Fig 1: Proteomics analysis comparing responders with non-responders in TED. (A) Volcano plot showing differential protein abundance between ivGC responders and non-responders. The y-axis represents -log10 unadjusted p-values. Multiple testing correction was performed using the Benjamini-Hochberg method; however, no proteins remained significant after FDR correction (q < 0.05). Therefore, unadjusted P values are shown for hypothesis-generating visualization. Candidate biomarkers are highlighted with circles. (B) PPI network analysis centered on TGF-β. Interactions were analyzed using the STRING database, illustrating the key regulatory nodes and connections among the DAPs. The network showed significant enrichment of protein-protein interactions (PPI enrichment p = 4.48 × 10− 7). ARPC5, actin-related protein 2/3 complex subunit 5; CAH1, carbonic anhydrase 1; CAH2, carbonic anhydrase 2; CMGA, chromogranin A; DAPs, differentially abundant proteins; DIA, data-independent acquisition; FGL2, fibrinogen-like protein 2; PGN, progranulin; HPA, heparanase; INHBE, inhibin beta E chain; MBL2, mannose-binding lectin 2; MGAT1, alpha-1,3-mannosyl-glycoprotein; PPI, protein–protein interaction; PRDX6, peroxiredoxin 6; TED, thyroid eye disease; TGF-β, transforming growth factor beta; UROM, uromodulin; VTN, vitronectin