Abstract
OBJECTIVE: To evaluate eosinophil peroxidase (EPX) as a biomarker for tissue levels of eosinophilia, cytokines, and chemokines within chronic rhinosinusitis (CRS). METHODS: Twenty-eight subjects undergoing sinonasal surgery were prospectively enrolled. Ethmoid tissue was analyzed with an in-house EPX immunoassay and a 48-plex cytokine-chemokine array. Clinical severity was assessed using SNOT-22 and Lund-Mackay scores. Subjects were grouped as follows: controls, polyp status (CRS with [CRSwNP] and without nasal polyps [CRSsNP]), tissue eosinophilia (eosinophilic CRS [eCRS], non-eosinophilic CRS [neCRS]), or combinations thereof (eCRSwNP, eCRSsNP, neCRSsNP). eCRS was defined as >10 eosinophils per high power field (HPF). Subjects without CRS or asthma were enrolled as controls. RESULTS: EPX was elevated in CRSwNP compared to control (p = 0.007), in eCRS compared to neCRS (p = 0.002), and in eCRSwNP along with eCRSsNP compared to neCRSsNP (p = 0.023, p = 0.015, respectively). eCRS displayed elevated IL-5 compared to neCRS (p = 0.005). No significant differences in EPX or IL-5 were observed between eCRSwNP and eCRSsNP. IL-5 was elevated in eCRSwNP (p = 0.019) compared neCRSsNP. Area under the receiver operator characteristic curve was 0.938 (95% CI, 0.835-1.00) for EPX and tissue eosinophilia, with an optimal cut-point of 470 ng/mL being 100% specific and 81.25% sensitive for tissue eosinophilia. Linear regression revealed a strong correlation between EPX and IL-5 (R(2) = 0.64, p < 0.001). Comparing EPX and IL-5, only EPX displayed significant correlation with SNOT-22 (p = 0.04) and Lund-Mackay score (p = 0.004). CONCLUSION: EPX is associated with tissue eosinophilia in CRS patients regardless of polyp status. EPX correlates with IL-5 and could be potentially considered a biomarker for anti-IL-5 therapies. LEVEL OF EVIDENCE: 3 Laryngoscope, 134:69-78, 2024.