Oncofusion

start / graph / exp-q147-paired-irx1-rna-and-oncoprotein-occupancy-in-non-aff1-kmt2a-r-blasts

Experimentmoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q147-hoxa-irx1-across-partners

Paired IRX1 RNA and KMT2A-N occupancy in the same non-AFF1 blasts: does the expression subgroup exist where the occupancy subgroup does not?

We know the fusion protein does not sit on IRX1 outside one fusion type. We do not know whether IRX1 is switched on there anyway. Measuring both in the same cells answers it in one go.

The same thing, for a scientist

**The gap.** GSE252378 has no RNA-seq, so no sample in it can be assigned to the published iALL-HOXA9 / iALL-IRX1 subgroup by the published criterion, which is a DESeq2 expression contrast. This beat's 0-of-18 is an occupancy negative and cannot distinguish *IRX1 is off* from *IRX1 is on and the fusion is not driving it* - and the second is the more interesting state, because IRX1 expression is separately reported in AML subsets [E1]. **Material.** 12-20 KMT2A-rearranged blasts spanning ENL, AF9, AFDN, AF10 and EPS15, with at least 3 per partner where possible, plus 6 KMT2A::AFF1 as the positive arm. Primary or PDX; the same nuclei preparation split for both assays so the pairing is real. **Assays on the same material.** (a) RNA-seq or IRX1 + HOXA9 + HOXA10 + MEIS1 RT-qPCR; (b) AutoCUT&RUN for KMT2A-N1/N2, KMT2A-C1/C2, H3K4me3 and IgG. **The 2x2 that is the whole result:** | | fusion at IRX1 | no fusion at IRX1 | |---|---|---| | IRX1 RNA high | the AFF1 state, seen outside AFF1 => **the axis is general** | **IRX1 is fusion-independent** => the expression subgroup and the occupancy subgroup are different objects | | IRX1 RNA low | fusion at a silent promoter (not observed anywhere yet) | IRX1 simply off => this beat's simplest reading | **Pre-declared thresholds.** IRX1 RNA high = above the 75th percentile of IRX1 across the cohort AND above the cohort's own negative-control gene set; occupancy as in this beat, >= 2.0x in both N antibodies and above the width-matched floor. **Declare both thresholds before unblinding either assay**, because the interesting cells of the 2x2 are the off-diagonal ones and a floating threshold can move a sample into either.

Go deeper
Why the negative-control arm has to be in the RNA assay too

An expression 'high' call has the same defect as a rank correlation: it is relative. Symeonidou's split is a clustering of two cohorts, so 'IRX1-high' means high within that cohort. A 12-sample cohort of AML-weighted KMT2A-r material will produce an IRX1-high quartile whatever the absolute levels are. So the RNA arm needs an absolute anchor too - the same eight tissue-restricted genes used as the chromatin floor work as an RNA floor, and a normal pro-B reference is the biologically meaningful comparator given Nagel's placement of IRX1 in normal pro-B cells.

Pre-declare the controls this beat used and do not substitute IgG for them: MEIS1 and CDK6 as within-sample positives in KMT2A-N (PAX5 only in B-lineage material); the eight-locus width-matched cannot-be-occupied panel (ALB, RHO, MYH7, MYOD1, NEUROD1, KRT14, GFAP, INS) as the per-library floor; and a depth report as the promoter double-zero probability p_dz with a stated cut. GSE252378's IgG libraries are 57-94% zero at promoters and reach 13.1x an average promoter at a locus that cannot be occupied, so IgG alone will not do.

Facts and figures
designpaired transcript and chromatin measurement on the same nuclei preparation
cohort12-20 non-AFF1 KMT2A-r blasts across ENL, AF9, AFDN, AF10, EPS15, plus 6 KMT2A::AFF1 as the positive arm
rna thresholdIRX1 above the 75th cohort percentile AND above an absolute tissue-restricted-gene floor
occupancy threshold>= 2.0x in both KMT2A-N antibodies and above the width-matched floor
key cellIRX1 RNA high with no fusion at IRX1 => the expression subgroup and the occupancy subgroup are different objects
assays
[
  "RNA-seq or IRX1/HOXA9/HOXA10/MEIS1 RT-qPCR",
  "AutoCUT&RUN KMT2A-N1/N2, KMT2A-C1/C2, H3K4me3, IgG"
]
tests hypotheses
[
  "hyp-q147-the-irx1-address-requires-the-aff1-partner-not-the-menin-interface"
]
analyses
[
  "ana-q147-irx1-fusion-occupancy-is-confined-to-kmt2a-aff1",
  "ana-q147-irx1-expression-is-reported-in-aml-while-irx1-fusion-occupancy-is-not"
]
How this could bite us back
  • If IRX1 turns out to be fusion-independent in non-AFF1 KMT2A-r, then menin inhibition will not touch it there and an IRX1-high patient on a menin inhibitor has an untreated programme by construction. That is a testable prediction about who relapses, and it is the clinically load-bearing cell of the 2x2.
Still open
  1. Is there an existing KMT2A-r cohort with paired RNA and chromatin on the same aliquots? TARGET and St. Jude have RNA at scale, but the chromatin and the RNA are usually from different aliquots, which breaks the pairing this design needs.
Every source behind this entry (1)
  1. E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset

    GSE252378 contains no RNA-seq for any of its 27 KMT2A-fusion samples, so none can be assigned to the published iALL-HOXA9 / iALL-IRX1 expression subgroup by the published Kallisto-tximport-DESeq2 criterion; and IRX1 expression is separately reported as aberrantly active in AML subsets, which 11 of the 27 samples are.

    What this evidence cannot show

    This is a statement about what the deposit contains and about the published record, not a new measurement. The published subgroup criterion is itself a within-cohort clustering, so even with RNA in hand the assignment would be cohort-relative unless an absolute anchor is added - which is why the design specifies one. Every number is a WITHIN-SAMPLE share: an interval's mean coverage divided by that same library's mean coverage over all 19,404 hg38 RefSeq promoters, so 1.0 = an average promoter in that library. GSE252378 has no spike-in, so no between-library absolute statement is licensed; the only cross-sample quantity used is a ratio of two loci measured inside one library, where the unknown per-library scale cancels.

    Evidence that argues against this (1)
    • Symeonidou & Ottersbach's two subgroups were derived in infant MLL-AF4 ALL specifically; applying their labels to KMT2A-AF9 or -AFDN AML is an extension the authors do not make, and the subgroup may simply not be defined outside t(4;11). Symeonidou V et al. 2021
Connected to