Ocular and systemic autoimmune diseases impair the proteome patterns of tear fluid. To learn more about the complex pathological processes in autoimmune thyroid eye disease (TED) it is essential to get detailed information on these proteins. Therefore, the purpose of this prospective and controlled study was to detect and evaluate possible changes in the proteomic patterns in tear fluid of patients with TED. Tear samples from TED patients with various disease severity and activity, and healthy controls were analyzed with the SELDI-TOF-MS-technology using arrays with different chromatographic surfaces (CM10 cation exchange, H50 reversed-phase). Data were analyzed by multivariate statistical techniques and artificial neural networks. The discriminate analysis revealed significant changes (p<0.05) in the protein profiles of TED patients compared to controls. We obtained a set of protein biomarkers that allowed us to clearly discriminate between patients and controls with a very high sensitivity and specificity (ROC curve, r=0.99). All possible biomarkers found in this study had a molecular weight between 3000 and 20,000 Da. The majority of the proteins was downregulated in the patient group, with only few proteins overexpressed in comparison to healthy controls. The SELDI-TOF-MS is an accurate method for proteome analysis in tear fluid of TED patients. These proteins may serve as biomarkers for diagnosis and follow-up during treatment.