Targeted therapies reshape extracellular matrix remodeling and microenvironmental regulation in pediatric acute myeloid leukemia

TL;DR

This study explores how targeted therapies affect the extracellular matrix and gene activity in pediatric acute myeloid leukemia, revealing new insights into treatment response.

Contribution

The study identifies a shared transcription factor-driven regulatory axis linking ECM remodeling and therapy response in pediatric AML.

Findings

Targeted therapies alter ECM-related genes and pathways, including upregulation of MMP10 and ABCB5.

PRMT5 inhibition and NID1 knockdown influence RNA processing and ECM genes, respectively.

A shared regulatory axis involving NF-κB, CEBPB, SP1, and STAT3 connects different therapeutic responses.

Abstract

Pediatric acute myeloid leukemia (AML) is a biologically distinct and aggressive malignancy with limited effective therapies. While emerging targeted agents, including menin, FLT3, and PRMT5 inhibitors, show promise, the transcriptomic effects and the role of extracellular matrix (ECM) regulators, such as nidogen-1 (NID1), in modulating therapeutic responses remain unclear.

We performed integrative transcriptome analyses of pediatric AML datasets (TARGET-AML, GSE246783: MI3454; GSE292324: PRMT5 inhibition; GSE292050: NID1 knockdown) and incorporated adult AML data from TCGA and GTEx. Differential expression was evaluated using DESeq2 and limma, functional enrichment using ClusterProfiler, Metascape, and GSEA, and protein–protein interaction networks using STRING, Cytoscape, NetworkAnalyst, and GeneMANIA. Transcription factor (TF) and upstream regulator inference were conducted using DoRothEA/decoupleR, Enrichr, and TRRUST v2.

MI3454 treatment altered 166 genes, including upregulation of MMP10, ABCB5, and RND3, and downregulation of CD38 and ANGPT1, indicating ECM remodeling and immunomodulation. PRMT5 inhibition affected 22 RNA-processing genes, while NID1 knockdown influenced 36 ECM-related genes. GSEA consistently highlighted ECM/MMP pathway enrichment. PPI analysis identified FN1, MMP10, and EPHA7 as central hubs. TF and upstream regulator analyses revealed a shared NF-κB, CEBPB, SP1, and STAT3 axis connecting MI3454 response, PRMT5 inhibition, and NID1 disruption. Compared to prior pediatric and adult AML transcriptomic studies, this work provides novel mechanistic insights into ECM remodeling as a central mediator of therapy response.

Targeted therapies in pediatric AML converge on ECM remodeling and a shared TF-driven regulatory axis, highlighting combinatorial strategies targeting microenvironmental and signaling vulnerabilities to overcome treatment resistance.

The online version contains supplementary material available at 10.1007/s12672-026-04617-w.

Figures