PubMed HealthSearch

Biomedical subjects

Yitian Zhao

Publications and source records attributed to Yitian Zhao.

1 recordsLinked to original sources

Aqueous humour extracellular vesicle membrane protein profiling reveals pathological features of refractory macular edema.

BACKGROUND: Both diabetic macular edema (DME) and retinal vein occlusion-related macular edema (RVO-ME) can become refractory to anti-vascular endothelial growth factor (anti-VEGF) therapy, but the underlying mechanisms are unclear. Molecular discrimination of refractory disease could guide personalized treatment. This study examined whether aqueous humor-derived extracellular vesicle (EV) membrane proteins can characterize refractoriness and reveal etiology-specific pathways. METHODS: This prospective cohort study included 28 patients with DME or RVO-ME (14 each), further divided into treatment-naïve and refractory subgroups. Aqueous humour samples were collected before intravitreal anti-VEGF injection. EV membrane proteins were profiled using an EV Array chip targeting 435 antibodies. Differentially expressed proteins were analyzed by bioinformatics, including Gene Ontology, Kyoto encyclopaedia of genes and genomes (KEGG) pathway enrichment, Gene set enrichment analysis (GSEA), and cell-of-origin mapping using public single-cell RNA-seq data. RESULTS: VEGF/VEGFR2 were elevated in treatment-naïve DME and RVO-ME. Refractory DME showed upregulation of C5 and CD34 (complement/immune activation). Refractory RVO-ME exhibited increased CD68 and Annexin A1 with decreased PDGFR (chronic inflammation, vascular dysregulation). RANTES was commonly upregulated in refractory disease. Several EV proteins discriminated refractory cases with high accuracy (AUC 0.898-0.980). Cellular origin suggested immune cell-derived EVs in DME, retinal cell-derived EVs in RVO-ME. External validation confirmed key differences. CONCLUSION: Refractory ME involves distinct pathways: immune-inflammatory activation in DME versus chronic inflammation with vascular dysregulation in RVO-ME. EV membrane proteins from aqueous humor provide insights into therapeutic resistance and hold promise as biomarkers for personalized treatment decisions.

artificial intelligence