Non-Eosinophilic Nasal Polyps Shows Increased Epithelial Proliferation and Localized Disease Pattern in the Early Stage

BACKGROUND Non-eosinophilic nasal polyps (NPs) show less inflammatory changes and are less commonly associated with lower airway inflammatory disorders such as asthma, compared with eosinophilic NPs. However, the development of non-eosinophilic NPs which is a predominant subtype in Asian population still remains unclear. METHODS A total of 81 patients (45 with non-eosinophilic NPs and 36 with eosinophilic NPs) were enrolled. Clinical information and computed tomography (CT), endoscopic, and histological findings were investigated. Tissue samples were analyzed for total IgE levels and for mRNA expression levels of interleukin (IL)-4, IL-5, IL-13, interferon (IFN)-γ, tumor necrosis factor (TNF)-α, IL-17A, IL-22, IL-23p19, transforming growth factor (TGF)-β1, TGF-β2, TGF-β3, and periostin. Immunostaining assessment of Ki-67 as a proliferation marker was performed. RESULTS We found that epithelial in-growing patterns such as pseudocysts were more frequently observed in histological and endoscopic evaluations of non-eosinophilic NPs, which was linked to increase epithelial staining of Ki-67, a proliferating marker. Eosinophilic NPs were characterized by high infiltration of inflammatory cells, compared with non-eosinophilic NPs. To investigate the developmental course of each subtype, CT was analyzed according to CT scores and subtypes. Non-eosinophilic NPs showed more localized pattern and maxillary sinus involvement, but lesser olfactory involvement in early stage whereas eosinophilic NPs were characterized by diffuse ethmoidal and olfactory involvement. In addition, high ethmoidal/maxillary (E/M) CT scores, indicating ethmoidal dominant involvement, were one of surrogate markers for eosinophilic NP. E/M CT scores was positively correlated with levels of TH2 inflammatory markers, including IL-4, IL-5, periostin mRNA expression and total IgE levels in NPs, whereas levels of the TH1 cytokine, IFN- γ were inversely correlated. Moreover, if the combinatorial algorithm meet the three of the four markers, including IL-5 (<2.379), periostin (<3.889), IFN-γ (>0.316), and E/M ratio (<2.167), non-eosinophilic CRSwNP are diagnosed with a sensitivity of 84.4% and a specificity of 84.8%. CONCLUSION Histologic, immunologic and clinical data suggest that non-eosinophilic NPs showed enhanced epithelial alteration and more localized maxillary involvement. Combination of cutoff value on IL-5, periostin, IFN-γ, and E/M scores may be one of surrogate markers for non-eosinophil NP subtype.