Background: Chronic rhinosinusitis (CRS) has traditionally been classified phenotypically according to the presence (CRSwNP) or absence (CRSsNP) of nasal polyps. However, the phenotypic dichotomy does not represent the complexity of the disease. Current research thus focuses on identifying underlying inflammatory mechanisms and distinguishing different endotypes. The objective of this study was to evaluate the best combination of commonly used non-invasive biomarkers to distinguish between the phenotypes CRSwNP and CRSsNP.
Methods: IRB approved study of n= 103 CRS (n=37 CRSsNP, n=66 CRSwNP) patients. Nasal mucus was collected using merocel sponges after a 3- week washout period from steroids. The nasal mucus was then examined for twelve cytokines/inflammatory protein biomarkers including IFN-γ, IL-4, -5 -17A, -22, IgE, CST-2, ECP, MMP-9, PAPP-A, periostin, and serpin E1). Protein concentrations were determined by ELISAs and Luminex assays. For phenotype classification, different artificial intelligence algorithms including t-SNE, Adaboost and XGBoostwere were applied to the data from the biomarker analysis.
Results: The analysis showed that Il-5 is the most suitable non-invasive marker to distinguish between the two phenotypic clusters (94.6%). For the inflammatory protein biomarkers, the classification becomes most accurate with a combination of four different biomarkers including periostin, CST-2, ECP, and PAPP-A (67.6%).
Conclusions: Il-5 as well as the combination of four protein biomarkers including periostin, CST-2, ECP, and PAPP-A were able to cluster the phenotypes CRSwNP and CRSsNP in nasal mucus best. Thus, those proteins may be investigated further to gain more information about the pathophysiology of CRS.