Discriminant analysis followed by unsupervised cluster analysis including exosomal cystatins predict presence of chronic rhinosinusitis, phenotype, and disease severity

Background Cystatins are epithelial protease inhibitors that participate in sinonasal immunity and inflammation. Nasal mucus-derived exosomes (NMDEs) are small vesicles secreted by epithelial cells that carry protein cargo reflective of their host cell. NMDEs have been used as a noninvasive biomarker source to study chronic rhinosinusitis with nasal polyps (CRSwNP) proteomics with superior sensitivity to whole mucus. The purpose of this study was to noninvasively quantify exosomal cystatins in a heterogenous population to determine their utility in predicting phenotype and disease severity. Methods This was an Institutional Review Board-approved study in which NMDEs were purified from 105 patients undergoing sinonasal surgery by ultracentrifugation. Demographic and clinical variables were collected and phenotypes were assigned a priori. Linear discriminant analysis was executed based on normalized Cystatin values as phenotype predictor variables. Unsupervised cluster analysis was performed using Ward's linkage followed by Duda/Hart Je(2)/Je(1) index cluster stopping rules. Analysis of variance (ANOVA), Welch's test, and Fisher's exact tests were used for continuous and categorical variables. Results NMDE Cystatin-2 expression segregated by phenotype (mean +/- standard error [SEM]): control (23.4 +/- 4.2 pg/mu g, n = 32); CRS without NP (CRSsNP) (56.6 +/- 8.3 pg/mu g, n = 33); and CRSwNP (130.5 +/- 16.7 pg/mu g, n = 40) (p < 0.0001). Seven clusters were identified among patients where the highest NMDE Cystatin-2 levels clustered with asthma, tissue eosinophilia, and aspirin-exacerbated respiratory disease (AERD). Conclusion Cystatin levels in NMDEs predict CRS phenotype and disease severity. As a liquid biopsy, noninvasive NMDE collection offers a promising opportunity to study disease pathophysiology, discriminate disease states, and potentially reveal novel therapeutic targets.