DEK::NUP214 is the sixth lesion on the XPO1 address - and the only one where the drug response was measured in primary patient cells
also called t(6;9) AML XPO1 dependence · DEK-NUP214 FOXC1 HOXA HOXB · DEK::NUP214 XPO1-dependent transcriptional activator
A rare AML caused by fusing DEK to the same nuclear-pore protein found in SET::NUP214 turns out to sit on the FOXC1 and HOX genes and to need the export receptor XPO1 to stay there. Among 57 patients' cells screened against 500 drugs, the ones with this fusion responded selectively to XPO1 blockers.
The same thing, for a scientist
**The result, in the systems it was measured in.** A multi-omics comparison of **57 primary AML samples** (whole-genome, targeted sequencing, transcriptomics and ex vivo screening of >500 compounds) found that **t(6;9)/DEK::NUP214** cases show a **selective response to selinexor and eltanexor** and a distinct transcriptomic signature of **FOXC1 and HOX** overexpression. **CUT&RUN** showed DEK::NUP214 binding directly to the promoters of **FOXC1 and the HOXA/HOXB clusters**, and both the expression of those genes and the fusion's binding to their regulatory regions were **selectively reduced by XPO1 inhibition in t(6;9) cells** [E1]. **Why this is the best evidence in the whole export axis.** Every other instance of the address is a cell line or a mouse. This one has **primary patient samples**, a **drug screen with a proper comparator population**, and a **CUT&RUN occupancy readout that moves in the right direction under the drug**. It is also the only instance where the fusion is shown to be removed from chromatin by XPO1 inhibition in the patient's own genotype. **And it adds a new output gene.** `FOXC1` is not in this graph anywhere. It is a forkhead transcription factor whose aberrant expression is an established co-dependency in HOXA-high AML, and here it is a **direct DEK::NUP214 target that falls under XPO1 inhibition**. If FOXC1 is part of the shared address's output, the axis is not purely a HOX story. **The lineage and age frame.** t(6;9) AML is a WHO-recognised entity of **younger adults**. This is not infant B-lineage KMT2A::AFF1, it is not lymphoid, and the abstract does not say how many of the 57 samples carried t(6;9) - in a rare entity that number could be very small.
Two NUP214 fusions, two pieces of the same argument
This graph already holds fusion-set-nup214 and its CRM1 story. DEK::NUP214 completes a symmetry that is worth stating explicitly:
| SET::NUP214 | DEK::NUP214 | |
|---|---|---|
| 5' partner | SET/TAF-I, no DNA-binding domain | DEK, a chromatin-binding nuclear factor |
| XPO1 contact | NUP214 FG repeats | NUP214 FG repeats |
| occupancy readout | ChIP-seq, cell lines (LOUCY, MEGAL) | CUT&RUN, t(6;9) cells |
| drug response | KPT-330 in cell lines | selinexor and eltanexor, primary samples, screen |
| outputs | HOXA, HOXB (HOXC in one title-level source) | FOXC1, HOXA, HOXB |
The interesting asymmetry is the 5' partner. SET has no DNA-binding domain, which is why mechanism-set-nup214-crm1-is-the-address-not-the-fusion concludes CRM1 supplies the address. DEK is a chromatin factor. So DEK::NUP214 has two plausible ways to reach a promoter, and the XPO1-inhibitor result says the NUP214-XPO1 half is nevertheless required. That is a stronger form of the address claim than SET::NUP214 can make, because here the fusion had an alternative and still needed XPO1.
Also relevant: localization-set-nup214-crm1-nuclear-bodies records that DEK-Nup214 forms the same CRM1-containing nuclear bodies as SET-Nup214 and tethers XPO1 onto them.
What it does and does not say about KMT2A-rearranged disease
Does not: contain a single KMT2A-rearranged sample, or any claim about them.
Does, indirectly: it is the strongest demonstration that an XPO1 inhibitor can strip a fusion off its target promoters in patient-derived cells, which is the mechanistic step every KMT2A-adjacent XPO1 argument needs and none of them has. If someone wants to argue XPO1 inhibition in KMT2A-rearranged leukaemia, this is the experiment to copy: CUT&RUN for the fusion at its target loci, plus and minus selinexor, in the genotype of interest.
It also sharpens the negative: the 57-sample screen had a comparator population of non-t(6;9) AML, which will have included KMT2A-rearranged cases, and the XPO1-inhibitor response was reported as selective for t(6;9). That is the closest thing this beat found to a direct comparison, and it points away from a KMT2A-rearranged XPO1 sensitivity - though the abstract gives no KMT2A-rearranged subgroup number, so it cannot be treated as a measurement of that.
| mechanism class | chromatin targeting by receptor capture |
|---|---|
| lesion | DEK::NUP214, t(6;9)(p22.3;q34.1), a WHO-recognised AML entity of younger adults |
| grip module | NUP214 FG repeats |
| receptor | gene-xpo1 |
| occupancy assay | CUT&RUN |
| drug effect | XPO1 inhibition selectively reduces both the expression of these genes and DEK::NUP214 binding at their regulatory regions in t(6;9) cells |
| drug screen | ex vivo, >500 compounds, 57 primary AML samples; selinexor and eltanexor selectively active in t(6;9) |
| new output gene for this graph | FOXC1 - not previously recorded anywhere in this graph |
| contrast with set nup214 | DEK is itself a chromatin-binding nuclear factor, so this fusion had an alternative route to a promoter and still required XPO1 - a stronger form of the address claim than SET::NUP214 can make |
| not present | no KMT2A-rearranged, lymphoid, paediatric or infant content |
loci bound
[ "FOXC1 promoter", "HOXA cluster", "HOXB cluster" ]
systems
[ "57 primary AML samples", "t(6;9) cells (line not stated in the abstract)" ]
- A selective ex vivo drug response in a rare entity within a 57-sample screen is a hypothesis-generating ranking, not a demonstrated therapeutic window, and XPO1 is common-essential so every sample would die at a high enough dose.
- If FOXC1 rather than HOX is the load-bearing output, then an intervention designed around HOX read-down would be monitoring the wrong gene.
- t(6;9) AML has a poor prognosis and is usually taken to transplant; an XPO1 inhibitor would have to fit into a schedule that does not compromise conditioning or engraftment, and no such study exists.
- Is FOXC1 also an output of the SET::NUP214 and NPM1c addresses? Nobody has looked, and if it is, the axis is a FOXC1/HOX programme rather than a HOX programme.
- How many of the 57 primary AML samples were t(6;9)? The strength of the primary-sample claim depends entirely on that number and the abstract does not give it.
- Did the screen's comparator population include KMT2A-rearranged AML, and how did those samples rank on selinexor? That subgroup ranking may exist in the paper's supplementary data and would be the closest thing to a direct answer for this graph's genotype.
- Would the same CUT&RUN-plus-selinexor design detect a KMT2A::AFF1 occupancy change? That is the experiment this beat proposes and it is copied directly from this paper.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In a multi-omics comparison of 57 primary AML samples including whole-genome sequencing, targeted sequencing, transcriptomics and ex vivo drug screening with more than 500 compounds, t(6;9)/DEK::NUP214 cases displayed a selective response to the XPO1 inhibitors selinexor and eltanexor and a transcriptomic signature of FOXC1 and HOX overexpression; CUT&RUN demonstrated direct binding of DEK::NUP214 to the promoters of FOXC1 and the HOXA/HOXB clusters; and both the expression of these genes and the binding of DEK::NUP214 to their regulatory regions were selectively reduced upon XPO1 inhibition in t(6;9) cells.
DEK::NUP214 acts as an XPO1-dependent transcriptional activator of essential leukemia genes. · Kaya F, Bewicke-Copley F, Miettinen JJ, et al. · Leukemia 2025What this evidence cannot show
Abstract-level for this beat - the CUT&RUN figures were not opened, and rule 80 applies to anything the abstract does not name. t(6;9) is a rare entity and the abstract does not state how many of the 57 samples carried it, so the primary-sample strength could rest on a handful of cases. Ex vivo drug screening measures viability in culture and is not a clinical response rate. Younger-adult AML; no paediatric, lymphoid or KMT2A-rearranged content. 'Selective response' is a between-genotype comparison within one screen, not a controlled experiment. SCOPE CORRECTED 2026-09-21. The abstract of PMID 40204893 was re-read in full. The 57 primary AML samples are the MULTI-OMICS AND DRUG-SCREEN arm only - whole genome sequencing, targeted sequencing, transcriptomics and >500-compound screening - and that arm is what shows the selective t(6;9) response to selinexor and eltanexor. The CUT&RUN is a SEPARATE experiment with no n stated in the abstract, and the key result that DEK::NUP214 binding falls under XPO1 inhibition is stated as occurring 'in t(6;9) cells', not in the 57 primary samples. Attributing the occupancy readout to 57 primary patient samples makes it a much stronger claim than the paper supports. The conflation entered via a hand-back summary and was then repeated in a dispatch brief before being caught by the beat that received it.
Evidence that argues against this (1)
- XPO1 is common-essential (DepMap mean gene effect -2.14 across 1,208 lines), so a 'selective response' in an ex vivo screen is a statement about degree, not about a private dependency; every sample in the screen would have died at a high enough concentration. The claim's strength comes from the CUT&RUN occupancy change, not from the viability ranking. Oncofusion Knowledge Graph project 2026
- E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
SET-Nup214 and DEK-Nup214 both interact with XPO1/CRM1 and with NXF1/TAP, decrease XPO1-mediated nuclear export of NES-containing proteins by tethering XPO1 onto the nuclear dots where the fusions localise, and inhibit NF-kB-mediated transcription by tethering a p65/IkB complex in the nucleus.
Leukemia-Associated Nup214 Fusion Proteins Disturb the XPO1-Mediated Nuclear-Cytoplasmic Transport Pathway and Thereby the NF-κB Signaling Pathway. · Saito S, et al. · Molecular and Cellular Biology 2016What this evidence cannot show
Abstract-level; recorded in full in `localization-set-nup214-crm1-nuclear-bodies`. Cited here to establish that DEK-Nup214 shares the XPO1-tethering behaviour of SET-Nup214, which is the prior that makes the 2025 CUT&RUN result expected rather than surprising. The system is an expression system rather than the endogenous fusion.
Evidence that argues against this (1)
- PUBLISHED CONFLICTING RESULT on the ORDERING, which is the whole content of the prior this item is cited to establish. The tethering account has the fusion capturing XPO1 and DECREASING export - the fusion is upstream and XPO1 is the captured cargo receptor. Conway et al. report the reverse for the closely related CALM-AF10: CRM1 localises to HOXA loci and RECRUITS the NES-containing fusion there, and disrupting the interaction prevents fusion enrichment at HOXA chromatin with immediate loss of transcription. Both cannot be the general rule, and this entry cites the tethering paper as the prior that makes the 2025 CUT&RUN 'expected rather than surprising' - under the tethering model, a fusion that sequesters XPO1 into nuclear dots is not obviously a fusion whose promoter occupancy should fall when XPO1 is inhibited. Conway AE et al. 2015
- E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
Targeted CRM1 inhibition disperses SET-NUP214 and DEK-NUP214 nuclear bodies and re-localises both fusions throughout the nucleoplasm, and cell viability, metabolism and proliferation of leukaemia cell lines harbouring these fusions are compromised by CRM1 inhibition.
Targeted CRM1-inhibition perturbs leukemogenic NUP214 fusion proteins and exerts anti-cancer effects in leukemia cell lines with NUP214 rearrangements. · Mendes A, Jühlen R, Martinelli V, et al. · Oncotarget 2020What this evidence cannot show
Abstract-level; cell lines only, no in vivo work, and viability readouts under a common-essential target's inhibitor cannot distinguish fusion-directed killing from general CRM1 dependence without a lineage-matched fusion-negative control. Recorded in full in `localization-set-nup214-crm1-nuclear-bodies`; cited here as the cell-line precedent for the 2025 primary-sample result.
Evidence that argues against this (1)
- NAMED METHODOLOGICAL DEFEATER, and the control is cheap and exists. Neggers et al. engineered XPO1 C528S into multiple leukaemia cell lines by CRISPR/Cas9 and showed that a single heterozygous substitution is sufficient and dominant for SINE resistance. That makes an isogenic C528S line the correct specificity control for any claim that a SINE's effect on a NUP214-fusion line runs through XPO1. Body dispersal plus loss of viability in a parental line under a common-essential target's inhibitor is compatible with fusion-directed killing, with general CRM1 dependence, and with an off-target liability of the compound; the caveat names the first two and the resistant allele separates all three. Neggers JE et al. 2016
DEK::NUP214 instance - the only one with primary samples
XPO1-dependent chromatin binding at FOXC1 and HOXA/HOXB
the FG-repeat half of the fusion
the other NUP214 fusion, same grip module, weaker occupancy evidence
- ← treated byEltanexor (KPT-8602) - the second-generation SINE whose whole point is that you can give it more often
selective ex vivo response of t(6;9) primary AML to selinexor and eltanexor
DEK-Nup214 forms the same CRM1-containing nuclear bodies
this entry supplies the DEK-side biology that mechanism presumes
DEK::NUP214 binds the FOXC1 promoter by CUT&RUN; binding and expression fall under XPO1 inhibition
the tethering model has the fusion capturing XPO1 and decreasing export; Conway et al. (PMID 25027513) have CRM1 pre-localised at HOXA and RECRUITING the NES-bearing fusion. Both cannot be the general rule, and this entry cites tethering as the prior that makes the 2025 CUT&RUN unsurprising