that this axis is live in t(4;11) lymphoid cells. Cited as support for the retention mechanism, it would be a misattribution — the paper measures leukaemia survival under IFN-beta and contains no chromatin measurement.

What would close the gap, at low cost

One experiment in SEM or RS4;11 (KMT2A::AFF1 B-ALL lines): IKK inhibitor or NEMO disruption, then ChIP-qPCR for KMT2A::AFF1 at HOXA9 and MEIS1, plus wild-type KMT2A with a C-terminal antibody, plus H3K79me2 and H3K4me3, with total fusion protein by western as the stability control. That is the 2013 figure 8 run in the index lineage, and it is the single highest-value missing measurement this beat found.

Facts and figures
mechanism classevidence-base-limitation
index disease statusuntested - neither supported nor refuted in infant KMT2A::AFF1 B-ALL
missing measurementIKK inhibition plus KMT2A::AFF1 ChIP at HOXA9/MEIS1 in a B-ALL line
myeloid systems
[
  "MLL-AF1p",
  "MLL-AF6",
  "MLL-AF9",
  "MLL-AF10",
  "MLL-ENL",
  "KMT2A::MLLT3",
  "MV4;11",
  "ML-2"
]
lymphoid systems in nfkb literature
[
  "murine t(4;11) ALL under IFN-beta (resistance only)",
  "KMT2A::AFF1 ALL lines and transgenic mice (TNF resistance only)"
]
insensitive comparators
[
  "E2A-PBX1",
  "E2A-HLF",
  "NUP98-HOXA9",
  "K562"
]
How this could bite us back
  • The whole therapeutic case in this folder rests on myeloid experiments. Proposing an IKK-directed trial in infant B-ALL on this evidence would be proposing a trial in a lineage where the target's requirement has never been demonstrated.
  • If the sign of the non-canonical arm inverts with lineage, a therapy that helps in KMT2A-rearranged AML could accelerate KMT2A-rearranged B-ALL.
  • The one t(4;11) result is a resistance mechanism under a therapy nobody uses. Building on it means building on a single, non-translated model.
Still open
  1. Is IKK/NF-kB required for KMT2A::AFF1 retention on chromatin in a B-lymphoid cell? This is the one experiment that would make this folder apply to the index disease.
  2. E2A-PBX1 and E2A-HLF transduced progenitors were insensitive to IKK inhibition. Is that because the requirement is MLL-specific, or because the assay was myeloid and those are lymphoid oncogenes? The paper reads it the first way and the design cannot distinguish them.
  3. Does the non-canonical arm invert its sign in B lineage, as the mature B-cell malignancy genetics would predict?
  4. Are there published RNA-seq datasets from infant KMT2A::AFF1 blasts in which an NF-kB signature could be scored without a new experiment? That would be the cheapest possible partial answer.
Every source behind this entry (4)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    The evidence base is myeloid throughout: the functional screen used MLL-oncogene-transformed myeloid progenitors; clonogenicity assays used MLL-AF6 and MLL-AF9 transduced myeloid progenitors with E2A-PBX1 and E2A-HLF as insensitive comparators; the Rela-/- assay used fetal liver myeloid progenitors transduced with MLL-AF9 or MLL-ENL, where Rela-/- cells could not sustain colonies past the second plating while non-MLL oncogenes replated normally; in vivo work used MLL-AF10 and MLL-AF9 AML and MLL-AF6-transduced cells; human lines were MV4;11 and ML-2 with K562 as a non-MLL comparator; the differentiation read-out was Mac-1 and Gr-1 with myeloid morphological maturation; and the occupancy experiment was in mouse MLL-AF10 AML cells. Toll-like receptor pathway transcript enrichment was significant in MLL-AF1p and MLL-AF10 leukaemia cells but did not achieve statistical significance in MLL-AF9 cells.

    Epigenetic roles of MLL oncoproteins are dependent on NF-κB. · Kuo HP, Wang Z, Lee DF, Iwasaki M, Duque-Afonso J, Wong SHK, Lin CH, Figueroa ME, Su J, Lemischka IR, Cleary ML · Cancer Cell 2013
    PMID 24054986DOI
    What this evidence cannot show

    MV4;11 carries t(4;11)/KMT2A::AFF1 and is therefore the one cell line in the paper that shares the index fusion - but it is a biphenotypic/AML-classified line, not a B-ALL line, and it was used for growth and phosphoflow, never for the chromatin experiment. Reading MV4;11 as 'the infant disease was covered' would be wrong. Full text read this session (PMC3816582).

    Evidence that argues against this (1)
    • A B-lineage, KMT2A::AFF1-specific literature touching this axis does exist and is not empty - amlexanox downregulates S100A6, enhances tumour immunity and lowers leukaemia penetrance in KMT2A/AFF1 transgenic mice and extends survival in a human KMT2A/AFF1 xenograft. So the claim here is not that nothing lymphoid has been done, but that nothing lymphoid tests pathway dependence or fusion retention. Tamai H et al. 2017
  2. E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    The non-canonical NF-kB work in this disease is also entirely myeloid: NIK stabilisation was tested in MLL-AF9-transduced HSPCs and MLL-AF9 AML mice, with LSCs defined by the MLL-AF9 immunophenotype lin-/c-kit+/Sca1-/CD16+/CD34+, and the authors frame their proposal as developing small molecules to stabilise NIK 'specifically in myeloid leukemias'.

    Stabilization of NF-κB-Inducing Kinase Suppresses MLL-AF9-Induced Acute Myeloid Leukemia. · Xiu Y, Dong Q, Li Q, Li F, Borcherding N, Zhang W, Boyce B, Xue HH, Zhao C · Cell Reports 2018
    PMID 29320732DOI
    What this evidence cannot show

    The authors' own lineage restriction is explicit in their concluding sentence, which is unusually helpful and should be taken at face value rather than generalised past. Full text read this session (PMC5810947).

    Evidence that argues against this (1)
    • Recording that the non-canonical work is entirely myeloid is correct and useful, but it makes the lineage split a statement about where people have LOOKED rather than about the biology. The lymphoid silence is an absence of studies, and this entry should not be read as evidence that the axis is inactive in B-lineage disease. Tracey L et al. 2010
  3. E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    In a murine model of t(4;11) acute lymphoblastic leukaemia, continuous AAV-mediated delivery of interferon beta produced complete remission; leukaemic cells eventually became resistant and relapsed; activation of NF-kappaB was identified as the mechanism of that resistance; and inhibition of NF-kappaB activity in resistant cells resensitised them to interferon beta.

    PMID 20130599DOI
    What this evidence cannot show

    Abstract fetched and read this session (not copied from a sibling entry). A murine xenograft-style model with a gene-therapy intervention that is not in clinical use; the NF-kB inhibition is not specified by agent or arm in the abstract. It is a resistance result under a specific therapy, not a measurement of baseline pathway dependence, and it contains no chromatin data.

    Evidence that argues against this (1)
    • This is the one lymphoid datapoint in the beat and it is a murine t(4;11) model under a non-clinical continuous-interferon exposure, so it carries more weight in the argument than its design supports. A single mouse study is a thin basis for asserting that the lymphoid side behaves differently rather than simply being unstudied. Xiu Y et al. 2018
  4. E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database

    Counted absence. (KMT2A OR MLL) AND "infant"[tiab] AND ("NF-kappaB"[tiab] OR "NF-kB"[tiab] OR NFKB[tiab]) returns 0 PubMed records. Separately, ("KMT2A"[tiab] OR "MLL-AF4"[tiab] OR "t(4;11)"[tiab]) AND ("NF-kappaB"[tiab] OR "NF-kB"[tiab]) returns 12 records, of which title-checking leaves three on topic: the two IKKb/NR4A1 PROTAC versions (AML), the TIFAB/KMT2A::MLLT3 paper (AML), and the t(4;11) interferon-beta resistance paper. No paper in that set measures NF-kB dependence in a KMT2A-rearranged B-ALL.

    What this evidence cannot show

    querytranslation read on both queries; no [UID] artefact. Note that the t(4;11) token was written WITHOUT other parentheses in the query, per the brief's rule about parenthesised karyotype strings - PubMed rendered it as the phrase 't 4 11'. 'Infant' in title/abstract is a proxy for the disease and would miss a KMT2A::AFF1 B-ALL paper that never uses the word, so the zero bounds explicitly infant-framed work rather than all lymphoid work.

    Evidence that argues against this (1)
    • A zero at the symbol level is not a zero at the pathway level, and on this beat that is demonstrable rather than hypothetical: PMID 29320732 studies the non-canonical arm directly in MLL-AF9 leukaemia and is unreachable from any of these subunit symbol queries, because it names the kinase and the pathway rather than the subunit. The count is honest about what was searched and should not be read as evidence that the biology is unstudied - only that it is not indexed under this symbol. Xiu Y et al. 2018
Connected to