Compartment-specific inflammation-metabolism remodeling across eutopic and ectopic endometriosis tissues with cross-context evaluation in TCGA-UCEC.

BACKGROUND: Endometriosis exhibits tissue- and cell-compartment heterogeneity in inflammatory and metabolic activity. We evaluated compartment-specific remodeling in primary endometriosis tissues and used TCGA-UCEC as an independent cancer comparison. METHODS: We reanalyzed GSE179640 single-cell RNA-sequencing data (14 donors, 24 libraries; 21,600 cells) and bulk RNA-sequencing data (24 biopsies). Analyses included Harmony integration, library-level program scoring, DESeq2, matched TCGA-UCEC tumor-normal comparisons, survival analysis, and CD45/CD24 flow cytometry. RESULTS: Inflammatory and metabolic programs varied by tissue and cellular compartment. In myeloid summaries, inflammation correlated with glycolysis (rho = 0.52, P = 0.009), whereas the bulk-wide association was weak (rho = 0.21, P = 0.320); the myeloid M2-like/iron-redox association was not significant (rho = 0.40, P = 0.054). Bulk analysis identified 31, 4480, and 4632 significant genes in eutopic endometrium, ectopic peritoneal lesions, and ectopic ovarian lesions versus controls, respectively. In TCGA-UCEC, paired tumor-normal glycolysis and oxidative-phosphorylation effects were significant, but the survival composite was null (HR = 1.00, 95% CI 0.82-1.22; P = 0.968), and five-program effects were not correlated across endometriosis and TCGA contrasts (rho = -0.30, P = 0.624). Flow cytometry detected 4.73% CD24-positive cells in endometriotic tissue versus 1.92% in control endometrium (P < 0.01). CONCLUSIONS: Endometriosis shows compartment- and tissue-specific inflammatory and metabolic remodeling. The myeloid inflammation-glycolysis relationship was not reproduced in bulk biopsies. TCGA-UCEC provides an independent cancer comparison but does not support an endometriosis-to-cancer progression mechanism.