In current study, patients were first categorized according to tissue eosinophilia, defined as eosinophils accounting for more than 20% of total inflammatory cells when viewed at high magnification. Although this method is only semiquantitative and may depend on the pathologist’s subjective judgment, it is intuitive and swift. In this study, we analyzed our results from an immunological perspective with a panel of multiple cytokines. Eosinophilic polyps had elevated levels of IL-5 (type 2 inflammation), whereas non-eosinophilic polyps had elevated levels of MPO, IL-1b, IL-6, IL-8, and TNF-α (non-type 2 inflammation). Therefore, defining tissue eosinophilia based on histological analysis seems to incorporate immunological characteristics as well.
Although immunologically distinct, tissue eosinophilia status alone does not seem to incorporate prognosis, since treatment outcomes of patients with eosinophilic CRSwNP were similar to those with non-eosinophilic CRSwNP 14. Therefore, for Koreans population, the dichotomous classification into eosinophilic and non-eosinophilic subtypes is not enough and subclassification beyond tissue eosinophilia is necessary. Our previous study demonstrated that several phenotypes of CRSwNP can be classified using routinely available clinical markers. According to our previous study, among patients with non-eosinophilic polyps, patients less than 35 years of age had worse outcomes compared to patients older than 35 years of age. Therefore, the goal of this study was to validate whether age can be used as a factor that could clinically and immunologically differentiate patients with CRSwNP.
First, age seems to be associated with the tissue eosinophilia status. In patients less than 35 years of age, almost 80% were non-eosinophilic CRSwNP suggesting that CRSwNP in younger age is more associated with non-type 2 inflammation. In addition, within patients with non-eosinophilic CRSwNP, younger patient (NE-Y) seems to be clinically and immunological distinct compared to older patients (NE-O).
Compared to NE-O, NE-Y was characterized by lower levels of IL-5 and A1AT. From clinical perspective, mLK scores measured at three months and six months postoperatively were significantly higher in NE-Y patients than in NE-O patients. As mLK score reflects disease outcome after the surgery 15, NE-Y patients had worse postoperative short-term outcomes compared to NE-O patients, which is in accordance with the results of our previous study 11. Therefore, decreased levels of IL-5 and A1AT in the tissue samples and worsening short-term outcomes result in NE-Y being a distinguishing phenotype in CRSwNP.
In non-eosinophilic CRSwNP, one of the possible factors responsible for disease refractoriness is tissue neutrophilia 4. IL-5, which is a typical type 2 inflammatory cytokine, is known to suppress neutrophil activity 16. Α1ΑΤ exerts its anti-inflammatory action by inhibiting inflammatory HNE, a serine protease stored in the azurophilic granules of neutrophils and released during neutrophil degranulation 17. Low Α1ΑΤ is known to be associated with disease refractoriness and tissue neutrophilia in nasal polyps 10. Therefore, characteristics of NE-Y may be associated with increased neutrophilic inflammation. However, this needs to be further evaluated since there was no significant difference in expression level neutrophilic markers, such as MPO, HNE, and IL-8 between NE-Y and NE-O.
There was a significant increase in proportion of eosinophilic CRSwNP in older patients compared to younger patients (20.8–43.4%). This may also mean that type 2 inflammation is more associated with older age that there is a higher proportion of eosinophilic CRSwNP compared to younger patients. Age association to eosinophilic CRSwNP can be partially explained by physiological and functional changes in the nasal mucosa, epithelial barrier dysfunction, or interaction with microbiome manifested as IgE responses to Staphylococcal enterotoxin B hich is also known to be associated with adult onset asthma 18–20.
However, there were no significant clinical or immunological differences between young and old-aged eosinophilic polyps. Our study showed somewhat contradictory results from other studies. Other studies had shown that elderly patients with CRS showed significantly higher CT scores, which were associated with nasal polyps 21,22. From immunological perspective, age affected tissue cytokine profile differently in eosinophilic polyps. Among eosinophilic polyps, levels of IL-17 and chemokine C-X-C motif ligand (CXCL1), both of which are type 3 cytokines related to neutrophilic inflammation, positively correlated with age 23. The discrepancy between current study and others may be due to arbitrarily determined age cutoff values among patients with eosinophilic polyps and the limited number of patients, especially those with younger age.
Our study has several limitations. This study was conducted at a single institution, and there may be a risk of selection bias. Postoperative endoscopy scores were the only outcome measured in our study. Since the endoscopy scores tend to reach a nadir around three to six months postoperatively and rise afterward 24, a longer follow-up duration is necessary to observe differences between the patient groups. For clinical validation, we analyzed an independent patient cohort, in which tissue cytokine analysis was not performed. Although various cytokines were measured in tissue samples for immunological validation, only a few patients were analyzed to identify differences in treatment outcomes. Therefore, a larger cohort, perhaps a multicenter study, with long-term follow-up is warranted.
In summary, age is associated the tissue eosinophilia status in CRSwNP. Especially, for younger patients, there was a significantly higher proportion of non-eosniphilic CRSwNP. Age appears to have a significant effect in patients with non-eosinophilic CRSwNP. Younger patients had worse treatment outcomes and were characterized by low IL-5 and A1AT compared to older patients with non-eosinophilic CRSwNP. Therefore, age may be an important parameter for phenotyping CRSwNP.