Abstract
Introduction
The emergency of biologics and surgical techniques targeting the specific inflammatory endotype in chronic rhinosinusitis with nasal polyps (CRSwNP) asks for efficient identification of patients with different endotypes. Although mucosal IL-4, IL-5, IL-13, and IgE have been used to define type 2 (T2) inflammation, the optimal one remains unclear. In this study, we aimed to determine the optimal anchor for T2 inflammation and identify clinical characteristics and nasal secretion biomarkers predicting different endotypes in CRSwNP.
Methods
Six mediators in sinonasal tissue and 36 mediators in nasal secretion samples were detected by the Bio-Plex suspension array system. Mucosal IFN-γ and IL-17A levels were used to define the T1 and T3 endotype, respectively. The efficacy of mucosal IL-4, IL-5, IL-13, and IgE to define the T2 endotype was compared. The power of clinical characteristics and nasal secretion biomarkers to predict the T1, T2, and T3 endotype was analyzed.
Results
Among mucosal IL-4, IL-5, IL-13, and IgE, IL-13 was the best one to coincide with the expression of other T2 biomarkers. A combination of atopy, facial pain symptom score, ethmoid/maxillary computed tomography score ratio, and blood eosinophil percentage had a moderate predictive performance for T2 endotype (area under the receiver operating curve [AUC] = 0.815), comparable to that of nasal secretion IL-5 (AUC = 0.819). For the T3 endotype, nasal secretion IL-1Rα identified it with an AUC value of 0.756. No efficient marker for the T1 endotype was found.
Conclusion
IL-13 is a primary anchor for the T2 endotype in CRSwNP. Clinical characteristics and nasal secretion biomarkers are helpful for identifying the T2 and T3 endotype of CRSwNP.
Keywords: Biomarker, Chronic rhinosinusitis with nasal polyps, Clinical characteristic, Inflammatory endotype, Nasal secretion
Introduction
Chronic rhinosinusitis with nasal polyps (CRSwNP) is a heterogeneous disease characterized by persistent inflammation of sinonasal mucosa with the formation of edematous nasal polyps (NPs) [1]. Different inflammatory endotypes have been observed under the term CRSwNP, making the tailored disease treatment a great challenge [2]. Type 2 (T2) polarized inflammatory endotype is characterized by enhanced production of IL-4, IL-5, IL-13, and IgE in NPs, while T1 and T3 endotype refer to inflammation driven by IFN-γ and IL-17A, respectively [3–6]. Recently, phase 3 clinical trials demonstrated the efficacy of biologics targeting T2 inflammation such as mepolizumab (anti-IL-5), benralizumab (anti-IL-5R), dupilumab (anti-IL-4Rα), and omalizumab (anti-IgE) on severe uncontrolled CRSwNP [7–10]. However, 30–50% of CRSwNP patients reported no improvement after those biologic treatments [11]. Mechanistically, these biologics should work more efficiently on patients with T2 endotype. Preliminary data also suggested that CRSwNP patients with severe T2 endotype might benefit from more extensive sinus surgery [12]. Therefore, there is an urgent need to identify the T2 endotype with clinically accessible markers. On the other hand, although there are currently no treatments specifically targeting the T1 or T3 endotype in CRSwNP, distinguishing patients with the T1 or T3 endotype would facilitate the development of therapeutic strategies targeting corresponding inflammation.
Various mucosal biomarkers, including tissue eosinophil cationic protein, IL-4, IL-5, and IL-13, have been used to define the T2 endotype in CRS [12–14]. However, the best anchor for the T2 endotype has not been investigated. Besides, instead of tissue-based biomarkers, it is critical to develop noninvasive and clinically accessible markers to identify the endotypes of CRSwNP. A previous study conducted in the USA found that the T2 endotype was associated with the presence of NPs, asthma comorbidity, smell loss, and allergic mucin, while the T1 endotype was more common in females and the T3 endotype was associated with pus in patients with CRS [4]. However, the predictive performance of clinical features for a certain endotype was unexplored in the study [4]. In addition, whether the noninvasive biomarkers such as those in nasal secretions could facilitate the endotype identification in patients with CRSwNP remains unclear. In this study, we determined the associations of clinical features and nasal secretion biomarkers with mucosal T1, T2, and T3 endotype in patients with CRSwNP and evaluated the predictive efficiency of those clinically accessible markers.
Materials and Methods
Ethical Approval
This study was approved by the Ethics Committee of Tongji Hospital of Huazhong University of Science and Technology (TJ-C20170301) and was conducted with written informed consent from all subjects.
Subjects
A total of 249 CRSwNP patients and 43 control subjects were included in this single-center, prospective, and observational study (online suppl. Table E1; for all online suppl. material, see https://doi.org/10.1159/000530193). CRSwNP was diagnosed according to the European Position Paper on Rhinosinusitis and Nasal Polyps 2020 [1]. All patients had failed medical treatments and underwent endoscopic sinus surgery. Subjects undergoing septoplasty due to anatomic variations and without other sinonasal diseases were enrolled as control subjects [13]. Patients with acute asthma episodes or upper respiratory tract infections within 4 weeks before surgery and those with antrochoanal polyps, cystic fibrosis, primary ciliary dyskinesia, fungal sinusitis, immunodeficiency, systemic vasculitis, or under immunotherapy were excluded. Patients who received oral glucocorticoids within 3 months, or intranasal glucocorticoids or antileukotrienes within 2 weeks before surgery were also excluded.
Baseline Assessment and Sample Collection
Demographics and clinical information were collected before surgery. The baseline severity of nasal symptoms including nasal obstruction, rhinorrhea, headache, facial pain or pressure, and hyposmia was rated by the visual analogue scale (VAS). Endoscopic physical findings including NP size, edema, discharge, scarring, and crusting were scored with the Lund-Kennedy scoring system [1]. Computed tomography (CT) scans of sinuses were graded with the Lund-Mackay scoring system. The ethmoid to maxillary sinus CT score ratio (E/M) and posterior ethmoid to anterior ethmoid sinus CT score ratio were calculated as previously mentioned [14, 15]. Atopic status was diagnosed through a skin prick test with a standard inhalant allergen panel (Macro-Union Pharmaceutical, Beijing, China) or by detecting serum allergen-specific IgE antibodies with ImmunoCAP (Phadia, Uppsala, Sweden) [16]. The diagnosis of allergic rhinitis was based on the presence of a typical history of allergic symptoms and the atopy test results [17]. The diagnosis of asthma was based on guidelines recommended by the Global Initiative for Asthma (GINA) [18]. Two to 3 days before surgery, peripheral blood samples were collected on an empty stomach in the morning and a complete peripheral blood cell count with differential was performed by automated analysis (Sysmex XE-5000, Sysmex Corporation, Kobe, Japan).
Nasal secretion samples were collected under general anesthesia before the beginning of the surgical procedure [19, 20]. Briefly, sponges were inserted into the bilateral nasal cavities for 5 min. Then, sponges were put in an Eppendorf tube containing 2 mL of sterile 0.9% NaCl for 30 min on ice to mobilize the nasal secretions. Sponges were transferred into a sterile syringe, and the liquid was squeezed out into the Eppendorf tube. After that, the Eppendorf tube was centrifuged at 1,500 g for 10 min and the supernatants were collected and frozen at −80°C until analysis. Only a portion of CRSwNP patients (n = 46) provided enough nasal secretion samples. NP samples and inferior turbinate mucosal tissues were obtained from all CRSwNP patients and control subjects during surgery, respectively. NP was classified as eosinophilic when the percentage of tissue eosinophils was greater than 10% of total infiltrated inflammatory cells [21]. Tissue samples were weighed and 1 mL of 0.9% NaCl supplemented with 10 μL of 100 mm phenylmethylsulfonyl fluoride was added for every 0.1 g tissue [22]. Then, the tissue samples were homogenized on ice and centrifuged at 3,000 rpm for 10 min at 4°C. Supernatants were harvested and stored at −80°C for further analysis.
Biological Mediator Measurement
The protein levels of IFN-γ, IL-4, IL-5, IL-13, IgE, and IL-17A in tissue homogenates and protein levels of 36 biomarkers including cytokines, chemokines, tissue remodeling molecules, and immunoglobulins in nasal secretions were detected using the Bio-Plex suspension array system (Bio-Rad, Hercules, CA, USA) according to the manufacturer’s instructions [23]. The lower detection limits were listed in online supplementary Table E2 in the online repository. The concentrations of protein levels in nasal tissues and secretions were normalized to total protein concentrations.
Stratifying Endotypes according to Mucosal Cytokine Levels
CRSwNP was stratified into the T1 or T3 endotype when the protein levels of IFN-γ or IL-17A in NPs were above the 95th percentile of corresponding cytokine levels in control tissues, respectively [24, 25]. T2 endotype was determined as the levels of IL-4, IL-5, IL-13, or IgE in NPs higher than the 95th percentile of corresponding cytokine levels in control tissues. Then, IL-4, IL-5, IL-13, and IgE were compared for their respective efficacy to classify T2 inflammation to select the primary anchor for the T2 endotype in CRSwNP. When a sample showed levels of 2 or 3 signature cytokines above the thresholds, it was considered the double or triple mixed endotype. The sample which exhibited all 3 signature cytokines below the thresholds was defined as all negative.
Statistical Analysis
Statistical analyses were performed using GraphPad Prism 8.0 (GraphPad Software, La Jolla, CA) and IBM SPSS 22.0 (SPSS Inc., Chicago, IL). For continuous variables, the normality of data distribution was tested with the Kolmogorov-Smirnov test or the Shapiro-Wilk test. Since variables were not normally distributed, Mann-Whitney U 2-tailed test was used for between-group comparisons. For categorical variables, the χ2 test was used for between-group comparisons. In the univariate logistic regression analyses, associations between the predictors and a particular endotype were evaluated after controlling confounding variates including age, gender, atopy, allergic rhinitis, and asthma. For age, atopy, allergic rhinitis, or asthma as a predictor, it was adjusted for the rest of the confounding variates. Factors significantly associated with the T1, T2, or T3 endotype were further introduced into multivariate logistic regression analyses to develop a predictive model. The corresponding odds ratio, 95% confidence interval, and p value were calculated. The receiver operating characteristic (ROC) curve was plotted for each model and the best cut-off value was calculated [26]. The diagnostic performance was evaluated based on the area under the ROC curve (AUC), with an AUC value >0.9, 0.7–0.9, and 0.5–0.7 indicating high, moderate, and low predictive ability, respectively [27]. p < 0.05 was considered significant.
Results
Comparison of Classifying T2 Endotype by IL-4, IL-5, IL-13, and IgE
A total of 249 Chinese CRSwNP were segregated into high or low T2 inflammation based on the levels of IL-4, IL-5, IL-13, or IgE in NPs in comparison to those in control tissues (online suppl. Fig. E1). As illustrated in Table 1 and online supplementary Figure E2, the majority of IL-13-high CRSwNP patients also had high IL-5 (90.16%), and more than half of them were also high for IL-4 (54.92%) and IgE (64.75%) expression. Lower proportions of patients with high mucosal IL-5 levels had high IL-13 (85.93%) and IL-4 (53.13%) levels. Among CRSwNP patients with high mucosal IL-4 expression, 80.9%, 79.76%, and 51.19% demonstrated high IL-5, IL-13, and IgE levels, respectively. However, about 30% of patients with low mucosal IL-4 levels also demonstrated high IL-5, IL-13, and IgE levels, which was worse than the situation in IL-13- and IL-5-low patients. Similar situations were found for the classification based on IgE and IL-4. These data suggest that defining T2 endotype by high IL-4 or high IgE would have low sensitivity, leaving out many patients with high levels of other T2 biomarkers. As shown in Table 1, around 60–70% of patients with high T2 cytokine expression had tissue eosinophilia. Taken together, these observations drove the selection of mucosal IL-13 expression as the primary anchor for the T2 endotype.
Table 1.
Segregation of T2-high versus T2-low status in CRSwNP by mucosal IL-4, IL-5, IL-13, and IgE protein levels*
| Group (n, %) | IL-13 high | IL-5 high | IL-4 high | IgE high | Eos CRSwNP |
|---|---|---|---|---|---|
| IL-13 high (122/249, 48.99) | 122 (100) | 110 (90.16) | 67 (54.92) | 79 (64.75) | 75 (61.48) |
| IL-13 low (127/249, 51.01) | 0 (0) | 18 (14.17) | 17 (13.39) | 16 (12.60) | 40 (31.49) |
| IL-5 high (128/249, 51.40) | 110 (85.93) | 128 (100) | 68 (53.13) | 83 (64.84) | 81 (63.28) |
| IL-5 low (121/249, 48.6) | 12 (9.92) | 0 (0) | 16 (13.22) | 12 (9.92) | 34 (28.10) |
| IL-4 high (84/249, 33.73) | 67 (79.76) | 68 (80.95) | 84 (100) | 43 (51.19) | 49 (58.33) |
| IL-4 low (165/249, 66.27) | 55 (33.33) | 60 (36.36) | 0 (0) | 52 (31.52) | 66 (40.00) |
| IgE high (95/249, 38.15) | 79 (83.16) | 83 (87.36) | 43 (45.26) | 95 (100) | 71 (74.74) |
| IgE low (154/249, 61.85) | 43 (27.92) | 45 (29.22) | 41 (26.62) | 0 (0) | 44 (28.57) |
IL, interleukin; Ig, immunoglobulin; Eos CRSwNP, eosinophilic chronic rhinosinusitis with nasal polyps.
*The frequency and percentage of CRSwNP subjects indicated in the rows with sinus mucosal high or low status based on biomarkers indicated in columns was shown in the table.
T1, T2, and T3 Endotype of CRSwNP Patients
Based on the mucosal IFN-γ, IL-13, and IL-17A levels, the frequency of total T1, T2, and T3 endotype was 33%, 49%, and 39% in patients with CRSwNP (n = 249), respectively (Fig. 1). Only 2%, 17%, and 5.6% of CRSwNP demonstrated single T1, T2, and T3 endotype, respectively (Fig. 1). About 37% of CRSwNP patients had mixed endotype with a combined elevation of IFN-γ, IL-13, and IL-17A. In addition, we found that 39% of CRSwNP patients had no elevated IFN-γ, IL-13, or IL-17A and thus were referred to as all negative endotypes (Fig. 1).
Fig. 1.
Composition of T1/T2/T3 inflammatory endotype in CRSwNP. CRSwNP, chronic rhinosinusitis with nasal polyps; T, type.
Clinical Characteristics of CRSwNP Patients with Different Inflammatory Endotypes
CRSwNP patients with the T1 endotype had higher facial pain VAS scores but lower NP scores than those with the non-T1 endotype (online suppl. Table E3). Patients with the T2 endotype had longer disease duration and were older and more common with atopy, allergic rhinitis, and asthma as compared to those with the non-T2 endotype (online suppl. Table E3). In addition, patients with the T2 endotype had higher facial pain VAS scores, loss of smell VAS scores, total symptom VAS scores, bilateral total CT scores, ethmoid sinus CT scores, E/M ratio, and peripheral blood eosinophil counts and percentages than those without the T2 endotype (online suppl. Table E3). In comparison with those without the T3 endotype, patients with the T3 endotype had longer disease duration and higher peripheral eosinophil ratio but lower scores of nasal obstruction, rhinorrhea, NP, and endoscopic discharge (online suppl. Table E3).
Predicting Mucosal Inflammatory Endotype by Clinical Characteristics in CRSwNP Patients
Univariate logistic regression analysis was performed to investigate associations between clinical features and a particular endotype after adjusting for confounding factors. Age, higher facial pain VAS scores, and lower NP scores were found to be associated with the T1 endotype (Table 2). However, ROC analysis revealed that the predictive ability of these factors alone for the T1 endotype was poor with AUC values <0.6 (online suppl. Table E4). NP score was negatively associated with T1 endotype, and its predictive ability for non-T1 was also very poor with an AUC value = 0.589 (online suppl. Table E4). T1 endotype-associated variables were then analyzed by multivariate logistic regression analysis, and the results showed that age, facial pain VAS scores, and NP scores were independent predictors for the T1 endotype. However, the predictive ability of the model which combined those variables was still very poor with an AUC value = 0.625 (Table 3).
Table 2.
Univariate logistic regression analysis of clinical factors associated with T1, T2, or T3 endotype of CRSwNP
| T1 endotype | T2 endotype | T3 endotype | ||||
|---|---|---|---|---|---|---|
| adjusted OR (95% CI) | p value | adjusted OR (95% CI) | p value | adjusted OR (95% CI) | p value | |
| Gender (male) | 0.947 (0.509–1.764) | 0.864 | 0.785 (0.431–1.431) | 0.430 | 0.821 (0.455–1.481) | 0.513 |
| Age (year) | 1.019 (1.001–1.038) | 0.041 | 1.028 (1.010–1.047) | 0.002 | 1.011 (0.994–1.028) | 0.219 |
| Atopy comorbidity | 1.722 (0.970–3.058) | 0.063 | 2.091 (1.093–4.001) | 0.026 | 1.042 (0.550–1.976) | 0.899 |
| AR comorbidity | 0.694 (0.268–1.798) | 0.451 | 1.664 (0.633–4.375) | 0.302 | 0.817 (0.319–2.091) | 0.674 |
| Asthma comorbidity | 0.678 (0.229–2.013) | 0.485 | 2.640 (0.920–7.576) | 0.071 | 0.441 (0.152–1.279) | 0.132 |
| Smoking habits | 1.381 (0.712–2.679) | 0.340 | 1.254 (0.651–2.414) | 0.499 | 1.308 (0.685–2.498) | 0.415 |
| Patients with prior surgery | 0.813 (0.455–1.454) | 0.864 | 1.028 (0.587–1.877) | 0.923 | 0.903 (0.521–1.563) | 0.821 |
| Disease duration (year) | 0.998 (0.960–1.037) | 0.922 | 1.030 (0.990–1.072) | 0.145 | 1.059 (1.016–1.013) | 0.006 |
| Nasal obstruction VAS score | 0.912 (0.829–1.004) | 0.060 | 1.017 (0.925–1.118) | 0.727 | 0.872 (0.793–0.960) | 0.005 |
| Rhinorrhea VAS score | 0.961 (0.880–1.049) | 0.371 | 1.011 (0.923–1.106) | 0.819 | 0.917 (0.839–0.989) | 0.045 |
| Headache VAS score | 1.041 (0.949–1.143) | 0.394 | 1.068 (0.972–1.173) | 0.170 | 1.108 (0.929–1.114) | 0.707 |
| Facial pain VAS score | 1.141 (1.020–1.275) | 0.021 | 1.118 (1.048–1.330) | 0.006 | 1.111 (0.996–1.239) | 0.047 |
| Loss of smell VAS score | 1.012 (0.939–1.090) | 0.076 | 1.094 (1.013–1.182) | 0.023 | 1.010 (0.938–1.088) | 0.785 |
| Total symptom VAS score | 1.003 (0.973–1.034) | 0.845 | 1.039 (1.006–1.076) | 0.018 | 0.989 (0.960–1.019) | 0.989 |
| Overall burden VAS score | 0.999 (0.885–1.129) | 0.102 | 1.057 (0.935–1.194) | 0.379 | 0.966 (0.858–1.087) | 0.563 |
| Bilateral CT score | 1.033 (0.963–1.046) | 0.879 | 1.041 (0.999–1.084) | 0.054 | 0.993 (0.954–1.033) | 0.720 |
| Ethmoid sinus CT score | 0.977 (0.774–1.083) | 0.730 | 1.163 (1.017–1.330) | 0.028 | 0.995 (0.878–1.128) | 0.937 |
| Maxillary sinus CT score | 0.909 (0.704–1.174) | 0.465 | 0.983 (0.759–1.272) | 0.896 | 0.839 (0.653–1.079) | 0.173 |
| E/M ratio | 1.097 (0.900–1.339) | 0.360 | 1.359 (1.072–1.724) | 0.011 | 1.142 (0.939–1.388) | 0.184 |
| PE/AE ratio | 1.579 (0.431–5.788) | 0.491 | 1.547 (0.448–5.349) | 0.490 | 1.452 (0.431–4.893) | 0.548 |
| Nasal polyp score | 0.784 (0.649–0.947) | 0.012 | 0.842 (0.701–1.012) | 0.067 | 0.832 (0.694–0.997) | 0.046 |
| Edema score | 1.089 (0.796–1.490) | 0.592 | 0.995 (0.737–1.342) | 0.972 | 1.047 (0.779–1.406) | 0.762 |
| Discharge score | 0.831 (0.625–1.106) | 0.204 | 0.927 (0.707–1.216) | 0.584 | 0.743 (0.563–0.982) | 0.037 |
| Scarring score | 1.009 (0.657–1.549) | 0.968 | 1.025 (0.677–1.552) | 0.907 | 1.363 (0.913–2.034) | 0.130 |
| Crusting score | 0.976 (0.578–1.648) | 0.927 | 0.923 (0.559–1.522) | 0.753 | 1.102 (0.677–1.794) | 0.696 |
| Total endoscopic score | 1.030 (0.941–1.128) | 0.516 | 1.004 (0.918–1.970) | 0.935 | 1.037 (0.950–1.132) | 0.412 |
| Blood leukocyte count | 0.982 (0.837–1.153) | 0.828 | 1.194 (0.932–1.409) | 0.135 | 1.033 (0.885–1.205) | 0.683 |
| Blood neutrophil count | 0.941 (0.778–1.138) | 0.530 | 1.095 (0.912–1.314) | 0.332 | 0.959 (0.801–1.147) | 0.646 |
| Blood neutrophil ratio | 1.011 (0.980–1.043) | 0.478 | 0.994 (0.964–1.024) | 0.994 | 0.985 (0.956–1.016) | 0.340 |
| Blood lymphocyte count | 1.106 (0.745–1.641) | 0.617 | 1.284 (0.833–1.978) | 0.257 | 1.077 (0.736–1.576) | 0.701 |
| Blood lymphocyte ratio | 1.016 (0.986–1.047) | 0.308 | 0.998 (0.969–1.027) | 0.873 | 1.007 (0.978–1.037) | 0.636 |
| Blood eosinophil count | 1.496 (0.512–4.374) | 0.462 | 27.949 (6.662–117.246) | <0.001 | 5.207 (1.715–15.814) | 0.004 |
| Blood eosinophil ratio | 1.040 (0.962–1.123) | 0.326 | 1.289 (1.161–1.431) | <0.001 | 1.187 (1.090–1.292) | <0.001 |
| Blood monocyte count | 1.606 (0.810–3.181) | 0.175 | 2.144 (0.913–4.539) | 0.076 | 2.289 (1.090–4.805) | 0.029 |
| Blood monocyte ratio | 1.018 (0.974–1.065) | 0.425 | 1.015 (0.970–1.062) | 0.528 | 1.048 (0.999–1.100) | 0.055 |
Results were adjusted for confounding factors, including age, gender, atopy, allergic rhinitis, and asthma.
AR, allergic rhinitis; CI, confidence interval; CT, computed tomography; E/M, ethmoid sinus/maxillary sinus CT score; PE/AE, posterior ethmoid sinus/anterior ethmoid sinus CT score; VAS, visual analogue scale.
Table 3.
Multivariate logistic regression analysis of clinical factors for predicting endotypes of CRSwNP
| Predictor | OR (95% CI) | p value | AUC | Sensitivity | Specificity | Overall predictive accuracy |
|---|---|---|---|---|---|---|
| T1 endotype model | ||||||
| Age (year) | 1.019 (1.001–1.038) | 0.045 | 0.625 | 0.458 | 0.801 | 68.7 |
| Nasal polyp score | 0.793 (0.655–0.960) | 0.017 | ||||
| Facial pain VAS score | 1.118 (1.001–1.250) | 0.049 | ||||
| T2 endotype model | ||||||
| Atopy comorbidity | 2.458 (1.322–4.568) | 0.014 | 0.815 | 0.733 | 0.845 | 78.1 |
| Facial pain VAS score | 1.246 (1.074–1.445) | 0.004 | ||||
| E/M ratio | 1.389 (1.101–1.753) | 0.006 | ||||
| Blood eosinophil ratio | 1.263 (1.148–1.390) | 0.000 | ||||
| T3 endotype model | ||||||
| Disease duration (year) | 1.041 (1.003–1.080) | 0.033 | 0.703 | 0.740 | 0.573 | 65.7 |
| Rhinorrhea VAS score | 0.874 (0.797–0.959) | 0.004 | ||||
| Facial pain VAS score | 1.194 (1.063–1.340) | 0.003 | ||||
| Blood eosinophil ratio | 1.085 (1.017–1.157) | 0.013 | ||||
AUC, area under the curve; CI, confidence interval; E/M, ethmoid sinus/maxillary sinus CT score; OR, odds ratio; VAS, visual analogue scale.
For the T2 endotype, we found that age, atopy, higher scores of facial pain, loss of smell and total symptom, increased ethmoid sinus CT scores and E/M ratio, and higher blood eosinophil counts and ratios were associated with the T2 endotype (Table 2). The predictive values of these factors were shown in online supplementary Table E4, and blood eosinophil counts achieved the highest AUC value of 0.721. Factors other than blood eosinophil number and percentage had poor predictive accuracy with AUC values lower than 0.7. These T2-associated variables were further analyzed by multivariate logistic regression analysis. We found that atopy, facial pain VAS score, E/M ratio, and blood eosinophil percentage were independently associated with T2 endotype and the predictive performance of the combination model was markedly improved with an AUC value = 0.815 (Table 3).
Regarding the T3 endotype, longer disease duration, higher facial pain scores, increased blood eosinophil counts and percentages, lower scores of nasal obstruction and rhinorrhea, and lower endoscopic NP and discharge scores were found to be associated with the T3 endotype in the univariate logistic regression analysis (Table 2). However, these factors had poor performance to predict T3 or non-T3 endotype with AUC values <0.6 (online suppl. Table E4). In multivariate logistic regression analysis, disease duration, rhinorrhea, facial pain VAS scores, and blood eosinophil ratios were independent predictors for the T3 endotype. The predictive performance of the combination model was moderate with an AUC value = 0.703 (Table 3).
Predicting Mucosal Inflammatory Endotype by Nasal Secretion Biomarkers in CRSwNP Patients
Given the limited capacity of clinical characteristics to identify endotypes of CRSwNP, particularly the T1 and T3 endotype, we further explored the value of biomarkers in noninvasive nasal secretion samples for predicting mucosal endotypes in a subset of 46 patients providing enough nasal secretion samples. There was no significant difference in demographic and clinical characteristics between patients involved in the nasal secretion biomarker study and total CRSwNP patients (online suppl. Table E5). Compared to patients with the non-T1 endotype, the levels of eotaxin and MIP-1β in nasal secretions were significantly reduced in those with a T1 endotype (online suppl. Table E6). However, after adjusting for confounding factors, no mediator in nasal secretions was found to be associated with the T1 endotype by univariate logistic regression analysis (Table 4).
Table 4.
Univariate logistic regression analysis of nasal secretion mediators associated with T1, T2, or T3 endotypes of CRSwNP
| T1 endotype | T2 endotype | T3 endotype | ||||
|---|---|---|---|---|---|---|
| adjusted OR (95% CI) | p value | adjusted OR (95% CI) | p value | adjusted OR (95% CI) | p value | |
| IL-1β (pg/mg) | 0.968 (0.927–1.010) | 0.129 | 0.961 (0.923–0.999) | 0.045 | 1.003 (0.988–1.018) | 0.712 |
| IL-1Ra (pg/mg) | 1.000 (1.000–1.000) | 0.708 | 1.000 (0.999–1.001) | 0.452 | 1.002 (1.001–1.004) | 0.014 |
| IL-2 (pg/mg) | 0.982 (0.909–1.061) | 0.649 | 1.026 (0.933–1.129) | 0.591 | 1.034 (0.913–1.170) | 0.602 |
| IL-4 (pg/mg) | 0.737 (0.227–2.391) | 0.612 | 1.054 (0.885–1.256) | 0.554 | 1.083 (0.818–1.434) | 0.579 |
| IL-5 (pg/mg) | 0.973 (0.869–1.089) | 0.636 | 2.489 (1.194–5.188) | 0.015 | 1.117 (0.803–1.554) | 0.512 |
| IL-6 (pg/mg) | 0.999 (0.992–1.006) | 0.846 | 1.006 (0.995–1.018) | 0.299 | 1.000 (0.993–1.007) | 0.998 |
| IL-7 (pg/mg) | 0.929 (0.819–1.054) | 0.251 | 1.001 (0.976–1.026) | 0.936 | 1.008 (0.980–1.036) | 0.590 |
| IL-8 (pg/mg) | 1.000 (0.999–1.000) | 0.332 | 1.000 (0.999–1.000) | 0.341 | 1.000 (1.000–1.000) | 0.744 |
| IL-9 (pg/mg) | 0.991 (0.953–1.030) | 0.644 | 1.016 (0.974–1.060) | 0.454 | 1.005 (0.976–1.036) | 0.720 |
| IL-10 (pg/mg) | 1.004 (0.985–1.023) | 0.686 | 1.009 (0.981–1.038) | 0.526 | 1.004 (0.985–1.024) | 0.653 |
| IL-12 (pg/mg) | 1.002 (0.998–1.006) | 0.384 | 1.002 (0.995–1.010) | 0.559 | 1.000 (0.996–1.005) | 0.935 |
| IL-13 (pg/mg) | 1.006 (0.994–1.018) | 0.346 | 1.064 (0.986–1.149) | 0.196 | 0.996 (0.983–1.010) | 0.598 |
| IL-15 (pg/mg) | 0.956 (0.807–1.132) | 0.600 | 1.014 (0.951–1.081) | 0.669 | 1.097 (0.934–1.289) | 0.261 |
| IL-17A (pg/mg) | 0.990 (0.892–1.099) | 0.847 | 1.006 (0.980–1.033) | 0.655 | 1.006 (0.984–1.029) | 0.598 |
| IL-22 (pg/mg) | 1.001 (0.997–1.005) | 0.625 | 1.001 (0.994–1.008) | 0.698 | 1.001 (0.997–1.005) | 0.625 |
| IL-25 (pg/mg) | 1.002 (0.994–1.010) | 0.674 | 1.002 (0.994–1.009) | 0.690 | 1.001 (1.997–1.006) | 0.620 |
| IL-33 (pg/mg) | 0.999 (0.997–1.000) | 0.150 | 1.000 (1.000–1.000) | 0.656 | 1.000 (1.000–1.000) | 0.861 |
| Eotaxin (pg/mg) | 0.999 (0.989–1.009) | 0.835 | 1.000 (0.990–1.010) | 0.990 | 0.991 (0.975–1.007) | 0.256 |
| bFGF (pg/mg) | 1.001 (0.993–1.009) | 0.828 | 1.009 (0.997–1.020) | 0.135 | 1.005 (0.997–1.013) | 0.238 |
| G-CSF (pg/mg) | 0.999 (0.999–1.000) | 0.071 | 0.998 (0.996–0.999) | 0.005 | 1.000 (0.999–1.001) | 0.141 |
| GM-CSF (pg/mg) | 1.002 (0.996–1.008) | 0.480 | 1.003 (0.996–1.011) | 0.348 | 1.005 (0.998–1.013) | 0.145 |
| IFN-γ (pg/mg) | 0.977 (0.935–1.021) | 0.300 | 1.002 (0.996–1.007) | 0.551 | 1.003 (0.993–1.013) | 0.607 |
| IP-10 (ng/mg) | 1.000 (1.000–1.000) | 0.345 | 1.000 (1.000–1.000) | 0.259 | 1.000 (1.000–1.000) | 0.363 |
| MCP-1 (pg/mg) | 0.998 (0.981–1.016) | 0.825 | 1.014 (0.989–1.039) | 0.277 | 1.004 (0.988–1.021) | 0.635 |
| MIP-1α (pg/mg) | 0.837 (0.657–1.039) | 0.103 | 0.964 (0.904–1.029) | 0.271 | 0.987 (0.948–1.028) | 0.521 |
| PDGF-BB (pg/mg) | 1.001 (0.999–1.003) | 0.400 | 1.002 (0.996–1.007) | 0.525 | 1.002 (0.998–1.005) | 0.372 |
| MIP-1β (pg/mg) | 0.985 (0.968–1.002) | 0.089 | 0.992 (0.981–1.004) | 0.199 | 0.996 (0.986–1.006) | 0.455 |
| RANTES (pg/mg) | 1.001 (0.999–1.003) | 0.283 | 1.001 (0.999–1.003) | 0.477 | 1.001 (0.999–1.002) | 0.457 |
| TNF-α (pg/mg) | 0.943 (0.858–1.035) | 0.217 | 1.002 (0.995–1.010) | 0.549 | 1.003 (0.995–1.011) | 0.491 |
| VEGF (pg/mg) | 0.999 (0.998–1.001) | 0.349 | 0.998 (0.996–0.999) | 0.038 | 0.999 (0.997–1.001) | 0.184 |
| IgG1 (ng/mg) | 1.000 (1.000–1.000) | 0.075 | 1.000 (1.000–1.000) | 0.322 | 1.000 (1.000–1.000) | 0.514 |
| IgG2 (ng/mg) | 1.000 (0.999–1.003) | 0.134 | 1.000 (1.000–1.000) | 0.074 | 1.000 (1.000–1.000) | 0.174 |
| IgG3 (ng/mg) | 0.999 (0.998–1.000) | 0.078 | 1.000 (1.000–1.000) | 0.388 | 1.000 (1.000–1.000) | 0.144 |
| IgG4 (ng/mg) | 1.000 (0.999–1.000) | 0.438 | 1.000 (1.000–1.000) | 0.925 | 1.000 (1.000–1.000) | 0.699 |
| IgM (ng/mg) | 1.000 (1.000–1.000) | 0.722 | 1.000 (1.000–1.000) | 0.953 | 1.000 (1.000–1.000) | 0.720 |
| IgE (pg/mg) | 1.008 (0.981–1.035) | 0.562 | 1.144 (0.749–1.749) | 0.533 | 0.985 (0.946–1.026) | 0.463 |
Results were adjusted for age, gender, atopy, allergic rhinitis, and asthma.
CI, confidence interval; IL, interleukin; IL-1Ra, IL-1 receptor antagonist; bFGF, basic fibroblast growth factor; G-CSF, granulocyte colony-stimulating factor; GM-CSF, granulocyte-macrophage colony-stimulating factor; IFN-γ, interferon-γ; IP-10, IFN-γ-induced protein 10; MCP, monocyte chemoattractant protein; MIP, macrophage inflammatory protein; PDGF-BB, platelet-derived growth factor-BB; RANTES, regulated upon activation normal T cell expressed and secreted; TNF, tumor necrosis factor; VEGF, vascular endothelial growth factor; Ig, immunoglobulin; OR, odds ratio.
Compared to patients with non-T2 endotype, IL-5 and IL-13 levels were significantly elevated, whereas IL-1β, IL-7, IL-10, G-CSF, and VEGF were significantly decreased in those with the T2 endotype (online suppl. Table E6). The univariate logistic regression analysis showed that increased IL-5 levels and decreased IL-1β, G-CSF, and VEGF levels were significantly associated with the T2 endotype (Table 4). As shown in Table 5, the predictive ability of nasal secretion IL-5 for the T2 endotype was moderate with an AUC value of 0.819, which was similar to the efficacy of the combined clinical model. IL-1β, G-CSF, and VEGF might be useful for identifying patients with non-T2 endotype, achieving an AUC value of 0.75, 0.869, and 0.633, respectively (Table 5). However, we failed to generate a model with a combination of several biomarkers by multivariate logistic regression analysis for predicting the T2 endotype.
Table 5.
Cut-off values of nasal secretion mediators for predicting endotypes of CRSwNP identified by univariate logistic regression analysis
| Predictors | Cut-off value | AUC | Sensitivity | Specificity | Overall predictive accuracy |
|---|---|---|---|---|---|
| For T2 endotype | |||||
| IL-5 (pg/mg) | 1.53 | 0.819 | 0.615 | 0.900 | 71.7 |
| For non-T2 endotype | |||||
| IL-1β (pg/mg) | 18.73 | 0.750 | 0.600 | 0.769 | 67.4 |
| G-CSF (pg/mg) | 752.39 | 0.869 | 0.750 | 0.885 | 82.6 |
| VEGF (pg/mg) | 476.38 | 0.633 | 0.464 | 0.767 | 62.5 |
| For T3 endotype | |||||
| IL-1Ra (pg/mg) | 570 | 0.756 | 0.714 | 0.800 | 73.9 |
AUC, area under the curve; IL, interleukin; G-CSF, granulocyte colony-stimulating factor; VEGF, vascular endothelial growth factor.
Regarding the T3 endotype, only IL-1Ra in nasal secretions differed significantly between T3 and non-T3 endotypes and was associated with the T3 endotype (online suppl. Table E6; Table 4). ROC analysis showed that IL-1Ra had a moderate capacity to predict T3 endotype in CRSwNP patients (an AUC value = 0.756) (Table 5).
Discussion
With the development of biologics and surgical interventions targeting different endotypes, a precision medicine approach for CRS is becoming possible [28–31]. Thus, identifying patients with distinct endotypes is important for clinical practice [32]. To the best of our knowledge, this is the first study to establish predictive models for mucosal T1, T2, and T3 endotypes in CRSwNP using clinical features and nasal secretion biomarkers.
Tissue IgE, IL-4, IL-5, and IL-13 levels have been widely used to define the T2 endotype in CRS [12–14]. However, these biomarkers are not fully coincident with each other. To determine the most representative marker for mucosal T2 inflammation in CRSwNP, we compared the coincidence between these biomarkers in classifying T2 and non-T2 inflammation. Mucosal IL-13 expression provided the most robust segregation for sinonasal mucosa T2 status in our dataset. Mucosal IL-13-high status had high concordance with other T2 biomarkers, with 90% of them being “IL-5-high” and more than 60% being eosinophilic CRSwNP. Also, importantly, low proportions of mucosal IL-13-low status were high for other T2 markers and eosinophilic inflammation. These observations affirmed our preference for mucosal IL-13 expression as the anchor for T2 status. Of note, T2 endotype and eosinophilic inflammation were not fully consistent with each other, suggesting the necessity to identify the predictors for T2 endotype and eosinophilic inflammation separately in CRSwNP patients.
Similar to our previous reports [6], around 50% of Chinese CRSwNP patients were T2 endotype (Fig. 1), which was significantly lower than that of patients in America (87%) [4, 6]. In contrast, 33% and 39% of patients in the present study had T1 and T3 endotype, respectively, which were higher than those in America (17% and 18%, respectively) [4]. Consistent with previous reports that Chinese patients were more likely to have mixed endotype, we found that more than 30% of CRSwNP patients had T1, T2, and/or T3 mixed endotype, which raises the difficulty in identifying patients with different endotypes by clinical factors [5, 22, 33].
By assessing associations between clinical factors and T1 endotype, older age, lower NP scores, and higher facial pain VAS scores were identified. In the current study, facial pain VAS scores were found to be associated with the T1, T2, and T3 endotypes, indicating that facial pain might be more likely a reflection of inflammation status rather than a unique presentation of any endotype. The multivariate clinical model for the T1 endotype was generated but with poor predictive performance (an AUC value = 0.625), suggesting the difficulty to identify the T1 endotype in patients with CRSwNP by clinical features. Consistently, although female gender and pus were found more common in American CRSsNP patients with the T1 endotype, no clinical indicator was identified for the T1 endotype in American CRSwNP patients [4].
Previously, Stevens et al. [4] explored the associations between demographic characteristics and symptoms and T2 endotype in American CRSwNP patients. They revealed that atopy and loss of smell were associated with the T2 endotype. In the present study, we had a similar finding in Chinese CRSwNP patients. Additionally, we explored the associations of blood leukocytes and endoscopic and CT findings with endotypes in Chinese CRSwNP patients. We found that blood eosinophil count and percentage, ethmoid sinus CT score, and E/M ratio were the risk factors for the T2 endotype. Peripheral blood eosinophil has been revealed to be a surrogate marker for mucosal eosinophil in CRSwNP patients, and eosinophilic CRSwNP patients had been reported to have a higher E/M ratio [20, 34, 35]. By multivariate logistic regression analysis, we found that atopy, facial pain score, E/M ratio, and blood eosinophil percentage were independently associated with T2 endotype in CRSwNP patients and their combination achieved clinically acceptable predictive value with an AUC value = 0.815.
Chinese CRSwNP patients with the T3 endotype had significantly decreased nasal obstruction VAS scores, rhinorrhea VAS scores, NP scores, and endoscopic discharge scores; longer disease duration; higher facial pain VAS scores; and increased blood eosinophil levels compared to non-T3 endotype. In contrast, rhinorrhea and nasal obstruction symptoms were not associated with any endotype, and T3 inflammation was associated with the intraoperative findings of pus in the American populations [4, 36]. Our CRSwNP patients had a higher frequency of T3 endotype compared to that reported in American patients, and most of them were mixed with other endotypes, which may drive different clinical manifestations of T3 endotype across different geographical areas and populations [4]. Of note, 76% of patients with the T3 endotype were also mixed with the T2 endotype, which may partially explain the increased blood eosinophil level observed in the T3 endotype. Besides, despite a positive correlation between systemic and local eosinophil numbers, previous studies also demonstrated that around 38% of CRSwNP patients had discordant blood and tissue eosinophilia [37, 38]. Those studies suggest that blood eosinophil levels cannot fully represent local T2 inflammation. In the multivariate logistic regression analysis, four factors (longer disease duration, higher facial pain VAS score, higher blood eosinophil ratio, and lower rhinorrhea VAS score) were independently associated with T3 status and the AUC value of the combined model was 0.703, suggesting an insufficient predictive power of clinical features for T3 endotype in Chinese CRSwNP patients.
Since the predictive performance of clinical factors for the T1 and T3 endotype was sub-optimal, we tried to seek noninvasive biomarkers to identify mucosal endotypes. Our previous study found that plasma biomarkers were unable to identify mucosal eosinophilic and neutrophilic inflammation in patients with CRSwNP [39]. We then focused on biomarkers in nasal secretions [20, 40]. IL-5 in nasal secretions achieved favorable predictive performance (an AUC value = 0.819) for the T2 endotype defined by mucosal IL-13 levels, which is in line with the good agreement between IL-13 and IL-5 in NPs. Besides, we noticed that G-CSF in nasal secretions was significantly negatively associated with the T2 endotype and could predict the non-T2 endotype with high accuracy (an AUC value = 0.869). In line with our study, Van Nevel et al. [41] reported that compared with T2-high CRS patients, G-CSF was indicated as the only marker significantly upregulated in secretion fluid of T2-low patients. G-CSF promotes neutrophil survival and activation and therefore may contribute to neutrophil infiltration in T2-low CRS [41, 42].
In our patient cohort, IL-1Ra in nasal secretions predicted the T3 endotype with an AUC value = 0.756. IL-1Ra was found to be significantly upregulated in neutrophil-high NPs which was associated with T1 and T3 endotype [42]. For patients with the T1 endotype, only eotaxin and MIP-1β in nasal secretions were significantly altered compared with the non-T1 endotype. However, none was identified as a risk factor for the T1 endotype after adjusting for confounding factors. In this study, we failed to identify nasal secretion IFN-γ, IL-13, or IL-17A as the predictor of the mucosal T1, T2, or T3 endotype, respectively, possibly due to the weak or no correlation between their levels in nasal secretions and NP tissues (data not shown).
Our study had several limitations. First, the predictive power of clinical models for the T1 and T3 endotype was not sufficient. Some clinical presentations such as intraoperative pus and allergic mucin were not included in our present study. Second, the number of patients providing nasal secretion samples was limited and our findings of nasal secretion biomarkers still need to be validated in a larger study cohort. Third, we did not further subgroup the patients into those with single or mixed endotype given the limited sample size of patients with single endotype. Fourth, regarding the difference in the composition of different endotypes in distinct geographic areas and populations, the extrapolation of our findings to other patient populations should be considered with caution. Nevertheless, our findings provide a valuable reference for future investigations. Fifth, we defined mucosal T2 endotype by IL-13 levels, which coincided best with the expression of other T2 biomarkers. However, different predictors may be identified for the conditions defined by other cytokines. Sixth, in this study, we collected whole nasal secretions. Secretions from the middle meatus may represent sinus conditions more precisely than whole nasal secretions; however, collecting middle meatus secretions is more complicated and requires the assistance of endoscopy. The value of middle meatus secretions in endotype identification in patients with CRSwNP needs further investigation.
Conclusion
As shown in Figure 2, we found that mucosal T2 endotype in Chinese patients with CRSwNP can be identified by the combination of atopy, facial pain score, E/M ratio and blood eosinophil percentage, or nasal secretion IL-5 levels. T3 endotype is more accurately identified by nasal secretion IL-1Rα than the combination of clinical factors. Unfortunately, no efficient clinical factor or nasal secretion biomarker was identified to predict the T1 endotype. The efforts to identify mucosal endotypes by noninvasive markers are important for establishing individualized therapeutic strategies for CRSwNP patients.
Fig. 2.
Summary of predictors for T1/T2/T3 inflammatory endotype in CRSwNP. AUC, area under the receiver operating characteristic curve; CRSwNP, chronic rhinosinusitis with nasal polyps; E/M, ethmoid sinus/maxillary sinus computed tomography score; IL, interleukin; IL-1Ra, IL-1 receptor antagonist; T, type; VAS, visual analogue scale.
Statement of Ethics
This study protocol was reviewed and approved by the Ethics Committee of Tongji Hospital of Huazhong University of Science and Technology, approval number TJ-C20170301, and this study was conducted with written informed consent from all participants to participate in the study.
Conflict of Interest Statement
The authors have no conflicts of interest to declare.
Funding Sources
This study was supported by the Key Research and Development Program of Hubei Province 2021BCA119 (Z.L.) and National Natural Science Foundation of China (NSFC) grants 81920108011 and 82130030 (Z.L.) and 82071026 (B.L.).
Author Contributions
Cui-Lian Guo performed a literature search, data collection, and statistical analysis and prepared the manuscript. Ruo-Yu Lu, Chong-Shu Wang, Jie-Fang Zhao, and Li Pan performed data acquisition. Hui-Cheng Liu participated in the data analysis. Bo Liao performed data collection and manuscript editing. Zheng Liu and Bo Liao designed the study, did endoscopic examination and surgery, and prepared the manuscript.
Funding Statement
This study was supported by the Key Research and Development Program of Hubei Province 2021BCA119 (Z.L.) and National Natural Science Foundation of China (NSFC) grants 81920108011 and 82130030 (Z.L.) and 82071026 (B.L.).
Data Availability Statement
All data generated or analyzed during this study are included in this article and its supplementary material files. Further inquiries can be directed to the corresponding author.
Supplementary Material
Supplementary Material
Supplementary Material
Supplementary Material
Supplementary Material
Supplementary Material
Supplementary Material
Supplementary Material
References
- 1. 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;58(Suppl S29):1–464. 10.4193/Rhin20.600. [DOI] [PubMed] [Google Scholar]
- 2. Bachert C, Akdis CA. Phenotypes and emerging endotypes of chronic rhinosinusitis. J Allergy Clin Immunol Pract. 2016;4:621–8. 10.1016/j.jaip.2016.05.004. [DOI] [PubMed] [Google Scholar]
- 3. Staudacher AG, Peters AT, Kato A, Stevens WW. Use of endotypes, phenotypes, and inflammatory markers to guide treatment decisions in chronic rhinosinusitis. Ann Allergy Asthma Immunol. 2020;124(4):318–25. 10.1016/j.anai.2020.01.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Stevens WW, Peters AT, Tan BK, Klingler AI, Poposki JA, Hulse KE, et al. Associations between inflammatory endotypes and clinical presentations in chronic rhinosinusitis. J Allergy Clin Immunol Pract. 2019;7(8):2812–20.e3. 10.1016/j.jaip.2019.05.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Wang X, Zhang N, Bo M, Holtappels G, Zheng M, Lou H, et al. Diversity of TH cytokine profiles in patients with chronic rhinosinusitis: a multicenter study in Europe, Asia, and Oceania. J Allergy Clin Immunol. 2016;138(5):1344–53. 10.1016/j.jaci.2016.05.041. [DOI] [PubMed] [Google Scholar]
- 6. Wang H, Li ZY, Jiang WX, Liao B, Zhai GT, Wang N, et al. The activation and function of IL-36γ in neutrophilic inflammation in chronic rhinosinusitis. J Allergy Clin Immunol. 2018;141(5):1646–58. 10.1016/j.jaci.2017.12.972. [DOI] [PubMed] [Google Scholar]
- 7. Bachert C, Zhang N, Cavaliere C, Weiping W, Gevaert E, Krysko O. Biologics for chronic rhinosinusitis with nasal polyps. J Allergy Clin Immunol. 2020;145(3):725–39. 10.1016/j.jaci.2020.01.020. [DOI] [PubMed] [Google Scholar]
- 8. Han JK, Bachert C, Fokkens W, Desrosiers M, Wagenmann M, Lee SE, et al. Mepolizumab for chronic rhinosinusitis with nasal polyps (SYNAPSE): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Respir Med. 2021;9(10):1141–53. 10.1016/S2213-2600(21)00097-7. [DOI] [PubMed] [Google Scholar]
- 9. Gevaert P, Omachi TA, Corren J, Mullol J, Han J, Lee SE, et al. Efficacy and safety of omalizumab in nasal polyposis: 2 randomized phase 3 trials. J Allergy Clin Immunol. 2020;146:595–605. 10.1016/j.jaci.2020.05.032. [DOI] [PubMed] [Google Scholar]
- 10. Bachert C, Han JK, Desrosiers M, Hellings PW, Amin N, Lee SE, et al. Efficacy and safety of dupilumab in patients with severe chronic rhinosinusitis with nasal polyps (LIBERTY NP SINUS-24 and LIBERTY NP SINUS-52): results from two multicentre, randomised, double-blind, placebo-controlled, parallel-group phase 3 trials. Lancet. 2019;394(10209):1638–50. 10.1016/S0140-6736(19)31881-1. [DOI] [PubMed] [Google Scholar]
- 11. Guo CL, Wang CS, Liu Z. Clinical and biological markers in disease and biologics to treat chronic rhinosinusitis. Curr Opin Allergy Clin Immunol. 2022;22(1):16–23. 10.1097/ACI.0000000000000799. [DOI] [PubMed] [Google Scholar]
- 12. Alsharif S, Jonstam K, van Zele T, Gevaert P, Holtappels G, Bachert C. Endoscopic sinus surgery for type-2 CRSwNP: an endotype-based retrospective study. Laryngoscope. 2019;129(6):1286–92. 10.1002/lary.27815. [DOI] [PubMed] [Google Scholar]
- 13. Wang ZC, Yao Y, Chen CL, Guo CL, Ding HX, Song J, et al. Extrafollicular PD-1(high)CXCR5(-)CD4(+) T cells participate in local immunoglobulin production in nasal polyps. J Allergy Clin Immunol. 2022;149(2):610–23. 10.1016/j.jaci.2021.06.023. [DOI] [PubMed] [Google Scholar]
- 14. Tao X, Chen F, Sun Y, Wu S, Hong H, Shi J, et al. Prediction models for postoperative uncontrolled chronic rhinosinusitis in daily practice. Laryngoscope. 2018;128(12):2673–80. 10.1002/lary.27267. [DOI] [PubMed] [Google Scholar]
- 15. Sakuma Y, Ishitoya J, Komatsu M, Shiono O, Hirama M, Yamashita Y, et al. New clinical diagnostic criteria for eosinophilic chronic rhinosinusitis. Auris Nasus Larynx. 2011;38(5):583–8. 10.1016/j.anl.2011.01.007. [DOI] [PubMed] [Google Scholar]
- 16. Wang Y, Chen H, Zhu R, Liu G, Huang N, Li W, et al. Allergic Rhinitis Control Test questionnaire-driven stepwise strategy to improve allergic rhinitis control: a prospective study. Allergy. 2016;71(11):1612–9. 10.1111/all.12963. [DOI] [PubMed] [Google Scholar]
- 17. Brożek JL, Bousquet J, Agache I, Agarwal A, Bachert C, Bosnic-Anticevich S, et al. Allergic rhinitis and its impact on asthma (ARIA) guidelines-2016 revision. J Allergy Clin Immunol. 2017;140(4):950–8. 10.1016/j.jaci.2017.03.050. [DOI] [PubMed] [Google Scholar]
- 18. Reddel HK, Bacharier LB, Bateman ED, Brightling CE, Brusselle GG, Buhl R, et al. Global initiative for asthma strategy 2021: executive summary and rationale for Key changes. Am J Respir Crit Care Med. 2022;205(1):17–35. 10.1164/rccm.202109-2205PP. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Liu JX, Liao B, Yu QH, Wang H, Liu YB, Guo CL, et al. The IL-37-Mex3B-Toll-like receptor 3 axis in epithelial cells in patients with eosinophilic chronic rhinosinusitis with nasal polyps. J Allergy Clin Immunol. 2020;145(1):160–72. 10.1016/j.jaci.2019.07.009. [DOI] [PubMed] [Google Scholar]
- 20. Guo CL, Liao B, Liu JX, Pan L, Liu Z. Predicting difficult-to-treat chronic rhinosinusitis by noninvasive biological markers. Rhinology. 2021;59(1):81–90. 10.4193/Rhin20.103. [DOI] [PubMed] [Google Scholar]
- 21. Cao PP, Li HB, Wang BF, Wang SB, You XJ, Cui YH, et al. Distinct immunopathologic characteristics of various types of chronic rhinosinusitis in adult Chinese. J Allergy Clin Immunol. 2009;124(3):478–84, 484.e1–2. 10.1016/j.jaci.2009.05.017. [DOI] [PubMed] [Google Scholar]
- 22. 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;73(7):1459–69. 10.1111/all.13411. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Houser B. Bio-Rad’s Bio-Plex® suspension array system, xMAP technology overview. Arch Physiol Biochem. 2012;118(4):192–6. 10.3109/13813455.2012.705301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Tan BK, Klingler AI, Poposki JA, Stevens WW, Peters AT, Suh LA, et al. Heterogeneous inflammatory patterns in chronic rhinosinusitis without nasal polyps in Chicago, Illinois. J Allergy Clin Immunol. 2017;139(2):699–703.e7. 10.1016/j.jaci.2016.06.063. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Chen CL, Yao Y, Pan L, Hu ST, Ma J, Wang ZC, et al. Common fibrin deposition and tissue plasminogen activator downregulation in nasal polyps with distinct inflammatory endotypes. J Allergy Clin Immunol. 2020;146(3):677–81. 10.1016/j.jaci.2020.02.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Fischer JE, Bachmann LM, Jaeschke R. A readers’ guide to the interpretation of diagnostic test properties: clinical example of sepsis. Intensive Care Med. 2003;29(7):1043–51. 10.1007/s00134-003-1761-8. [DOI] [PubMed] [Google Scholar]
- 27. 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;6(8):812–9. 10.1002/alr.21749. [DOI] [PubMed] [Google Scholar]
- 28. Ye LR, Yan BX, Chen XY, Chen SQ, Chen JQ, Man XY, et al. Extended dosing intervals of ixekizumab for psoriasis: a single-center, uncontrolled, prospective study. J Am Acad Dermatol. 2022;86(6):1348–50. 10.1016/j.jaad.2021.04.093. [DOI] [PubMed] [Google Scholar]
- 29. Reich K, Warren RB, Lebwohl M, Gooderham M, Strober B, Langley RG, et al. Bimekizumab versus secukinumab in plaque psoriasis. N Engl J Med. 2021;385(2):142–52. 10.1056/NEJMoa2102383. [DOI] [PubMed] [Google Scholar]
- 30. Wei JCC, Kim TH, Kishimoto M, Ogusu N, Jeong H, Kobayashi S, et al. Efficacy and safety of brodalumab, an anti-IL17RA monoclonal antibody, in patients with axial spondyloarthritis: 16-week results from a randomised, placebo-controlled, phase 3 trial. Ann Rheum Dis. 2021;80(8):1014–21. 10.1136/annrheumdis-2020-219406. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Bachert C, Desrosiers MY, Hellings PW, Laidlaw TM. The role of biologics in chronic rhinosinusitis with nasal polyps. J Allergy Clin Immunol Pract. 2021;9(3):1099–106. 10.1016/j.jaip.2020.11.017. [DOI] [PubMed] [Google Scholar]
- 32. Bachert C, Marple B, Hosemann W, Cavaliere C, Wen W, Zhang N. Endotypes of chronic rhinosinusitis with nasal polyps: pathology and possible therapeutic implications. J Allergy Clin Immunol Pract. 2020;8(5):1514–9. 10.1016/j.jaip.2020.03.007. [DOI] [PubMed] [Google Scholar]
- 33. Delemarre T, Holtappels G, De Ruyck N, Zhang N, Nauwynck H, Bachert C, et al. Type 2 inflammation in chronic rhinosinusitis without nasal polyps: another relevant endotype. J Allergy Clin Immunol. 2020;146:337–43.e6. 10.1016/j.jaci.2020.04.040. [DOI] [PubMed] [Google Scholar]
- 34. Hu Y, Cao PP, Liang GT, Cui YH, Liu Z. Diagnostic significance of blood eosinophil count in eosinophilic chronic rhinosinusitis with nasal polyps in Chinese adults. Laryngoscope. 2012;122(3):498–503. 10.1002/lary.22507. [DOI] [PubMed] [Google Scholar]
- 35. 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;70(8):995–1003. 10.1111/all.12644. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Yu S, Cao C, Li Q, Wen X, Guo X, Bao Q, et al. Local IL-17 positive T cells are functionally associated with neutrophil infiltration and their development is regulated by mucosal microenvironment in nasal polyps. Inflamm Res. 2021;70(1):139–49. 10.1007/s00011-020-01424-z. [DOI] [PubMed] [Google Scholar]
- 37. Pan L, Liao B, Guo CL, Liu JX, Wang H, Long XB, et al. Inflammatory features and predictors for postsurgical outcomes in patients with nasal polyps stratified by local and systemic eosinophilia. Int Forum Allergy Rhinol. 2021;11(5):846–56. 10.1002/alr.22702. [DOI] [PubMed] [Google Scholar]
- 38. Wang K, Deng J, Yang M, Chen Y, Chen F, Gao WX, et al. Concordant systemic and local eosinophilia relates to poorer disease control in patients with nasal polyps. World Allergy Organ J. 2019;12(8):100052. 10.1016/j.waojou.2019.100052. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Wang H, Guo CL, Xiao Q, Liu Z. Association between plasma inflammatory mediators and histological endotypes of nasal polyps. Zhonghua Er Bi Yan Hou Tou Jing Wai Ke Za Zhi. 2022;57(2):153–60. 10.3760/cma.j.cn115330-20210829-00579. [DOI] [PubMed] [Google Scholar]
- 40. Turner JH, Li P, Chandra RK. Mucus T helper 2 biomarkers predict chronic rhinosinusitis disease severity and prior surgical intervention. Int Forum Allergy Rhinol. 2018;8(10):1175–83. 10.1002/alr.22160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Van Nevel S, Declercq J, Holtappels G, Lambrecht BN, Bachert C. Granulocyte-colony stimulating factor: missing link for stratification of type 2-high and type 2-low chronic rhinosinusitis patients. J Allergy Clin Immunol. 2022;149(5):1655–65.e5. 10.1016/j.jaci.2022.02.019. [DOI] [PubMed] [Google Scholar]
- 42. Ruan JW, Zhao JF, Li XL, Liao B, Pan L, Zhu KZ, et al. Characterizing the neutrophilic inflammation in chronic rhinosinusitis with nasal polyps. Front Cell Dev Biol. 2021;9:793073. 10.3389/fcell.2021.793073. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
All data generated or analyzed during this study are included in this article and its supplementary material files. Further inquiries can be directed to the corresponding author.
