Identification of biomarker candidates for exfoliative glaucoma from autoimmunity profiling

Abstract

Abstract Background Exfoliative glaucoma (XFG) is a subtype of open-angle glaucoma characterized by distinctive extracellular fibrils and a yet unknown pathogenesis potentially involving immune-related factors. The aim of this exploratory study was to identify biomarkers for XFG using data from autoimmunity profiling performed on blood samples from a Scandinavian cohort of patients. Methods Autoantibody screening was analyzed against 258 different protein fragments in blood samples taken from 30 patients diagnosed with XFG and 30 healthy donors. The 258 protein fragments were selected based on a preliminary study performed on 3072 randomly selected antigens and antigens associated with the eye. The “limma” package was used to perform moderated t-tests on the proteomic data to identify differentially expressed reactivity between the groups. Results Multiple associated genes were highlighted as possible biomarker candidates including FUT2, CDH5, and the LOX family genes. Using seven variables, our binary logistic regression model was able to classify the cases from the controls with an AUC of 0.85, and our reduced model using only one variable corresponding to the FUT2 gene provided an AUC of 0.75, based on LOOCV. Furthermore, over-representation gene analysis was performed to identify pathways that were associated with antigens differentially bound to self-antibodies. This highlighted the enrichment of pathways related to collagen fibril formation and the regulatory molecules mir-3176 and mir-876-5p. Conclusions This study suggests several potential biomarkers that may be useful in developing further models of the pathology of XFG. In particular, CDH5, FUT2, and the LOX family seem to have a relationship which merits additional exploration.