The lineage-switched AML state of infant KMT2A::AFF1, described by chromatin rather than by flow cytometry - fusion retained on PAX5 and MEIS1, lost from EBF1 and HOXA, myeloid promoters gained by DOT1L and ENL
After an infant's lymphoid leukaemia turned myeloid under treatment, the fusion protein stayed on some genes, left others, and the cells' switched-on genes changed. This is what that state looks like measured on the DNA rather than by surface markers.
The same thing, for a scientist
**What this state is.** `p318` in GSE252378: the AML into which the infant KMT2A::AFF1 B-ALL `p279` lineage-switched under ALL-directed chemotherapy, from the same patient [E2]. Chromatin was profiled by AutoCUT&RUN/CUT&Tag for KMT2A N and C termini, AFF1 (two antibodies), DOT1L, ENL, H3K4me3 and IgG [E1]. Every value below is a within-sample share - there is no spike-in - and n is one patient. **Fusion occupancy, retained:** PAX5 (N/C 8.47, the highest of the three states), MEIS1 (N 17.4x mean promoter), CDK6 (13.1x), FLT3 (14.5x), JMJD1C/KDM3C (16.4x, N/C 26.8), PROM1 (2.2x, N/C 4.01) [E1]. **Fusion occupancy, lost relative to its own B-ALL:** EBF1 (N/C 3.09 to 1.02), and the HOXA9/HOXA10 region, which falls to KMT2A-N 0.15-0.24x mean promoter with H3K4me3 at 0.04x - i.e. off, not merely unoccupied [E1]. **Promoters gained by the elongation machinery:** LYZ carries DOT1L at 65.5x and ENL at 16.3x its own mean promoter in this state, against DOT1L 0.0x and H3K4me3 0.09x in the B-ALL of the same patient. LYZ is **not** called a fusion target by the N/C criterion (N/C 1.59), so the myeloid gene gains elongation machinery without gaining a selective fusion signal [E1]. **Surface phenotype, as chromatin sees it:** the CD19 promoter loses H3K4me3 from 2.15x to 0.06x. CD19 was never a fusion target in either state [E1]. **How this state differs from its own B-ALL more than two B-ALLs differ from each other:** fusion-target promoter Jaccard with its own prior state is 0.117, against 0.149 between two unrelated infant B-ALLs [E1].
Why a chromatin description of this state is worth having
Lineage switch is currently defined in this corpus by immunophenotype and by clinical course. A chromatin description is a different and complementary definition, and it is the one a mechanistic or pharmacological argument needs - flow cytometry cannot say whether the fusion is still on PAX5. It also makes the state falsifiable in a new way: a second lineage-switched KMT2A::AFF1 patient either shows the same retained/lost split or it does not.
What is confounded and cannot be separated here
Therapy exposure, clonal selection and lineage identity all change between p279 and p318. The source paper's own model is that oncoprotein LEVELS fall in lineage-switching samples and that the redistribution follows from the level change, which would make part of this description a dose effect rather than a lineage effect. Nothing in this deposit separates the two.
| location | bone marrow, relapse after ALL-directed chemotherapy |
|---|
markers
[ "CD19 promoter H3K4me3 0.06x mean promoter (lost)", "LYZ promoter DOT1L 65.5x, ENL 16.3x (gained)", "PAX5 promoter N/C 8.47 (fusion retained)", "EBF1 promoter N/C 1.02 (fusion lost)", "HOXA9/HOXA10 KMT2A-N 0.15-0.24x, H3K4me3 0.04x (off)" ]
atlas refs
[ "GSE252378 GSM8000497-GSM8000501, GSM8371041-GSM8371048" ]
- A therapy chosen on the diagnostic B-ALL chromatin map would still hit PAX5, MEIS1, CDK6 and FLT3 in this state but would have been chosen partly for EBF1 and HOXA targets that are no longer there.
- The state is defined here on one patient; using it to stratify patients would be building a biomarker on an n of 1.
- What is the blast fraction of p279 and p318? Without it, every within-sample share is partly a purity statement.
- Do the KMT2A::ENL and KMT2A::EPS15 switched states in the same deposit share the retained/lost split seen here?
- E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset
In the lineage-switched AML state of one infant KMT2A::AFF1 patient, PAX5, MEIS1, CDK6, FLT3, KDM3C and PROM1 remain fusion-occupied while EBF1 and the HOXA9/HOXA10 region do not, LYZ carries DOT1L at 65.5x mean promoter without being a fusion target, and the CD19 promoter loses H3K4me3 to 0.06x.
GSE252378 - AutoCUT&RUN and AutoCUT&Tag profiles of 36 MLL-rearranged leukemia samples (171 GSMs, hg38) · Janssens DH, Henikoff S (Fred Hutchinson Cancer Center / Van Andel Institute) · NCBI Gene Expression Omnibus 2024What this evidence cannot show
One patient. Promoter windows and one 20 kb HOXA window; within-sample shares only. p279 DOT1L fails this beat's replicate guard (rho 0.357), so the LYZ DOT1L contrast rests on the p318 side of the comparison being sound and the p279 side being weak.
Evidence that argues against this (1)
- Blast purity is not reported for either sample. A myeloid relapse specimen with a different normal-cell admixture would shift every within-sample share, and LYZ in particular is expressed by contaminating normal monocytes - so the LYZ gain could be a purity artefact rather than a leukaemic state change. Janssens DH et al. 2024
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
p318 is the AML resulting from the lineage switch of p279, the same patient's infant B-ALL, under ALL-directed chemotherapy.
MLL oncoprotein levels influence leukemia lineage identities · Janssens DH, Duran M, Otto DJ, Wu W, Xu Y, Kirkey D, Mullighan CG, Yi JS, Meshinchi S, Sarthy JF, Ahmad K, Henikoff S · Nature Communications 15:9341 2024What this evidence cannot show
From the paper text and GEO descriptions; no clinical record consulted.
Evidence that argues against this (1)
- If the two samples are not the same patient, this entry describes a myeloid KMT2A::AFF1 leukaemia rather than a switched state, which is a different and much less interesting object. Janssens DH et al. 2024
the occupancy comparison that defines the state
the AML a KMT2A::AFF1 infant B-ALL became
- supports →Lineage switch in KMT2A-rearranged leukemia — the same clone changes what kind of blood cell it is
a chromatin-level description of the switched state
the retained/lost split
the untreated lineage-switched AML, re-measured from fragment counts
p318 is the chromatin-profiled post-switch AML state
54.6% of the B-ALL target promoters are still called in this AML state