Exosome-Enriched Hub Gene Networks Identify Diagnostic Biomarkers and Repurposable Therapeutic Targets in Endometriosis

Endometriosis is a heterogeneous chronic inflammatory disorder associated with substantial diagnostic delay and limited therapeutic options, highlighting the need of robust non-invasive biomarkers and actionable molecular targets to complement existing low-sensitivity tests. To identify conserved pathogenic mechanisms with translational potential, here, we uniformly reprocessed three independent the Gene Expression Omnibus (GEO) microarray cohorts (GSE7305, GSE25628, and GSE11691) and applied a strict, directionally consistent intersection strategy to identify conserved transcriptional signals. We identified 262 consensus differentially expressed genes enriched for immunity/inflammation, cell adhesion and migration, and angiogenesis, consistent with key biological hallmarks of lesion establishment and persistence. Protein-protein interaction topology prioritized 11 highly connected hub genes (<i>VCAM1</i>, <i>CCL2</i>, <i>MCAM</i>, <i>CD14</i>, <i>CD24</i>, <i>FGFR1</i>, <i>SIRPA</i>, <i>CSF1R</i>, <i>S100A9</i>, <i>S100A8</i>, and <i>LY96</i>) that likely act as an integrated immune-adhesion-angiogenesis axis. Notably, 63/262 (24%) of the consensus genes were annotated to the extracellular exosome compartment, supporting their translational relevance as liquid-biopsy candidates. Finally, connectivity mapping using the LINCS L1000 framework nominated small-molecule perturbagens predicted to reverse the endometriosis-associated signature, providing a rational starting point for drug-repurposing experiments. In conclusion, this study elucidates a conserved immune-adhesion-angiogenesis axis driven by an 11-gene hub network in endometriosis. These core regulators represent promising candidates for the development of non-invasive liquid biopsies and precision, non-hormonal therapeutics.