Background: RNA sequencing (RNAseq) is being used to study inflammatory pathways in chronic rhinosinusitis (CRS). Our goal was to probe validity of tissue eosinophilia as a metric to study RNAseq in CRS.
Methods: The study was conducted on prospectively-enrolled subjects undergoing sinonasal surgery. Subjects were categorized as control, CRS, CRSwNP and CRSsNP. CRS was also categorized by tissue eosinophil levels per high power field (EOS/HPF) as ≥10 EOS/HPF or <10 EOS/HPF. Ethmoidal tissue samples were processed, differentially expressed (DE) genes were calculated and mechanistic pathway analysis performed.
Results: After removing low quality samples and incomplete metadata from 28 subjects, 3 controls and CRS samples ([Analysis 1: 5 CRSwNP, 11 CRSsNP; Analysis 2: 7 CRS <10 EOS/HPF, 8 ≥10 EOS/HPF)] were analyzed. Controls separated clearly from CRS by both study criteria (polyp status, EOS/HPF). In both analyses, CRS differentiated into two distinct CRS subgroups. However, heatmaps showed greater homogeneity within each CRS subtype when studied by eosinophilia vs. polyp status. Overall, higher gene expression was found in controls vs. CRS. When comparing CRS with control, 545 DE genes were found, with 269 being common between CRSwNP and CRSsNP. When analyzing by tissue EOS/HPF, 206 DE genes were common between CRS <10 EOS/HPF and CRS ≥10 EOS/HPF analyses. Tissue eosinophilia as a metric was further validated by finding DE of IL 17 signalling pathway between <10EOS/HPF (increased) versus ≥10 EOS/HPF samples.
Conclusions: As a metric, tissue eosinophilia is at least as effective as analysis by polyp status for RNA sequencing and may potentially offer superior insights into mechanistic pathways.