A combinatorial EVs-miRNA signature mediates the anti-tumoral activity of NFAT3-regulated extracellular vesicles in aggressive cancers

Abstract Aggressive cancers such as triple-negative breast cancer (TNBC) and pancreatic cancer remain difficult to treat because their malignant behavior is driven by complex gene networks rather than single oncogenic targets. Here, we identify an extracellular vesicle (EV)-associated miRNA signature functionally linked to NFAT3 activity and demonstrate its ability to suppress tumor aggressiveness. Functional analyses revealed that a combination of fifteen miRNAs (miR-Comb 15) was required to fully reproduce the anti-tumoral effects of NFAT3-regulated EVs across TNBC and pancreatic cancer models, whereas individual miRNAs showed only partial activity. These effects were associated with coordinated regulation of validated target genes controlling proliferation and invasion, supporting a network-modulating mechanism of action. To facilitate therapeutic translation, we used EVs derived from HEK 293T cells, a non-tumoral, scalable, and readily engineerable EV source. Using an optimized exogenous pH-gradient loading strategy, miR-Comb 15 was efficiently incorporated into EVs without affecting vesicle integrity or intrinsic bioactivity. HEK 293T EVs loaded with miR-Comb 15 consistently showed the strongest anti-tumoral activity in vitro and in vivo among the delivery platforms evaluated. Together, these findings identify a functional NFAT3-dependent EV-miRNA program and support EV-mediated delivery of combinatorial miRNA therapeutics as a promising strategy for aggressive cancers. Competing Interest Statement SJ is the holder of a patent related to the NFAT3-expressing cells derived EVs use in therapy and uses thereof (US11154598B2) and a patent related to the miRNAs combination use for therapy (US20240084300A1) filed by Institut National de la Sante et de la Recherche Medicale INSERM, Universite Paris Cite. Footnotes This revised version incorporates extensive changes following peer review. We added new experiments to strengthen the mechanistic and translational aspects of the study, including additional functional validation, transcriptomic analyses, target enrichment analyses, optimization and characterization of EV loading, and additional in vivo data. The manuscript has been extensively revised for clarity, several figures have been updated, and the supplementary information has been expanded. These revisions substantially strengthen the biological, mechanistic, and translational conclusions of the study.