The nasal microbiome in patients suffering from non-steroidal anti-inflammatory drugs-exacerbated respiratory disease in absence of corticosteroids

Abstract

Chronic rhinosinusitis (CRS) is a chronic inflammatory disease phenotypically classified by the absence (CRSsNP) or presence of nasal polyps (CRSwNP). The latter may also be associated with asthma and hypersensitivity towards non-steroidal anti-inflammatory drugs (NSAID) as a triad termed NSAID-exacerbated respiratory disease (N-ERD). The role of the microbiome in these different disease entities with regard to the underlying inflammatory process and disease burden is yet not fully understood. To address this question, we measured clinical parameters and collected nasal samples (nasal mucosal fluids, microbiome swabs from middle meatus and anterior naris) of patients suffering from CRSsNP (n=20), CRSwNP (n=20) or N-ERD (n=20) as well as from patients without CRS (=disease controls, n=20). Importantly, all subjects refrained from taking local or systemic corticosteroids or immunosuppressants for at least two weeks prior to sampling. The nasal microbiome was analyzed using 16S rRNA gene amplicon sequencing, and levels of 33 inflammatory cytokines were determined in nasal mucosal fluids using the MSD platform. Patients suffering from N-ERD and CRSwNP showed significantly worse smell perception and significantly higher levels of type 2 associated cytokines IL-5, IL-9, Eotaxin and CCL17. Across all 4 patient groups, Corynebacteria and Staphylococci showed the highest relative abundances. Although no significant difference in alpha and beta diversity was observed between the control and the CRS groups, pairwise testing revealed a higher relative abundance of Staphylococci in the middle meatus in N-ERD patients as compared to CRSwNP (p<0.001), CRSsNP (p<0.01) and disease controls (p<0.05) and of Lawsonella in patients suffering from CRSwNP in middle meatus and anterior naris in comparison to CRSsNP (p<0.0001 for both locations) and disease controls (p<0.01 and p<0.0001). Furthermore, we observed a positive correlation of Staphylococci with IL-5 (Pearson r=0.548) and a negative correlation for Corynebacteria and Eotaxin-3 (r=-0.540). Thus, in patients refraining from oral and nasal corticosteroid therapy for at least two weeks known to alter microbiome composition, we did not observe differences in microbiome alpha or beta diversity between various CRS entities and disease controls. However, our data suggest a close association between increased bacterial colonization with Staphylococci and decreased colonization by Corynebacteria as well as increased type 2 inflammation.