Differences in Clinical and Immunological Characteristics According to the Various Criteria for Tissue Eosinophilia in Chronic Rhinosinusitis With Nasal Polyps

Article information

Clin Exp Otorhinolaryngol. 2023;16(4):359-368

Publication date (electronic) : 2023 August 23

doi : https://doi.org/10.21053/ceo.2023.00542

Seung Koo Yang ¹^,^*

, Jeong-Whun Kim ¹^,²^,^*

, Tae-Bin Won ¹

, Chae-Seo Rhee ²^,³

, Yeon Bi Han^,⁴

, Sung-Woo Cho^,¹

¹Department of Otorhinolaryngology-Head and Neck Surgery, Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seongnam, Korea

²Sensory Organ Research Institute, Seoul National University Medical Research Center, Seoul, Korea

³Department of Otorhinolaryngology-Head and Neck Surgery, Seoul National University Hospital, Seoul National University College of Medicine, Seoul, Korea

⁴Department of Pathology and Translational Medicine, Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seongnam, Korea

Corresponding author: Sung-Woo Cho Department of Otorhinolaryngology-Head and Neck Surgery, Seoul National University Bundang Hospital, Seoul National University College of Medicine, 82 Gumi-ro 173beon-gil, Bundang-gu, Seongnam 13620, Korea Tel: +82-31-787-7413, Fax: +82-31-787-4057 Email: iamsungu@gmail.com

Co-Corresponding author: Yeon Bi Han Department of Pathology and Translational Medicine, Seoul National University Bundang Hospital, Seoul National University College of Medicine, 82 Gumi-ro 173beon-gil, Bundang-gu, Seongnam 13620, Korea Tel: +82-31-787-7723, Fax: +82-31-787-4012 Email: yeonbimoya@gmail.com

These authors contributed equally to this work.

Received 2023 April 19; Revised 2023 August 9; Accepted 2023 August 23.

Abstract

Objectives

Several criteria exist for classifying chronic rhinosinusitis with nasal polyps (CRSwNP) as eosinophilic or non-eosinophilic. This study attempted to evaluate several criteria for defining eosinophilic CRSwNP from clinical and immunological perspectives.

Methods

A cohort of 84 patients (73 patients with CRSwNP and 11 control patients) was retrospectively analyzed. Patients were divided into eosinophilic and non-eosinophilic CRSwNP based on four different criteria: eosinophils (EOS) accounting for more than 20% of the total inflammatory cells; ≥70 EOS per high-power field (HPF); >55 EOS/HPF; and ≥10 EOS/HPF. Preoperative clinical characteristics, the immunological profiles of 14 cytokines from nasal tissue, and postoperative outcomes were compared between eosinophilic and non-eosinophilic CRSwNP based on each criterion. These criteria were immunologically validated by using 14 cytokines to predict the performance of tissue eosinophilia with a random forest model.

Results

Patients with eosinophilic CRSwNP were significantly older when the criterion of ≥10 EOS/HPF or EOS >20% was used. The number of patients with aspirin intolerance was significantly higher in eosinophilic CRSwNP based on the criterion of EOS >20%. From an immunological perspective, non-type 2 inflammatory cytokines were significantly higher in non-eosinophilic CRSwNP with the criterion of EOS >20% of the total inflammatory cells. In addition, the criterion of EOS >20% of the total inflammatory cells resulted in the best prediction of eosinophilic CRSwNP, with an accuracy of 88.10% and area under the curve of 0.94.

Conclusion

Clinical and immunological characteristics were different between eosinophilic and non-eosinophilic CRSwNP depending on a variety of criteria, and the results of this study should be taken into account when choosing the criterion for defining eosinophilic CRSwNP and interpreting the data accordingly.

Keywords: Sinusitis; Nasal Polyp; Eosinophils; Classification

INTRODUCTION

Chronic rhinosinusitis with nasal polyps (CRSwNP) is a common inflammatory upper airway disorder affecting approximately 1%–4% of the population worldwide [1,2]. CRSwNP is a heterogeneous disorder composed of many subtypes, and subtyping based on the inflammatory endotypes—type 2 or non-type 2 inflammation—has been widely accepted, since clinical characteristics and treatment outcomes are often different. For instance, patients with high interleukin (IL)-5 levels have a higher prevalence of comorbid asthma and a higher rate of recurrence [1,3-5]. Biologics targeting type 2 inflammation have been widely used to treat a select population of patients with CRSwNP, especially characterized by type 2 inflammation. The proportions of type 2 and non-type 2 inflammation CRSwNP vary among races, with Asian CRS patients presenting more prominent neutrophilic inflammation than patients from Western countries [6]. In addition, non-type 2 inflammation is known to be more prevalent in younger age groups in Asia, and younger patients with non-type 2 CRSwNP tend to have poorer treatment outcomes than their elderly counterparts [7].

Tissue eosinophilia status is a well-known biomarker indicating whether CRSwNP is type 2 or non-type 2 inflammation. Previous studies have demonstrated the importance of tissue eosinophilia in relation to treatment outcomes after endoscopic sinus surgery [8]. There are several criteria for defining tissue eosinophilia, yet there is no consensus on a definite criterion. Japan and China reported ≥70 eosinophils per high-power field (EOS/HPF) and >55 EOS/HPF, respectively, as cut-offs corresponding to a higher likelihood of recurrence [9,10]. The European position paper on rhinosinusitis with nasal polyps (EPOS) 2020 suggests ≥10 EOS/HPF as the cut-off point when defining evidence of type 2 inflammation for the indication of biological therapy in CRSwNP [11]. Kim et al. [12] employed a novel criterion of EOS accounting for more than 20% of the total inflammatory cells when describing the clinical characteristics and treatment outcomes of eosinophilic CRSwNP in South Korea. Although previous studies have focused on varying treatment outcomes according to the criteria of tissue eosinophilia, no studies have analyzed the clinical and immunological differences between eosinophilic and non-eosinophilic CRSwNP depending on which criteria are used. In this study, the clinical and immunological characteristics and postoperative outcomes were analyzed for eosinophilic and non-eosinophilic CRSwNP based on the aforementioned criteria of tissue eosinophilia.

MATERIALS AND METHODS

Study population

A cohort of 73 patients with bilateral CRSwNP and 11 control patients were recruited from the Department of Otolaryngology Head and Neck Surgery of Seoul National University Bundang Hospital from July 2019 to July 2020. The diagnosis of CRSwNP was based on the criteria described by the European position paper on rhinosinusitis with nasal polyps [11]. All recruited patients with CRSwNP had not responded to medical therapy and underwent functional endoscopic sinus surgery (FESS). Patients who had been treated with preoperative systemic corticosteroids (SCSs) within the 4‐week surgery period were excluded. Nasal polyp tissue samples were harvested during the endoscopic sinus surgery, and for control patients, the ethmoid sinus mucosa samples were harvested during the endoscopic endonasal approach for skull base tumors. All control patients had no evidence of chronic rhinosinusitis at the time of the surgery. Patients who provided written informed consent for participation were analyzed. Patients with unilateral disease, fungal disease, antrochoanal polyps, cystic fibrosis, primary ciliary dyskinesia, or other sinonasal tumorous conditions, such as sinonasal inverted papilloma, were excluded.

This study was approved by the Institutional Review Board at Seoul National University Bundang Hospital (No. B-2211-790-301). Written informed consent was obtained from patients.

Patient classification and evaluation of clinical parameters

CRSwNP patients were divided into eosinophilic and non-eosinophilic groups based on the following four criteria: (1) ≥70 EOS/HPF [10]; (2) >55 EOS/HPF [9]; (3) ≥10 EOS/HPF [13]; and (4) EOS accounting for approximately more than 20% of the total inflammatory cells [12]. A detailed histological review was performed at 400× magnification under the microscope. The number of EOS in the mucosa was counted at HPF in three densest areas with cellular infiltrate beneath the epithelial surface, and the mean number of EOS count was calculated. Whether the percentage of EOS exceeded 20% of the total inflammatory cells was determined according to the pathologist’s judgment instead of directly counting all the inflammatory cells. All nasal polyps have been divided into eosinophilic and non-eosinophilic groups by the pathologists based on this criterion from 2006 at our institution. The advantage of this method is its intuitiveness and reproducibility. All cases were reviewed by one experienced pathologist (YBH).

The clinical characteristics including age, gender, blood EOS count, Lund-Mackay (LM) computed tomography (CT) score [14], the ratio of CT scores for ethmoid sinus and maxillary sinus (E/M ratio), and comorbid asthma status were compared between eosinophilic and non-eosinophilic groups. The presence of asthma was assessed by the institution’s allergists after a thorough review of previous medical history and lung function tests, such as spirometry, methacholine challenge test, or bronchodilator response test.

Cytokine measurement from tissue homogenates

All tissue samples were snap-frozen in liquid nitrogen and were stored at –80 ºC. Samples were thawed at room temperature and homogenization was done with a mechanical homogenizer. After homogenization, the homogenates were centrifuged for 20 minutes at 4,000 rpm at 4 ºC. The supernatants were then collected and were stored at −80 °C. These samples were thawed at room temperature, and thorough mixing was done with a vortex shaker. Protein concentrations of tissue extracts were determined using the Pierce 660 nm Protein Assay Kits (Thermo Fisher Scientific). Tissue homogenates were assayed using multiple cytokine analysis kits for detecting macrophage inflammatory protein (MIP)-1β/CCL4, interferon (IFN)-γ, IL-1β/IL-1F2, IL-5, IL-6, IL-8/CXCL8, IL-12 (p70), IL-33, tumor necrosis factor (TNF)-α (Luminex Human Magnetic Assay [9-Plex]; LXSAHM-9; R&D Systems), myeloperoxidase (MPO) (Luminex Human Magnetic Assay [LXSAHM-01]; R&D Systems), tumor growth factor (TGF) β1 (TGFBMAG-64K-01), α1 anti-trypsin (A1AT) (HNDG2MAG36K-01; Millipore), human neutrophil elastase (HNE) (HSP3MAG-63K-01; Millipore), and IL-22 (HTH17MAG-14K-01; Millipore). Data were collected using a Luminex 100 reader (Luminex), and data analysis was performed using MasterPlex QT version 2.0 software (MiraiBio). All assays were performed in duplicates according to the manufacturer’s protocol. All protein levels in the tissue homogenates were normalized to total protein concentration level. The sensitivity for each cytokine is as follows : MIP-1β/CCL4 (5.8 pg/mL), IFN-γ (0.4 pg/mL), IL-1β/IL-1F2 (0.8 pg/mL), IL-5 (0.5 pg/mL), IL-6 (1.7 pg/mL), IL-8/CXCL8 (1.8 pg/mL), IL-12 (p70) (20.2 pg/mL), IL-33 (1.8 pg/mL), TNF-α (1.2 pg/mL), MPO (26.2 pg/mL), TGFβ1 (11.4 pg/mL), α1 anti-trypsin (0.085 ng/mL), HNE (6.2 pg/mL), and IL-22 (0.032 pg/mL). The values less than the detection limit were given equally to half of the lower limit [5].

Evaluation of treatment outcomes

After standard FESS, oral antibiotics were routinely prescribed for five-seven days, and intranasal corticosteroid spray was used afterward. Patients were encouraged to perform saline nasal irrigation for 3 months. Short courses of antibiotics or SCSs were prescribed when necessary. For postoperative outcome, modified Lund-Kennedy (MLK) score, response to SCS use, and exacerbation were compared between eosinophilic and non-eosinophilic groups based on each criterion. MLK score was calculated by adding scores of polyps (0–2), edema (0–2), and discharge (0–2) for both sides [15]. MLK scores were recorded at 3, 6, and 12 months after the operation. For analysis, the MLK scores from 6–12 months after the operation were used. The response to SCS use was analyzed only in patients who were prescribed with SCS at least 1 month after the operation. A patient was considered to be responsive to SCS if the MLK score decreased by at least 50% after the SCS use. Last, exacerbation was defined as an increase of at least 1 point of MLK score during postoperative course along with subjective worsening of the patient’s symptoms.

Statistical analysis

A random forest model was constructed using the normalized values of 14 cytokines measured in tissues from CRSwNP or control with the input variables for immunological validation of patient classification based on tissue eosinophilia status defined by aforementioned criteria. A five-fold cross-validation was used, and accuracy, precision, recall, and area under the curve (AUC) were used to assess the classification performance. In order to identify cytokines that are strongly associated with tissue eosinophilia, the filter method (InfoGainAttributeEval) as the attribute evaluator was used for feature selection, in which features with high scores were identified. The machine learning was performed by Weka [16].

For continuous variables, Student t-test was used for comparison between the two groups, and the three groups were compared using the Kruskal-Wallis test and Dunn’s multiple comparison test after confirming the nonparametric distribution using the Shapiro test. Pearson’s chi-square test or Fisher’s exact test was used for categorical variables. Results are presented as mean± standard deviation. Statistical analyses were performed using IBM SPSS ver. 22.0 software (IBM Corp.). Values of P<0.05 were considered statistically significant.

RESULTS

Preoperative clinical characteristics

Among patients with CRSwNP (n=73), the mean age was 47.4±15.0 years and the male-to-female ratio was 47:26. Fifteen patients were revision cases, 13 patients had comorbid asthma, and four patients had aspirin intolerance. The control group comprised 11 individuals, whose clinical characteristics are described in Supplementary Table 1.

The clinical characteristics of patients with eosinophilic or non-eosinophilic CRSwNP according to different criteria for tissue eosinophilia are presented in Table 1. Blood eosinophilia, age, and the E/M ratio were significantly different between patients with eosinophilic and non-eosinophilic CRSwNP when using criteria with relatively low cut-off values: ≥10 EOS/HPF and EOS >20% of the total inflammatory cells. In addition, aspirin intolerance was associated with eosinophilic CRSwNP when the criterion of EOS >20% of the total inflammatory cells was used. On the contrary, no significant differences were observed in any of the clinical characteristics when the criteria with relatively high cut-off values—≥70 EOS/HPF or >55 EOS/HPF—were used. In addition, the LM CT score, male-to-female ratio, and proportion of patients with comorbid asthma were not significantly different between eosinophilic and non-eosinophilic CRSwNP for all criteria.

Table 1.

Comparison of clinical characteristics of eosinophilic and non-eosinophilic polyps based on different criteria

Immunological characteristics based on tissue cytokine levels

All mean cytokine levels are listed in Supplementary Table 2, and several important cytokines are described in Fig. 1. The mean IL-5 level was significantly higher in eosinophilic CRSwNP than in non-eosinophilic CRSwNP for all criteria. The mean IL-1β level was significantly lower in eosinophilic CRSwNP than in non-eosinophilic CRSwNP for all criteria except for the criterion of ≥70 EOS/HPF. Interestingly, TNF-α, IL-6, IL-8, and IL-12 levels were significantly lower in eosinophilic CRSwNP than in non-eosinophilic CRSwNP only when the criterion of EOS >20% of the total inflammatory cells was used.

Fig. 1.

Immunological comparison of eosinophilic and non-eosinophilic polyps based on different criteria. The level of interleukin (IL)-5 (A), a cytokine representative of type 2 inflammation, is significantly different between eosinophilic and non-eosinophilic chronic rhinosinusitis with nasal polyps (CRSwNP) regardless of the diagnostic criteria. On the contrary, IL-1β (B), tumor necrosis factor (TNF)-α (C), IL-6 (D), IL-8 (E), and IL-12 (F) were significantly different between eosinophilic and non-eosinophilic CRSwNP only using the criterion of eosinophils (EOS) >20% of total inflammatory cells. E, eosinophilic; NE, non-eosinophilic; HPF, high-power field. ^*P<0.05, ^**P<0.01, ^***P<0.001.

The classification performance of the 3-class classification (eosinophilic polyp, non-eosinophilic polyp, and control) was the highest when the criterion of EOS >20% of the total inflammatory cells was used to define tissue eosinophilia (Table 2). The accuracy was 88.10%, and the mean AUC was 0.94. Heatmaps of protein concentrations among patient groups according to each criterion are shown in Fig. 2. The segregation between eosinophilic and non-eosinophilic polyps seemed to be the highest with the criterion of EOS >20% of the total inflammatory cells. The filter method for feature selection identified the top five features with high scores: IL-5, IL-8, IL-1β, TGF-β, MPO. Performing cross-validation with the exclusion of each of these features showed changes in accuracy and AUC, with the exclusion of IL-5 resulting in the largest change in both accuracy (88.10–75.00) and AUC (0.94–0.85) (Supplementary Table 3).

Table 2.

Immunological validation of tissue eosinophilia using a random forest model based on the levels of 14 cytokines from nasal tissues

Fig. 2.

Heatmaps of protein concentrations among the classes based on different criteria for eosinophilia. (A) ≥70 eosinophils (EOS)/highpower field (HPF), (B) >55 EOS/HPF, (C) ≥10 EOS/HPF, (D) EOS >20% of the total inflammatory cells. Color bars present Z-scores of tissue concentration for each measured cytokine. HNE, human neutrophil elastase; TNF, tumor necrosis factor; IL, interleukin; IFN, interferon; MIP, macrophage inflammatory protein; MPO, myeloperoxidase; TGF, tumor growth factor.

Postoperative outcomes

Postoperative outcomes were analyzed in 44 (60.2%) patients, who visited the outpatient clinic for at least 6 months after the operation. As shown in Table 3, the MLK scores and the exacerbation rates were not significantly different between eosinophilic and non-eosinophilic groups in all criteria. With regard to the response to SCS, a total of 12 patients were prescribed SCS at least a month after the operation, and 10 patients (83.3%) were responsive to SCS. The criterion of EOS accounting for more than 20% of total inflammatory cells was the only criterion in which there was a significant difference in the number of patients responsive to SCS between the eosinophilic and non-eosinophilic groups (Table 3).

Table 3.

Comparison of postoperative outcome of eosinophilic and non-eosinophilic polyps based on different criteria

DISCUSSION

In this study, we analyzed the differences in clinical parameters and immunological characteristics between eosinophilic and non-eosinophilic CRSwNP based on various diagnostic cut-off values for tissue eosinophilia. Both clinical and immunological parameters were most distinguishable when tissue eosinophilia is defined by the criterion of EOS comprising more than 20% of the total inflammatory cells.

Non-type 2 CRSwNP is known to be predominant at younger ages, and the level of IL-5 is positively correlated with age in CRSwNP patients in Korea [17]. Regarding other clinical parameters, the E/M ratio and the level of blood EOS are both known to be associated with type 2 CRSwNP [18-20]. These findings are in accordance with the current results when the criteria with low cut-off values (e.g., ≥10 EOS/HPF or EOS >20% of the total inflammatory cells) were used. However, these clinical characteristics were not different between eosinophilic and non-eosinophilic CRSwNP when high cut-off values (e.g., ≥70 or >55 EOS/HPF) were used. Patients who are regarded as having eosinophilic CRSwNP using the criteria with low cut-off values may be classified as having non-eosinophilic CRSwNP using the criteria with high cut-off values (Fig. 3A), obscuring the distinction between the two conditions.

Fig. 3.

Possible underestimation or overestimation of the expected characteristics of tissue eosinophilia depending on the criteria, as demonstrated by two samples of nasal polyps in hematoxylin and eosin sections (magnification, ×200). (A) 47 Eosinophils (EOS)/high-power field (HPF) may be classified as non-eosinophilic according to the criteria with high cut-off values, while the criterion of EOS >20% of total inflammatory cells classifies it as eosinophilic. (B) 11.66 EOS/HPF may be classified as eosinophilic according to the criterion of ≥10 EOS/HPF, while the criterion of EOS >20% of total inflammatory cells classifies it as non-eosinophilic. NE, non-eosinophilic; E, eosinophilic.

Aspirin intolerance is a risk factor for disease recurrence and is associated with disease refractoriness [21]. Consequently, one might anticipate a higher proportion of aspirin-exacerbated respiratory disease (AERD) among patients categorized as eosinophilic CRSwNP with high cut-off values (e.g., ≥70 or >55 EOS/HPF), as these values are linked to disease recurrence [10,22]. However, our study found no significant difference in the proportion of aspirin intolerance between eosinophilic and non-eosinophilic CRSwNP when using the criteria with high cut-off values. This can be explained by previous findings that there is no significant difference in tissue EOS levels between CRSwNP and AERD [23]. Instead, nasal polyp tissue from AERD patients is known to exhibit high levels of EOS cationic protein [20], which accumulates due to the activation and degranulation of EOS, and may not be detected by traditional hematoxylin and eosin staining.

Comorbid asthma is known to be associated with type 2 CRSwNP, as was also demonstrated in our recent publication [17]. However, this association was not observed in the current study. This discrepancy may be attributed to the small sample size of the study and potential under-diagnosis of comorbid asthma. It is possible that patients with asymptomatic asthma did not seek consultation from allergists. From an immunological standpoint, the level of IL-5, a cytokine indicative of type 2 inflammation, showed a significant difference between eosinophilic and non-eosinophilic CRSwNP, irrespective of the diagnostic criteria used. When IL-5 was excluded, identified via the filter method as the feature with the highest score, there was a substantial shift in both accuracy and AUC in the immunological validation of tissue eosinophilia, as defined by the criterion of EOS >20% of total inflammatory cells. This outcome suggests that IL-5 could serve as a potential biomarker for tissue eosinophilia, aligning with previous research findings [5,18].

Contrarily, significant differences were observed between eosinophilic and non-eosinophilic CRSwNP in terms of IL-1β, TNF-α, IL-6, IL-8, and IL-12 when the sole criterion was EOS >20% of total inflammatory cells. When other criteria were considered, fewer cytokines showed differences between eosinophilic and non-eosinophilic CRSwNP.

IL-1β, a pro-inflammatory cytokine that induces the expression of IL-17, is recognized to be elevated in non-eosinophilic CRSwNP [24]. One potential mechanism in the pathogenesis of chronic rhinosinusitis involves the invasion of an external pathogen into the nasal epithelium. This invasion further stimulates the production of IL-6 and IL-8, markers of neutrophil inflammation, as well as TNF-α, a Th1 marker [25]. A cluster analysis study has shown an increase in IL-6, IL-8, and TNF-α across several non-type 2 endotypes in CRSwNP [5,26]. Furthermore, IL-12 is known to boost the production of IFN-γ by CD4+ T cells [27], so the observed elevation of this cytokine in non-eosinophilic CRSwNP is not unexpected.

The mean values of each aforementioned cytokine in eosinophilic or non-eosinophilic CRSwNP appear to vary based on the criteria used for tissue eosinophilia. Non-eosinophilic CRSwNP, as defined by criteria using high cut-off values, may include patients who could be classified as eosinophilic under different criteria, which could potentially lower the average levels of non-type 2 inflammatory cytokines. Furthermore, a recent study has shown high levels of neutrophils in severe eosinophilic CRSwNP, accompanied by increased Charcot-Leyden crystal deposition and EOS extracellular trap cell death [28]. Therefore, when defined by criteria using high cut-off values, some patients with severe eosinophilic CRSwNP may also exhibit high levels of non-type 2 inflammatory cytokines, such as IL-6 and IL-8, in eosinophilic CRSwNP. Consequently, several non-type 2 inflammatory cytokines may not show significant differences between eosinophilic and non-eosinophilic CRSwNP when high cut-off value criteria are applied.

For postoperative outcomes, no criterion was successful in distinguishing between eosinophilic and non-eosinophilic groups, as evidenced by the MLK score. The definition of eosinophilic CRSwNP as a predictor of disease recurrence may be less impactful for the Korean population, given the prevalence of non-type 2 CRSwNP in Korea. A significant portion of non-type 2 CRSwNP may recur and become resistant cases [29]. The consideration of neutrophilic phenotypes has been suggested in relation to disease severity or steroid resistance in Asian CRS [30]. As such, it may be more practical to categorize eosinophilic CRSwNP patients based on their response to biologics targeting type 2 inflammation. If the eosinophilia cut-off value is set too high, the prediction of an individual’s response to biological therapy could be underestimated. Furthermore, the presence of non-type 2 inflammation in severe type 2 CRSwNP, as defined by high cut-off values, could potentially explain the insensitivity to biological therapy against type 2 inflammation [28]. While EPOS 2020 proposed ≥10 EOS/HPF as the minimum cut-off value to confirm the presence of type 2 inflammation [11], our study found that a 20% cut-off value provided the best performance for immunologic differentiation between eosinophilic and non-eosinophilic CRSwNP. This suggests that this criterion is more representative of the tissue cytokine signature than other criteria.

The response to SCSs is acknowledged as a sign that patients may also respond positively to biological therapy. As such, the current study suggests that the criterion of EOS comprising more than 20% of total inflammatory cells is the most effective predictor of an individual’s response to SCS. This may also imply its potential role in predicting an individual’s response to biologics [31]. However, whether the 20% cut-off value is a superior indicator for evaluating response to biologic therapy requires further validation in future studies.

Eosinophilic CRSwNP, as defined by the criterion of ≥10 EOS/ HPF, may encompass patients who would otherwise be classified as non-eosinophilic according to different criteria, thereby elevating the average levels of non-type 2 inflammatory cytokines. For example, non-eosinophilic CRSwNP, determined by EOS comprising less than 20% of the total inflammatory cells, might actually contain more than 10 EOS/HPF. This would categorize them as eosinophilic CRSwNP under the ≥10 EOS/HPF criterion (Fig. 3B). Consequently, the proportion of EOS, rather than the absolute EOS count, as assessed by a seasoned pathologist, appears to yield a more precise classification that accurately reflects anticipated clinical and immunological profiles. For reference, the classification of tissue eosinophilia based on all criteria, and the number of EOS/HPF for each patient, as well as the comparison of EOS/HPF in each criterion, are presented in Supplementary Table 4 and Table 1, respectively.

This study has several limitations. Its design—as a study conducted at a single institution with a small patient sample—may introduce selection bias, and it does not allow external validation. Additionally, variations in genetic or environmental backgrounds, such as those found in Western countries, could lead to different conclusions due to potential shifts in the proportions of type 2 and non-type 2 CRSwNP. The distribution of patients with eosinophilic and non-eosinophilic CRSwNP is dependent on different criteria, which could result in overfitting or underfitting. This study also includes patients who have previously undergone surgery and experienced recurrence. Patients with recurrent disease are known to have a different immunological profile, often with a higher proportion of mixed-type inflammation [32]. Furthermore, the study population includes patients with a history of sinus surgery and/or use of intranasal corticosteroids. The history of sinus surgery could influence the cytokine profile, as the immunological profile is known to differ between patients with primary nasal polyps and refractory nasal polyps—those who have undergone endoscopic sinus surgery two or more times [32]. Intranasal steroid spray has been reported to decrease intra-epithelial EOS [33]. Finally, the pathologists’ judgments in determining the 20% cut-off value for tissue eosinophilia may vary from center to center. Therefore, a prospective, large-scale, multi-center cohort study with external validation is necessary.

In summary, the clinical and immunological distinctions between eosinophilic and non-eosinophilic CRSwNP may vary based on the criteria used. Eosinophilic CRSwNP is most accurately identified, both clinically and immunologically, when the criterion of EOS >20% of the total inflammatory cells is applied. Further research is required to confirm whether this criterion can serve as a biomarker for predicting an individual’s response to biological therapy targeting type 2 inflammation.

HIGHLIGHTS

▪ Although a variety of criteria for tissue eosinophilia exists, there is no consensus.

▪ Eosinophilic nasal polyp may be best identified when the criterion of eosinophils accounting for more than 20% of the total inflammatory cells is used.

▪ Future studies should be done concerning a criterion which can best predict one’s response to biological therapy.

Notes

No potential conflict of interest relevant to this article was reported.

AUTHOR CONTRIBUTIONS

Conceptualization: SWC. Methodology: SWC. Formal analysis: SWC. Data curation: SWC, YBH. Visualization: JWK, TBW, CSR. Project administration: JWK, TBW, CSR. Funding acquisition: SWC. Writing–original draft: SWC, SKY. Writing–review & editing: SWC, SKY, YBH.

Acknowledgements

This study was supported by a National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (No. NRF-2019R1C1C1009886).

Supplementary materials

Supplementary materials can be found online at https://doi.org/10.21053/ceo.2023.00542.

Supplementary Table 1.

Clinical characteristics of the control group

ceo-2023-00542-Supplementary-Table-1.pdf

Supplementary Table 2.

Comparison of immunological characteristics of eosinophilic polyps, non-eosinophilic polyps, and control based on different criteria

ceo-2023-00542-Supplementary-Table-2.pdf

Supplementary Table 3.

Immunological validation of tissue eosinophilia using a random forest model with the exclusion of each of top five ranked features

ceo-2023-00542-Supplementary-Table-3.pdf

Supplementary Table 4.

Classification of enrolled patients based on different criteria for tissue eosinophilia and the number of tissue EOS

ceo-2023-00542-Supplementary-Table-4.pdf

References

1. Ahn JC, Kim JW, Lee CH, Rhee CS. Prevalence and risk factors of chronic rhinosinusitus, allergic rhinitis, and nasal septal deviation: results of the Korean National Health and Nutrition Survey 2008- 2012. JAMA Otolaryngol Head Neck Surg 2016;Feb. 142(2):162–7.

2. Larsen K, Tos M. The estimated incidence of symptomatic nasal polyps. Acta Otolaryngol 2002;Mar. 122(2):179–82.

3. Calus L, Van Bruaene N, Bosteels C, Dejonckheere S, Van Zele T, Holtappels G, et al. Twelve-year follow-up study after endoscopic sinus surgery in patients with chronic rhinosinusitis with nasal polyposis. Clin Transl Allergy 2019;9:30.

4. Liao B, Liu JX, Li ZY, Zhen Z, Cao PP, Yao Y, et al. Multidimensional endotypes of chronic rhinosinusitis and their association with treatment outcomes. Allergy 2018;Jul. 73(7):1459–69.

5. Tomassen P, Vandeplas G, Van Zele T, Cardell LO, Arebro J, Olze H, et al. Inflammatory endotypes of chronic rhinosinusitis based on cluster analysis of biomarkers. J Allergy Clin Immunol 2016;May. 137(5):1449–56.

6. Yao Y, Zeng M, Liu Z. Revisiting Asian chronic rhinosinusitis in the era of type 2 biologics. Clin Exp Allergy 2022;Feb. 52(2):231–43.

7. Kim JW, Huh G, Rhee CS, Lee CH, Lee J, Chung JH, et al. Unsupervised cluster analysis of chronic rhinosinusitis with nasal polyp using routinely available clinical markers and its implication in treatment outcomes. Int Forum Allergy Rhinol 2019;Jan. 9(1):79–86.

8. Lou H, Zhang N, Bachert C, Zhang L. Highlights of eosinophilic chronic rhinosinusitis with nasal polyps in definition, prognosis, and advancement. Int Forum Allergy Rhinol 2018;Nov. 8(11):1218–25.

9. Lou H, Meng Y, Piao Y, Wang C, Zhang L, Bachert C. Predictive significance of tissue eosinophilia for nasal polyp recurrence in the Chinese population. Am J Rhinol Allergy 2015;29(5):350–6.

10. Tokunaga T, Sakashita M, Haruna T, Asaka D, Takeno S, Ikeda H, et al. Novel scoring system and algorithm for classifying chronic rhinosinusitis: the JESREC Study. Allergy 2015;Aug. 70(8):995–1003.

11. Fokkens WJ, Lund VJ, Hopkins C, Hellings PW, Kern R, Reitsma S, et al. European position paper on rhinosinusitis and nasal polyps 2020. Rhinology 2020;Feb. 58(Suppl S29):1–464.

12. Kim SY, Park JH, Rhee CS, Chung JH, Kim JW. Does eosinophilic inflammation affect the outcome of endoscopic sinus surgery in chronic rhinosinusitis in Koreans. Am J Rhinol Allergy 2013;27(6):e166–9.

13. Soler ZM, Sauer D, Mace J, Smith TL. Impact of mucosal eosinophilia and nasal polyposis on quality-of-life outcomes after sinus surgery. Otolaryngol Head Neck Surg 2010;Jan. 142(1):64–71.

14. Lund VJ, Kennedy DW. Staging for rhinosinusitis. Otolaryngol Head Neck Surg 1997;Sep. 117(3 Pt 2):S35–40.

15. Psaltis AJ, Li G, Vaezeafshar R, Cho KS, Hwang PH. Modification of the Lund-Kennedy endoscopic scoring system improves its reliability and correlation with patient-reported outcome measures. Laryngoscope 2014;Oct. 124(10):2216–23.

16. Witten I, Frank E, Hall M, Pal C. Data mining. 4th ed. Morgan Kaufmann; 2016.

17. Kim JW, Won TB, Woo H, Yang SK, Phannikul C, Ko SA, et al. Age related non-type 2 inflammation and its association with treatment outcome in patients with chronic rhinosinusitis with nasal polyp in Korea. Sci Rep 2022;Jan. 12(1):1671.

18. Gevaert P, Han JK, Smith SG, Sousa AR, Howarth PH, Yancey SW, et al. The roles of eosinophils and interleukin-5 in the pathophysiology of chronic rhinosinusitis with nasal polyps. Int Forum Allergy Rhinol 2022;Nov. 12(11):1413–23.

19. Meng Y, Lou H, Wang C, Zhang L. Predictive significance of computed tomography in eosinophilic chronic rhinosinusitis with nasal polyps. Int Forum Allergy Rhinol 2016;Aug. 6(8):812–9.

20. Stevens WW, Schleimer RP. Aspirin-exacerbated respiratory disease as an endotype of chronic rhinosinusitis. Immunol Allergy Clin North Am 2016;Nov. 36(4):669–80.

21. Rosati D, Rosato C, Pagliuca G, Cerbelli B, Della Rocca C, Di Cristofano C, et al. Predictive markers of long-term recurrence in chronic rhinosinusitis with nasal polyps. Am J Otolaryngol 2020;41(1):102286.

22. McHugh T, Snidvongs K, Xie M, Banglawala S, Sommer D. High tissue eosinophilia as a marker to predict recurrence for eosinophilic chronic rhinosinusitis: a systematic review and meta-analysis. Int Forum Allergy Rhinol 2018;Dec. 8(12):1421–9.

23. Stevens WW, Ocampo CJ, Berdnikovs S, Sakashita M, Mahdavinia M, Suh L, et al. Cytokines in chronic rhinosinusitis. Role in eosinophilia and aspirin-exacerbated respiratory disease. Am J Respir Crit Care Med 2015;Sep. 192(6):682–94.

24. Kim DK, Eun KM, Kim MK, Cho D, Han SA, Han SY, et al. Comparison between signature cytokines of nasal tissues in subtypes of chronic rhinosinusitis. Allergy Asthma Immunol Res 2019;Mar. 11(2):201–11.

25. Ramezanpour M, Bolt H, Psaltis AJ, Wormald PJ, Vreugde S. Primary human nasal epithelial cells: a source of poly (I:C) LMW-induced IL-6 production. Sci Rep 2018;Jul. 8(1):11325.

26. Wei B, Liu F, Zhang J, Liu Y, Du J, Liu S, et al. Multivariate analysis of inflammatory endotypes in recurrent nasal polyposis in a Chinese population. Rhinology 2018;Sep. 56(3):216–26.

27. Xiao L, Jia L, Zhang Y, Yu S, Wu X, Yang B, et al. Human IL-21+IFN-γ+CD4+ T cells in nasal polyps are regulated by IL-12. Sci Rep 2015;Aug. 5:12781.

28. Delemarre T, Holtappels G, De Ruyck N, Zhang N, Nauwynck H, Bachert C, et al. A substantial neutrophilic inflammation as regular part of severe type 2 chronic rhinosinusitis with nasal polyps. J Allergy Clin Immunol 2021;Jan. 147(1):179–88.

29. Kim DK, Kim JY, Han YE, Kim JK, Lim HS, Eun KM, et al. Elastase-positive neutrophils are associated with refractoriness of chronic rhinosinusitis with nasal polyps in an Asian population. Allergy Asthma Immunol Res 2020;Jan. 12(1):42–55.

30. Kim DW. Can neutrophils be a cellular biomarker in asian chronic rhinosinusitis. Clin Exp Otorhinolaryngol 2019;Nov. 12(4):325–6.

31. Cardell LO, Stjarne P, Jonstam K, Bachert C. Endotypes of chronic rhinosinusitis: impact on management. J Allergy Clin Immunol 2020;Mar. 145(3):752–6.

32. Ryu G, Kim DK, Dhong HJ, Eun KM, Lee KE, Kong IG, et al. Immunological characteristics in refractory chronic rhinosinusitis with nasal polyps undergoing revision surgeries. Allergy Asthma Immunol Res 2019;Sep. 11(5):664–76.

33. Burgel PR, Cardell LO, Ueki IF, Nadel JA. Intranasal steroids decrease eosinophils but not mucin expression in nasal polyps. Eur Respir J 2004;Oct. 24(4):594–600.

Article information Continued

This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Characteristics	EOS >20%			≥70 EOS/HPF			>55 EOS/HPF			≥10 EOS/HPF
Characteristics	E (n=35)	NE (n=38)	P-value	E (n=13)	NE (n=60)	P-value	E (n=22)	NE (n=51)	P-value	E (n=51)	NE (n=22)	P-value
Age (yr)	51.4±11.6	43.7±16.8	0.024^a)	51.1±10.1	46.6±15.8	0.329	51.0±12.4	45.8±15.8	0.171	51.2±11.5	38.5±18.3	0.005^a)
Male:female	21:14	26:12	0.453	9:4	38:22	0.687	12:10	35:16	0.249	34:17	13:9	0.535
LM CT score	18.3±5.0	18.3±4.0	0.958	18.8±5.7	18.2±4.2	0.747	19.4±5.2	17.9±4.1	0.236	18.4±4.8	18.0±3.9	0.739
Asthma	9	4	0.090	3	10	0.584	5	8	0.471	11	2	0.201
Aspirin intolerance	4	0	0.048^a)	0	4	0.338	3	1	0.079	4	0	0.177
E/M ratio	2.4±0.7	2.0±0.8	0.038^a)	2.2±0.6	2.2±0.8	0.827	2.4±0.7	2.1±0.8	0.135	2.3±0.8	1.9±0.6	0.027^a)
% Blood EOS	6.0±3.3	3.1±2.8	<0.001^a)	5.3±3.6	4.4±3.3	0.375	5.6±3.1	4.0±3.4	0.062	5.3±3.5	2.6±2.1	0.001^a)
EOS/HPF	63.36±35.05	14.11±18.90	<0.001^a)	102.40±20.90	23.71±21.71	<0.001^a)	85.30±26.37	17.20±16.31	<0.001^a)	52.75±34.89	2.90±2.61	<0.001^a)

Criteria	Accuracy (%)	Recall	Precision	AUC
EOS >20%	88.10	0.88	0.89	0.94
≥70 EOS/HPF	78.57	0.79	0.75	0.81
>55 EOS/HPF	77.38	0.77	0.77	0.85
≥10 EOS/HPF	79.76	0.80	0.80	0.93

Postoperative outcome	EOS >20%			≥70 EOS/HPF			>55 EOS/HPF			≥10 EOS/HPF
Postoperative outcome	E	NE	P-value	E	NE	P-value	E	NE	P-value	E	NE	P-value
MLK score	1.87±1.74 (n=23)	1.81±2.23 (n=21)	0.921	2.50±1.60 (n=8)	1.69±2.03 (n=36)	0.299	2.13±1.63 (n=16)	1.68±2.14 (n=28)	0.475	1.90±2.02 (n=29)	1.73±1.91 (n=15)	0.318
Exacerbation	60.0 (20)	30.0 (20)	0.057	42.9 (7)	45.4 (33)	>0.999	57.1 (14)	38.5 (26)	0.257	53.8 (26)	28.6 (14)	0.186
Response to SCS	100.0 (9)	33.3 (3)	0.046^a)	100.0 (2)	80.0 (10)	>0.999	100.0 (5)	71.4 (7)	0.470	90.0 (10)	50.0 (2)	0.318